From b80d5ccca3a012e91ca64a2a0b13049163a6a698 Mon Sep 17 00:00:00 2001 From: Haiyan Hu Date: Tue, 10 Sep 2013 11:25:37 +0800 Subject: [PATCH 001/596] NVMe: Avoid shift operation when writing cq head doorbell Changes the type of dev->db_stride to unsigned and changes the value stored there to be 1 << the current value. Then there is less calculation to be done at completion time. Signed-off-by: Haiyan Hu Signed-off-by: Matthew Wilcox --- drivers/block/nvme-core.c | 10 +++++----- include/linux/nvme.h | 2 +- 2 files changed, 6 insertions(+), 6 deletions(-) diff --git a/drivers/block/nvme-core.c b/drivers/block/nvme-core.c index 26d03fa0bf26..073aec913c78 100644 --- a/drivers/block/nvme-core.c +++ b/drivers/block/nvme-core.c @@ -775,7 +775,7 @@ static int nvme_process_cq(struct nvme_queue *nvmeq) if (head == nvmeq->cq_head && phase == nvmeq->cq_phase) return 0; - writel(head, nvmeq->q_db + (1 << nvmeq->dev->db_stride)); + writel(head, nvmeq->q_db + nvmeq->dev->db_stride); nvmeq->cq_head = head; nvmeq->cq_phase = phase; @@ -1113,7 +1113,7 @@ static struct nvme_queue *nvme_alloc_queue(struct nvme_dev *dev, int qid, init_waitqueue_head(&nvmeq->sq_full); init_waitqueue_entry(&nvmeq->sq_cong_wait, nvme_thread); bio_list_init(&nvmeq->sq_cong); - nvmeq->q_db = &dev->dbs[qid << (dev->db_stride + 1)]; + nvmeq->q_db = &dev->dbs[qid * 2 * dev->db_stride]; nvmeq->q_depth = depth; nvmeq->cq_vector = vector; nvmeq->q_suspended = 1; @@ -1149,7 +1149,7 @@ static void nvme_init_queue(struct nvme_queue *nvmeq, u16 qid) nvmeq->sq_tail = 0; nvmeq->cq_head = 0; nvmeq->cq_phase = 1; - nvmeq->q_db = &dev->dbs[qid << (dev->db_stride + 1)]; + nvmeq->q_db = &dev->dbs[qid * 2 * dev->db_stride]; memset(nvmeq->cmdid_data, 0, extra); memset((void *)nvmeq->cqes, 0, CQ_SIZE(nvmeq->q_depth)); nvme_cancel_ios(nvmeq, false); @@ -1741,7 +1741,7 @@ static int set_queue_count(struct nvme_dev *dev, int count) static size_t db_bar_size(struct nvme_dev *dev, unsigned nr_io_queues) { - return 4096 + ((nr_io_queues + 1) << (dev->db_stride + 3)); + return 4096 + ((nr_io_queues + 1) * 8 * dev->db_stride); } static int nvme_setup_io_queues(struct nvme_dev *dev) @@ -1958,7 +1958,7 @@ static int nvme_dev_map(struct nvme_dev *dev) if (!dev->bar) goto disable; - dev->db_stride = NVME_CAP_STRIDE(readq(&dev->bar->cap)); + dev->db_stride = 1 << NVME_CAP_STRIDE(readq(&dev->bar->cap)); dev->dbs = ((void __iomem *)dev->bar) + 4096; return 0; diff --git a/include/linux/nvme.h b/include/linux/nvme.h index 26ebcf41c213..8119a476cb29 100644 --- a/include/linux/nvme.h +++ b/include/linux/nvme.h @@ -80,7 +80,7 @@ struct nvme_dev { struct dma_pool *prp_small_pool; int instance; int queue_count; - int db_stride; + u32 db_stride; u32 ctrl_config; struct msix_entry *entry; struct nvme_bar __iomem *bar; From 481e5bad84239b01415b888df2040c1860ae3cfd Mon Sep 17 00:00:00 2001 From: Michael Opdenacker Date: Sat, 12 Oct 2013 06:23:29 +0200 Subject: [PATCH 002/596] NVMe: remove deprecated IRQF_DISABLED This patch proposes to remove the use of the IRQF_DISABLED flag It's a NOOP since 2.6.35 and it will be removed one day. Signed-off-by: Michael Opdenacker Signed-off-by: Matthew Wilcox --- drivers/block/nvme-core.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/drivers/block/nvme-core.c b/drivers/block/nvme-core.c index 073aec913c78..df3daeb6227f 100644 --- a/drivers/block/nvme-core.c +++ b/drivers/block/nvme-core.c @@ -1134,11 +1134,10 @@ static int queue_request_irq(struct nvme_dev *dev, struct nvme_queue *nvmeq, { if (use_threaded_interrupts) return request_threaded_irq(dev->entry[nvmeq->cq_vector].vector, - nvme_irq_check, nvme_irq, - IRQF_DISABLED | IRQF_SHARED, + nvme_irq_check, nvme_irq, IRQF_SHARED, name, nvmeq); return request_irq(dev->entry[nvmeq->cq_vector].vector, nvme_irq, - IRQF_DISABLED | IRQF_SHARED, name, nvmeq); + IRQF_SHARED, name, nvmeq); } static void nvme_init_queue(struct nvme_queue *nvmeq, u16 qid) From a6f089e95b1e08cdea9633d50ad20aa5d44ba64d Mon Sep 17 00:00:00 2001 From: Lior Amsalem Date: Mon, 25 Nov 2013 17:26:44 +0100 Subject: [PATCH 003/596] irqchip: armada-370-xp: fix IPI race condition In the Armada 370/XP driver, when we receive an IRQ 0, we read the list of doorbells that caused the interrupt from register ARMADA_370_XP_IN_DRBEL_CAUSE_OFFS. This gives the list of IPIs that were generated. However, instead of acknowledging only the IPIs that were generated, we acknowledge *all* the IPIs, by writing ~IPI_DOORBELL_MASK in the ARMADA_370_XP_IN_DRBEL_CAUSE_OFFS register. This creates a race condition: if a new IPI that isn't part of the ones read into the temporary "ipimask" variable is fired before we acknowledge all IPIs, then we will simply loose it. This is causing scheduling hangs on SMP intensive workloads. It is important to mention that this ARMADA_370_XP_IN_DRBEL_CAUSE_OFFS register has the following behavior: "A CPU write of 0 clears the bits in this field. A CPU write of 1 has no effect". This is what allows us to simply write ~ipimask to acknoledge the handled IPIs. Notice that the same problem is present in the MSI implementation, but it will be fixed as a separate patch, so that this IPI fix can be pushed to older stable versions as appropriate (all the way to 3.8), while the MSI code only appeared in 3.13. Signed-off-by: Lior Amsalem Signed-off-by: Thomas Petazzoni Cc: stable@vger.kernel.org # v3.8+ Fixes: 344e873e5657e8dc0 'arm: mvebu: Add IPI support via doorbells' Cc: Thomas Gleixner Signed-off-by: Jason Cooper --- drivers/irqchip/irq-armada-370-xp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/irqchip/irq-armada-370-xp.c b/drivers/irqchip/irq-armada-370-xp.c index 433cc8568dec..f5e49a2d8e5a 100644 --- a/drivers/irqchip/irq-armada-370-xp.c +++ b/drivers/irqchip/irq-armada-370-xp.c @@ -407,7 +407,7 @@ armada_370_xp_handle_irq(struct pt_regs *regs) ARMADA_370_XP_IN_DRBEL_CAUSE_OFFS) & IPI_DOORBELL_MASK; - writel(~IPI_DOORBELL_MASK, per_cpu_int_base + + writel(~ipimask, per_cpu_int_base + ARMADA_370_XP_IN_DRBEL_CAUSE_OFFS); /* Handle all pending doorbells */ From c7f7bd4a136e4b02dd2a66bf95aec545bd93e8db Mon Sep 17 00:00:00 2001 From: Lior Amsalem Date: Mon, 25 Nov 2013 17:26:45 +0100 Subject: [PATCH 004/596] irqchip: armada-370-xp: fix MSI race condition In the Armada 370/XP driver, when we receive an IRQ 1, we read the list of doorbells that caused the interrupt from register ARMADA_370_XP_IN_DRBEL_CAUSE_OFFS. This gives the list of MSIs that were generated. However, instead of acknowledging only the MSIs that were generated, we acknowledge *all* the MSIs, by writing ~MSI_DOORBELL_MASK in the ARMADA_370_XP_IN_DRBEL_CAUSE_OFFS register. This creates a race condition: if a new MSI that isn't part of the ones read into the temporary "msimask" variable is fired before we acknowledge all MSIs, then we will simply loose it. It is important to mention that this ARMADA_370_XP_IN_DRBEL_CAUSE_OFFS register has the following behavior: "A CPU write of 0 clears the bits in this field. A CPU write of 1 has no effect". This is what allows us to simply write ~msimask to acknoledge the handled MSIs. Notice that the same problem is present in the IPI implementation, but it is fixed as a separate patch, so that this IPI fix can be pushed to older stable versions as appropriate (all the way to 3.8), while the MSI code only appeared in 3.13. Signed-off-by: Lior Amsalem Signed-off-by: Thomas Petazzoni Cc: Thomas Gleixner Signed-off-by: Jason Cooper --- drivers/irqchip/irq-armada-370-xp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/irqchip/irq-armada-370-xp.c b/drivers/irqchip/irq-armada-370-xp.c index f5e49a2d8e5a..3fac063b4a78 100644 --- a/drivers/irqchip/irq-armada-370-xp.c +++ b/drivers/irqchip/irq-armada-370-xp.c @@ -381,7 +381,7 @@ armada_370_xp_handle_irq(struct pt_regs *regs) ARMADA_370_XP_IN_DRBEL_CAUSE_OFFS) & PCI_MSI_DOORBELL_MASK; - writel(~PCI_MSI_DOORBELL_MASK, per_cpu_int_base + + writel(~msimask, per_cpu_int_base + ARMADA_370_XP_IN_DRBEL_CAUSE_OFFS); for (msinr = PCI_MSI_DOORBELL_START; From 320a382746e0ab1304476ea7e986a8d416ab99db Mon Sep 17 00:00:00 2001 From: Keith Busch Date: Wed, 23 Oct 2013 13:07:34 -0600 Subject: [PATCH 005/596] NVMe: compat SG_IO ioctl For 32-bit versions of sg3-utils running on a 64-bit system. This is mostly a copy from the relevent portions of fs/compat_ioctl.c, with slight modifications for going through block_device_operations. Signed-off-by: Keith Busch Reviewed-by: Vishal Verma [fixed up CONFIG_COMPAT=n build problems] Signed-off-by: Matthew Wilcox --- drivers/block/nvme-core.c | 18 ++++- drivers/block/nvme-scsi.c | 147 ++++++++++++++++++++++++++++++++++++++ include/linux/nvme.h | 1 + 3 files changed, 165 insertions(+), 1 deletion(-) diff --git a/drivers/block/nvme-core.c b/drivers/block/nvme-core.c index df3daeb6227f..e44350de8a21 100644 --- a/drivers/block/nvme-core.c +++ b/drivers/block/nvme-core.c @@ -1568,10 +1568,26 @@ static int nvme_ioctl(struct block_device *bdev, fmode_t mode, unsigned int cmd, } } +#ifdef CONFIG_COMPAT +static int nvme_compat_ioctl(struct block_device *bdev, fmode_t mode, + unsigned int cmd, unsigned long arg) +{ + struct nvme_ns *ns = bdev->bd_disk->private_data; + + switch (cmd) { + case SG_IO: + return nvme_sg_io32(ns, arg); + } + return nvme_ioctl(bdev, mode, cmd, arg); +} +#else +#define nvme_compat_ioctl NULL +#endif + static const struct block_device_operations nvme_fops = { .owner = THIS_MODULE, .ioctl = nvme_ioctl, - .compat_ioctl = nvme_ioctl, + .compat_ioctl = nvme_compat_ioctl, }; static void nvme_resubmit_bios(struct nvme_queue *nvmeq) diff --git a/drivers/block/nvme-scsi.c b/drivers/block/nvme-scsi.c index 4a4ff4eb8e23..4a0ceb64e269 100644 --- a/drivers/block/nvme-scsi.c +++ b/drivers/block/nvme-scsi.c @@ -25,6 +25,7 @@ #include #include #include +#include #include #include #include @@ -3038,6 +3039,152 @@ int nvme_sg_io(struct nvme_ns *ns, struct sg_io_hdr __user *u_hdr) return retcode; } +#ifdef CONFIG_COMPAT +typedef struct sg_io_hdr32 { + compat_int_t interface_id; /* [i] 'S' for SCSI generic (required) */ + compat_int_t dxfer_direction; /* [i] data transfer direction */ + unsigned char cmd_len; /* [i] SCSI command length ( <= 16 bytes) */ + unsigned char mx_sb_len; /* [i] max length to write to sbp */ + unsigned short iovec_count; /* [i] 0 implies no scatter gather */ + compat_uint_t dxfer_len; /* [i] byte count of data transfer */ + compat_uint_t dxferp; /* [i], [*io] points to data transfer memory + or scatter gather list */ + compat_uptr_t cmdp; /* [i], [*i] points to command to perform */ + compat_uptr_t sbp; /* [i], [*o] points to sense_buffer memory */ + compat_uint_t timeout; /* [i] MAX_UINT->no timeout (unit: millisec) */ + compat_uint_t flags; /* [i] 0 -> default, see SG_FLAG... */ + compat_int_t pack_id; /* [i->o] unused internally (normally) */ + compat_uptr_t usr_ptr; /* [i->o] unused internally */ + unsigned char status; /* [o] scsi status */ + unsigned char masked_status; /* [o] shifted, masked scsi status */ + unsigned char msg_status; /* [o] messaging level data (optional) */ + unsigned char sb_len_wr; /* [o] byte count actually written to sbp */ + unsigned short host_status; /* [o] errors from host adapter */ + unsigned short driver_status; /* [o] errors from software driver */ + compat_int_t resid; /* [o] dxfer_len - actual_transferred */ + compat_uint_t duration; /* [o] time taken by cmd (unit: millisec) */ + compat_uint_t info; /* [o] auxiliary information */ +} sg_io_hdr32_t; /* 64 bytes long (on sparc32) */ + +typedef struct sg_iovec32 { + compat_uint_t iov_base; + compat_uint_t iov_len; +} sg_iovec32_t; + +static int sg_build_iovec(sg_io_hdr_t __user *sgio, void __user *dxferp, u16 iovec_count) +{ + sg_iovec_t __user *iov = (sg_iovec_t __user *) (sgio + 1); + sg_iovec32_t __user *iov32 = dxferp; + int i; + + for (i = 0; i < iovec_count; i++) { + u32 base, len; + + if (get_user(base, &iov32[i].iov_base) || + get_user(len, &iov32[i].iov_len) || + put_user(compat_ptr(base), &iov[i].iov_base) || + put_user(len, &iov[i].iov_len)) + return -EFAULT; + } + + if (put_user(iov, &sgio->dxferp)) + return -EFAULT; + return 0; +} + +int nvme_sg_io32(struct nvme_ns *ns, unsigned long arg) +{ + sg_io_hdr32_t __user *sgio32 = (sg_io_hdr32_t __user *)arg; + sg_io_hdr_t __user *sgio; + u16 iovec_count; + u32 data; + void __user *dxferp; + int err; + int interface_id; + + if (get_user(interface_id, &sgio32->interface_id)) + return -EFAULT; + if (interface_id != 'S') + return -EINVAL; + + if (get_user(iovec_count, &sgio32->iovec_count)) + return -EFAULT; + + { + void __user *top = compat_alloc_user_space(0); + void __user *new = compat_alloc_user_space(sizeof(sg_io_hdr_t) + + (iovec_count * sizeof(sg_iovec_t))); + if (new > top) + return -EINVAL; + + sgio = new; + } + + /* Ok, now construct. */ + if (copy_in_user(&sgio->interface_id, &sgio32->interface_id, + (2 * sizeof(int)) + + (2 * sizeof(unsigned char)) + + (1 * sizeof(unsigned short)) + + (1 * sizeof(unsigned int)))) + return -EFAULT; + + if (get_user(data, &sgio32->dxferp)) + return -EFAULT; + dxferp = compat_ptr(data); + if (iovec_count) { + if (sg_build_iovec(sgio, dxferp, iovec_count)) + return -EFAULT; + } else { + if (put_user(dxferp, &sgio->dxferp)) + return -EFAULT; + } + + { + unsigned char __user *cmdp; + unsigned char __user *sbp; + + if (get_user(data, &sgio32->cmdp)) + return -EFAULT; + cmdp = compat_ptr(data); + + if (get_user(data, &sgio32->sbp)) + return -EFAULT; + sbp = compat_ptr(data); + + if (put_user(cmdp, &sgio->cmdp) || + put_user(sbp, &sgio->sbp)) + return -EFAULT; + } + + if (copy_in_user(&sgio->timeout, &sgio32->timeout, + 3 * sizeof(int))) + return -EFAULT; + + if (get_user(data, &sgio32->usr_ptr)) + return -EFAULT; + if (put_user(compat_ptr(data), &sgio->usr_ptr)) + return -EFAULT; + + err = nvme_sg_io(ns, sgio); + if (err >= 0) { + void __user *datap; + + if (copy_in_user(&sgio32->pack_id, &sgio->pack_id, + sizeof(int)) || + get_user(datap, &sgio->usr_ptr) || + put_user((u32)(unsigned long)datap, + &sgio32->usr_ptr) || + copy_in_user(&sgio32->status, &sgio->status, + (4 * sizeof(unsigned char)) + + (2 * sizeof(unsigned short)) + + (3 * sizeof(int)))) + err = -EFAULT; + } + + return err; +} +#endif + int nvme_sg_get_version_num(int __user *ip) { return put_user(sg_version_num, ip); diff --git a/include/linux/nvme.h b/include/linux/nvme.h index 8119a476cb29..5bc29197d90e 100644 --- a/include/linux/nvme.h +++ b/include/linux/nvme.h @@ -165,6 +165,7 @@ int nvme_set_features(struct nvme_dev *dev, unsigned fid, unsigned dword11, struct sg_io_hdr; int nvme_sg_io(struct nvme_ns *ns, struct sg_io_hdr __user *u_hdr); +int nvme_sg_io32(struct nvme_ns *ns, unsigned long arg); int nvme_sg_get_version_num(int __user *ip); #endif /* _LINUX_NVME_H */ From 0a8d44cb33377969337fa6c1961b631605d5f453 Mon Sep 17 00:00:00 2001 From: Matthew Wilcox Date: Tue, 15 Oct 2013 15:01:10 -0400 Subject: [PATCH 006/596] NVMe: Fix lockdep warnings During the initialisation path, the queue lock is taken without interrupt protection. It's perfectly safe to do so, because the interrupt handler can't run at this point, but it confuses lockdep. Signed-off-by: Matthew Wilcox --- drivers/block/nvme-core.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/drivers/block/nvme-core.c b/drivers/block/nvme-core.c index e44350de8a21..0e9c5dcb2bd7 100644 --- a/drivers/block/nvme-core.c +++ b/drivers/block/nvme-core.c @@ -1172,9 +1172,9 @@ static int nvme_create_queue(struct nvme_queue *nvmeq, int qid) if (result < 0) goto release_sq; - spin_lock(&nvmeq->q_lock); + spin_lock_irq(&nvmeq->q_lock); nvme_init_queue(nvmeq, qid); - spin_unlock(&nvmeq->q_lock); + spin_unlock_irq(&nvmeq->q_lock); return result; @@ -1290,9 +1290,9 @@ static int nvme_configure_admin_queue(struct nvme_dev *dev) if (result) return result; - spin_lock(&nvmeq->q_lock); + spin_lock_irq(&nvmeq->q_lock); nvme_init_queue(nvmeq, 0); - spin_unlock(&nvmeq->q_lock); + spin_unlock_irq(&nvmeq->q_lock); return result; } @@ -1836,9 +1836,9 @@ static int nvme_setup_io_queues(struct nvme_dev *dev) for (i = dev->queue_count - 1; i > nr_io_queues; i--) { struct nvme_queue *nvmeq = dev->queues[i]; - spin_lock(&nvmeq->q_lock); + spin_lock_irq(&nvmeq->q_lock); nvme_cancel_ios(nvmeq, false); - spin_unlock(&nvmeq->q_lock); + spin_unlock_irq(&nvmeq->q_lock); nvme_free_queue(nvmeq); dev->queue_count--; From 68608c268bf265b039daeaafee6c902dfef32024 Mon Sep 17 00:00:00 2001 From: Matthew Wilcox Date: Fri, 21 Jun 2013 14:36:34 -0400 Subject: [PATCH 007/596] NVMe: Cache dev->pci_dev in a local pointer Helps with line-length issues Signed-off-by: Matthew Wilcox --- drivers/block/nvme-core.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/block/nvme-core.c b/drivers/block/nvme-core.c index 0e9c5dcb2bd7..300766973d76 100644 --- a/drivers/block/nvme-core.c +++ b/drivers/block/nvme-core.c @@ -1891,6 +1891,7 @@ static int nvme_setup_io_queues(struct nvme_dev *dev) */ static int nvme_dev_add(struct nvme_dev *dev) { + struct pci_dev *pdev = dev->pci_dev; int res; unsigned nn, i; struct nvme_ns *ns; @@ -1900,8 +1901,7 @@ static int nvme_dev_add(struct nvme_dev *dev) dma_addr_t dma_addr; int shift = NVME_CAP_MPSMIN(readq(&dev->bar->cap)) + 12; - mem = dma_alloc_coherent(&dev->pci_dev->dev, 8192, &dma_addr, - GFP_KERNEL); + mem = dma_alloc_coherent(&pdev->dev, 8192, &dma_addr, GFP_KERNEL); if (!mem) return -ENOMEM; @@ -1919,8 +1919,8 @@ static int nvme_dev_add(struct nvme_dev *dev) memcpy(dev->firmware_rev, ctrl->fr, sizeof(ctrl->fr)); if (ctrl->mdts) dev->max_hw_sectors = 1 << (ctrl->mdts + shift - 9); - if ((dev->pci_dev->vendor == PCI_VENDOR_ID_INTEL) && - (dev->pci_dev->device == 0x0953) && ctrl->vs[3]) + if ((pdev->vendor == PCI_VENDOR_ID_INTEL) && + (pdev->device == 0x0953) && ctrl->vs[3]) dev->stripe_size = 1 << (ctrl->vs[3] + shift); id_ns = mem; From 9a6b94584de1a0467d85b435df9c744c5c45a270 Mon Sep 17 00:00:00 2001 From: Keith Busch Date: Tue, 10 Dec 2013 13:10:36 -0700 Subject: [PATCH 008/596] NVMe: Device resume error handling Adds controller error handling on resume power management. If the device fails to initialize, the device is queued for a reset. If the reset fails, a thread is spawned to remove the pci device. If the device resumes as "busy", the device is responding to admin commands but will not create IO queues. In this case, we need to remove the gendisks and free the IO queues since they can't be used and may be holding bios in their lists. From testing, the dma pools require a pci device so this had to change the pci driver 'remove' to release the dma resources in line with that call instead of after all references to the device are released. Signed-off-by: Keith Busch Signed-off-by: Matthew Wilcox --- drivers/block/nvme-core.c | 108 ++++++++++++++++++++++++++++++++------ include/linux/nvme.h | 1 + 2 files changed, 94 insertions(+), 15 deletions(-) diff --git a/drivers/block/nvme-core.c b/drivers/block/nvme-core.c index 300766973d76..000bca43c23b 100644 --- a/drivers/block/nvme-core.c +++ b/drivers/block/nvme-core.c @@ -58,6 +58,7 @@ module_param(use_threaded_interrupts, int, 0); static DEFINE_SPINLOCK(dev_list_lock); static LIST_HEAD(dev_list); static struct task_struct *nvme_thread; +static struct workqueue_struct *nvme_workq; /* * An NVM Express queue. Each device has at least two (one for admin @@ -1968,7 +1969,6 @@ static int nvme_dev_map(struct nvme_dev *dev) dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(32))) goto disable; - pci_set_drvdata(pdev, dev); dev->bar = ioremap(pci_resource_start(pdev, 0), 8192); if (!dev->bar) goto disable; @@ -1995,9 +1995,9 @@ static void nvme_dev_unmap(struct nvme_dev *dev) if (dev->bar) { iounmap(dev->bar); dev->bar = NULL; + pci_release_regions(dev->pci_dev); } - pci_release_regions(dev->pci_dev); if (pci_is_enabled(dev->pci_dev)) pci_disable_device(dev->pci_dev); } @@ -2085,11 +2085,6 @@ static void nvme_release_instance(struct nvme_dev *dev) static void nvme_free_dev(struct kref *kref) { struct nvme_dev *dev = container_of(kref, struct nvme_dev, kref); - nvme_dev_remove(dev); - nvme_dev_shutdown(dev); - nvme_free_queues(dev); - nvme_release_instance(dev); - nvme_release_prp_pools(dev); kfree(dev->queues); kfree(dev->entry); kfree(dev); @@ -2161,6 +2156,70 @@ static int nvme_dev_start(struct nvme_dev *dev) return result; } +static int nvme_remove_dead_ctrl(void *arg) +{ + struct nvme_dev *dev = (struct nvme_dev *)arg; + struct pci_dev *pdev = dev->pci_dev; + + if (pci_get_drvdata(pdev)) + pci_stop_and_remove_bus_device(pdev); + kref_put(&dev->kref, nvme_free_dev); + return 0; +} + +static void nvme_remove_disks(struct work_struct *ws) +{ + int i; + struct nvme_dev *dev = container_of(ws, struct nvme_dev, reset_work); + + nvme_dev_remove(dev); + spin_lock(&dev_list_lock); + for (i = dev->queue_count - 1; i > 0; i--) { + BUG_ON(!dev->queues[i] || !dev->queues[i]->q_suspended); + nvme_free_queue(dev->queues[i]); + dev->queue_count--; + dev->queues[i] = NULL; + } + spin_unlock(&dev_list_lock); +} + +static int nvme_dev_resume(struct nvme_dev *dev) +{ + int ret; + + ret = nvme_dev_start(dev); + if (ret && ret != -EBUSY) + return ret; + if (ret == -EBUSY) { + spin_lock(&dev_list_lock); + INIT_WORK(&dev->reset_work, nvme_remove_disks); + queue_work(nvme_workq, &dev->reset_work); + spin_unlock(&dev_list_lock); + } + return 0; +} + +static void nvme_dev_reset(struct nvme_dev *dev) +{ + nvme_dev_shutdown(dev); + if (nvme_dev_resume(dev)) { + dev_err(&dev->pci_dev->dev, "Device failed to resume\n"); + kref_get(&dev->kref); + if (IS_ERR(kthread_run(nvme_remove_dead_ctrl, dev, "nvme%d", + dev->instance))) { + dev_err(&dev->pci_dev->dev, + "Failed to start controller remove task\n"); + kref_put(&dev->kref, nvme_free_dev); + } + } +} + +static void nvme_reset_failed_dev(struct work_struct *ws) +{ + struct nvme_dev *dev = container_of(ws, struct nvme_dev, reset_work); + nvme_dev_reset(dev); +} + static int nvme_probe(struct pci_dev *pdev, const struct pci_device_id *id) { int result = -ENOMEM; @@ -2180,7 +2239,7 @@ static int nvme_probe(struct pci_dev *pdev, const struct pci_device_id *id) INIT_LIST_HEAD(&dev->namespaces); dev->pci_dev = pdev; - + pci_set_drvdata(pdev, dev); result = nvme_set_instance(dev); if (result) goto free; @@ -2232,7 +2291,19 @@ static int nvme_probe(struct pci_dev *pdev, const struct pci_device_id *id) static void nvme_remove(struct pci_dev *pdev) { struct nvme_dev *dev = pci_get_drvdata(pdev); + + spin_lock(&dev_list_lock); + list_del_init(&dev->node); + spin_unlock(&dev_list_lock); + + pci_set_drvdata(pdev, NULL); + flush_work(&dev->reset_work); misc_deregister(&dev->miscdev); + nvme_dev_remove(dev); + nvme_dev_shutdown(dev); + nvme_free_queues(dev); + nvme_release_instance(dev); + nvme_release_prp_pools(dev); kref_put(&dev->kref, nvme_free_dev); } @@ -2256,13 +2327,12 @@ static int nvme_resume(struct device *dev) { struct pci_dev *pdev = to_pci_dev(dev); struct nvme_dev *ndev = pci_get_drvdata(pdev); - int ret; - ret = nvme_dev_start(ndev); - /* XXX: should remove gendisks if resume fails */ - if (ret) - nvme_free_queues(ndev); - return ret; + if (nvme_dev_resume(ndev) && !work_busy(&ndev->reset_work)) { + INIT_WORK(&ndev->reset_work, nvme_reset_failed_dev); + queue_work(nvme_workq, &ndev->reset_work); + } + return 0; } static SIMPLE_DEV_PM_OPS(nvme_dev_pm_ops, nvme_suspend, nvme_resume); @@ -2303,9 +2373,14 @@ static int __init nvme_init(void) if (IS_ERR(nvme_thread)) return PTR_ERR(nvme_thread); + result = -ENOMEM; + nvme_workq = create_singlethread_workqueue("nvme"); + if (!nvme_workq) + goto kill_kthread; + result = register_blkdev(nvme_major, "nvme"); if (result < 0) - goto kill_kthread; + goto kill_workq; else if (result > 0) nvme_major = result; @@ -2316,6 +2391,8 @@ static int __init nvme_init(void) unregister_blkdev: unregister_blkdev(nvme_major, "nvme"); + kill_workq: + destroy_workqueue(nvme_workq); kill_kthread: kthread_stop(nvme_thread); return result; @@ -2325,6 +2402,7 @@ static void __exit nvme_exit(void) { pci_unregister_driver(&nvme_driver); unregister_blkdev(nvme_major, "nvme"); + destroy_workqueue(nvme_workq); kthread_stop(nvme_thread); } diff --git a/include/linux/nvme.h b/include/linux/nvme.h index 5bc29197d90e..eed81cc56d7e 100644 --- a/include/linux/nvme.h +++ b/include/linux/nvme.h @@ -87,6 +87,7 @@ struct nvme_dev { struct list_head namespaces; struct kref kref; struct miscdevice miscdev; + struct work_struct reset_work; char name[12]; char serial[20]; char model[40]; From b8989db9d82465bf38a48a4d3ef32e7d8afc4d08 Mon Sep 17 00:00:00 2001 From: Alan Date: Mon, 20 Jan 2014 18:01:56 +0000 Subject: [PATCH 009/596] x86, doc, kconfig: Fix dud URL for Microcode data The actual data lives in the Intel download center, and that ought to also be a reliable way to continue to find it. Unfortunately the actual URL needed for doing it directly is about a foot long so give instructions. Signed-off-by: Alan Cox Link: http://lkml.kernel.org/r/20140120180056.7173.62222.stgit@alan.etchedpixels.co.uk Signed-off-by: H. Peter Anvin --- arch/x86/Kconfig | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 0952ecd60eca..64199bc08d66 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -1064,9 +1064,9 @@ config MICROCODE_INTEL This options enables microcode patch loading support for Intel processors. - For latest news and information on obtaining all the required - Intel ingredients for this driver, check: - . + For the current Intel microcode data package go to + and search for + 'Linux Processor Microcode Data File'. config MICROCODE_AMD bool "AMD microcode loading support" From 3aba55605326be6d7e624090858aa921ab519cda Mon Sep 17 00:00:00 2001 From: Magnus Damm Date: Wed, 4 Dec 2013 21:05:46 +0900 Subject: [PATCH 010/596] irqchip: renesas-irqc: Use lazy disable Set the ->irq_enable() and ->irq_disable() methods to NULL to enable lazy disable of interrupts. This by itself provides some level of optimization, but is mainly enabled as ground work for future Suspend-to-RAM wake up support. Signed-off-by: Magnus Damm Cc: rob.herring@calxeda.com Cc: grant.likely@secretlab.ca Cc: horms@verge.net.au Link: http://lkml.kernel.org/r/20131204120546.29642.15772.sendpatchset@w520 Signed-off-by: Thomas Gleixner --- drivers/irqchip/irq-renesas-irqc.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/drivers/irqchip/irq-renesas-irqc.c b/drivers/irqchip/irq-renesas-irqc.c index 2f404ba61c6c..8fdd7d68cd00 100644 --- a/drivers/irqchip/irq-renesas-irqc.c +++ b/drivers/irqchip/irq-renesas-irqc.c @@ -212,8 +212,6 @@ static int irqc_probe(struct platform_device *pdev) irq_chip->name = name; irq_chip->irq_mask = irqc_irq_disable; irq_chip->irq_unmask = irqc_irq_enable; - irq_chip->irq_enable = irqc_irq_enable; - irq_chip->irq_disable = irqc_irq_disable; irq_chip->irq_set_type = irqc_irq_set_type; irq_chip->flags = IRQCHIP_SKIP_SET_WAKE; From 43881ec7a88a3d3b2fd6da58168173e135b41fba Mon Sep 17 00:00:00 2001 From: Magnus Damm Date: Wed, 4 Dec 2013 21:05:56 +0900 Subject: [PATCH 011/596] irqchip: renesas-irqc: Enable mask on suspend Now when lazy interrupt disable has been enabled in the driver then extend the code to set IRQCHIP_MASK_ON_SUSPEND which tells the core that only IRQs marked as wakeups need to stay enabled during Suspend-to-RAM. Signed-off-by: Magnus Damm Cc: rob.herring@calxeda.com Cc: grant.likely@secretlab.ca Cc: horms@verge.net.au Link: http://lkml.kernel.org/r/20131204120556.29642.27021.sendpatchset@w520 Signed-off-by: Thomas Gleixner --- drivers/irqchip/irq-renesas-irqc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/irqchip/irq-renesas-irqc.c b/drivers/irqchip/irq-renesas-irqc.c index 8fdd7d68cd00..082d95cb5528 100644 --- a/drivers/irqchip/irq-renesas-irqc.c +++ b/drivers/irqchip/irq-renesas-irqc.c @@ -213,7 +213,7 @@ static int irqc_probe(struct platform_device *pdev) irq_chip->irq_mask = irqc_irq_disable; irq_chip->irq_unmask = irqc_irq_enable; irq_chip->irq_set_type = irqc_irq_set_type; - irq_chip->flags = IRQCHIP_SKIP_SET_WAKE; + irq_chip->flags = IRQCHIP_SKIP_SET_WAKE | IRQCHIP_MASK_ON_SUSPEND; p->irq_domain = irq_domain_add_simple(pdev->dev.of_node, p->number_of_irqs, From 397e7b515785cad6e10b29f3001fd80c3f519bb8 Mon Sep 17 00:00:00 2001 From: Daniel Tang Date: Thu, 5 Dec 2013 17:12:17 +1100 Subject: [PATCH 012/596] irqchip: Add support for TI-NSPIRE irqchip This patch adds support for the interrupt controllers found in some TI-Nspire models. FIQ support was taken out to simplify the driver code and may be added in later. Since Linux on this platform doesn't really use FIQs, this wasn't really that important in the first place. [ tglx: Made zevio_handle_irq static and reordered __init functions ] Signed-off-by: Daniel Tang Acked-by: Grant Likely Link: http://lkml.kernel.org/r/1386223937-12189-1-git-send-email-dt.tangr@gmail.com Signed-off-by: Thomas Gleixner --- .../interrupt-controller/lsi,zevio-intc.txt | 18 +++ drivers/irqchip/Makefile | 1 + drivers/irqchip/irq-zevio.c | 127 ++++++++++++++++++ 3 files changed, 146 insertions(+) create mode 100644 Documentation/devicetree/bindings/interrupt-controller/lsi,zevio-intc.txt create mode 100644 drivers/irqchip/irq-zevio.c diff --git a/Documentation/devicetree/bindings/interrupt-controller/lsi,zevio-intc.txt b/Documentation/devicetree/bindings/interrupt-controller/lsi,zevio-intc.txt new file mode 100644 index 000000000000..aee38e7c13e7 --- /dev/null +++ b/Documentation/devicetree/bindings/interrupt-controller/lsi,zevio-intc.txt @@ -0,0 +1,18 @@ +TI-NSPIRE interrupt controller + +Required properties: +- compatible: Compatible property value should be "lsi,zevio-intc". + +- reg: Physical base address of the controller and length of memory mapped + region. + +- interrupt-controller : Identifies the node as an interrupt controller + +Example: + +interrupt-controller { + compatible = "lsi,zevio-intc"; + interrupt-controller; + reg = <0xDC000000 0x1000>; + #interrupt-cells = <1>; +}; diff --git a/drivers/irqchip/Makefile b/drivers/irqchip/Makefile index c60b9010b152..292b10645fd8 100644 --- a/drivers/irqchip/Makefile +++ b/drivers/irqchip/Makefile @@ -20,5 +20,6 @@ obj-$(CONFIG_SIRF_IRQ) += irq-sirfsoc.o obj-$(CONFIG_RENESAS_INTC_IRQPIN) += irq-renesas-intc-irqpin.o obj-$(CONFIG_RENESAS_IRQC) += irq-renesas-irqc.o obj-$(CONFIG_VERSATILE_FPGA_IRQ) += irq-versatile-fpga.o +obj-$(CONFIG_ARCH_NSPIRE) += irq-zevio.o obj-$(CONFIG_ARCH_VT8500) += irq-vt8500.o obj-$(CONFIG_TB10X_IRQC) += irq-tb10x.o diff --git a/drivers/irqchip/irq-zevio.c b/drivers/irqchip/irq-zevio.c new file mode 100644 index 000000000000..8ed04c4a43ee --- /dev/null +++ b/drivers/irqchip/irq-zevio.c @@ -0,0 +1,127 @@ +/* + * linux/drivers/irqchip/irq-zevio.c + * + * Copyright (C) 2013 Daniel Tang + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2, as + * published by the Free Software Foundation. + * + */ + +#include +#include +#include +#include +#include + +#include +#include + +#include "irqchip.h" + +#define IO_STATUS 0x000 +#define IO_RAW_STATUS 0x004 +#define IO_ENABLE 0x008 +#define IO_DISABLE 0x00C +#define IO_CURRENT 0x020 +#define IO_RESET 0x028 +#define IO_MAX_PRIOTY 0x02C + +#define IO_IRQ_BASE 0x000 +#define IO_FIQ_BASE 0x100 + +#define IO_INVERT_SEL 0x200 +#define IO_STICKY_SEL 0x204 +#define IO_PRIORITY_SEL 0x300 + +#define MAX_INTRS 32 +#define FIQ_START MAX_INTRS + +static struct irq_domain *zevio_irq_domain; +static void __iomem *zevio_irq_io; + +static void zevio_irq_ack(struct irq_data *irqd) +{ + struct irq_chip_generic *gc = irq_data_get_irq_chip_data(irqd); + struct irq_chip_regs *regs = + &container_of(irqd->chip, struct irq_chip_type, chip)->regs; + + readl(gc->reg_base + regs->ack); +} + +static asmlinkage void __exception_irq_entry zevio_handle_irq(struct pt_regs *regs) +{ + int irqnr; + + while (readl(zevio_irq_io + IO_STATUS)) { + irqnr = readl(zevio_irq_io + IO_CURRENT); + irqnr = irq_find_mapping(zevio_irq_domain, irqnr); + handle_IRQ(irqnr, regs); + }; +} + +static void __init zevio_init_irq_base(void __iomem *base) +{ + /* Disable all interrupts */ + writel(~0, base + IO_DISABLE); + + /* Accept interrupts of all priorities */ + writel(0xF, base + IO_MAX_PRIOTY); + + /* Reset existing interrupts */ + readl(base + IO_RESET); +} + +static int __init zevio_of_init(struct device_node *node, + struct device_node *parent) +{ + unsigned int clr = IRQ_NOREQUEST | IRQ_NOPROBE | IRQ_NOAUTOEN; + struct irq_chip_generic *gc; + int ret; + + if (WARN_ON(zevio_irq_io || zevio_irq_domain)) + return -EBUSY; + + zevio_irq_io = of_iomap(node, 0); + BUG_ON(!zevio_irq_io); + + /* Do not invert interrupt status bits */ + writel(~0, zevio_irq_io + IO_INVERT_SEL); + + /* Disable sticky interrupts */ + writel(0, zevio_irq_io + IO_STICKY_SEL); + + /* We don't use IRQ priorities. Set each IRQ to highest priority. */ + memset_io(zevio_irq_io + IO_PRIORITY_SEL, 0, MAX_INTRS * sizeof(u32)); + + /* Init IRQ and FIQ */ + zevio_init_irq_base(zevio_irq_io + IO_IRQ_BASE); + zevio_init_irq_base(zevio_irq_io + IO_FIQ_BASE); + + zevio_irq_domain = irq_domain_add_linear(node, MAX_INTRS, + &irq_generic_chip_ops, NULL); + BUG_ON(!zevio_irq_domain); + + ret = irq_alloc_domain_generic_chips(zevio_irq_domain, MAX_INTRS, 1, + "zevio_intc", handle_level_irq, + clr, 0, IRQ_GC_INIT_MASK_CACHE); + BUG_ON(ret); + + gc = irq_get_domain_generic_chip(zevio_irq_domain, 0); + gc->reg_base = zevio_irq_io; + gc->chip_types[0].chip.irq_ack = zevio_irq_ack; + gc->chip_types[0].chip.irq_mask = irq_gc_mask_disable_reg; + gc->chip_types[0].chip.irq_unmask = irq_gc_unmask_enable_reg; + gc->chip_types[0].regs.mask = IO_IRQ_BASE + IO_ENABLE; + gc->chip_types[0].regs.enable = IO_IRQ_BASE + IO_ENABLE; + gc->chip_types[0].regs.disable = IO_IRQ_BASE + IO_DISABLE; + gc->chip_types[0].regs.ack = IO_IRQ_BASE + IO_RESET; + + set_handle_irq(zevio_handle_irq); + + pr_info("TI-NSPIRE classic IRQ controller\n"); + return 0; +} + +IRQCHIP_DECLARE(zevio_irq, "lsi,zevio-intc", zevio_of_init); From 1aa9578c1a9450fb21501c4f549f5b1edb557e6d Mon Sep 17 00:00:00 2001 From: Sarah Sharp Date: Fri, 17 Jan 2014 15:38:12 -0800 Subject: [PATCH 013/596] xhci: Fix resume issues on Renesas chips in Samsung laptops Don Zickus writes: Some co-workers of mine bought Samsung laptops that had mostly usb3 ports. Those ports did not resume correctly (the driver would timeout communicating and fail). This led to frustration as suspend/resume is a common use for laptops. Poking around, I applied the reset on resume quirk to this chipset and the resume started working. Reloading the xhci_hcd module had been the temporary workaround. Signed-off-by: Sarah Sharp Reported-by: Don Zickus Tested-by: Prarit Bhargava Cc: stable # 2.6.37 --- drivers/usb/host/xhci-pci.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/usb/host/xhci-pci.c b/drivers/usb/host/xhci-pci.c index 3c898c12a06b..04f986d9234f 100644 --- a/drivers/usb/host/xhci-pci.c +++ b/drivers/usb/host/xhci-pci.c @@ -142,6 +142,11 @@ static void xhci_pci_quirks(struct device *dev, struct xhci_hcd *xhci) "QUIRK: Resetting on resume"); xhci->quirks |= XHCI_TRUST_TX_LENGTH; } + if (pdev->vendor == PCI_VENDOR_ID_RENESAS && + pdev->device == 0x0015 && + pdev->subsystem_vendor == PCI_VENDOR_ID_SAMSUNG && + pdev->subsystem_device == 0xc0cd) + xhci->quirks |= XHCI_RESET_ON_RESUME; if (pdev->vendor == PCI_VENDOR_ID_VIA) xhci->quirks |= XHCI_RESET_ON_RESUME; } From 140e3026a57ab7d830dab2f2c57796c222db0ea9 Mon Sep 17 00:00:00 2001 From: Sarah Sharp Date: Wed, 22 Jan 2014 13:35:02 -0800 Subject: [PATCH 014/596] Revert "usbcore: set lpm_capable field for LPM capable root hubs" Commit 9df89d85b407690afa46ddfbccc80bec6869971d "usbcore: set lpm_capable field for LPM capable root hubs" was created under the assumption that all USB host controllers should have USB 3.0 Link PM enabled for all devices under the hosts. Unfortunately, that's not the case. The xHCI driver relies on knowledge of the host hardware scheduler to calculate the LPM U1/U2 timeout values, and it only sets lpm_capable to one for Intel host controllers (that have the XHCI_LPM_SUPPORT quirk set). When LPM is enabled for some Fresco Logic hosts, it causes failures with a AgeStar 3UBT USB 3.0 hard drive dock: Jan 11 13:59:03 sg-laptop kernel: usb 3-1: new SuperSpeed USB device number 2 using xhci_hcd Jan 11 13:59:03 sg-laptop kernel: usb 3-1: Set SEL for device-initiated U1 failed. Jan 11 13:59:08 sg-laptop kernel: usb 3-1: Set SEL for device-initiated U2 failed. Jan 11 13:59:08 sg-laptop kernel: usb-storage 3-1:1.0: USB Mass Storage device detected Jan 11 13:59:08 sg-laptop mtp-probe[613]: checking bus 3, device 2: "/sys/devices/pci0000:00/0000:00:1c.3/0000:04:00.0/usb3/3-1" Jan 11 13:59:08 sg-laptop mtp-probe[613]: bus: 3, device: 2 was not an MTP device Jan 11 13:59:08 sg-laptop kernel: scsi6 : usb-storage 3-1:1.0 Jan 11 13:59:13 sg-laptop kernel: usb 3-1: Set SEL for device-initiated U1 failed. Jan 11 13:59:18 sg-laptop kernel: usb 3-1: Set SEL for device-initiated U2 failed. Jan 11 13:59:18 sg-laptop kernel: usbcore: registered new interface driver usb-storage Jan 11 13:59:40 sg-laptop kernel: usb 3-1: reset SuperSpeed USB device number 2 using xhci_hcd Jan 11 13:59:41 sg-laptop kernel: usb 3-1: device descriptor read/8, error -71 Jan 11 13:59:41 sg-laptop kernel: usb 3-1: reset SuperSpeed USB device number 2 using xhci_hcd Jan 11 13:59:46 sg-laptop kernel: usb 3-1: device descriptor read/8, error -110 Jan 11 13:59:46 sg-laptop kernel: scsi 6:0:0:0: Device offlined - not ready after error recovery Jan 11 13:59:46 sg-laptop kernel: usb 3-1: USB disconnect, device number 2 lspci for the affected host: 04:00.0 0c03: 1b73:1000 (rev 04) (prog-if 30 [XHCI]) Subsystem: 1043:1039 Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx- Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- SERR- Reported-by: Sergey Galanov Cc: stable@vger.kernel.org --- drivers/usb/core/hcd.c | 1 - drivers/usb/core/hub.c | 7 +------ drivers/usb/core/usb.h | 1 - 3 files changed, 1 insertion(+), 8 deletions(-) diff --git a/drivers/usb/core/hcd.c b/drivers/usb/core/hcd.c index 199aaea6bfe0..2518c3250750 100644 --- a/drivers/usb/core/hcd.c +++ b/drivers/usb/core/hcd.c @@ -1032,7 +1032,6 @@ static int register_root_hub(struct usb_hcd *hcd) dev_name(&usb_dev->dev), retval); return retval; } - usb_dev->lpm_capable = usb_device_supports_lpm(usb_dev); } retval = usb_new_device (usb_dev); diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c index babba885978d..64ea21971be2 100644 --- a/drivers/usb/core/hub.c +++ b/drivers/usb/core/hub.c @@ -128,7 +128,7 @@ struct usb_hub *usb_hub_to_struct_hub(struct usb_device *hdev) return usb_get_intfdata(hdev->actconfig->interface[0]); } -int usb_device_supports_lpm(struct usb_device *udev) +static int usb_device_supports_lpm(struct usb_device *udev) { /* USB 2.1 (and greater) devices indicate LPM support through * their USB 2.0 Extended Capabilities BOS descriptor. @@ -149,11 +149,6 @@ int usb_device_supports_lpm(struct usb_device *udev) "Power management will be impacted.\n"); return 0; } - - /* udev is root hub */ - if (!udev->parent) - return 1; - if (udev->parent->lpm_capable) return 1; diff --git a/drivers/usb/core/usb.h b/drivers/usb/core/usb.h index c49383669cd8..823857767a16 100644 --- a/drivers/usb/core/usb.h +++ b/drivers/usb/core/usb.h @@ -35,7 +35,6 @@ extern int usb_get_device_descriptor(struct usb_device *dev, unsigned int size); extern int usb_get_bos_descriptor(struct usb_device *dev); extern void usb_release_bos_descriptor(struct usb_device *dev); -extern int usb_device_supports_lpm(struct usb_device *udev); extern char *usb_cache_string(struct usb_device *udev, int index); extern int usb_set_configuration(struct usb_device *dev, int configuration); extern int usb_choose_configuration(struct usb_device *udev); From 23e21ddf6c3d9cb37aa69da496f80e8481e2c8bd Mon Sep 17 00:00:00 2001 From: Linus Walleij Date: Thu, 23 Jan 2014 15:20:01 +0100 Subject: [PATCH 015/596] regulator: ab3100: cast fix MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The AB3100 regulator driver emits a warning when compiled on 64bit systems like this: drivers/regulator/ab3100.c: In function ‘ab3100_regulator_of_probe’: srivers/regulator/ab3100.c:649:4: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast] As the int is a different size than the 64bit pointer used to pass regulator data. Switch to using an unsigned long as ID passed for the regulator to get rid of the warning. Reported-by: Lee Jones Signed-off-by: Linus Walleij Signed-off-by: Mark Brown --- drivers/regulator/ab3100.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/regulator/ab3100.c b/drivers/regulator/ab3100.c index 77b46d0b37a6..e10febe9ec34 100644 --- a/drivers/regulator/ab3100.c +++ b/drivers/regulator/ab3100.c @@ -498,7 +498,7 @@ static int ab3100_regulator_register(struct platform_device *pdev, struct ab3100_platform_data *plfdata, struct regulator_init_data *init_data, struct device_node *np, - int id) + unsigned long id) { struct regulator_desc *desc; struct ab3100_regulator *reg; @@ -646,7 +646,7 @@ ab3100_regulator_of_probe(struct platform_device *pdev, struct device_node *np) err = ab3100_regulator_register( pdev, NULL, ab3100_regulator_matches[i].init_data, ab3100_regulator_matches[i].of_node, - (int) ab3100_regulator_matches[i].driver_data); + (unsigned long)ab3100_regulator_matches[i].driver_data); if (err) { ab3100_regulators_remove(pdev); return err; From fb53a1ab88d14848dc292842e35c3bda3a665997 Mon Sep 17 00:00:00 2001 From: Aravind Gopalakrishnan Date: Thu, 23 Jan 2014 16:13:32 -0600 Subject: [PATCH 016/596] x86/quirks: Add workaround for AMD F16h Erratum792 The workaround for this Erratum is included in AGESA. But BIOSes spun only after Jan2014 will have the fix (atleast server versions of the chip). The erratum affects both embedded and server platforms and since we cannot say with certainity that ALL BIOSes on systems out in the field will have the fix, we should probably insulate ourselves in case BIOS does not do the right thing or someone is using old BIOSes. Refer to Revision Guide for AMD F16h models 00h-0fh, document 51810 Rev. 3.04, November2013 for details on the Erratum. Tested the patch on Fam16h server platform and it works fine. Signed-off-by: Aravind Gopalakrishnan Cc: Cc: Cc: Cc: Cc: Link: http://lkml.kernel.org/r/1390515212-1824-1-git-send-email-Aravind.Gopalakrishnan@amd.com [ Minor edits. ] Signed-off-by: Ingo Molnar --- arch/x86/kernel/quirks.c | 37 +++++++++++++++++++++++++++++++++++++ 1 file changed, 37 insertions(+) diff --git a/arch/x86/kernel/quirks.c b/arch/x86/kernel/quirks.c index 04ee1e2e4c02..7c6acd4b8995 100644 --- a/arch/x86/kernel/quirks.c +++ b/arch/x86/kernel/quirks.c @@ -571,3 +571,40 @@ DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_15H_NB_F5, quirk_amd_nb_node); #endif + +#ifdef CONFIG_PCI +/* + * Processor does not ensure DRAM scrub read/write sequence + * is atomic wrt accesses to CC6 save state area. Therefore + * if a concurrent scrub read/write access is to same address + * the entry may appear as if it is not written. This quirk + * applies to Fam16h models 00h-0Fh + * + * See "Revision Guide" for AMD F16h models 00h-0fh, + * document 51810 rev. 3.04, Nov 2013 + */ +static void amd_disable_seq_and_redirect_scrub(struct pci_dev *dev) +{ + u32 val; + + /* + * Suggested workaround: + * set D18F3x58[4:0] = 00h and set D18F3x5C[0] = 0b + */ + pci_read_config_dword(dev, 0x58, &val); + if (val & 0x1F) { + val &= ~(0x1F); + pci_write_config_dword(dev, 0x58, val); + } + + pci_read_config_dword(dev, 0x5C, &val); + if (val & BIT(0)) { + val &= ~BIT(0); + pci_write_config_dword(dev, 0x5c, val); + } +} + +DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_16H_NB_F3, + amd_disable_seq_and_redirect_scrub); + +#endif From 2993ae3305ad10b41e0d0bc2662f7754ee8e30fa Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Tue, 21 Jan 2014 10:22:09 +0300 Subject: [PATCH 017/596] x86/AMD/NB: Fix amd_set_subcaches() parameter type This is under CAP_SYS_ADMIN, but Smatch complains that mask comes from the user and the test for "mask > 0xf" can underflow. The fix is simple: amd_set_subcaches() should hand down not an 'int' but an 'unsigned long' like it was originally indended to do. Signed-off-by: Dan Carpenter Acked-by: Borislav Petkov Cc: Borislav Petkov Cc: Daniel J Blueman Link: http://lkml.kernel.org/r/20140121072209.GA22095@elgon.mountain Signed-off-by: Ingo Molnar --- arch/x86/include/asm/amd_nb.h | 2 +- arch/x86/kernel/amd_nb.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/x86/include/asm/amd_nb.h b/arch/x86/include/asm/amd_nb.h index a54ee1d054d9..aaac3b2fb746 100644 --- a/arch/x86/include/asm/amd_nb.h +++ b/arch/x86/include/asm/amd_nb.h @@ -19,7 +19,7 @@ extern int amd_cache_northbridges(void); extern void amd_flush_garts(void); extern int amd_numa_init(void); extern int amd_get_subcaches(int); -extern int amd_set_subcaches(int, int); +extern int amd_set_subcaches(int, unsigned long); struct amd_l3_cache { unsigned indices; diff --git a/arch/x86/kernel/amd_nb.c b/arch/x86/kernel/amd_nb.c index 59554dca96ec..dec8de4e1663 100644 --- a/arch/x86/kernel/amd_nb.c +++ b/arch/x86/kernel/amd_nb.c @@ -179,7 +179,7 @@ int amd_get_subcaches(int cpu) return (mask >> (4 * cuid)) & 0xf; } -int amd_set_subcaches(int cpu, int mask) +int amd_set_subcaches(int cpu, unsigned long mask) { static unsigned int reset, ban; struct amd_northbridge *nb = node_to_amd_nb(amd_get_nb_id(cpu)); From ec65993443736a5091b68e80ff1734548944a4b8 Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Tue, 21 Jan 2014 14:33:16 -0800 Subject: [PATCH 018/596] mm, x86: Account for TLB flushes only when debugging Bisection between 3.11 and 3.12 fingered commit 9824cf97 ("mm: vmstats: tlb flush counters") to cause overhead problems. The counters are undeniably useful but how often do we really need to debug TLB flush related issues? It does not justify taking the penalty everywhere so make it a debugging option. Signed-off-by: Mel Gorman Tested-by: Davidlohr Bueso Reviewed-by: Rik van Riel Signed-off-by: Andrew Morton Cc: Hugh Dickins Cc: Alex Shi Cc: Linus Torvalds Cc: Peter Zijlstra Link: http://lkml.kernel.org/n/tip-XzxjntugxuwpxXhcrxqqh53b@git.kernel.org Signed-off-by: Ingo Molnar --- arch/x86/include/asm/tlbflush.h | 6 +++--- arch/x86/kernel/cpu/mtrr/generic.c | 4 ++-- arch/x86/mm/tlb.c | 14 +++++++------- include/linux/vm_event_item.h | 4 +++- include/linux/vmstat.h | 8 ++++++++ mm/vmstat.c | 4 +++- 6 files changed, 26 insertions(+), 14 deletions(-) diff --git a/arch/x86/include/asm/tlbflush.h b/arch/x86/include/asm/tlbflush.h index e6d90babc245..04905bfc508b 100644 --- a/arch/x86/include/asm/tlbflush.h +++ b/arch/x86/include/asm/tlbflush.h @@ -62,7 +62,7 @@ static inline void __flush_tlb_all(void) static inline void __flush_tlb_one(unsigned long addr) { - count_vm_event(NR_TLB_LOCAL_FLUSH_ONE); + count_vm_tlb_event(NR_TLB_LOCAL_FLUSH_ONE); __flush_tlb_single(addr); } @@ -93,13 +93,13 @@ static inline void __flush_tlb_one(unsigned long addr) */ static inline void __flush_tlb_up(void) { - count_vm_event(NR_TLB_LOCAL_FLUSH_ALL); + count_vm_tlb_event(NR_TLB_LOCAL_FLUSH_ALL); __flush_tlb(); } static inline void flush_tlb_all(void) { - count_vm_event(NR_TLB_LOCAL_FLUSH_ALL); + count_vm_tlb_event(NR_TLB_LOCAL_FLUSH_ALL); __flush_tlb_all(); } diff --git a/arch/x86/kernel/cpu/mtrr/generic.c b/arch/x86/kernel/cpu/mtrr/generic.c index ce2d0a2c3e4f..0e25a1bc5ab5 100644 --- a/arch/x86/kernel/cpu/mtrr/generic.c +++ b/arch/x86/kernel/cpu/mtrr/generic.c @@ -683,7 +683,7 @@ static void prepare_set(void) __acquires(set_atomicity_lock) } /* Flush all TLBs via a mov %cr3, %reg; mov %reg, %cr3 */ - count_vm_event(NR_TLB_LOCAL_FLUSH_ALL); + count_vm_tlb_event(NR_TLB_LOCAL_FLUSH_ALL); __flush_tlb(); /* Save MTRR state */ @@ -697,7 +697,7 @@ static void prepare_set(void) __acquires(set_atomicity_lock) static void post_set(void) __releases(set_atomicity_lock) { /* Flush TLBs (no need to flush caches - they are disabled) */ - count_vm_event(NR_TLB_LOCAL_FLUSH_ALL); + count_vm_tlb_event(NR_TLB_LOCAL_FLUSH_ALL); __flush_tlb(); /* Intel (P6) standard MTRRs */ diff --git a/arch/x86/mm/tlb.c b/arch/x86/mm/tlb.c index ae699b3bbac8..05446c1cccfe 100644 --- a/arch/x86/mm/tlb.c +++ b/arch/x86/mm/tlb.c @@ -103,7 +103,7 @@ static void flush_tlb_func(void *info) if (f->flush_mm != this_cpu_read(cpu_tlbstate.active_mm)) return; - count_vm_event(NR_TLB_REMOTE_FLUSH_RECEIVED); + count_vm_tlb_event(NR_TLB_REMOTE_FLUSH_RECEIVED); if (this_cpu_read(cpu_tlbstate.state) == TLBSTATE_OK) { if (f->flush_end == TLB_FLUSH_ALL) local_flush_tlb(); @@ -131,7 +131,7 @@ void native_flush_tlb_others(const struct cpumask *cpumask, info.flush_start = start; info.flush_end = end; - count_vm_event(NR_TLB_REMOTE_FLUSH); + count_vm_tlb_event(NR_TLB_REMOTE_FLUSH); if (is_uv_system()) { unsigned int cpu; @@ -151,7 +151,7 @@ void flush_tlb_current_task(void) preempt_disable(); - count_vm_event(NR_TLB_LOCAL_FLUSH_ALL); + count_vm_tlb_event(NR_TLB_LOCAL_FLUSH_ALL); local_flush_tlb(); if (cpumask_any_but(mm_cpumask(mm), smp_processor_id()) < nr_cpu_ids) flush_tlb_others(mm_cpumask(mm), mm, 0UL, TLB_FLUSH_ALL); @@ -215,7 +215,7 @@ void flush_tlb_mm_range(struct mm_struct *mm, unsigned long start, /* tlb_flushall_shift is on balance point, details in commit log */ if ((end - start) >> PAGE_SHIFT > act_entries >> tlb_flushall_shift) { - count_vm_event(NR_TLB_LOCAL_FLUSH_ALL); + count_vm_tlb_event(NR_TLB_LOCAL_FLUSH_ALL); local_flush_tlb(); } else { if (has_large_page(mm, start, end)) { @@ -224,7 +224,7 @@ void flush_tlb_mm_range(struct mm_struct *mm, unsigned long start, } /* flush range by one by one 'invlpg' */ for (addr = start; addr < end; addr += PAGE_SIZE) { - count_vm_event(NR_TLB_LOCAL_FLUSH_ONE); + count_vm_tlb_event(NR_TLB_LOCAL_FLUSH_ONE); __flush_tlb_single(addr); } @@ -262,7 +262,7 @@ void flush_tlb_page(struct vm_area_struct *vma, unsigned long start) static void do_flush_tlb_all(void *info) { - count_vm_event(NR_TLB_REMOTE_FLUSH_RECEIVED); + count_vm_tlb_event(NR_TLB_REMOTE_FLUSH_RECEIVED); __flush_tlb_all(); if (this_cpu_read(cpu_tlbstate.state) == TLBSTATE_LAZY) leave_mm(smp_processor_id()); @@ -270,7 +270,7 @@ static void do_flush_tlb_all(void *info) void flush_tlb_all(void) { - count_vm_event(NR_TLB_REMOTE_FLUSH); + count_vm_tlb_event(NR_TLB_REMOTE_FLUSH); on_each_cpu(do_flush_tlb_all, NULL, 1); } diff --git a/include/linux/vm_event_item.h b/include/linux/vm_event_item.h index c557c6d096de..3a712e2e7d76 100644 --- a/include/linux/vm_event_item.h +++ b/include/linux/vm_event_item.h @@ -71,12 +71,14 @@ enum vm_event_item { PGPGIN, PGPGOUT, PSWPIN, PSWPOUT, THP_ZERO_PAGE_ALLOC, THP_ZERO_PAGE_ALLOC_FAILED, #endif +#ifdef CONFIG_DEBUG_TLBFLUSH #ifdef CONFIG_SMP NR_TLB_REMOTE_FLUSH, /* cpu tried to flush others' tlbs */ NR_TLB_REMOTE_FLUSH_RECEIVED,/* cpu received ipi for flush */ -#endif +#endif /* CONFIG_SMP */ NR_TLB_LOCAL_FLUSH_ALL, NR_TLB_LOCAL_FLUSH_ONE, +#endif /* CONFIG_DEBUG_TLBFLUSH */ NR_VM_EVENT_ITEMS }; diff --git a/include/linux/vmstat.h b/include/linux/vmstat.h index e4b948080d20..80ebba9c2e87 100644 --- a/include/linux/vmstat.h +++ b/include/linux/vmstat.h @@ -83,6 +83,14 @@ static inline void vm_events_fold_cpu(int cpu) #define count_vm_numa_events(x, y) do { (void)(y); } while (0) #endif /* CONFIG_NUMA_BALANCING */ +#ifdef CONFIG_DEBUG_TLBFLUSH +#define count_vm_tlb_event(x) count_vm_event(x) +#define count_vm_tlb_events(x, y) count_vm_events(x, y) +#else +#define count_vm_tlb_event(x) do {} while (0) +#define count_vm_tlb_events(x, y) do { (void)(y); } while (0) +#endif + #define __count_zone_vm_events(item, zone, delta) \ __count_vm_events(item##_NORMAL - ZONE_NORMAL + \ zone_idx(zone), delta) diff --git a/mm/vmstat.c b/mm/vmstat.c index 72496140ac08..def5dd2fbe61 100644 --- a/mm/vmstat.c +++ b/mm/vmstat.c @@ -851,12 +851,14 @@ const char * const vmstat_text[] = { "thp_zero_page_alloc", "thp_zero_page_alloc_failed", #endif +#ifdef CONFIG_DEBUG_TLBFLUSH #ifdef CONFIG_SMP "nr_tlb_remote_flush", "nr_tlb_remote_flush_received", -#endif +#endif /* CONFIG_SMP */ "nr_tlb_local_flush_all", "nr_tlb_local_flush_one", +#endif /* CONFIG_DEBUG_TLBFLUSH */ #endif /* CONFIG_VM_EVENTS_COUNTERS */ }; From 15aa368255f249df0b2af630c9487bb5471bd7da Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Tue, 21 Jan 2014 14:33:18 -0800 Subject: [PATCH 019/596] x86/mm: Clean up inconsistencies when flushing TLB ranges NR_TLB_LOCAL_FLUSH_ALL is not always accounted for correctly and the comparison with total_vm is done before taking tlb_flushall_shift into account. Clean it up. Signed-off-by: Mel Gorman Tested-by: Davidlohr Bueso Reviewed-by: Alex Shi Reviewed-by: Rik van Riel Signed-off-by: Andrew Morton Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Hugh Dickins Link: http://lkml.kernel.org/n/tip-Iz5gcahrgskIldvukulzi0hh@git.kernel.org Signed-off-by: Ingo Molnar --- arch/x86/mm/tlb.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/arch/x86/mm/tlb.c b/arch/x86/mm/tlb.c index 05446c1cccfe..5176526ddd59 100644 --- a/arch/x86/mm/tlb.c +++ b/arch/x86/mm/tlb.c @@ -189,6 +189,7 @@ void flush_tlb_mm_range(struct mm_struct *mm, unsigned long start, { unsigned long addr; unsigned act_entries, tlb_entries = 0; + unsigned long nr_base_pages; preempt_disable(); if (current->active_mm != mm) @@ -210,18 +211,17 @@ void flush_tlb_mm_range(struct mm_struct *mm, unsigned long start, tlb_entries = tlb_lli_4k[ENTRIES]; else tlb_entries = tlb_lld_4k[ENTRIES]; + /* Assume all of TLB entries was occupied by this task */ - act_entries = mm->total_vm > tlb_entries ? tlb_entries : mm->total_vm; + act_entries = tlb_entries >> tlb_flushall_shift; + act_entries = mm->total_vm > act_entries ? act_entries : mm->total_vm; + nr_base_pages = (end - start) >> PAGE_SHIFT; /* tlb_flushall_shift is on balance point, details in commit log */ - if ((end - start) >> PAGE_SHIFT > act_entries >> tlb_flushall_shift) { + if (nr_base_pages > act_entries || has_large_page(mm, start, end)) { count_vm_tlb_event(NR_TLB_LOCAL_FLUSH_ALL); local_flush_tlb(); } else { - if (has_large_page(mm, start, end)) { - local_flush_tlb(); - goto flush_all; - } /* flush range by one by one 'invlpg' */ for (addr = start; addr < end; addr += PAGE_SIZE) { count_vm_tlb_event(NR_TLB_LOCAL_FLUSH_ONE); From 71b54f8263860a37dd9f50f81880a9d681fd9c10 Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Tue, 21 Jan 2014 14:33:19 -0800 Subject: [PATCH 020/596] x86/mm: Eliminate redundant page table walk during TLB range flushing When choosing between doing an address space or ranged flush, the x86 implementation of flush_tlb_mm_range takes into account whether there are any large pages in the range. A per-page flush typically requires fewer entries than would covered by a single large page and the check is redundant. There is one potential exception. THP migration flushes single THP entries and it conceivably would benefit from flushing a single entry instead of the mm. However, this flush is after a THP allocation, copy and page table update potentially with any other threads serialised behind it. In comparison to that, the flush is noise. It makes more sense to optimise balancing to require fewer flushes than to optimise the flush itself. This patch deletes the redundant huge page check. Signed-off-by: Mel Gorman Tested-by: Davidlohr Bueso Reviewed-by: Rik van Riel Signed-off-by: Andrew Morton Cc: Alex Shi Cc: Linus Torvalds Cc: Peter Zijlstra Link: http://lkml.kernel.org/n/tip-sgei1drpOcburujPsfh6ovmo@git.kernel.org Signed-off-by: Ingo Molnar --- arch/x86/mm/tlb.c | 28 +--------------------------- 1 file changed, 1 insertion(+), 27 deletions(-) diff --git a/arch/x86/mm/tlb.c b/arch/x86/mm/tlb.c index 5176526ddd59..dd8dda167a24 100644 --- a/arch/x86/mm/tlb.c +++ b/arch/x86/mm/tlb.c @@ -158,32 +158,6 @@ void flush_tlb_current_task(void) preempt_enable(); } -/* - * It can find out the THP large page, or - * HUGETLB page in tlb_flush when THP disabled - */ -static inline unsigned long has_large_page(struct mm_struct *mm, - unsigned long start, unsigned long end) -{ - pgd_t *pgd; - pud_t *pud; - pmd_t *pmd; - unsigned long addr = ALIGN(start, HPAGE_SIZE); - for (; addr < end; addr += HPAGE_SIZE) { - pgd = pgd_offset(mm, addr); - if (likely(!pgd_none(*pgd))) { - pud = pud_offset(pgd, addr); - if (likely(!pud_none(*pud))) { - pmd = pmd_offset(pud, addr); - if (likely(!pmd_none(*pmd))) - if (pmd_large(*pmd)) - return addr; - } - } - } - return 0; -} - void flush_tlb_mm_range(struct mm_struct *mm, unsigned long start, unsigned long end, unsigned long vmflag) { @@ -218,7 +192,7 @@ void flush_tlb_mm_range(struct mm_struct *mm, unsigned long start, nr_base_pages = (end - start) >> PAGE_SHIFT; /* tlb_flushall_shift is on balance point, details in commit log */ - if (nr_base_pages > act_entries || has_large_page(mm, start, end)) { + if (nr_base_pages > act_entries) { count_vm_tlb_event(NR_TLB_LOCAL_FLUSH_ALL); local_flush_tlb(); } else { From f98b7a772ab51b52ca4d2a14362fc0e0c8a2e0f3 Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Tue, 21 Jan 2014 14:33:21 -0800 Subject: [PATCH 021/596] x86: mm: change tlb_flushall_shift for IvyBridge There was a large performance regression that was bisected to commit 611ae8e3 ("x86/tlb: enable tlb flush range support for x86"). This patch simply changes the default balance point between a local and global flush for IvyBridge. In the interest of allowing the tests to be reproduced, this patch was tested using mmtests 0.15 with the following configurations configs/config-global-dhp__tlbflush-performance configs/config-global-dhp__scheduler-performance configs/config-global-dhp__network-performance Results are from two machines Ivybridge 4 threads: Intel(R) Core(TM) i3-3240 CPU @ 3.40GHz Ivybridge 8 threads: Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz Page fault microbenchmark showed nothing interesting. Ebizzy was configured to run multiple iterations and threads. Thread counts ranged from 1 to NR_CPUS*2. For each thread count, it ran 100 iterations and each iteration lasted 10 seconds. Ivybridge 4 threads 3.13.0-rc7 3.13.0-rc7 vanilla altshift-v3 Mean 1 6395.44 ( 0.00%) 6789.09 ( 6.16%) Mean 2 7012.85 ( 0.00%) 8052.16 ( 14.82%) Mean 3 6403.04 ( 0.00%) 6973.74 ( 8.91%) Mean 4 6135.32 ( 0.00%) 6582.33 ( 7.29%) Mean 5 6095.69 ( 0.00%) 6526.68 ( 7.07%) Mean 6 6114.33 ( 0.00%) 6416.64 ( 4.94%) Mean 7 6085.10 ( 0.00%) 6448.51 ( 5.97%) Mean 8 6120.62 ( 0.00%) 6462.97 ( 5.59%) Ivybridge 8 threads 3.13.0-rc7 3.13.0-rc7 vanilla altshift-v3 Mean 1 7336.65 ( 0.00%) 7787.02 ( 6.14%) Mean 2 8218.41 ( 0.00%) 9484.13 ( 15.40%) Mean 3 7973.62 ( 0.00%) 8922.01 ( 11.89%) Mean 4 7798.33 ( 0.00%) 8567.03 ( 9.86%) Mean 5 7158.72 ( 0.00%) 8214.23 ( 14.74%) Mean 6 6852.27 ( 0.00%) 7952.45 ( 16.06%) Mean 7 6774.65 ( 0.00%) 7536.35 ( 11.24%) Mean 8 6510.50 ( 0.00%) 6894.05 ( 5.89%) Mean 12 6182.90 ( 0.00%) 6661.29 ( 7.74%) Mean 16 6100.09 ( 0.00%) 6608.69 ( 8.34%) Ebizzy hits the worst case scenario for TLB range flushing every time and it shows for these Ivybridge CPUs at least that the default choice is a poor on. The patch addresses the problem. Next was a tlbflush microbenchmark written by Alex Shi at http://marc.info/?l=linux-kernel&m=133727348217113 . It measures access costs while the TLB is being flushed. The expectation is that if there are always full TLB flushes that the benchmark would suffer and it benefits from range flushing There are 320 iterations of the test per thread count. The number of entries is randomly selected with a min of 1 and max of 512. To ensure a reasonably even spread of entries, the full range is broken up into 8 sections and a random number selected within that section. iteration 1, random number between 0-64 iteration 2, random number between 64-128 etc This is still a very weak methodology. When you do not know what are typical ranges, random is a reasonable choice but it can be easily argued that the opimisation was for smaller ranges and an even spread is not representative of any workload that matters. To improve this, we'd need to know the probability distribution of TLB flush range sizes for a set of workloads that are considered "common", build a synthetic trace and feed that into this benchmark. Even that is not perfect because it would not account for the time between flushes but there are limits of what can be reasonably done and still be doing something useful. If a representative synthetic trace is provided then this benchmark could be revisited and the shift values retuned. Ivybridge 4 threads 3.13.0-rc7 3.13.0-rc7 vanilla altshift-v3 Mean 1 10.50 ( 0.00%) 10.50 ( 0.03%) Mean 2 17.59 ( 0.00%) 17.18 ( 2.34%) Mean 3 22.98 ( 0.00%) 21.74 ( 5.41%) Mean 5 47.13 ( 0.00%) 46.23 ( 1.92%) Mean 8 43.30 ( 0.00%) 42.56 ( 1.72%) Ivybridge 8 threads 3.13.0-rc7 3.13.0-rc7 vanilla altshift-v3 Mean 1 9.45 ( 0.00%) 9.36 ( 0.93%) Mean 2 9.37 ( 0.00%) 9.70 ( -3.54%) Mean 3 9.36 ( 0.00%) 9.29 ( 0.70%) Mean 5 14.49 ( 0.00%) 15.04 ( -3.75%) Mean 8 41.08 ( 0.00%) 38.73 ( 5.71%) Mean 13 32.04 ( 0.00%) 31.24 ( 2.49%) Mean 16 40.05 ( 0.00%) 39.04 ( 2.51%) For both CPUs, average access time is reduced which is good as this is the benchmark that was used to tune the shift values in the first place albeit it is now known *how* the benchmark was used. The scheduler benchmarks were somewhat inconclusive. They showed gains and losses and makes me reconsider how stable those benchmarks really are or if something else might be interfering with the test results recently. Network benchmarks were inconclusive. Almost all results were flat except for netperf-udp tests on the 4 thread machine. These results were unstable and showed large variations between reboots. It is unknown if this is a recent problems but I've noticed before that netperf-udp results tend to vary. Based on these results, changing the default for Ivybridge seems like a logical choice. Signed-off-by: Mel Gorman Tested-by: Davidlohr Bueso Reviewed-by: Alex Shi Reviewed-by: Rik van Riel Signed-off-by: Andrew Morton Cc: Linus Torvalds Cc: Peter Zijlstra Link: http://lkml.kernel.org/n/tip-cqnadffh1tiqrshthRj3Esge@git.kernel.org Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/intel.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c index ea04b342c026..bbe1b8b1f1c4 100644 --- a/arch/x86/kernel/cpu/intel.c +++ b/arch/x86/kernel/cpu/intel.c @@ -628,7 +628,7 @@ static void intel_tlb_flushall_shift_set(struct cpuinfo_x86 *c) tlb_flushall_shift = 5; break; case 0x63a: /* Ivybridge */ - tlb_flushall_shift = 1; + tlb_flushall_shift = 2; break; default: tlb_flushall_shift = 6; From b9a3b4c976c1209957326537ad5c0bb633dfd764 Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Tue, 21 Jan 2014 14:33:22 -0800 Subject: [PATCH 022/596] mm, x86: Revisit tlb_flushall_shift tuning for page flushes except on IvyBridge There was a large ebizzy performance regression that was bisected to commit 611ae8e3 (x86/tlb: enable tlb flush range support for x86). The problem was related to the tlb_flushall_shift tuning for IvyBridge which was altered. The problem is that it is not clear if the tuning values for each CPU family is correct as the methodology used to tune the values is unclear. This patch uses a conservative tlb_flushall_shift value for all CPU families except IvyBridge so the decision can be revisited if any regression is found as a result of this change. IvyBridge is an exception as testing with one methodology determined that the value of 2 is acceptable. Details are in the changelog for the patch "x86: mm: Change tlb_flushall_shift for IvyBridge". One important aspect of this to watch out for is Xen. The original commit log mentioned large performance gains on Xen. It's possible Xen is more sensitive to this value if it flushes small ranges of pages more frequently than workloads on bare metal typically do. Signed-off-by: Mel Gorman Tested-by: Davidlohr Bueso Reviewed-by: Rik van Riel Signed-off-by: Andrew Morton Cc: Alex Shi Cc: Linus Torvalds Cc: Peter Zijlstra Link: http://lkml.kernel.org/n/tip-dyzMww3fqugnhbhgo6Gxmtkw@git.kernel.org Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/amd.c | 5 +---- arch/x86/kernel/cpu/intel.c | 10 +++------- 2 files changed, 4 insertions(+), 11 deletions(-) diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index 59bfebc8c805..96abccaada33 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -768,10 +768,7 @@ static unsigned int amd_size_cache(struct cpuinfo_x86 *c, unsigned int size) static void cpu_set_tlb_flushall_shift(struct cpuinfo_x86 *c) { - tlb_flushall_shift = 5; - - if (c->x86 <= 0x11) - tlb_flushall_shift = 4; + tlb_flushall_shift = 6; } static void cpu_detect_tlb_amd(struct cpuinfo_x86 *c) diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c index bbe1b8b1f1c4..d358a3928b8f 100644 --- a/arch/x86/kernel/cpu/intel.c +++ b/arch/x86/kernel/cpu/intel.c @@ -615,21 +615,17 @@ static void intel_tlb_flushall_shift_set(struct cpuinfo_x86 *c) case 0x61d: /* six-core 45 nm xeon "Dunnington" */ tlb_flushall_shift = -1; break; + case 0x63a: /* Ivybridge */ + tlb_flushall_shift = 2; + break; case 0x61a: /* 45 nm nehalem, "Bloomfield" */ case 0x61e: /* 45 nm nehalem, "Lynnfield" */ case 0x625: /* 32 nm nehalem, "Clarkdale" */ case 0x62c: /* 32 nm nehalem, "Gulftown" */ case 0x62e: /* 45 nm nehalem-ex, "Beckton" */ case 0x62f: /* 32 nm Xeon E7 */ - tlb_flushall_shift = 6; - break; case 0x62a: /* SandyBridge */ case 0x62d: /* SandyBridge, "Romely-EP" */ - tlb_flushall_shift = 5; - break; - case 0x63a: /* Ivybridge */ - tlb_flushall_shift = 2; - break; default: tlb_flushall_shift = 6; } From a85eba8814631d0d48361c8b9a7ee0984e80c03c Mon Sep 17 00:00:00 2001 From: Toshi Kani Date: Tue, 21 Jan 2014 14:33:15 -0800 Subject: [PATCH 023/596] arch/x86/mm/srat: Skip NUMA_NO_NODE while parsing SLIT When ACPI SLIT table has an I/O locality (i.e. a locality unique to an I/O device), numa_set_distance() emits this warning message: NUMA: Warning: node ids are out of bound, from=-1 to=-1 distance=10 acpi_numa_slit_init() calls numa_set_distance() with pxm_to_node(), which assumes that all localities have been parsed with SRAT previously. SRAT does not list I/O localities, where as SLIT lists all localities including I/Os. Hence, pxm_to_node() returns NUMA_NO_NODE (-1) for an I/O locality. I/O localities are not supported and are ignored today, but emitting such warning message leads to unnecessary confusion. Change acpi_numa_slit_init() to avoid calling numa_set_distance() with NUMA_NO_NODE. Signed-off-by: Toshi Kani Acked-by: David Rientjes Signed-off-by: Andrew Morton Cc: Yinghai Lu Link: http://lkml.kernel.org/n/tip-dSvpjjvp8aMzs1ybkftxohlh@git.kernel.org Signed-off-by: Ingo Molnar --- arch/x86/mm/srat.c | 16 +++++++++++++--- 1 file changed, 13 insertions(+), 3 deletions(-) diff --git a/arch/x86/mm/srat.c b/arch/x86/mm/srat.c index 266ca912f62e..5ecf65117e6f 100644 --- a/arch/x86/mm/srat.c +++ b/arch/x86/mm/srat.c @@ -42,15 +42,25 @@ static __init inline int srat_disabled(void) return acpi_numa < 0; } -/* Callback for SLIT parsing */ +/* + * Callback for SLIT parsing. pxm_to_node() returns NUMA_NO_NODE for + * I/O localities since SRAT does not list them. I/O localities are + * not supported at this point. + */ void __init acpi_numa_slit_init(struct acpi_table_slit *slit) { int i, j; - for (i = 0; i < slit->locality_count; i++) - for (j = 0; j < slit->locality_count; j++) + for (i = 0; i < slit->locality_count; i++) { + if (pxm_to_node(i) == NUMA_NO_NODE) + continue; + for (j = 0; j < slit->locality_count; j++) { + if (pxm_to_node(j) == NUMA_NO_NODE) + continue; numa_set_distance(pxm_to_node(i), pxm_to_node(j), slit->entry[slit->locality_count * i + j]); + } + } } /* Callback for Proximity Domain -> x2APIC mapping */ From edc6bc784028f2ef80dff0ff697363ff5a06e3a3 Mon Sep 17 00:00:00 2001 From: David Cohen Date: Tue, 21 Jan 2014 10:41:39 -0800 Subject: [PATCH 024/596] x86/intel/mid: Fix X86_INTEL_MID dependencies This patch fixes the following warning: warning: (X86_INTEL_MID) selects INTEL_SCU_IPC which has unmet direct dependencies (X86 && X86_PLATFORM_DEVICES && X86_INTEL_MID) It happens because when selected, X86_INTEL_MID tries to select INTEL_SCU_IPC regardless all its dependencies are met or not. This patch fixes it by adding the missing X86_PLATFORM_DEVICES dependency to X86_INTEL_MID. Reported-by: kbuild test robot Signed-off-by: David Cohen Link: http://lkml.kernel.org/r/1390329699-20782-1-git-send-email-david.a.cohen@linux.intel.com Signed-off-by: Ingo Molnar --- arch/x86/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index cb9af474dfca..3b6922ebf170 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -443,6 +443,7 @@ config X86_INTEL_MID bool "Intel MID platform support" depends on X86_32 depends on X86_EXTENDED_PLATFORM + depends on X86_PLATFORM_DEVICES depends on PCI depends on PCI_GOANY depends on X86_IO_APIC From 1b5e1c6e6e728774ba7a1e8286d4833e4a16e918 Mon Sep 17 00:00:00 2001 From: Matthias Brugger Date: Fri, 17 Jan 2014 18:02:38 +0100 Subject: [PATCH 025/596] power: ds2782_battery: Typo in comment Change missleading comment to actual shift value provided by the chip. Signed-off-by: Matthias Brugger --- drivers/power/ds2782_battery.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/power/ds2782_battery.c b/drivers/power/ds2782_battery.c index 563174891c90..041f9b638d28 100644 --- a/drivers/power/ds2782_battery.c +++ b/drivers/power/ds2782_battery.c @@ -192,7 +192,7 @@ static int ds2786_get_voltage(struct ds278x_info *info, int *voltage_uV) /* * Voltage is measured in units of 1.22mV. The voltage is stored as - * a 10-bit number plus sign, in the upper bits of a 16-bit register + * a 12-bit number plus sign, in the upper bits of a 16-bit register */ err = ds278x_read_reg16(info, DS278x_REG_VOLT_MSB, &raw); if (err) From 4fa99230eade164137d94178b12e1e5f1c3d2bdd Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Tue, 7 Jan 2014 21:12:29 +0800 Subject: [PATCH 026/596] isp1704_charger: remove useless check in isp1704_charger_probe() Neither devm_usb_get_phy_by_phandle() nor devm_usb_get_phy() can return a NULL result, so remove the useless !isp->phy check. Signed-off-by: Wei Yongjun --- drivers/power/isp1704_charger.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/drivers/power/isp1704_charger.c b/drivers/power/isp1704_charger.c index 80edb7d8cb54..0b4cf9d63291 100644 --- a/drivers/power/isp1704_charger.c +++ b/drivers/power/isp1704_charger.c @@ -444,8 +444,6 @@ static int isp1704_charger_probe(struct platform_device *pdev) ret = PTR_ERR(isp->phy); goto fail0; } - if (!isp->phy) - goto fail0; isp->dev = &pdev->dev; platform_set_drvdata(pdev, isp); From 4f8dd2c10bec8a6c1928c3ce00ac68dba5a5d667 Mon Sep 17 00:00:00 2001 From: Sebastian Reichel Date: Sat, 11 Jan 2014 01:47:48 +0100 Subject: [PATCH 027/596] dt: binding documentation for bq2415x charger Add devicetree binding documentation for bq2415x charger. Signed-off-by: Sebastian Reichel Acked-by: Pavel Machek --- .../devicetree/bindings/power/bq2415x.txt | 47 +++++++++++++++++++ 1 file changed, 47 insertions(+) create mode 100644 Documentation/devicetree/bindings/power/bq2415x.txt diff --git a/Documentation/devicetree/bindings/power/bq2415x.txt b/Documentation/devicetree/bindings/power/bq2415x.txt new file mode 100644 index 000000000000..d0327f0b59ad --- /dev/null +++ b/Documentation/devicetree/bindings/power/bq2415x.txt @@ -0,0 +1,47 @@ +Binding for TI bq2415x Li-Ion Charger + +Required properties: +- compatible: Should contain one of the following: + * "ti,bq24150" + * "ti,bq24150" + * "ti,bq24150a" + * "ti,bq24151" + * "ti,bq24151a" + * "ti,bq24152" + * "ti,bq24153" + * "ti,bq24153a" + * "ti,bq24155" + * "ti,bq24156" + * "ti,bq24156a" + * "ti,bq24158" +- reg: integer, i2c address of the device. +- ti,current-limit: integer, initial maximum current charger can pull + from power supply in mA. +- ti,weak-battery-voltage: integer, weak battery voltage threshold in mV. + The chip will use slow precharge if battery voltage + is below this value. +- ti,battery-regulation-voltage: integer, maximum charging voltage in mV. +- ti,charge-current: integer, maximum charging current in mA. +- ti,termination-current: integer, charge will be terminated when current in + constant-voltage phase drops below this value (in mA). +- ti,resistor-sense: integer, value of sensing resistor in milliohm. + +Optional properties: +- ti,usb-charger-detection: phandle to usb charger detection device. + (required for auto mode) + +Example from Nokia N900: + +bq24150a { + compatible = "ti,bq24150a"; + reg = <0x6b>; + + ti,current-limit = <100>; + ti,weak-battery-voltage = <3400>; + ti,battery-regulation-voltage = <4200>; + ti,charge-current = <650>; + ti,termination-current = <100>; + ti,resistor-sense = <68>; + + ti,usb-charger-detection = <&isp1704>; +}; From 0d4dd797564cddc1f71ab0b239e9ea50ddd40b2a Mon Sep 17 00:00:00 2001 From: Baruch Siach Date: Sat, 25 Jan 2014 23:50:23 +0200 Subject: [PATCH 028/596] perf/doc: Remove mention of non-existent set_perf_event_pending() from design.txt set_perf_event_pending() was removed in e360adbe ("irq_work: Add generic hardirq context callbacks"). Signed-off-by: Baruch Siach Cc: Peter Zijlstra Cc: Paul Mackerras Cc: Arnaldo Carvalho de Melo Link: http://lkml.kernel.org/r/4c54761865d40210be0628cb84701afc5d57b5d8.1390686193.git.baruch@tkos.co.il Signed-off-by: Ingo Molnar --- tools/perf/design.txt | 1 - 1 file changed, 1 deletion(-) diff --git a/tools/perf/design.txt b/tools/perf/design.txt index 67e5d0cace85..63a0e6f04a01 100644 --- a/tools/perf/design.txt +++ b/tools/perf/design.txt @@ -454,7 +454,6 @@ So to start with, in order to add HAVE_PERF_EVENTS to your Kconfig, you will need at least this: - asm/perf_event.h - a basic stub will suffice at first - support for atomic64 types (and associated helper functions) - - set_perf_event_pending() implemented If your architecture does have hardware capabilities, you can override the weak stub hw_perf_event_init() to register hardware counters. From 950b8354716eb1f9c0b39777d379efa5f4125c04 Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Wed, 22 Jan 2014 21:58:46 +0200 Subject: [PATCH 029/596] perf tools: Demangle kernel and kernel module symbols too Some kernels contain C++ code, and thus their symbols need to be demangled. This allows 'perf kvm top' to generate readable output. Signed-off-by: Avi Kivity Cc: Ingo Molnar Cc: Paul Mackerras Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/26f71bf5bf7ee1408e3f1a803556d5df18223ef1.1390420726.git.avi@cloudius-systems.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/symbol-elf.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/perf/util/symbol-elf.c b/tools/perf/util/symbol-elf.c index 759456728703..8f12f0f5101d 100644 --- a/tools/perf/util/symbol-elf.c +++ b/tools/perf/util/symbol-elf.c @@ -922,6 +922,7 @@ int dso__load_sym(struct dso *dso, struct map *map, (u64)shdr.sh_offset); sym.st_value -= shdr.sh_addr - shdr.sh_offset; } +new_symbol: /* * We need to figure out if the object was created from C++ sources * DWARF DW_compile_unit has this, but we don't always have access @@ -933,7 +934,6 @@ int dso__load_sym(struct dso *dso, struct map *map, if (demangled != NULL) elf_name = demangled; } -new_symbol: f = symbol__new(sym.st_value, sym.st_size, GELF_ST_BIND(sym.st_info), elf_name); free(demangled); From 6e5f52674ff0756e61a8879f6232b9ac33735cba Mon Sep 17 00:00:00 2001 From: Baruch Siach Date: Sat, 25 Jan 2014 22:36:13 +0200 Subject: [PATCH 030/596] spi: spi.h: clarify the documentation of transfer_one Explicitly note the transfer_one and transfer_one_message are mutually exclusive, to make the text a little more newcomers friendly. Signed-off-by: Baruch Siach Signed-off-by: Mark Brown --- include/linux/spi/spi.h | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/include/linux/spi/spi.h b/include/linux/spi/spi.h index a1d4ca290862..ad050f0dd39d 100644 --- a/include/linux/spi/spi.h +++ b/include/linux/spi/spi.h @@ -287,7 +287,10 @@ static inline void spi_unregister_driver(struct spi_driver *sdrv) * - return 1 if the transfer is still in progress. When * the driver is finished with this transfer it must * call spi_finalize_current_transfer() so the subsystem - * can issue the next transfer + * can issue the next transfer. Note: transfer_one and + * transfer_one_message are mutually exclusive; when both + * are set, the generic subsystem does not call your + * transfer_one callback. * @unprepare_message: undo any work done by prepare_message(). * @cs_gpios: Array of GPIOs to use as chip select lines; one per CS * number. Any individual value may be -ENOENT for CS lines that From 18cc0adb132b390d026c574deec29bf182aede51 Mon Sep 17 00:00:00 2001 From: Baruch Siach Date: Sat, 25 Jan 2014 22:36:14 +0200 Subject: [PATCH 031/596] spi: document the transfer_one spi_master callback Signed-off-by: Baruch Siach Signed-off-by: Mark Brown --- Documentation/spi/spi-summary | 15 +++++++++++++++ 1 file changed, 15 insertions(+) diff --git a/Documentation/spi/spi-summary b/Documentation/spi/spi-summary index f72e0d1e0da8..dcaad476789c 100644 --- a/Documentation/spi/spi-summary +++ b/Documentation/spi/spi-summary @@ -545,6 +545,21 @@ SPI MASTER METHODS spi_finalize_current_message() so the subsystem can issue the next transfer. This may sleep. + master->transfer_one(struct spi_master *master, struct spi_device *spi, + struct spi_transfer *transfer) + The subsystem calls the driver to transfer a single transfer while + queuing transfers that arrive in the meantime. When the driver is + finished with this transfer, it must call + spi_finalize_current_transfer() so the subsystem can issue the next + transfer. This may sleep. Note: transfer_one and transfer_one_message + are mutually exclusive; when both are set, the generic subsystem does + not call your transfer_one callback. + + Return values: + negative errno: error + 0: transfer is finished + 1: transfer is still in progress + DEPRECATED METHODS master->transfer(struct spi_device *spi, struct spi_message *message) From e9305331f64ab9e5210baa4307355635c5e849f4 Mon Sep 17 00:00:00 2001 From: Baruch Siach Date: Sat, 25 Jan 2014 22:36:15 +0200 Subject: [PATCH 032/596] spi: correct the transfer_one_message documentation wording The transfer_one_message callback handles messages, not transfers. Signed-off-by: Baruch Siach Signed-off-by: Mark Brown --- Documentation/spi/spi-summary | 2 +- include/linux/spi/spi.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/Documentation/spi/spi-summary b/Documentation/spi/spi-summary index dcaad476789c..7982bcc4d151 100644 --- a/Documentation/spi/spi-summary +++ b/Documentation/spi/spi-summary @@ -543,7 +543,7 @@ SPI MASTER METHODS queuing transfers that arrive in the meantime. When the driver is finished with this message, it must call spi_finalize_current_message() so the subsystem can issue the next - transfer. This may sleep. + message. This may sleep. master->transfer_one(struct spi_master *master, struct spi_device *spi, struct spi_transfer *transfer) diff --git a/include/linux/spi/spi.h b/include/linux/spi/spi.h index ad050f0dd39d..4203c66d8803 100644 --- a/include/linux/spi/spi.h +++ b/include/linux/spi/spi.h @@ -273,7 +273,7 @@ static inline void spi_unregister_driver(struct spi_driver *sdrv) * message while queuing transfers that arrive in the meantime. When the * driver is finished with this message, it must call * spi_finalize_current_message() so the subsystem can issue the next - * transfer + * message * @unprepare_transfer_hardware: there are currently no more messages on the * queue so the subsystem notifies the driver that it may relax the * hardware by issuing this call From d4b4ff8e28b474fac0fbfa9cfc40f88b9e41e380 Mon Sep 17 00:00:00 2001 From: Keith Busch Date: Tue, 10 Dec 2013 13:10:37 -0700 Subject: [PATCH 033/596] NVMe: Schedule reset for failed controllers Schedules a controller reset when it indicates it has a failed status. If the device does not become ready after a reset, the pci device will be scheduled for removal. Signed-off-by: Keith Busch [fixed checkpatch issue] Signed-off-by: Matthew Wilcox --- drivers/block/nvme-core.c | 21 +++++++++++++++++++-- include/linux/nvme.h | 1 + 2 files changed, 20 insertions(+), 2 deletions(-) diff --git a/drivers/block/nvme-core.c b/drivers/block/nvme-core.c index 000bca43c23b..2f5b9f5f5a21 100644 --- a/drivers/block/nvme-core.c +++ b/drivers/block/nvme-core.c @@ -60,6 +60,8 @@ static LIST_HEAD(dev_list); static struct task_struct *nvme_thread; static struct workqueue_struct *nvme_workq; +static void nvme_reset_failed_dev(struct work_struct *ws); + /* * An NVM Express queue. Each device has at least two (one for admin * commands and one for I/O commands). @@ -1612,13 +1614,25 @@ static void nvme_resubmit_bios(struct nvme_queue *nvmeq) static int nvme_kthread(void *data) { - struct nvme_dev *dev; + struct nvme_dev *dev, *next; while (!kthread_should_stop()) { set_current_state(TASK_INTERRUPTIBLE); spin_lock(&dev_list_lock); - list_for_each_entry(dev, &dev_list, node) { + list_for_each_entry_safe(dev, next, &dev_list, node) { int i; + if (readl(&dev->bar->csts) & NVME_CSTS_CFS && + dev->initialized) { + if (work_busy(&dev->reset_work)) + continue; + list_del_init(&dev->node); + dev_warn(&dev->pci_dev->dev, + "Failed status, reset controller\n"); + INIT_WORK(&dev->reset_work, + nvme_reset_failed_dev); + queue_work(nvme_workq, &dev->reset_work); + continue; + } for (i = 0; i < dev->queue_count; i++) { struct nvme_queue *nvmeq = dev->queues[i]; if (!nvmeq) @@ -2006,6 +2020,7 @@ static void nvme_dev_shutdown(struct nvme_dev *dev) { int i; + dev->initialized = 0; for (i = dev->queue_count - 1; i >= 0; i--) nvme_disable_queue(dev, i); @@ -2196,6 +2211,7 @@ static int nvme_dev_resume(struct nvme_dev *dev) queue_work(nvme_workq, &dev->reset_work); spin_unlock(&dev_list_lock); } + dev->initialized = 1; return 0; } @@ -2269,6 +2285,7 @@ static int nvme_probe(struct pci_dev *pdev, const struct pci_device_id *id) if (result) goto remove; + dev->initialized = 1; kref_init(&dev->kref); return 0; diff --git a/include/linux/nvme.h b/include/linux/nvme.h index eed81cc56d7e..117d877e8be5 100644 --- a/include/linux/nvme.h +++ b/include/linux/nvme.h @@ -95,6 +95,7 @@ struct nvme_dev { u32 max_hw_sectors; u32 stripe_size; u16 oncs; + u8 initialized; }; /* From c30341dc3c436cf43508cd44cdfbb3810c38c195 Mon Sep 17 00:00:00 2001 From: Keith Busch Date: Tue, 10 Dec 2013 13:10:38 -0700 Subject: [PATCH 034/596] NVMe: Abort timed out commands Send nvme abort command to io requests that have timed out on an initialized device. If the command is not returned after another timeout, schedule the controller for reset. Signed-off-by: Keith Busch [fix endianness issues] Signed-off-by: Matthew Wilcox --- drivers/block/nvme-core.c | 68 ++++++++++++++++++++++++++++++++++++++- include/linux/nvme.h | 1 + include/uapi/linux/nvme.h | 11 +++++++ 3 files changed, 79 insertions(+), 1 deletion(-) diff --git a/drivers/block/nvme-core.c b/drivers/block/nvme-core.c index 2f5b9f5f5a21..75049532df96 100644 --- a/drivers/block/nvme-core.c +++ b/drivers/block/nvme-core.c @@ -83,6 +83,7 @@ struct nvme_queue { u16 sq_head; u16 sq_tail; u16 cq_head; + u16 qid; u8 cq_phase; u8 cqe_seen; u8 q_suspended; @@ -100,6 +101,7 @@ static inline void _nvme_check_size(void) BUILD_BUG_ON(sizeof(struct nvme_delete_queue) != 64); BUILD_BUG_ON(sizeof(struct nvme_features) != 64); BUILD_BUG_ON(sizeof(struct nvme_format_cmd) != 64); + BUILD_BUG_ON(sizeof(struct nvme_abort_cmd) != 64); BUILD_BUG_ON(sizeof(struct nvme_command) != 64); BUILD_BUG_ON(sizeof(struct nvme_id_ctrl) != 4096); BUILD_BUG_ON(sizeof(struct nvme_id_ns) != 4096); @@ -114,6 +116,7 @@ struct nvme_cmd_info { nvme_completion_fn fn; void *ctx; unsigned long timeout; + int aborted; }; static struct nvme_cmd_info *nvme_cmd_info(struct nvme_queue *nvmeq) @@ -157,6 +160,7 @@ static int alloc_cmdid(struct nvme_queue *nvmeq, void *ctx, info[cmdid].fn = handler; info[cmdid].ctx = ctx; info[cmdid].timeout = jiffies + timeout; + info[cmdid].aborted = 0; return cmdid; } @@ -175,6 +179,7 @@ static int alloc_cmdid_killable(struct nvme_queue *nvmeq, void *ctx, #define CMD_CTX_COMPLETED (0x310 + CMD_CTX_BASE) #define CMD_CTX_INVALID (0x314 + CMD_CTX_BASE) #define CMD_CTX_FLUSH (0x318 + CMD_CTX_BASE) +#define CMD_CTX_ABORT (0x31C + CMD_CTX_BASE) static void special_completion(struct nvme_dev *dev, void *ctx, struct nvme_completion *cqe) @@ -183,6 +188,10 @@ static void special_completion(struct nvme_dev *dev, void *ctx, return; if (ctx == CMD_CTX_FLUSH) return; + if (ctx == CMD_CTX_ABORT) { + ++dev->abort_limit; + return; + } if (ctx == CMD_CTX_COMPLETED) { dev_warn(&dev->pci_dev->dev, "completed id %d twice on queue %d\n", @@ -1004,6 +1013,56 @@ int nvme_set_features(struct nvme_dev *dev, unsigned fid, unsigned dword11, return nvme_submit_admin_cmd(dev, &c, result); } +/** + * nvme_abort_cmd - Attempt aborting a command + * @cmdid: Command id of a timed out IO + * @queue: The queue with timed out IO + * + * Schedule controller reset if the command was already aborted once before and + * still hasn't been returned to the driver, or if this is the admin queue. + */ +static void nvme_abort_cmd(int cmdid, struct nvme_queue *nvmeq) +{ + int a_cmdid; + struct nvme_command cmd; + struct nvme_dev *dev = nvmeq->dev; + struct nvme_cmd_info *info = nvme_cmd_info(nvmeq); + + if (!nvmeq->qid || info[cmdid].aborted) { + if (work_busy(&dev->reset_work)) + return; + list_del_init(&dev->node); + dev_warn(&dev->pci_dev->dev, + "I/O %d QID %d timeout, reset controller\n", cmdid, + nvmeq->qid); + INIT_WORK(&dev->reset_work, nvme_reset_failed_dev); + queue_work(nvme_workq, &dev->reset_work); + return; + } + + if (!dev->abort_limit) + return; + + a_cmdid = alloc_cmdid(dev->queues[0], CMD_CTX_ABORT, special_completion, + ADMIN_TIMEOUT); + if (a_cmdid < 0) + return; + + memset(&cmd, 0, sizeof(cmd)); + cmd.abort.opcode = nvme_admin_abort_cmd; + cmd.abort.cid = cmdid; + cmd.abort.sqid = cpu_to_le16(nvmeq->qid); + cmd.abort.command_id = a_cmdid; + + --dev->abort_limit; + info[cmdid].aborted = 1; + info[cmdid].timeout = jiffies + ADMIN_TIMEOUT; + + dev_warn(nvmeq->q_dmadev, "Aborting I/O %d QID %d\n", cmdid, + nvmeq->qid); + nvme_submit_cmd(dev->queues[0], &cmd); +} + /** * nvme_cancel_ios - Cancel outstanding I/Os * @queue: The queue to cancel I/Os on @@ -1027,7 +1086,12 @@ static void nvme_cancel_ios(struct nvme_queue *nvmeq, bool timeout) continue; if (info[cmdid].ctx == CMD_CTX_CANCELLED) continue; - dev_warn(nvmeq->q_dmadev, "Cancelling I/O %d\n", cmdid); + if (timeout && nvmeq->dev->initialized) { + nvme_abort_cmd(cmdid, nvmeq); + continue; + } + dev_warn(nvmeq->q_dmadev, "Cancelling I/O %d QID %d\n", cmdid, + nvmeq->qid); ctx = cancel_cmdid(nvmeq, cmdid, &fn); fn(nvmeq->dev, ctx, &cqe); } @@ -1119,6 +1183,7 @@ static struct nvme_queue *nvme_alloc_queue(struct nvme_dev *dev, int qid, nvmeq->q_db = &dev->dbs[qid * 2 * dev->db_stride]; nvmeq->q_depth = depth; nvmeq->cq_vector = vector; + nvmeq->qid = qid; nvmeq->q_suspended = 1; dev->queue_count++; @@ -1929,6 +1994,7 @@ static int nvme_dev_add(struct nvme_dev *dev) ctrl = mem; nn = le32_to_cpup(&ctrl->nn); dev->oncs = le16_to_cpup(&ctrl->oncs); + dev->abort_limit = ctrl->acl + 1; memcpy(dev->serial, ctrl->sn, sizeof(ctrl->sn)); memcpy(dev->model, ctrl->mn, sizeof(ctrl->mn)); memcpy(dev->firmware_rev, ctrl->fr, sizeof(ctrl->fr)); diff --git a/include/linux/nvme.h b/include/linux/nvme.h index 117d877e8be5..69ae03f6eb15 100644 --- a/include/linux/nvme.h +++ b/include/linux/nvme.h @@ -95,6 +95,7 @@ struct nvme_dev { u32 max_hw_sectors; u32 stripe_size; u16 oncs; + u16 abort_limit; u8 initialized; }; diff --git a/include/uapi/linux/nvme.h b/include/uapi/linux/nvme.h index 989c04e0c563..e5ab62201119 100644 --- a/include/uapi/linux/nvme.h +++ b/include/uapi/linux/nvme.h @@ -350,6 +350,16 @@ struct nvme_delete_queue { __u32 rsvd11[5]; }; +struct nvme_abort_cmd { + __u8 opcode; + __u8 flags; + __u16 command_id; + __u32 rsvd1[9]; + __le16 sqid; + __u16 cid; + __u32 rsvd11[5]; +}; + struct nvme_download_firmware { __u8 opcode; __u8 flags; @@ -384,6 +394,7 @@ struct nvme_command { struct nvme_download_firmware dlfw; struct nvme_format_cmd format; struct nvme_dsm_cmd dsm; + struct nvme_abort_cmd abort; }; }; From 0e53d18051725da46cbccfb7874a6422d4d4f274 Mon Sep 17 00:00:00 2001 From: Keith Busch Date: Tue, 10 Dec 2013 13:10:39 -0700 Subject: [PATCH 035/596] NVMe: Surprise removal handling This adds checks to see if the nvme pci device was removed. The check reads the status register for the value of -1, which it should never be unless the device is no longer present. If a user performs a surprise removal on an nvme device, the driver will be notified either by the pci driver remove callback if the platform's slot is capable of this event, or via reading the device BAR status register, which will indicate controller failure and trigger a reset. Either way, the device is not present so all outstanding commands would timeout. This will not send queue deletion commands to a drive that isn't present and fail after ioremap, significantly speeding up surprise removal; previously this took over 2 minutes per IO queue pair created, but this will complete removing the device within a few seconds. Signed-off-by: Keith Busch Signed-off-by: Matthew Wilcox --- drivers/block/nvme-core.c | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) diff --git a/drivers/block/nvme-core.c b/drivers/block/nvme-core.c index 75049532df96..a51126129784 100644 --- a/drivers/block/nvme-core.c +++ b/drivers/block/nvme-core.c @@ -1140,8 +1140,9 @@ static void nvme_disable_queue(struct nvme_dev *dev, int qid) irq_set_affinity_hint(vector, NULL); free_irq(vector, nvmeq); - /* Don't tell the adapter to delete the admin queue */ - if (qid) { + /* Don't tell the adapter to delete the admin queue. + * Don't tell a removed adapter to delete IO queues. */ + if (qid && readl(&dev->bar->csts) != -1) { adapter_delete_sq(dev, qid); adapter_delete_cq(dev, qid); } @@ -2052,12 +2053,18 @@ static int nvme_dev_map(struct nvme_dev *dev) dev->bar = ioremap(pci_resource_start(pdev, 0), 8192); if (!dev->bar) goto disable; - + if (readl(&dev->bar->csts) == -1) { + result = -ENODEV; + goto unmap; + } dev->db_stride = 1 << NVME_CAP_STRIDE(readq(&dev->bar->cap)); dev->dbs = ((void __iomem *)dev->bar) + 4096; return 0; + unmap: + iounmap(dev->bar); + dev->bar = NULL; disable: pci_release_regions(pdev); disable_pci: From 4d115420707afcabe77d2535e092356df6664b70 Mon Sep 17 00:00:00 2001 From: Keith Busch Date: Tue, 10 Dec 2013 13:10:40 -0700 Subject: [PATCH 036/596] NVMe: Async IO queue deletion This attempts to delete all IO queues at the same time asynchronously on shutdown. This is necessary for a present device that is not responding; a shutdown operation previously would take 2 minutes per queue-pair to timeout before moving on to the next queue, making a device removal appear to take a very long time or "hung" as reported by users. In the previous worst case, a removal may be stuck forever until a kill signal is given if there are more than 32 queue pairs since it would run out of admin command IDs after over an hour of timed out sync commands (admin queue depth is 64). This patch will wait for the admin command timeout for all commands to complete, so the worst case now for an unresponsive controller is 60 seconds, though that still seems like a long time. Since this adds another way to take queues offline, some duplicate code resulted so I moved these into more convienient functions. Signed-off-by: Keith Busch [make functions static, correct line length and whitespace issues] Signed-off-by: Matthew Wilcox --- drivers/block/nvme-core.c | 229 ++++++++++++++++++++++++++++++++++++-- 1 file changed, 217 insertions(+), 12 deletions(-) diff --git a/drivers/block/nvme-core.c b/drivers/block/nvme-core.c index a51126129784..1c8a82fbbc37 100644 --- a/drivers/block/nvme-core.c +++ b/drivers/block/nvme-core.c @@ -62,6 +62,14 @@ static struct workqueue_struct *nvme_workq; static void nvme_reset_failed_dev(struct work_struct *ws); +struct async_cmd_info { + struct kthread_work work; + struct kthread_worker *worker; + u32 result; + int status; + void *ctx; +}; + /* * An NVM Express queue. Each device has at least two (one for admin * commands and one for I/O commands). @@ -87,6 +95,7 @@ struct nvme_queue { u8 cq_phase; u8 cqe_seen; u8 q_suspended; + struct async_cmd_info cmdinfo; unsigned long cmdid_data[]; }; @@ -208,6 +217,15 @@ static void special_completion(struct nvme_dev *dev, void *ctx, dev_warn(&dev->pci_dev->dev, "Unknown special completion %p\n", ctx); } +static void async_completion(struct nvme_dev *dev, void *ctx, + struct nvme_completion *cqe) +{ + struct async_cmd_info *cmdinfo = ctx; + cmdinfo->result = le32_to_cpup(&cqe->result); + cmdinfo->status = le16_to_cpup(&cqe->status) >> 1; + queue_kthread_work(cmdinfo->worker, &cmdinfo->work); +} + /* * Called with local interrupts disabled and the q_lock held. May not sleep. */ @@ -898,12 +916,34 @@ int nvme_submit_sync_cmd(struct nvme_queue *nvmeq, struct nvme_command *cmd, return cmdinfo.status; } +static int nvme_submit_async_cmd(struct nvme_queue *nvmeq, + struct nvme_command *cmd, + struct async_cmd_info *cmdinfo, unsigned timeout) +{ + int cmdid; + + cmdid = alloc_cmdid_killable(nvmeq, cmdinfo, async_completion, timeout); + if (cmdid < 0) + return cmdid; + cmdinfo->status = -EINTR; + cmd->common.command_id = cmdid; + nvme_submit_cmd(nvmeq, cmd); + return 0; +} + int nvme_submit_admin_cmd(struct nvme_dev *dev, struct nvme_command *cmd, u32 *result) { return nvme_submit_sync_cmd(dev->queues[0], cmd, result, ADMIN_TIMEOUT); } +static int nvme_submit_admin_cmd_async(struct nvme_dev *dev, + struct nvme_command *cmd, struct async_cmd_info *cmdinfo) +{ + return nvme_submit_async_cmd(dev->queues[0], cmd, cmdinfo, + ADMIN_TIMEOUT); +} + static int adapter_delete_queue(struct nvme_dev *dev, u8 opcode, u16 id) { int status; @@ -1124,15 +1164,20 @@ static void nvme_free_queues(struct nvme_dev *dev) } } -static void nvme_disable_queue(struct nvme_dev *dev, int qid) +/** + * nvme_suspend_queue - put queue into suspended state + * @nvmeq - queue to suspend + * + * Returns 1 if already suspended, 0 otherwise. + */ +static int nvme_suspend_queue(struct nvme_queue *nvmeq) { - struct nvme_queue *nvmeq = dev->queues[qid]; - int vector = dev->entry[nvmeq->cq_vector].vector; + int vector = nvmeq->dev->entry[nvmeq->cq_vector].vector; spin_lock_irq(&nvmeq->q_lock); if (nvmeq->q_suspended) { spin_unlock_irq(&nvmeq->q_lock); - return; + return 1; } nvmeq->q_suspended = 1; spin_unlock_irq(&nvmeq->q_lock); @@ -1140,17 +1185,33 @@ static void nvme_disable_queue(struct nvme_dev *dev, int qid) irq_set_affinity_hint(vector, NULL); free_irq(vector, nvmeq); + return 0; +} + +static void nvme_clear_queue(struct nvme_queue *nvmeq) +{ + spin_lock_irq(&nvmeq->q_lock); + nvme_process_cq(nvmeq); + nvme_cancel_ios(nvmeq, false); + spin_unlock_irq(&nvmeq->q_lock); +} + +static void nvme_disable_queue(struct nvme_dev *dev, int qid) +{ + struct nvme_queue *nvmeq = dev->queues[qid]; + + if (!nvmeq) + return; + if (nvme_suspend_queue(nvmeq)) + return; + /* Don't tell the adapter to delete the admin queue. * Don't tell a removed adapter to delete IO queues. */ if (qid && readl(&dev->bar->csts) != -1) { adapter_delete_sq(dev, qid); adapter_delete_cq(dev, qid); } - - spin_lock_irq(&nvmeq->q_lock); - nvme_process_cq(nvmeq); - nvme_cancel_ios(nvmeq, false); - spin_unlock_irq(&nvmeq->q_lock); + nvme_clear_queue(nvmeq); } static struct nvme_queue *nvme_alloc_queue(struct nvme_dev *dev, int qid, @@ -2089,20 +2150,164 @@ static void nvme_dev_unmap(struct nvme_dev *dev) pci_disable_device(dev->pci_dev); } +struct nvme_delq_ctx { + struct task_struct *waiter; + struct kthread_worker *worker; + atomic_t refcount; +}; + +static void nvme_wait_dq(struct nvme_delq_ctx *dq, struct nvme_dev *dev) +{ + dq->waiter = current; + mb(); + + for (;;) { + set_current_state(TASK_KILLABLE); + if (!atomic_read(&dq->refcount)) + break; + if (!schedule_timeout(ADMIN_TIMEOUT) || + fatal_signal_pending(current)) { + set_current_state(TASK_RUNNING); + + nvme_disable_ctrl(dev, readq(&dev->bar->cap)); + nvme_disable_queue(dev, 0); + + send_sig(SIGKILL, dq->worker->task, 1); + flush_kthread_worker(dq->worker); + return; + } + } + set_current_state(TASK_RUNNING); +} + +static void nvme_put_dq(struct nvme_delq_ctx *dq) +{ + atomic_dec(&dq->refcount); + if (dq->waiter) + wake_up_process(dq->waiter); +} + +static struct nvme_delq_ctx *nvme_get_dq(struct nvme_delq_ctx *dq) +{ + atomic_inc(&dq->refcount); + return dq; +} + +static void nvme_del_queue_end(struct nvme_queue *nvmeq) +{ + struct nvme_delq_ctx *dq = nvmeq->cmdinfo.ctx; + + nvme_clear_queue(nvmeq); + nvme_put_dq(dq); +} + +static int adapter_async_del_queue(struct nvme_queue *nvmeq, u8 opcode, + kthread_work_func_t fn) +{ + struct nvme_command c; + + memset(&c, 0, sizeof(c)); + c.delete_queue.opcode = opcode; + c.delete_queue.qid = cpu_to_le16(nvmeq->qid); + + init_kthread_work(&nvmeq->cmdinfo.work, fn); + return nvme_submit_admin_cmd_async(nvmeq->dev, &c, &nvmeq->cmdinfo); +} + +static void nvme_del_cq_work_handler(struct kthread_work *work) +{ + struct nvme_queue *nvmeq = container_of(work, struct nvme_queue, + cmdinfo.work); + nvme_del_queue_end(nvmeq); +} + +static int nvme_delete_cq(struct nvme_queue *nvmeq) +{ + return adapter_async_del_queue(nvmeq, nvme_admin_delete_cq, + nvme_del_cq_work_handler); +} + +static void nvme_del_sq_work_handler(struct kthread_work *work) +{ + struct nvme_queue *nvmeq = container_of(work, struct nvme_queue, + cmdinfo.work); + int status = nvmeq->cmdinfo.status; + + if (!status) + status = nvme_delete_cq(nvmeq); + if (status) + nvme_del_queue_end(nvmeq); +} + +static int nvme_delete_sq(struct nvme_queue *nvmeq) +{ + return adapter_async_del_queue(nvmeq, nvme_admin_delete_sq, + nvme_del_sq_work_handler); +} + +static void nvme_del_queue_start(struct kthread_work *work) +{ + struct nvme_queue *nvmeq = container_of(work, struct nvme_queue, + cmdinfo.work); + allow_signal(SIGKILL); + if (nvme_delete_sq(nvmeq)) + nvme_del_queue_end(nvmeq); +} + +static void nvme_disable_io_queues(struct nvme_dev *dev) +{ + int i; + DEFINE_KTHREAD_WORKER_ONSTACK(worker); + struct nvme_delq_ctx dq; + struct task_struct *kworker_task = kthread_run(kthread_worker_fn, + &worker, "nvme%d", dev->instance); + + if (IS_ERR(kworker_task)) { + dev_err(&dev->pci_dev->dev, + "Failed to create queue del task\n"); + for (i = dev->queue_count - 1; i > 0; i--) + nvme_disable_queue(dev, i); + return; + } + + dq.waiter = NULL; + atomic_set(&dq.refcount, 0); + dq.worker = &worker; + for (i = dev->queue_count - 1; i > 0; i--) { + struct nvme_queue *nvmeq = dev->queues[i]; + + if (nvme_suspend_queue(nvmeq)) + continue; + nvmeq->cmdinfo.ctx = nvme_get_dq(&dq); + nvmeq->cmdinfo.worker = dq.worker; + init_kthread_work(&nvmeq->cmdinfo.work, nvme_del_queue_start); + queue_kthread_work(dq.worker, &nvmeq->cmdinfo.work); + } + nvme_wait_dq(&dq, dev); + kthread_stop(kworker_task); +} + static void nvme_dev_shutdown(struct nvme_dev *dev) { int i; dev->initialized = 0; - for (i = dev->queue_count - 1; i >= 0; i--) - nvme_disable_queue(dev, i); spin_lock(&dev_list_lock); list_del_init(&dev->node); spin_unlock(&dev_list_lock); - if (dev->bar) + if (!dev->bar || (dev->bar && readl(&dev->bar->csts) == -1)) { + for (i = dev->queue_count - 1; i >= 0; i--) { + struct nvme_queue *nvmeq = dev->queues[i]; + nvme_suspend_queue(nvmeq); + nvme_clear_queue(nvmeq); + } + } else { + nvme_disable_io_queues(dev); nvme_shutdown_ctrl(dev); + nvme_disable_queue(dev, 0); + } nvme_dev_unmap(dev); } From 469071a37afc8a627b6b2ddf29db0a097d864845 Mon Sep 17 00:00:00 2001 From: Matthew Wilcox Date: Mon, 9 Dec 2013 12:58:46 -0500 Subject: [PATCH 037/596] NVMe: Dynamically allocate partition numbers MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Some users need more than 64 partitions per device. Rather than simply increasing the number of partitions, switch to the dynamic partition allocation scheme. This means that minor numbers are not stable across boots, but since major numbers aren't either, I cannot see this being a significant problem. Tested-by: Matias Bjørling Signed-off-by: Matthew Wilcox --- drivers/block/nvme-core.c | 36 +++--------------------------------- 1 file changed, 3 insertions(+), 33 deletions(-) diff --git a/drivers/block/nvme-core.c b/drivers/block/nvme-core.c index 1c8a82fbbc37..7b615a063e92 100644 --- a/drivers/block/nvme-core.c +++ b/drivers/block/nvme-core.c @@ -46,7 +46,6 @@ #define NVME_Q_DEPTH 1024 #define SQ_SIZE(depth) (depth * sizeof(struct nvme_command)) #define CQ_SIZE(depth) (depth * sizeof(struct nvme_completion)) -#define NVME_MINORS 64 #define ADMIN_TIMEOUT (60 * HZ) static int nvme_major; @@ -1780,33 +1779,6 @@ static int nvme_kthread(void *data) return 0; } -static DEFINE_IDA(nvme_index_ida); - -static int nvme_get_ns_idx(void) -{ - int index, error; - - do { - if (!ida_pre_get(&nvme_index_ida, GFP_KERNEL)) - return -1; - - spin_lock(&dev_list_lock); - error = ida_get_new(&nvme_index_ida, &index); - spin_unlock(&dev_list_lock); - } while (error == -EAGAIN); - - if (error) - index = -1; - return index; -} - -static void nvme_put_ns_idx(int index) -{ - spin_lock(&dev_list_lock); - ida_remove(&nvme_index_ida, index); - spin_unlock(&dev_list_lock); -} - static void nvme_config_discard(struct nvme_ns *ns) { u32 logical_block_size = queue_logical_block_size(ns->queue); @@ -1840,7 +1812,7 @@ static struct nvme_ns *nvme_alloc_ns(struct nvme_dev *dev, unsigned nsid, ns->dev = dev; ns->queue->queuedata = ns; - disk = alloc_disk(NVME_MINORS); + disk = alloc_disk(0); if (!disk) goto out_free_queue; ns->ns_id = nsid; @@ -1853,12 +1825,12 @@ static struct nvme_ns *nvme_alloc_ns(struct nvme_dev *dev, unsigned nsid, blk_queue_max_hw_sectors(ns->queue, dev->max_hw_sectors); disk->major = nvme_major; - disk->minors = NVME_MINORS; - disk->first_minor = NVME_MINORS * nvme_get_ns_idx(); + disk->first_minor = 0; disk->fops = &nvme_fops; disk->private_data = ns; disk->queue = ns->queue; disk->driverfs_dev = &dev->pci_dev->dev; + disk->flags = GENHD_FL_EXT_DEVT; sprintf(disk->disk_name, "nvme%dn%d", dev->instance, nsid); set_capacity(disk, le64_to_cpup(&id->nsze) << (ns->lba_shift - 9)); @@ -1876,9 +1848,7 @@ static struct nvme_ns *nvme_alloc_ns(struct nvme_dev *dev, unsigned nsid, static void nvme_ns_free(struct nvme_ns *ns) { - int index = ns->disk->first_minor / NVME_MINORS; put_disk(ns->disk); - nvme_put_ns_idx(index); blk_cleanup_queue(ns->queue); kfree(ns); } From a1a5ef9998b1379780790b878457b0e0f48b25db Mon Sep 17 00:00:00 2001 From: Keith Busch Date: Mon, 16 Dec 2013 13:50:00 -0500 Subject: [PATCH 038/596] NVMe: Disable admin queue on init failure Disable the admin queue if device fails during initialization so the queue's irq is freed. Signed-off-by: Keith Busch [rewritten to use nvme_free_queues] Signed-off-by: Matthew Wilcox --- drivers/block/nvme-core.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/drivers/block/nvme-core.c b/drivers/block/nvme-core.c index 7b615a063e92..89021d710ec2 100644 --- a/drivers/block/nvme-core.c +++ b/drivers/block/nvme-core.c @@ -1152,11 +1152,11 @@ static void nvme_free_queue(struct nvme_queue *nvmeq) kfree(nvmeq); } -static void nvme_free_queues(struct nvme_dev *dev) +static void nvme_free_queues(struct nvme_dev *dev, int lowest) { int i; - for (i = dev->queue_count - 1; i >= 0; i--) { + for (i = dev->queue_count - 1; i >= lowest; i--) { nvme_free_queue(dev->queues[i]); dev->queue_count--; dev->queues[i] = NULL; @@ -1991,7 +1991,7 @@ static int nvme_setup_io_queues(struct nvme_dev *dev) return 0; free_queues: - nvme_free_queues(dev); + nvme_free_queues(dev, 1); return result; } @@ -2411,6 +2411,7 @@ static int nvme_dev_start(struct nvme_dev *dev) return result; disable: + nvme_disable_queue(dev, 0); spin_lock(&dev_list_lock); list_del_init(&dev->node); spin_unlock(&dev_list_lock); @@ -2542,7 +2543,7 @@ static int nvme_probe(struct pci_dev *pdev, const struct pci_device_id *id) shutdown: nvme_dev_shutdown(dev); release_pools: - nvme_free_queues(dev); + nvme_free_queues(dev, 0); nvme_release_prp_pools(dev); release: nvme_release_instance(dev); @@ -2566,7 +2567,7 @@ static void nvme_remove(struct pci_dev *pdev) misc_deregister(&dev->miscdev); nvme_dev_remove(dev); nvme_dev_shutdown(dev); - nvme_free_queues(dev); + nvme_free_queues(dev, 0); nvme_release_instance(dev); nvme_release_prp_pools(dev); kref_put(&dev->kref, nvme_free_dev); From 09ece1424f52a2d3ad0ab715e96b0fb342af4f03 Mon Sep 17 00:00:00 2001 From: Keith Busch Date: Mon, 27 Jan 2014 11:29:40 -0500 Subject: [PATCH 039/596] NVMe: Add a pci_driver shutdown method We need to shut down the device cleanly when the system is being shut down. This was in an earlier patch but was inadvertently lost during a rewrite. Signed-off-by: Keith Busch Signed-off-by: Matthew Wilcox --- drivers/block/nvme-core.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/block/nvme-core.c b/drivers/block/nvme-core.c index 89021d710ec2..eaace767be49 100644 --- a/drivers/block/nvme-core.c +++ b/drivers/block/nvme-core.c @@ -2554,6 +2554,12 @@ static int nvme_probe(struct pci_dev *pdev, const struct pci_device_id *id) return result; } +static void nvme_shutdown(struct pci_dev *pdev) +{ + struct nvme_dev *dev = pci_get_drvdata(pdev); + nvme_dev_shutdown(dev); +} + static void nvme_remove(struct pci_dev *pdev) { struct nvme_dev *dev = pci_get_drvdata(pdev); @@ -2625,6 +2631,7 @@ static struct pci_driver nvme_driver = { .id_table = nvme_id_table, .probe = nvme_probe, .remove = nvme_remove, + .shutdown = nvme_shutdown, .driver = { .pm = &nvme_dev_pm_ops, }, From 3193f07bb7ae723f33ac8e8b9db317a4f68d7d18 Mon Sep 17 00:00:00 2001 From: Matthew Wilcox Date: Mon, 27 Jan 2014 15:57:22 -0500 Subject: [PATCH 040/596] NVMe: Include device and queue numbers in interrupt name On larger systems with many drives, it may help debugging to know which queue is tied to which interrupt, just by looking at /proc/interrupts. Signed-off-by: Matthew Wilcox --- drivers/block/nvme-core.c | 14 +++++++++----- 1 file changed, 9 insertions(+), 5 deletions(-) diff --git a/drivers/block/nvme-core.c b/drivers/block/nvme-core.c index eaace767be49..5ffc26937226 100644 --- a/drivers/block/nvme-core.c +++ b/drivers/block/nvme-core.c @@ -76,6 +76,7 @@ struct async_cmd_info { struct nvme_queue { struct device *q_dmadev; struct nvme_dev *dev; + char irqname[24]; /* nvme4294967295-65535\0 */ spinlock_t q_lock; struct nvme_command *sq_cmds; volatile struct nvme_completion *cqes; @@ -1235,6 +1236,8 @@ static struct nvme_queue *nvme_alloc_queue(struct nvme_dev *dev, int qid, nvmeq->q_dmadev = dmadev; nvmeq->dev = dev; + snprintf(nvmeq->irqname, sizeof(nvmeq->irqname), "nvme%dq%d", + dev->instance, qid); spin_lock_init(&nvmeq->q_lock); nvmeq->cq_head = 0; nvmeq->cq_phase = 1; @@ -1297,7 +1300,7 @@ static int nvme_create_queue(struct nvme_queue *nvmeq, int qid) if (result < 0) goto release_cq; - result = queue_request_irq(dev, nvmeq, "nvme"); + result = queue_request_irq(dev, nvmeq, nvmeq->irqname); if (result < 0) goto release_sq; @@ -1415,7 +1418,7 @@ static int nvme_configure_admin_queue(struct nvme_dev *dev) if (result) return result; - result = queue_request_irq(dev, nvmeq, "nvme admin"); + result = queue_request_irq(dev, nvmeq, nvmeq->irqname); if (result) return result; @@ -1873,6 +1876,7 @@ static size_t db_bar_size(struct nvme_dev *dev, unsigned nr_io_queues) static int nvme_setup_io_queues(struct nvme_dev *dev) { + struct nvme_queue *adminq = dev->queues[0]; struct pci_dev *pdev = dev->pci_dev; int result, cpu, i, vecs, nr_io_queues, size, q_depth; @@ -1899,7 +1903,7 @@ static int nvme_setup_io_queues(struct nvme_dev *dev) } /* Deregister the admin queue's interrupt */ - free_irq(dev->entry[0].vector, dev->queues[0]); + free_irq(dev->entry[0].vector, adminq); vecs = nr_io_queues; for (i = 0; i < vecs; i++) @@ -1937,9 +1941,9 @@ static int nvme_setup_io_queues(struct nvme_dev *dev) */ nr_io_queues = vecs; - result = queue_request_irq(dev, dev->queues[0], "nvme admin"); + result = queue_request_irq(dev, adminq, adminq->irqname); if (result) { - dev->queues[0]->q_suspended = 1; + adminq->q_suspended = 1; goto free_queues; } From 317b5684d52269b75b4ec6480f9dac056d0d4ba8 Mon Sep 17 00:00:00 2001 From: Mark Brown Date: Mon, 27 Jan 2014 17:34:07 +0000 Subject: [PATCH 041/596] regulator: core: Correct default return value for full constraints Once we have full constraints then all supply mappings should be known to the regulator API. This means that we should treat failed lookups as fatal rather than deferring in the hope of further registrations but this was broken by commit 9b92da1f1205bd25 "regulator: core: Fix default return value for _get()" which was targeted at DT systems but unintentionally broke non-DT systems by changing the default return value. Fix this by explicitly returning -EPROBE_DEFER from the DT lookup if we find a property but no corresponding regulator and by having the non-DT case default to -ENODEV when we have full constraints. Fixes: 9b92da1f1205bd25 "regulator: core: Fix default return value for _get()" Signed-off-by: Mark Brown Tested-by: Guenter Roeck Cc: stable@vger.kernel.org --- drivers/regulator/core.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/drivers/regulator/core.c b/drivers/regulator/core.c index b38a6b669e8c..16a309e5c024 100644 --- a/drivers/regulator/core.c +++ b/drivers/regulator/core.c @@ -1272,6 +1272,8 @@ static struct regulator_dev *regulator_dev_lookup(struct device *dev, if (r->dev.parent && node == r->dev.of_node) return r; + *ret = -EPROBE_DEFER; + return NULL; } else { /* * If we couldn't even get the node then it's @@ -1312,7 +1314,7 @@ static struct regulator *_regulator_get(struct device *dev, const char *id, struct regulator_dev *rdev; struct regulator *regulator = ERR_PTR(-EPROBE_DEFER); const char *devname = NULL; - int ret = -EPROBE_DEFER; + int ret; if (id == NULL) { pr_err("get() with no identifier\n"); @@ -1322,6 +1324,11 @@ static struct regulator *_regulator_get(struct device *dev, const char *id, if (dev) devname = dev_name(dev); + if (have_full_constraints()) + ret = -ENODEV; + else + ret = -EPROBE_DEFER; + mutex_lock(®ulator_list_mutex); rdev = regulator_dev_lookup(dev, id, &ret); From 1f802f8249a0da536877842c43c7204064c4de8b Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Tue, 28 Jan 2014 10:33:03 +0100 Subject: [PATCH 042/596] spi: Fix crash with double message finalisation on error handling This reverts commit e120cc0dcf2880a4c5c0a6cb27b655600a1cfa1d. It causes a NULL pointer dereference with drivers using the generic spi_transfer_one_message(), which always calls spi_finalize_current_message(), which zeroes master->cur_msg. Drivers implementing transfer_one_message() theirselves must always call spi_finalize_current_message(), even if the transfer failed: * @transfer_one_message: the subsystem calls the driver to transfer a single * message while queuing transfers that arrive in the meantime. When the * driver is finished with this message, it must call * spi_finalize_current_message() so the subsystem can issue the next * transfer Signed-off-by: Geert Uytterhoeven Signed-off-by: Mark Brown --- drivers/spi/spi.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/drivers/spi/spi.c b/drivers/spi/spi.c index 63613a96233c..26f4afd19faf 100644 --- a/drivers/spi/spi.c +++ b/drivers/spi/spi.c @@ -745,9 +745,7 @@ static void spi_pump_messages(struct kthread_work *work) ret = master->transfer_one_message(master, master->cur_msg); if (ret) { dev_err(&master->dev, - "failed to transfer one message from queue: %d\n", ret); - master->cur_msg->status = ret; - spi_finalize_current_message(master); + "failed to transfer one message from queue\n"); return; } } From 7de8246eba58c6f4c315b8c69fb29e445ccf3c80 Mon Sep 17 00:00:00 2001 From: Tony Luck Date: Tue, 28 Jan 2014 09:38:37 -0800 Subject: [PATCH 043/596] [IA64] Wire up new sched_setattr and sched_getattr syscalls New syscalls for v3.14 Signed-off-by: Tony Luck --- arch/ia64/include/asm/unistd.h | 2 +- arch/ia64/include/uapi/asm/unistd.h | 2 ++ arch/ia64/kernel/entry.S | 2 ++ 3 files changed, 5 insertions(+), 1 deletion(-) diff --git a/arch/ia64/include/asm/unistd.h b/arch/ia64/include/asm/unistd.h index afd45e0d552e..ae763d8bf55a 100644 --- a/arch/ia64/include/asm/unistd.h +++ b/arch/ia64/include/asm/unistd.h @@ -11,7 +11,7 @@ -#define NR_syscalls 312 /* length of syscall table */ +#define NR_syscalls 314 /* length of syscall table */ /* * The following defines stop scripts/checksyscalls.sh from complaining about diff --git a/arch/ia64/include/uapi/asm/unistd.h b/arch/ia64/include/uapi/asm/unistd.h index 34fd6fe46da1..715e85f858de 100644 --- a/arch/ia64/include/uapi/asm/unistd.h +++ b/arch/ia64/include/uapi/asm/unistd.h @@ -325,5 +325,7 @@ #define __NR_process_vm_writev 1333 #define __NR_accept4 1334 #define __NR_finit_module 1335 +#define __NR_sched_setattr 1336 +#define __NR_sched_getattr 1337 #endif /* _UAPI_ASM_IA64_UNISTD_H */ diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S index ddea607f948a..fa8d61a312a7 100644 --- a/arch/ia64/kernel/entry.S +++ b/arch/ia64/kernel/entry.S @@ -1773,6 +1773,8 @@ sys_call_table: data8 sys_process_vm_writev data8 sys_accept4 data8 sys_finit_module // 1335 + data8 sys_sched_setattr + data8 sys_sched_getattr .org sys_call_table + 8*NR_syscalls // guard against failures to increase NR_syscalls #endif /* __IA64_ASM_PARAVIRTUALIZED_NATIVE */ From dbb0dd021dc282fe65c4f66c86821419850c43cc Mon Sep 17 00:00:00 2001 From: Jose Alonso Date: Sun, 26 Jan 2014 08:54:18 -0200 Subject: [PATCH 044/596] s390/qdio: for_each macro correctness I observed that there are for_each macros that do an extra memory access beyond the defined area. Normally this does not cause problems. But, this can cause exceptions. For example: if the area is allocated at the end of a page and the next page is not accessible. For correctness, I suggest changing the arguments of the 'for loop' like others 'for_each' do in the kernel. Signed-off-by: Jose Alonso Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- drivers/s390/cio/qdio.h | 14 ++++++-------- 1 file changed, 6 insertions(+), 8 deletions(-) diff --git a/drivers/s390/cio/qdio.h b/drivers/s390/cio/qdio.h index 8acaae18bd11..a563e4c00590 100644 --- a/drivers/s390/cio/qdio.h +++ b/drivers/s390/cio/qdio.h @@ -359,14 +359,12 @@ static inline int multicast_outbound(struct qdio_q *q) #define need_siga_sync_out_after_pci(q) \ (unlikely(q->irq_ptr->siga_flag.sync_out_after_pci)) -#define for_each_input_queue(irq_ptr, q, i) \ - for (i = 0, q = irq_ptr->input_qs[0]; \ - i < irq_ptr->nr_input_qs; \ - q = irq_ptr->input_qs[++i]) -#define for_each_output_queue(irq_ptr, q, i) \ - for (i = 0, q = irq_ptr->output_qs[0]; \ - i < irq_ptr->nr_output_qs; \ - q = irq_ptr->output_qs[++i]) +#define for_each_input_queue(irq_ptr, q, i) \ + for (i = 0; i < irq_ptr->nr_input_qs && \ + ({ q = irq_ptr->input_qs[i]; 1; }); i++) +#define for_each_output_queue(irq_ptr, q, i) \ + for (i = 0; i < irq_ptr->nr_output_qs && \ + ({ q = irq_ptr->output_qs[i]; 1; }); i++) #define prev_buf(bufnr) \ ((bufnr + QDIO_MAX_BUFFERS_MASK) & QDIO_MAX_BUFFERS_MASK) From 0f308f4f9e3ddd85869bb128e37f8dd08f3f513c Mon Sep 17 00:00:00 2001 From: Ursula Braun Date: Tue, 28 Jan 2014 13:06:47 +0100 Subject: [PATCH 045/596] s390/qdio: correct program-controlled interruption checking Get rid of compile warning in qdio_int_handler_pci() when checking for program-controlled interruption on outbound queues. Signed-off-by: Ursula Braun Reviewed-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- drivers/s390/cio/qdio_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/s390/cio/qdio_main.c b/drivers/s390/cio/qdio_main.c index c883a085c059..77466c4faabb 100644 --- a/drivers/s390/cio/qdio_main.c +++ b/drivers/s390/cio/qdio_main.c @@ -996,7 +996,7 @@ static void qdio_int_handler_pci(struct qdio_irq *irq_ptr) } } - if (!pci_out_supported(q)) + if (!(irq_ptr->qib.ac & QIB_AC_OUTBOUND_PCI_SUPPORTED)) return; for_each_output_queue(irq_ptr, q, i) { From b7c5b1aa2836c933ab03f90391619ebdc9112e46 Mon Sep 17 00:00:00 2001 From: Gerald Schaefer Date: Tue, 28 Jan 2014 18:40:12 +0100 Subject: [PATCH 046/596] s390/appldata: restore missing init_virt_timer() Commit 27f6b416 "s390/vtimer: rework virtual timer interface" removed the call to init_virt_timer() by mistake, which is added again by this patch. Signed-off-by: Gerald Schaefer Signed-off-by: Martin Schwidefsky --- arch/s390/appldata/appldata_base.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/s390/appldata/appldata_base.c b/arch/s390/appldata/appldata_base.c index 4c4a1cef5208..47c8630c93cd 100644 --- a/arch/s390/appldata/appldata_base.c +++ b/arch/s390/appldata/appldata_base.c @@ -529,6 +529,7 @@ static int __init appldata_init(void) { int rc; + init_virt_timer(&appldata_timer); appldata_timer.function = appldata_timer_function; appldata_timer.data = (unsigned long) &appldata_work; From bdfd70fde389412be60f7e8aaed5732dc26fc8ac Mon Sep 17 00:00:00 2001 From: Matthew Wilcox Date: Wed, 29 Jan 2014 09:33:52 -0500 Subject: [PATCH 047/596] NVMe: Correct uses of INIT_WORK We need to initialise the work_struct when we initialise the rest of the struct nvme_dev, otherwise we'll hit a lockdep warning when we remove the device. Use PREPARE_WORK to change the function pointer instead of INIT_WORK. Signed-off-by: Matthew Wilcox --- drivers/block/nvme-core.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/drivers/block/nvme-core.c b/drivers/block/nvme-core.c index 5ffc26937226..23728099e36f 100644 --- a/drivers/block/nvme-core.c +++ b/drivers/block/nvme-core.c @@ -1075,7 +1075,7 @@ static void nvme_abort_cmd(int cmdid, struct nvme_queue *nvmeq) dev_warn(&dev->pci_dev->dev, "I/O %d QID %d timeout, reset controller\n", cmdid, nvmeq->qid); - INIT_WORK(&dev->reset_work, nvme_reset_failed_dev); + PREPARE_WORK(&dev->reset_work, nvme_reset_failed_dev); queue_work(nvme_workq, &dev->reset_work); return; } @@ -1757,7 +1757,7 @@ static int nvme_kthread(void *data) list_del_init(&dev->node); dev_warn(&dev->pci_dev->dev, "Failed status, reset controller\n"); - INIT_WORK(&dev->reset_work, + PREPARE_WORK(&dev->reset_work, nvme_reset_failed_dev); queue_work(nvme_workq, &dev->reset_work); continue; @@ -2460,7 +2460,7 @@ static int nvme_dev_resume(struct nvme_dev *dev) return ret; if (ret == -EBUSY) { spin_lock(&dev_list_lock); - INIT_WORK(&dev->reset_work, nvme_remove_disks); + PREPARE_WORK(&dev->reset_work, nvme_remove_disks); queue_work(nvme_workq, &dev->reset_work); spin_unlock(&dev_list_lock); } @@ -2507,6 +2507,7 @@ static int nvme_probe(struct pci_dev *pdev, const struct pci_device_id *id) goto free; INIT_LIST_HEAD(&dev->namespaces); + INIT_WORK(&dev->reset_work, nvme_reset_failed_dev); dev->pci_dev = pdev; pci_set_drvdata(pdev, dev); result = nvme_set_instance(dev); @@ -2605,7 +2606,7 @@ static int nvme_resume(struct device *dev) struct nvme_dev *ndev = pci_get_drvdata(pdev); if (nvme_dev_resume(ndev) && !work_busy(&ndev->reset_work)) { - INIT_WORK(&ndev->reset_work, nvme_reset_failed_dev); + PREPARE_WORK(&ndev->reset_work, nvme_reset_failed_dev); queue_work(nvme_workq, &ndev->reset_work); } return 0; From f428ebd184c82a7914b2aa7e9f868918aaf7ea78 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 24 Jan 2014 16:40:02 +0100 Subject: [PATCH 048/596] perf tools: Fix AAAAARGH64 memory barriers Someone got the load and store barriers mixed up for AAAAARGH64. Turn them the right side up. Reported-by: Will Deacon Signed-off-by: Peter Zijlstra Fixes: a94d342b9cb0 ("tools/perf: Add required memory barriers") Cc: Ingo Molnar Cc: Will Deacon Link: http://lkml.kernel.org/r/20140124154002.GF31570@twins.programming.kicks-ass.net Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/perf.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tools/perf/perf.h b/tools/perf/perf.h index 7daa806d9050..e84fa26bc1be 100644 --- a/tools/perf/perf.h +++ b/tools/perf/perf.h @@ -100,8 +100,8 @@ #ifdef __aarch64__ #define mb() asm volatile("dmb ish" ::: "memory") -#define wmb() asm volatile("dmb ishld" ::: "memory") -#define rmb() asm volatile("dmb ishst" ::: "memory") +#define wmb() asm volatile("dmb ishst" ::: "memory") +#define rmb() asm volatile("dmb ishld" ::: "memory") #define cpu_relax() asm volatile("yield" ::: "memory") #endif From 0e4b949e6620689b1d3142e20bca2c9d1f918fda Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Tue, 14 Jan 2014 11:44:09 +0100 Subject: [PATCH 049/596] can: flexcan: fix flexcan driver build for big endian on ARM and little endian on PowerPc MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit There is no reason to disallow building the driver on big-endian ARM kernels. Furthermore, the current behavior is actually broken on little-endian PowerPC as well. The choice of register accessor functions must purely depend on the CPU architecture, not which endianess the CPU is running on. Note that we nowadays allow both big-endian ARM and little-endian PowerPC kernels. With this patch applied, we will do the right thing in all four combinations. Signed-off-by: Arnd Bergmann Cc: Guenter Roeck Cc: Lothar Waßmann Signed-off-by: Marc Kleine-Budde --- drivers/net/can/Kconfig | 2 +- drivers/net/can/flexcan.c | 7 +++++-- 2 files changed, 6 insertions(+), 3 deletions(-) diff --git a/drivers/net/can/Kconfig b/drivers/net/can/Kconfig index d447b881bbde..9e7d95dae2c7 100644 --- a/drivers/net/can/Kconfig +++ b/drivers/net/can/Kconfig @@ -104,7 +104,7 @@ config CAN_JANZ_ICAN3 config CAN_FLEXCAN tristate "Support for Freescale FLEXCAN based chips" - depends on (ARM && CPU_LITTLE_ENDIAN) || PPC + depends on ARM || PPC ---help--- Say Y here if you want to support for Freescale FlexCAN. diff --git a/drivers/net/can/flexcan.c b/drivers/net/can/flexcan.c index aaed97bee471..320bef2dba42 100644 --- a/drivers/net/can/flexcan.c +++ b/drivers/net/can/flexcan.c @@ -235,9 +235,12 @@ static const struct can_bittiming_const flexcan_bittiming_const = { }; /* - * Abstract off the read/write for arm versus ppc. + * Abstract off the read/write for arm versus ppc. This + * assumes that PPC uses big-endian registers and everything + * else uses little-endian registers, independent of CPU + * endianess. */ -#if defined(__BIG_ENDIAN) +#if defined(CONFIG_PPC) static inline u32 flexcan_read(void __iomem *addr) { return in_be32(addr); From bb5ecb0c63ac88b6f39029f75c47f4be4e352e8d Mon Sep 17 00:00:00 2001 From: Rostislav Lisovy Date: Fri, 24 Jan 2014 13:17:37 +0100 Subject: [PATCH 050/596] can: Propagate SO_PRIORITY of raw sockets to skbs This allows controlling certain queueing disciplines by setting the socket's SO_PRIORITY option. For example, with the default pfifo_fast queueing discipline, which provides three priorities, socket priority TC_PRIO_CONTROL means higher than default and TC_PRIO_BULK means lower than default. Signed-off-by: Rostislav Lisovy Signed-off-by: Michal Sojka Acked-by: Oliver Hartkopp Signed-off-by: Marc Kleine-Budde --- net/can/raw.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/can/raw.c b/net/can/raw.c index 07d72d852324..8be757cca2ec 100644 --- a/net/can/raw.c +++ b/net/can/raw.c @@ -715,6 +715,7 @@ static int raw_sendmsg(struct kiocb *iocb, struct socket *sock, skb->dev = dev; skb->sk = sk; + skb->priority = sk->sk_priority; err = can_send(skb, ro->loopback); From 73615c8669ef1c497c942cae0d5079b95824332a Mon Sep 17 00:00:00 2001 From: "Ira W. Snyder" Date: Wed, 29 Jan 2014 09:58:25 -0800 Subject: [PATCH 051/596] can: janz-ican3: fix uninitialized variable warnings Analysis of the code shows that the struct ican3_msg variable cannot be used uninitialized. Error conditions are checked and the loop terminates before calling the ican3_handle_message() function with an uninitialized value. Signed-off-by: Ira W. Snyder Acked-by: Oliver Hartkopp Signed-off-by: Marc Kleine-Budde --- drivers/net/can/janz-ican3.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/can/janz-ican3.c b/drivers/net/can/janz-ican3.c index e24e6690d672..07f0ba03cd59 100644 --- a/drivers/net/can/janz-ican3.c +++ b/drivers/net/can/janz-ican3.c @@ -1322,7 +1322,7 @@ static int ican3_napi(struct napi_struct *napi, int budget) /* process all communication messages */ while (true) { - struct ican3_msg msg; + struct ican3_msg uninitialized_var(msg); ret = ican3_recv_msg(mod, &msg); if (ret) break; From 947174476701fbc84ea8c7ec9664270f9d80b076 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Tue, 28 Jan 2014 16:57:39 -0800 Subject: [PATCH 052/596] bcache: fix BUG_ON due to integer overflow with GC_SECTORS_USED The BUG_ON at the end of __bch_btree_mark_key can be triggered due to an integer overflow error: BITMASK(GC_SECTORS_USED, struct bucket, gc_mark, 2, 13); ... SET_GC_SECTORS_USED(g, min_t(unsigned, GC_SECTORS_USED(g) + KEY_SIZE(k), (1 << 14) - 1)); BUG_ON(!GC_SECTORS_USED(g)); In bcache.h, the SECTORS_USED bitfield is defined to be 13 bits wide. While the SET_ code tries to ensure that the field doesn't overflow by clamping it to (1<<14)-1 == 16383, this is incorrect because 16383 requires 14 bits. Therefore, if GC_SECTORS_USED() + KEY_SIZE() = 8192, the SET_ statement tries to store 8192 into a 13-bit field. In a 13-bit field, 8192 becomes zero, thus triggering the BUG_ON. Therefore, create a field width constant and a max value constant, and use those to create the bitfield and check the inputs to SET_GC_SECTORS_USED. Arguably the BITMASK() template ought to have BUG_ON checks for too-large values, but that's a separate patch. Signed-off-by: Darrick J. Wong --- drivers/md/bcache/bcache.h | 4 +++- drivers/md/bcache/btree.c | 2 +- 2 files changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/md/bcache/bcache.h b/drivers/md/bcache/bcache.h index 0c707e4f4eaf..a4c7306ff43d 100644 --- a/drivers/md/bcache/bcache.h +++ b/drivers/md/bcache/bcache.h @@ -210,7 +210,9 @@ BITMASK(GC_MARK, struct bucket, gc_mark, 0, 2); #define GC_MARK_RECLAIMABLE 0 #define GC_MARK_DIRTY 1 #define GC_MARK_METADATA 2 -BITMASK(GC_SECTORS_USED, struct bucket, gc_mark, 2, 13); +#define GC_SECTORS_USED_SIZE 13 +#define MAX_GC_SECTORS_USED (~(~0ULL << GC_SECTORS_USED_SIZE)) +BITMASK(GC_SECTORS_USED, struct bucket, gc_mark, 2, GC_SECTORS_USED_SIZE); BITMASK(GC_MOVE, struct bucket, gc_mark, 15, 1); #include "journal.h" diff --git a/drivers/md/bcache/btree.c b/drivers/md/bcache/btree.c index 98cc0a810a36..7d421180b6d2 100644 --- a/drivers/md/bcache/btree.c +++ b/drivers/md/bcache/btree.c @@ -1167,7 +1167,7 @@ uint8_t __bch_btree_mark_key(struct cache_set *c, int level, struct bkey *k) /* guard against overflow */ SET_GC_SECTORS_USED(g, min_t(unsigned, GC_SECTORS_USED(g) + KEY_SIZE(k), - (1 << 14) - 1)); + MAX_GC_SECTORS_USED)); BUG_ON(!GC_SECTORS_USED(g)); } From 3572324af0f4ef877545e5a17bd3e788551f166a Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 10 Jan 2014 18:53:02 -0800 Subject: [PATCH 053/596] bcache: Minor fixes from kbuild robot Signed-off-by: Kent Overstreet --- drivers/md/bcache/bset.c | 7 +++++-- drivers/md/bcache/btree.c | 2 +- drivers/md/bcache/extents.c | 2 +- drivers/md/bcache/sysfs.c | 2 +- 4 files changed, 8 insertions(+), 5 deletions(-) diff --git a/drivers/md/bcache/bset.c b/drivers/md/bcache/bset.c index 4f6b5940e609..3f74b4b0747b 100644 --- a/drivers/md/bcache/bset.c +++ b/drivers/md/bcache/bset.c @@ -23,7 +23,7 @@ void bch_dump_bset(struct btree_keys *b, struct bset *i, unsigned set) for (k = i->start; k < bset_bkey_last(i); k = next) { next = bkey_next(k); - printk(KERN_ERR "block %u key %zi/%u: ", set, + printk(KERN_ERR "block %u key %li/%u: ", set, (uint64_t *) k - i->d, i->keys); if (b->ops->key_dump) @@ -1185,9 +1185,12 @@ static void __btree_sort(struct btree_keys *b, struct btree_iter *iter, struct bset *out = (void *) __get_free_pages(__GFP_NOWARN|GFP_NOIO, order); if (!out) { + struct page *outp; + BUG_ON(order > state->page_order); - out = page_address(mempool_alloc(state->pool, GFP_NOIO)); + outp = mempool_alloc(state->pool, GFP_NOIO); + out = page_address(outp); used_mempool = true; order = state->page_order; } diff --git a/drivers/md/bcache/btree.c b/drivers/md/bcache/btree.c index 7d421180b6d2..5f9c2a665ca5 100644 --- a/drivers/md/bcache/btree.c +++ b/drivers/md/bcache/btree.c @@ -1805,7 +1805,7 @@ static bool btree_insert_key(struct btree *b, struct bkey *k, static size_t insert_u64s_remaining(struct btree *b) { - ssize_t ret = bch_btree_keys_u64s_remaining(&b->keys); + long ret = bch_btree_keys_u64s_remaining(&b->keys); /* * Might land in the middle of an existing extent and have to split it diff --git a/drivers/md/bcache/extents.c b/drivers/md/bcache/extents.c index c3ead586dc27..416d1a3e028e 100644 --- a/drivers/md/bcache/extents.c +++ b/drivers/md/bcache/extents.c @@ -194,7 +194,7 @@ static bool btree_ptr_bad_expensive(struct btree *b, const struct bkey *k) mutex_unlock(&b->c->bucket_lock); bch_extent_to_text(buf, sizeof(buf), k); btree_bug(b, -"inconsistent btree pointer %s: bucket %li pin %i prio %i gen %i last_gc %i mark %llu gc_gen %i", +"inconsistent btree pointer %s: bucket %zi pin %i prio %i gen %i last_gc %i mark %llu gc_gen %i", buf, PTR_BUCKET_NR(b->c, k, i), atomic_read(&g->pin), g->prio, g->gen, g->last_gc, GC_MARK(g), g->gc_gen); return true; diff --git a/drivers/md/bcache/sysfs.c b/drivers/md/bcache/sysfs.c index c6ab69333a6d..d8458d477a12 100644 --- a/drivers/md/bcache/sysfs.c +++ b/drivers/md/bcache/sysfs.c @@ -416,7 +416,7 @@ static int btree_bset_stats(struct btree_op *b_op, struct btree *b) return MAP_CONTINUE; } -int bch_bset_print_stats(struct cache_set *c, char *buf) +static int bch_bset_print_stats(struct cache_set *c, char *buf) { struct bset_stats_op op; int ret; From e3b4825b85eab879b618af6ea18529ca7ab9a64f Mon Sep 17 00:00:00 2001 From: Nicholas Swenson Date: Thu, 12 Dec 2013 12:53:28 -0800 Subject: [PATCH 054/596] bcache: bugfix - gc thread now gets woken when cache is full Signed-off-by: Nicholas Swenson --- drivers/md/bcache/request.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/md/bcache/request.c b/drivers/md/bcache/request.c index fcdb59f9ca91..675229b93718 100644 --- a/drivers/md/bcache/request.c +++ b/drivers/md/bcache/request.c @@ -354,14 +354,14 @@ static void bch_data_insert_start(struct closure *cl) struct data_insert_op *op = container_of(cl, struct data_insert_op, cl); struct bio *bio = op->bio, *n; - if (op->bypass) - return bch_data_invalidate(cl); - if (atomic_sub_return(bio_sectors(bio), &op->c->sectors_to_gc) < 0) { set_gc_sectors(op->c); wake_up_gc(op->c); } + if (op->bypass) + return bch_data_invalidate(cl); + /* * Journal writes are marked REQ_FLUSH; if the original write was a * flush, it'll wait on the journal write. From 477632dff5c7deaa165701f441e9a4bd33e22b18 Mon Sep 17 00:00:00 2001 From: Sarah Sharp Date: Wed, 29 Jan 2014 14:02:00 -0800 Subject: [PATCH 055/596] Revert "xhci: replace xhci_write_64() with writeq()" MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This reverts commit 7dd09a1af2c7150269350aaa567a11b06e831003. Many xHCI host controllers can only handle 32-bit addresses, and writing 64-bits at a time causes them to fail. Rafał reports that USB devices simply do not enumerate, and reverting this patch helps. Branimir reports that his host controller doesn't respond to an Enable Slot command and dies: [ 75.576160] xhci_hcd 0000:03:00.0: Timeout while waiting for a slot [ 88.991634] xhci_hcd 0000:03:00.0: Stopped the command ring failed, maybe the host is dead [ 88.991748] xhci_hcd 0000:03:00.0: Abort command ring failed [ 88.991845] xhci_hcd 0000:03:00.0: HC died; cleaning up [ 93.985489] xhci_hcd 0000:03:00.0: Timeout while waiting for a slot [ 93.985494] xhci_hcd 0000:03:00.0: Abort the command ring, but the xHCI is dead. [ 98.982586] xhci_hcd 0000:03:00.0: Timeout while waiting for a slot [ 98.982591] xhci_hcd 0000:03:00.0: Abort the command ring, but the xHCI is dead. [ 103.979696] xhci_hcd 0000:03:00.0: Timeout while waiting for a slot [ 103.979702] xhci_hcd 0000:03:00.0: Abort the command ring, but the xHCI is dead Signed-off-by: Sarah Sharp Reported-by: Rafał Miłecki Reported-by: Branimir Maksimovic Cc: Xenia Ragiadakou --- drivers/usb/host/xhci-mem.c | 8 ++++---- drivers/usb/host/xhci-ring.c | 8 +++++--- drivers/usb/host/xhci.c | 8 ++++---- drivers/usb/host/xhci.h | 29 ++++++++++++++++++++--------- 4 files changed, 33 insertions(+), 20 deletions(-) diff --git a/drivers/usb/host/xhci-mem.c b/drivers/usb/host/xhci-mem.c index 873c272b3ef5..4b87026f8a5a 100644 --- a/drivers/usb/host/xhci-mem.c +++ b/drivers/usb/host/xhci-mem.c @@ -1967,7 +1967,7 @@ static void xhci_set_hc_event_deq(struct xhci_hcd *xhci) xhci_dbg_trace(xhci, trace_xhci_dbg_init, "// Write event ring dequeue pointer, " "preserving EHB bit"); - writeq(((u64) deq & (u64) ~ERST_PTR_MASK) | temp, + xhci_write_64(xhci, ((u64) deq & (u64) ~ERST_PTR_MASK) | temp, &xhci->ir_set->erst_dequeue); } @@ -2269,7 +2269,7 @@ int xhci_mem_init(struct xhci_hcd *xhci, gfp_t flags) xhci_dbg_trace(xhci, trace_xhci_dbg_init, "// Device context base array address = 0x%llx (DMA), %p (virt)", (unsigned long long)xhci->dcbaa->dma, xhci->dcbaa); - writeq(dma, &xhci->op_regs->dcbaa_ptr); + xhci_write_64(xhci, dma, &xhci->op_regs->dcbaa_ptr); /* * Initialize the ring segment pool. The ring must be a contiguous @@ -2318,7 +2318,7 @@ int xhci_mem_init(struct xhci_hcd *xhci, gfp_t flags) xhci->cmd_ring->cycle_state; xhci_dbg_trace(xhci, trace_xhci_dbg_init, "// Setting command ring address to 0x%x", val); - writeq(val_64, &xhci->op_regs->cmd_ring); + xhci_write_64(xhci, val_64, &xhci->op_regs->cmd_ring); xhci_dbg_cmd_ptrs(xhci); xhci->lpm_command = xhci_alloc_command(xhci, true, true, flags); @@ -2399,7 +2399,7 @@ int xhci_mem_init(struct xhci_hcd *xhci, gfp_t flags) val_64 = readq(&xhci->ir_set->erst_base); val_64 &= ERST_PTR_MASK; val_64 |= (xhci->erst.erst_dma_addr & (u64) ~ERST_PTR_MASK); - writeq(val_64, &xhci->ir_set->erst_base); + xhci_write_64(xhci, val_64, &xhci->ir_set->erst_base); /* Set the event ring dequeue address */ xhci_set_hc_event_deq(xhci); diff --git a/drivers/usb/host/xhci-ring.c b/drivers/usb/host/xhci-ring.c index a0b248c34526..e19fae964f76 100644 --- a/drivers/usb/host/xhci-ring.c +++ b/drivers/usb/host/xhci-ring.c @@ -313,7 +313,8 @@ static int xhci_abort_cmd_ring(struct xhci_hcd *xhci) return 0; } xhci->cmd_ring_state = CMD_RING_STATE_ABORTED; - writeq(temp_64 | CMD_RING_ABORT, &xhci->op_regs->cmd_ring); + xhci_write_64(xhci, temp_64 | CMD_RING_ABORT, + &xhci->op_regs->cmd_ring); /* Section 4.6.1.2 of xHCI 1.0 spec says software should * time the completion od all xHCI commands, including @@ -2865,7 +2866,8 @@ irqreturn_t xhci_irq(struct usb_hcd *hcd) * the event ring should be empty. */ temp_64 = readq(&xhci->ir_set->erst_dequeue); - writeq(temp_64 | ERST_EHB, &xhci->ir_set->erst_dequeue); + xhci_write_64(xhci, temp_64 | ERST_EHB, + &xhci->ir_set->erst_dequeue); spin_unlock(&xhci->lock); return IRQ_HANDLED; @@ -2892,7 +2894,7 @@ irqreturn_t xhci_irq(struct usb_hcd *hcd) /* Clear the event handler busy flag (RW1C); event ring is empty. */ temp_64 |= ERST_EHB; - writeq(temp_64, &xhci->ir_set->erst_dequeue); + xhci_write_64(xhci, temp_64, &xhci->ir_set->erst_dequeue); spin_unlock(&xhci->lock); diff --git a/drivers/usb/host/xhci.c b/drivers/usb/host/xhci.c index ad364394885a..ebbe52cafaa1 100644 --- a/drivers/usb/host/xhci.c +++ b/drivers/usb/host/xhci.c @@ -769,11 +769,11 @@ static void xhci_restore_registers(struct xhci_hcd *xhci) { writel(xhci->s3.command, &xhci->op_regs->command); writel(xhci->s3.dev_nt, &xhci->op_regs->dev_notification); - writeq(xhci->s3.dcbaa_ptr, &xhci->op_regs->dcbaa_ptr); + xhci_write_64(xhci, xhci->s3.dcbaa_ptr, &xhci->op_regs->dcbaa_ptr); writel(xhci->s3.config_reg, &xhci->op_regs->config_reg); writel(xhci->s3.erst_size, &xhci->ir_set->erst_size); - writeq(xhci->s3.erst_base, &xhci->ir_set->erst_base); - writeq(xhci->s3.erst_dequeue, &xhci->ir_set->erst_dequeue); + xhci_write_64(xhci, xhci->s3.erst_base, &xhci->ir_set->erst_base); + xhci_write_64(xhci, xhci->s3.erst_dequeue, &xhci->ir_set->erst_dequeue); writel(xhci->s3.irq_pending, &xhci->ir_set->irq_pending); writel(xhci->s3.irq_control, &xhci->ir_set->irq_control); } @@ -792,7 +792,7 @@ static void xhci_set_cmd_ring_deq(struct xhci_hcd *xhci) xhci_dbg_trace(xhci, trace_xhci_dbg_init, "// Setting command ring address to 0x%llx", (long unsigned long) val_64); - writeq(val_64, &xhci->op_regs->cmd_ring); + xhci_write_64(xhci, val_64, &xhci->op_regs->cmd_ring); } /* diff --git a/drivers/usb/host/xhci.h b/drivers/usb/host/xhci.h index f8416639bf31..60def15d129c 100644 --- a/drivers/usb/host/xhci.h +++ b/drivers/usb/host/xhci.h @@ -28,15 +28,6 @@ #include #include -/* - * Registers should always be accessed with double word or quad word accesses. - * - * Some xHCI implementations may support 64-bit address pointers. Registers - * with 64-bit address pointers should be written to with dword accesses by - * writing the low dword first (ptr[0]), then the high dword (ptr[1]) second. - * xHCI implementations that do not support 64-bit address pointers will ignore - * the high dword, and write order is irrelevant. - */ #include /* Code sharing between pci-quirks and xhci hcd */ @@ -1614,6 +1605,26 @@ static inline struct usb_hcd *xhci_to_hcd(struct xhci_hcd *xhci) #define xhci_warn_ratelimited(xhci, fmt, args...) \ dev_warn_ratelimited(xhci_to_hcd(xhci)->self.controller , fmt , ## args) +/* + * Registers should always be accessed with double word or quad word accesses. + * + * Some xHCI implementations may support 64-bit address pointers. Registers + * with 64-bit address pointers should be written to with dword accesses by + * writing the low dword first (ptr[0]), then the high dword (ptr[1]) second. + * xHCI implementations that do not support 64-bit address pointers will ignore + * the high dword, and write order is irrelevant. + */ +static inline void xhci_write_64(struct xhci_hcd *xhci, + const u64 val, __le64 __iomem *regs) +{ + __u32 __iomem *ptr = (__u32 __iomem *) regs; + u32 val_lo = lower_32_bits(val); + u32 val_hi = upper_32_bits(val); + + writel(val_lo, ptr); + writel(val_hi, ptr + 1); +} + static inline int xhci_link_trb_quirk(struct xhci_hcd *xhci) { return xhci->quirks & XHCI_LINK_TRB_QUIRK; From aac5c4226e7136c331ed384c25d5560204da10a0 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Wed, 29 Jan 2014 16:16:39 +0300 Subject: [PATCH 056/596] KVM: return an error code in kvm_vm_ioctl_register_coalesced_mmio() If kvm_io_bus_register_dev() fails then it returns success but it should return an error code. I also did a little cleanup like removing an impossible NULL test. Cc: stable@vger.kernel.org Fixes: 2b3c246a682c ('KVM: Make coalesced mmio use a device per zone') Signed-off-by: Dan Carpenter Signed-off-by: Paolo Bonzini --- virt/kvm/coalesced_mmio.c | 8 ++------ 1 file changed, 2 insertions(+), 6 deletions(-) diff --git a/virt/kvm/coalesced_mmio.c b/virt/kvm/coalesced_mmio.c index 88b2fe3ddf42..00d86427af0f 100644 --- a/virt/kvm/coalesced_mmio.c +++ b/virt/kvm/coalesced_mmio.c @@ -154,17 +154,13 @@ int kvm_vm_ioctl_register_coalesced_mmio(struct kvm *kvm, list_add_tail(&dev->list, &kvm->coalesced_zones); mutex_unlock(&kvm->slots_lock); - return ret; + return 0; out_free_dev: mutex_unlock(&kvm->slots_lock); - kfree(dev); - if (dev == NULL) - return -ENXIO; - - return 0; + return ret; } int kvm_vm_ioctl_unregister_coalesced_mmio(struct kvm *kvm, From 0519e9ad89e5cd6e6b08398f57c6a71d9580564c Mon Sep 17 00:00:00 2001 From: Harald Freudenberger Date: Thu, 16 Jan 2014 16:01:11 +0100 Subject: [PATCH 057/596] crypto: s390 - fix concurrency issue in aes-ctr mode The aes-ctr mode uses one preallocated page without any concurrency protection. When multiple threads run aes-ctr encryption or decryption this can lead to data corruption. The patch introduces locking for the page and a fallback solution with slower en/decryption performance in concurrency situations. Cc: stable@vger.kernel.org Signed-off-by: Harald Freudenberger Signed-off-by: Herbert Xu --- arch/s390/crypto/aes_s390.c | 65 ++++++++++++++++++++++++++----------- 1 file changed, 46 insertions(+), 19 deletions(-) diff --git a/arch/s390/crypto/aes_s390.c b/arch/s390/crypto/aes_s390.c index b3feabd39f31..cf3c0089bef2 100644 --- a/arch/s390/crypto/aes_s390.c +++ b/arch/s390/crypto/aes_s390.c @@ -25,6 +25,7 @@ #include #include #include +#include #include "crypt_s390.h" #define AES_KEYLEN_128 1 @@ -32,6 +33,7 @@ #define AES_KEYLEN_256 4 static u8 *ctrblk; +static DEFINE_SPINLOCK(ctrblk_lock); static char keylen_flag; struct s390_aes_ctx { @@ -758,43 +760,67 @@ static int ctr_aes_set_key(struct crypto_tfm *tfm, const u8 *in_key, return aes_set_key(tfm, in_key, key_len); } +static unsigned int __ctrblk_init(u8 *ctrptr, unsigned int nbytes) +{ + unsigned int i, n; + + /* only use complete blocks, max. PAGE_SIZE */ + n = (nbytes > PAGE_SIZE) ? PAGE_SIZE : nbytes & ~(AES_BLOCK_SIZE - 1); + for (i = AES_BLOCK_SIZE; i < n; i += AES_BLOCK_SIZE) { + memcpy(ctrptr + i, ctrptr + i - AES_BLOCK_SIZE, + AES_BLOCK_SIZE); + crypto_inc(ctrptr + i, AES_BLOCK_SIZE); + } + return n; +} + static int ctr_aes_crypt(struct blkcipher_desc *desc, long func, struct s390_aes_ctx *sctx, struct blkcipher_walk *walk) { int ret = blkcipher_walk_virt_block(desc, walk, AES_BLOCK_SIZE); - unsigned int i, n, nbytes; - u8 buf[AES_BLOCK_SIZE]; - u8 *out, *in; + unsigned int n, nbytes; + u8 buf[AES_BLOCK_SIZE], ctrbuf[AES_BLOCK_SIZE]; + u8 *out, *in, *ctrptr = ctrbuf; if (!walk->nbytes) return ret; - memcpy(ctrblk, walk->iv, AES_BLOCK_SIZE); + if (spin_trylock(&ctrblk_lock)) + ctrptr = ctrblk; + + memcpy(ctrptr, walk->iv, AES_BLOCK_SIZE); while ((nbytes = walk->nbytes) >= AES_BLOCK_SIZE) { out = walk->dst.virt.addr; in = walk->src.virt.addr; while (nbytes >= AES_BLOCK_SIZE) { - /* only use complete blocks, max. PAGE_SIZE */ - n = (nbytes > PAGE_SIZE) ? PAGE_SIZE : - nbytes & ~(AES_BLOCK_SIZE - 1); - for (i = AES_BLOCK_SIZE; i < n; i += AES_BLOCK_SIZE) { - memcpy(ctrblk + i, ctrblk + i - AES_BLOCK_SIZE, - AES_BLOCK_SIZE); - crypto_inc(ctrblk + i, AES_BLOCK_SIZE); - } - ret = crypt_s390_kmctr(func, sctx->key, out, in, n, ctrblk); - if (ret < 0 || ret != n) + if (ctrptr == ctrblk) + n = __ctrblk_init(ctrptr, nbytes); + else + n = AES_BLOCK_SIZE; + ret = crypt_s390_kmctr(func, sctx->key, out, in, + n, ctrptr); + if (ret < 0 || ret != n) { + if (ctrptr == ctrblk) + spin_unlock(&ctrblk_lock); return -EIO; + } if (n > AES_BLOCK_SIZE) - memcpy(ctrblk, ctrblk + n - AES_BLOCK_SIZE, + memcpy(ctrptr, ctrptr + n - AES_BLOCK_SIZE, AES_BLOCK_SIZE); - crypto_inc(ctrblk, AES_BLOCK_SIZE); + crypto_inc(ctrptr, AES_BLOCK_SIZE); out += n; in += n; nbytes -= n; } ret = blkcipher_walk_done(desc, walk, nbytes); } + if (ctrptr == ctrblk) { + if (nbytes) + memcpy(ctrbuf, ctrptr, AES_BLOCK_SIZE); + else + memcpy(walk->iv, ctrptr, AES_BLOCK_SIZE); + spin_unlock(&ctrblk_lock); + } /* * final block may be < AES_BLOCK_SIZE, copy only nbytes */ @@ -802,14 +828,15 @@ static int ctr_aes_crypt(struct blkcipher_desc *desc, long func, out = walk->dst.virt.addr; in = walk->src.virt.addr; ret = crypt_s390_kmctr(func, sctx->key, buf, in, - AES_BLOCK_SIZE, ctrblk); + AES_BLOCK_SIZE, ctrbuf); if (ret < 0 || ret != AES_BLOCK_SIZE) return -EIO; memcpy(out, buf, nbytes); - crypto_inc(ctrblk, AES_BLOCK_SIZE); + crypto_inc(ctrbuf, AES_BLOCK_SIZE); ret = blkcipher_walk_done(desc, walk, 0); + memcpy(walk->iv, ctrbuf, AES_BLOCK_SIZE); } - memcpy(walk->iv, ctrblk, AES_BLOCK_SIZE); + return ret; } From adc3fcf1552b6e406d172fd9690bbd1395053d13 Mon Sep 17 00:00:00 2001 From: Harald Freudenberger Date: Wed, 22 Jan 2014 13:00:04 +0100 Subject: [PATCH 058/596] crypto: s390 - fix des and des3_ede cbc concurrency issue In s390 des and des3_ede cbc mode the iv value is not protected against concurrency access and modifications from another running en/decrypt operation which is using the very same tfm struct instance. This fix copies the iv to the local stack before the crypto operation and stores the value back when done. Cc: stable@vger.kernel.org Signed-off-by: Harald Freudenberger Signed-off-by: Herbert Xu --- arch/s390/crypto/des_s390.c | 26 ++++++++++++++------------ 1 file changed, 14 insertions(+), 12 deletions(-) diff --git a/arch/s390/crypto/des_s390.c b/arch/s390/crypto/des_s390.c index 200f2a1b599d..82a0a2ad2494 100644 --- a/arch/s390/crypto/des_s390.c +++ b/arch/s390/crypto/des_s390.c @@ -105,29 +105,35 @@ static int ecb_desall_crypt(struct blkcipher_desc *desc, long func, } static int cbc_desall_crypt(struct blkcipher_desc *desc, long func, - u8 *iv, struct blkcipher_walk *walk) + struct blkcipher_walk *walk) { + struct s390_des_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); int ret = blkcipher_walk_virt(desc, walk); unsigned int nbytes = walk->nbytes; + struct { + u8 iv[DES_BLOCK_SIZE]; + u8 key[DES3_KEY_SIZE]; + } param; if (!nbytes) goto out; - memcpy(iv, walk->iv, DES_BLOCK_SIZE); + memcpy(param.iv, walk->iv, DES_BLOCK_SIZE); + memcpy(param.key, ctx->key, DES3_KEY_SIZE); do { /* only use complete blocks */ unsigned int n = nbytes & ~(DES_BLOCK_SIZE - 1); u8 *out = walk->dst.virt.addr; u8 *in = walk->src.virt.addr; - ret = crypt_s390_kmc(func, iv, out, in, n); + ret = crypt_s390_kmc(func, ¶m, out, in, n); if (ret < 0 || ret != n) return -EIO; nbytes &= DES_BLOCK_SIZE - 1; ret = blkcipher_walk_done(desc, walk, nbytes); } while ((nbytes = walk->nbytes)); - memcpy(walk->iv, iv, DES_BLOCK_SIZE); + memcpy(walk->iv, param.iv, DES_BLOCK_SIZE); out: return ret; @@ -179,22 +185,20 @@ static int cbc_des_encrypt(struct blkcipher_desc *desc, struct scatterlist *dst, struct scatterlist *src, unsigned int nbytes) { - struct s390_des_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); struct blkcipher_walk walk; blkcipher_walk_init(&walk, dst, src, nbytes); - return cbc_desall_crypt(desc, KMC_DEA_ENCRYPT, ctx->iv, &walk); + return cbc_desall_crypt(desc, KMC_DEA_ENCRYPT, &walk); } static int cbc_des_decrypt(struct blkcipher_desc *desc, struct scatterlist *dst, struct scatterlist *src, unsigned int nbytes) { - struct s390_des_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); struct blkcipher_walk walk; blkcipher_walk_init(&walk, dst, src, nbytes); - return cbc_desall_crypt(desc, KMC_DEA_DECRYPT, ctx->iv, &walk); + return cbc_desall_crypt(desc, KMC_DEA_DECRYPT, &walk); } static struct crypto_alg cbc_des_alg = { @@ -327,22 +331,20 @@ static int cbc_des3_encrypt(struct blkcipher_desc *desc, struct scatterlist *dst, struct scatterlist *src, unsigned int nbytes) { - struct s390_des_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); struct blkcipher_walk walk; blkcipher_walk_init(&walk, dst, src, nbytes); - return cbc_desall_crypt(desc, KMC_TDEA_192_ENCRYPT, ctx->iv, &walk); + return cbc_desall_crypt(desc, KMC_TDEA_192_ENCRYPT, &walk); } static int cbc_des3_decrypt(struct blkcipher_desc *desc, struct scatterlist *dst, struct scatterlist *src, unsigned int nbytes) { - struct s390_des_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); struct blkcipher_walk walk; blkcipher_walk_init(&walk, dst, src, nbytes); - return cbc_desall_crypt(desc, KMC_TDEA_192_DECRYPT, ctx->iv, &walk); + return cbc_desall_crypt(desc, KMC_TDEA_192_DECRYPT, &walk); } static struct crypto_alg cbc_des3_alg = { From ee97dc7db4cbda33e4241c2d85b42d1835bc8a35 Mon Sep 17 00:00:00 2001 From: Harald Freudenberger Date: Wed, 22 Jan 2014 13:01:33 +0100 Subject: [PATCH 059/596] crypto: s390 - fix des and des3_ede ctr concurrency issue In s390 des and 3des ctr mode there is one preallocated page used to speed up the en/decryption. This page is not protected against concurrent usage and thus there is a potential of data corruption with multiple threads. The fix introduces locking/unlocking the ctr page and a slower fallback solution at concurrency situations. Cc: stable@vger.kernel.org Signed-off-by: Harald Freudenberger Signed-off-by: Herbert Xu --- arch/s390/crypto/des_s390.c | 69 ++++++++++++++++++++++++++----------- 1 file changed, 48 insertions(+), 21 deletions(-) diff --git a/arch/s390/crypto/des_s390.c b/arch/s390/crypto/des_s390.c index 82a0a2ad2494..0a5aac8a9412 100644 --- a/arch/s390/crypto/des_s390.c +++ b/arch/s390/crypto/des_s390.c @@ -25,6 +25,7 @@ #define DES3_KEY_SIZE (3 * DES_KEY_SIZE) static u8 *ctrblk; +static DEFINE_SPINLOCK(ctrblk_lock); struct s390_des_ctx { u8 iv[DES_BLOCK_SIZE]; @@ -368,54 +369,80 @@ static struct crypto_alg cbc_des3_alg = { } }; +static unsigned int __ctrblk_init(u8 *ctrptr, unsigned int nbytes) +{ + unsigned int i, n; + + /* align to block size, max. PAGE_SIZE */ + n = (nbytes > PAGE_SIZE) ? PAGE_SIZE : nbytes & ~(DES_BLOCK_SIZE - 1); + for (i = DES_BLOCK_SIZE; i < n; i += DES_BLOCK_SIZE) { + memcpy(ctrptr + i, ctrptr + i - DES_BLOCK_SIZE, DES_BLOCK_SIZE); + crypto_inc(ctrptr + i, DES_BLOCK_SIZE); + } + return n; +} + static int ctr_desall_crypt(struct blkcipher_desc *desc, long func, - struct s390_des_ctx *ctx, struct blkcipher_walk *walk) + struct s390_des_ctx *ctx, + struct blkcipher_walk *walk) { int ret = blkcipher_walk_virt_block(desc, walk, DES_BLOCK_SIZE); - unsigned int i, n, nbytes; - u8 buf[DES_BLOCK_SIZE]; - u8 *out, *in; + unsigned int n, nbytes; + u8 buf[DES_BLOCK_SIZE], ctrbuf[DES_BLOCK_SIZE]; + u8 *out, *in, *ctrptr = ctrbuf; - memcpy(ctrblk, walk->iv, DES_BLOCK_SIZE); + if (!walk->nbytes) + return ret; + + if (spin_trylock(&ctrblk_lock)) + ctrptr = ctrblk; + + memcpy(ctrptr, walk->iv, DES_BLOCK_SIZE); while ((nbytes = walk->nbytes) >= DES_BLOCK_SIZE) { out = walk->dst.virt.addr; in = walk->src.virt.addr; while (nbytes >= DES_BLOCK_SIZE) { - /* align to block size, max. PAGE_SIZE */ - n = (nbytes > PAGE_SIZE) ? PAGE_SIZE : - nbytes & ~(DES_BLOCK_SIZE - 1); - for (i = DES_BLOCK_SIZE; i < n; i += DES_BLOCK_SIZE) { - memcpy(ctrblk + i, ctrblk + i - DES_BLOCK_SIZE, - DES_BLOCK_SIZE); - crypto_inc(ctrblk + i, DES_BLOCK_SIZE); - } - ret = crypt_s390_kmctr(func, ctx->key, out, in, n, ctrblk); - if (ret < 0 || ret != n) + if (ctrptr == ctrblk) + n = __ctrblk_init(ctrptr, nbytes); + else + n = DES_BLOCK_SIZE; + ret = crypt_s390_kmctr(func, ctx->key, out, in, + n, ctrptr); + if (ret < 0 || ret != n) { + if (ctrptr == ctrblk) + spin_unlock(&ctrblk_lock); return -EIO; + } if (n > DES_BLOCK_SIZE) - memcpy(ctrblk, ctrblk + n - DES_BLOCK_SIZE, + memcpy(ctrptr, ctrptr + n - DES_BLOCK_SIZE, DES_BLOCK_SIZE); - crypto_inc(ctrblk, DES_BLOCK_SIZE); + crypto_inc(ctrptr, DES_BLOCK_SIZE); out += n; in += n; nbytes -= n; } ret = blkcipher_walk_done(desc, walk, nbytes); } - + if (ctrptr == ctrblk) { + if (nbytes) + memcpy(ctrbuf, ctrptr, DES_BLOCK_SIZE); + else + memcpy(walk->iv, ctrptr, DES_BLOCK_SIZE); + spin_unlock(&ctrblk_lock); + } /* final block may be < DES_BLOCK_SIZE, copy only nbytes */ if (nbytes) { out = walk->dst.virt.addr; in = walk->src.virt.addr; ret = crypt_s390_kmctr(func, ctx->key, buf, in, - DES_BLOCK_SIZE, ctrblk); + DES_BLOCK_SIZE, ctrbuf); if (ret < 0 || ret != DES_BLOCK_SIZE) return -EIO; memcpy(out, buf, nbytes); - crypto_inc(ctrblk, DES_BLOCK_SIZE); + crypto_inc(ctrbuf, DES_BLOCK_SIZE); ret = blkcipher_walk_done(desc, walk, 0); + memcpy(walk->iv, ctrbuf, DES_BLOCK_SIZE); } - memcpy(walk->iv, ctrblk, DES_BLOCK_SIZE); return ret; } From 2cac613be8d4d661edd359cdab3c474286c4f5f0 Mon Sep 17 00:00:00 2001 From: Imre Deak Date: Thu, 30 Jan 2014 16:50:42 +0200 Subject: [PATCH 060/596] drm/i915: vlv: fix DP PHY lockup due to invalid PP sequencer setup MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Atm we setup the HW panel power sequencer logic both for eDP and DP ports. On eDP we then go on and start the power on sequence and commence with link training when it's ready. On DP we don't do the power on sequencing but do the link training immediately. At this point the DP PHY block gets stuck, since - supposedly - it is waiting for the power on sequence to finish. The actual register write that seems to hold off the PHY is PIPEX_PP_ON_DELAYS[Panel Control Port Select]. Writing here a non-0 value eventually sets PIPEX_PP_STATUS[Require Asset Status] to 1 and blocks the PHY until the panel power on is ready. Fix this by not doing any PP sequencing setup for DP ports. Thanks to Ville Syrjälä, Jesse Barnes and Todd Previte for the help in tracking this down. Note that on older gmch platforms (where we have lvds instead of edp) we've hacked around this by writing the magic ABCD unlock key to PP registers, which disables the hw sanity checks. For edp all platforms thus far had the pch split, with the edp port in the north display complex and the PP registers on the pch the hw sanity checks (expressed through the "Require Asset Status" bit) was never functional, hence never a real issue. This regression has been introduce in commit bf13e81b904a37d94d83dd6c3b53a147719a3ead Author: Jani Nikula Date: Fri Sep 6 07:40:05 2013 +0300 drm/i915: add support for per-pipe power sequencing on vlv Signed-off-by: Imre Deak [danvet: Add note about the bigger story here.] Cc: stable@vger.kernel.org Signed-off-by: Daniel Vetter --- drivers/gpu/drm/i915/intel_dp.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c index 5ede4e8e290d..c00b6e352c2b 100644 --- a/drivers/gpu/drm/i915/intel_dp.c +++ b/drivers/gpu/drm/i915/intel_dp.c @@ -1869,10 +1869,12 @@ static void vlv_pre_enable_dp(struct intel_encoder *encoder) mutex_unlock(&dev_priv->dpio_lock); - /* init power sequencer on this pipe and port */ - intel_dp_init_panel_power_sequencer(dev, intel_dp, &power_seq); - intel_dp_init_panel_power_sequencer_registers(dev, intel_dp, - &power_seq); + if (is_edp(intel_dp)) { + /* init power sequencer on this pipe and port */ + intel_dp_init_panel_power_sequencer(dev, intel_dp, &power_seq); + intel_dp_init_panel_power_sequencer_registers(dev, intel_dp, + &power_seq); + } intel_enable_dp(encoder); From a4243402274343d8e596fb0b25674e52088a7488 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Mon, 20 Jan 2014 23:46:38 +0100 Subject: [PATCH 061/596] iwlwifi: mvm: make local pointer non-static The address pointer used in the function shouldn't be static since it's local data only. Having it static causes races if a single machine has two devices, as the pointer would be shared between instances. Signed-off-by: Johannes Berg Signed-off-by: Emmanuel Grumbach --- drivers/net/wireless/iwlwifi/mvm/sta.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/wireless/iwlwifi/mvm/sta.c b/drivers/net/wireless/iwlwifi/mvm/sta.c index ec1812133235..3397f59cd4e4 100644 --- a/drivers/net/wireless/iwlwifi/mvm/sta.c +++ b/drivers/net/wireless/iwlwifi/mvm/sta.c @@ -652,7 +652,7 @@ int iwl_mvm_send_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif, { struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif); static const u8 _baddr[] = {0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF}; - static const u8 *baddr = _baddr; + const u8 *baddr = _baddr; lockdep_assert_held(&mvm->mutex); From 0822afe8ebb9389997ef677447c7b08e08797de9 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Thu, 16 Jan 2014 09:18:07 +0100 Subject: [PATCH 062/596] iwlwifi: mvm: disable scheduled scan The iwlwifi scheduled scan implementation doesn't adhere to the userspace API correctly - the API assumes that any new incoming 'incompatible' request (like scan or remain-on-channel for this driver) will just cancel the scheduled scan. Instead our driver relies on userspace cancelling it, thus breaking existing wpa_s versions. Cc: stable@vger.kernel.org [3.13] Fixes: 35a000b7c1bb ("iwlwifi: mvm: support sched scan if supported by the fw") Signed-off-by: Johannes Berg Signed-off-by: Emmanuel Grumbach --- drivers/net/wireless/iwlwifi/mvm/mac80211.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/wireless/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/iwlwifi/mvm/mac80211.c index c49b5073c251..6bf9766e5982 100644 --- a/drivers/net/wireless/iwlwifi/mvm/mac80211.c +++ b/drivers/net/wireless/iwlwifi/mvm/mac80211.c @@ -246,7 +246,7 @@ int iwl_mvm_mac_setup_register(struct iwl_mvm *mvm) else hw->wiphy->flags &= ~WIPHY_FLAG_PS_ON_BY_DEFAULT; - if (mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_SCHED_SCAN) { + if (0 && mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_SCHED_SCAN) { hw->wiphy->flags |= WIPHY_FLAG_SUPPORTS_SCHED_SCAN; hw->wiphy->max_sched_scan_ssids = PROBE_OPTION_MAX; hw->wiphy->max_match_sets = IWL_SCAN_MAX_PROFILES; From b900a87b2eb90c0b9586496c82a323a1b8832d73 Mon Sep 17 00:00:00 2001 From: Emmanuel Grumbach Date: Thu, 23 Jan 2014 11:55:16 +0200 Subject: [PATCH 063/596] iwlwifi: mvm: print the version of the firmware when it asserts This can be useful to be able to spot the firmware version from the error reports without needing to fetch it from another place. Cc: [3.10+] Signed-off-by: Emmanuel Grumbach Reviewed-by: Johannes Berg Signed-off-by: Emmanuel Grumbach --- drivers/net/wireless/iwlwifi/mvm/utils.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/wireless/iwlwifi/mvm/utils.c b/drivers/net/wireless/iwlwifi/mvm/utils.c index a4a5e25623c3..86989df69356 100644 --- a/drivers/net/wireless/iwlwifi/mvm/utils.c +++ b/drivers/net/wireless/iwlwifi/mvm/utils.c @@ -411,6 +411,8 @@ void iwl_mvm_dump_nic_error_log(struct iwl_mvm *mvm) mvm->status, table.valid); } + IWL_ERR(mvm, "Loaded firmware version: %s\n", mvm->fw->fw_version); + trace_iwlwifi_dev_ucode_error(trans->dev, table.error_id, table.tsf_low, table.data1, table.data2, table.data3, table.blink1, table.blink2, table.ilink1, From f7690915ccce98553c5425b51e6b5a6c51e27f4e Mon Sep 17 00:00:00 2001 From: Oren Givon Date: Thu, 23 Jan 2014 01:19:33 +0200 Subject: [PATCH 064/596] iwlwifi: add more 7265 HW IDs Add 6 new HW IDs for the 7265 series. Cc: [3.13] Signed-off-by: Oren Givon Signed-off-by: Emmanuel Grumbach --- drivers/net/wireless/iwlwifi/pcie/drv.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/iwlwifi/pcie/drv.c b/drivers/net/wireless/iwlwifi/pcie/drv.c index 3040924f5f3c..f47bcbe2945a 100644 --- a/drivers/net/wireless/iwlwifi/pcie/drv.c +++ b/drivers/net/wireless/iwlwifi/pcie/drv.c @@ -359,20 +359,25 @@ static DEFINE_PCI_DEVICE_TABLE(iwl_hw_card_ids) = { /* 7265 Series */ {IWL_PCI_DEVICE(0x095A, 0x5010, iwl7265_2ac_cfg)}, {IWL_PCI_DEVICE(0x095A, 0x5110, iwl7265_2ac_cfg)}, + {IWL_PCI_DEVICE(0x095A, 0x5112, iwl7265_2ac_cfg)}, + {IWL_PCI_DEVICE(0x095A, 0x5100, iwl7265_2ac_cfg)}, + {IWL_PCI_DEVICE(0x095A, 0x510A, iwl7265_2ac_cfg)}, {IWL_PCI_DEVICE(0x095B, 0x5310, iwl7265_2ac_cfg)}, {IWL_PCI_DEVICE(0x095B, 0x5302, iwl7265_2ac_cfg)}, {IWL_PCI_DEVICE(0x095B, 0x5210, iwl7265_2ac_cfg)}, {IWL_PCI_DEVICE(0x095A, 0x5012, iwl7265_2ac_cfg)}, - {IWL_PCI_DEVICE(0x095A, 0x500A, iwl7265_2ac_cfg)}, {IWL_PCI_DEVICE(0x095A, 0x5410, iwl7265_2ac_cfg)}, {IWL_PCI_DEVICE(0x095A, 0x5400, iwl7265_2ac_cfg)}, {IWL_PCI_DEVICE(0x095A, 0x1010, iwl7265_2ac_cfg)}, {IWL_PCI_DEVICE(0x095A, 0x5000, iwl7265_2n_cfg)}, + {IWL_PCI_DEVICE(0x095A, 0x500A, iwl7265_2n_cfg)}, {IWL_PCI_DEVICE(0x095B, 0x5200, iwl7265_2n_cfg)}, {IWL_PCI_DEVICE(0x095A, 0x5002, iwl7265_n_cfg)}, {IWL_PCI_DEVICE(0x095B, 0x5202, iwl7265_n_cfg)}, {IWL_PCI_DEVICE(0x095A, 0x9010, iwl7265_2ac_cfg)}, + {IWL_PCI_DEVICE(0x095A, 0x9012, iwl7265_2ac_cfg)}, {IWL_PCI_DEVICE(0x095A, 0x9110, iwl7265_2ac_cfg)}, + {IWL_PCI_DEVICE(0x095A, 0x9112, iwl7265_2ac_cfg)}, {IWL_PCI_DEVICE(0x095A, 0x9210, iwl7265_2ac_cfg)}, {IWL_PCI_DEVICE(0x095A, 0x9510, iwl7265_2ac_cfg)}, {IWL_PCI_DEVICE(0x095A, 0x9310, iwl7265_2ac_cfg)}, From 6e0bbe5ee845e185d237ec4f266b7be495c50eb6 Mon Sep 17 00:00:00 2001 From: David Spinadel Date: Mon, 30 Dec 2013 09:59:45 +0200 Subject: [PATCH 065/596] iwlwifi: mvm: notify match found without filtering Configure scheduled scan to notify match found on every beacon or probe response if the scan request doesn't contain valid ssid list for filtering. Without this configuration the FW passes all beacons to the host but doesn't notify the stack that the scan results are ready for processing. Signed-off-by: David Spinadel Reviewed-by: Alexander Bondar Reviewed-by: Johannes Berg Signed-off-by: Emmanuel Grumbach --- drivers/net/wireless/iwlwifi/mvm/fw-api-scan.h | 4 +++- drivers/net/wireless/iwlwifi/mvm/scan.c | 2 ++ 2 files changed, 5 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/iwlwifi/mvm/fw-api-scan.h b/drivers/net/wireless/iwlwifi/mvm/fw-api-scan.h index 73cbba7424f2..9426905de6b2 100644 --- a/drivers/net/wireless/iwlwifi/mvm/fw-api-scan.h +++ b/drivers/net/wireless/iwlwifi/mvm/fw-api-scan.h @@ -504,6 +504,7 @@ struct iwl_scan_offload_profile { * @match_notify: clients waiting for match found notification * @pass_match: clients waiting for the results * @active_clients: active clients bitmap - enum scan_framework_client + * @any_beacon_notify: clients waiting for match notification without match */ struct iwl_scan_offload_profile_cfg { struct iwl_scan_offload_profile profiles[IWL_SCAN_MAX_PROFILES]; @@ -512,7 +513,8 @@ struct iwl_scan_offload_profile_cfg { u8 match_notify; u8 pass_match; u8 active_clients; - u8 reserved[3]; + u8 any_beacon_notify; + u8 reserved[2]; } __packed; /** diff --git a/drivers/net/wireless/iwlwifi/mvm/scan.c b/drivers/net/wireless/iwlwifi/mvm/scan.c index 0e0007960612..c35f35cc4391 100644 --- a/drivers/net/wireless/iwlwifi/mvm/scan.c +++ b/drivers/net/wireless/iwlwifi/mvm/scan.c @@ -807,6 +807,8 @@ int iwl_mvm_config_sched_scan_profiles(struct iwl_mvm *mvm, profile_cfg->active_clients = SCAN_CLIENT_SCHED_SCAN; profile_cfg->pass_match = SCAN_CLIENT_SCHED_SCAN; profile_cfg->match_notify = SCAN_CLIENT_SCHED_SCAN; + if (!req->n_match_sets || !req->match_sets[0].ssid.ssid_len) + profile_cfg->any_beacon_notify = SCAN_CLIENT_SCHED_SCAN; for (i = 0; i < req->n_match_sets; i++) { profile = &profile_cfg->profiles[i]; From 9bb0c1adc52ca3c7026811a6630a3c78eec1f135 Mon Sep 17 00:00:00 2001 From: Emmanuel Grumbach Date: Mon, 20 Jan 2014 15:21:26 +0200 Subject: [PATCH 066/596] iwlwifi: mvm: don't leak a station when we drain We had a bug that prevented us from removing a station after we entered the drain flow: We assign sta to be NULL if it was an error value. Then we tested it against -EBUSY, but forget to retrieve the value again from mvm->fw_id_to_mac_id[sta_id]. Due to this bug, we ended up never removing the STA from the firmware. This led to an firmware assert when we remove the GO vif. Reviewed-by: Johannes Berg Signed-off-by: Emmanuel Grumbach --- drivers/net/wireless/iwlwifi/mvm/tx.c | 73 ++++++++++++++------------- 1 file changed, 37 insertions(+), 36 deletions(-) diff --git a/drivers/net/wireless/iwlwifi/mvm/tx.c b/drivers/net/wireless/iwlwifi/mvm/tx.c index 90378c217bc7..4df12fa9d336 100644 --- a/drivers/net/wireless/iwlwifi/mvm/tx.c +++ b/drivers/net/wireless/iwlwifi/mvm/tx.c @@ -659,8 +659,14 @@ static void iwl_mvm_rx_tx_cmd_single(struct iwl_mvm *mvm, rcu_read_lock(); sta = rcu_dereference(mvm->fw_id_to_mac_id[sta_id]); + /* + * sta can't be NULL otherwise it'd mean that the sta has been freed in + * the firmware while we still have packets for it in the Tx queues. + */ + if (WARN_ON_ONCE(!sta)) + goto out; - if (!IS_ERR_OR_NULL(sta)) { + if (!IS_ERR(sta)) { mvmsta = iwl_mvm_sta_from_mac80211(sta); if (tid != IWL_TID_NON_QOS) { @@ -675,7 +681,6 @@ static void iwl_mvm_rx_tx_cmd_single(struct iwl_mvm *mvm, spin_unlock_bh(&mvmsta->lock); } } else { - sta = NULL; mvmsta = NULL; } @@ -683,42 +688,38 @@ static void iwl_mvm_rx_tx_cmd_single(struct iwl_mvm *mvm, * If the txq is not an AMPDU queue, there is no chance we freed * several skbs. Check that out... */ - if (txq_id < mvm->first_agg_queue && !WARN_ON(skb_freed > 1) && - atomic_sub_and_test(skb_freed, &mvm->pending_frames[sta_id])) { - if (mvmsta) { - /* - * If there are no pending frames for this STA, notify - * mac80211 that this station can go to sleep in its - * STA table. - */ - if (mvmsta->vif->type == NL80211_IFTYPE_AP) - ieee80211_sta_block_awake(mvm->hw, sta, false); - /* - * We might very well have taken mvmsta pointer while - * the station was being removed. The remove flow might - * have seen a pending_frame (because we didn't take - * the lock) even if now the queues are drained. So make - * really sure now that this the station is not being - * removed. If it is, run the drain worker to remove it. - */ - spin_lock_bh(&mvmsta->lock); - sta = rcu_dereference(mvm->fw_id_to_mac_id[sta_id]); - if (!sta || PTR_ERR(sta) == -EBUSY) { - /* - * Station disappeared in the meantime: - * so we are draining. - */ - set_bit(sta_id, mvm->sta_drained); - schedule_work(&mvm->sta_drained_wk); - } - spin_unlock_bh(&mvmsta->lock); - } else if (!mvmsta && PTR_ERR(sta) == -EBUSY) { - /* Tx response without STA, so we are draining */ - set_bit(sta_id, mvm->sta_drained); - schedule_work(&mvm->sta_drained_wk); - } + if (txq_id >= mvm->first_agg_queue) + goto out; + + /* We can't free more than one frame at once on a shared queue */ + WARN_ON(skb_freed > 1); + + /* If we have still frames from this STA nothing to do here */ + if (!atomic_sub_and_test(skb_freed, &mvm->pending_frames[sta_id])) + goto out; + + if (mvmsta && mvmsta->vif->type == NL80211_IFTYPE_AP) { + /* + * If there are no pending frames for this STA, notify + * mac80211 that this station can go to sleep in its + * STA table. + * If mvmsta is not NULL, sta is valid. + */ + ieee80211_sta_block_awake(mvm->hw, sta, false); } + if (PTR_ERR(sta) == -EBUSY || PTR_ERR(sta) == -ENOENT) { + /* + * We are draining and this was the last packet - pre_rcu_remove + * has been called already. We might be after the + * synchronize_net already. + * Don't rely on iwl_mvm_rm_sta to see the empty Tx queues. + */ + set_bit(sta_id, mvm->sta_drained); + schedule_work(&mvm->sta_drained_wk); + } + +out: rcu_read_unlock(); } From d835502f3dacad1638d516ab156d66f0ba377cf5 Mon Sep 17 00:00:00 2001 From: Shaohua Li Date: Tue, 31 Dec 2013 11:38:27 +0800 Subject: [PATCH 067/596] percpu_ida: fix a live lock steal_tags only happens when free tags is more than half of the total tags. This is too strict and can cause live lock. I found that if one cpu has free tags, but other cpu can't steal (thread is bound to specific cpus), threads which want to allocate tags are always sleeping. I found this when I run next patch, but this could happen without it I think. I did performance test too with null_blk. Two cases (each cpu has enough percpu tags, or total tags are limited), no performance changes were observed. Signed-off-by: Shaohua Li Signed-off-by: Jens Axboe --- lib/percpu_ida.c | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) diff --git a/lib/percpu_ida.c b/lib/percpu_ida.c index 9d054bf91d0f..85f43b1718d4 100644 --- a/lib/percpu_ida.c +++ b/lib/percpu_ida.c @@ -54,9 +54,7 @@ static inline void move_tags(unsigned *dst, unsigned *dst_nr, /* * Try to steal tags from a remote cpu's percpu freelist. * - * We first check how many percpu freelists have tags - we don't steal tags - * unless enough percpu freelists have tags on them that it's possible more than - * half the total tags could be stuck on remote percpu freelists. + * We first check how many percpu freelists have tags * * Then we iterate through the cpus until we find some tags - we don't attempt * to find the "best" cpu to steal from, to keep cacheline bouncing to a @@ -69,8 +67,7 @@ static inline void steal_tags(struct percpu_ida *pool, struct percpu_ida_cpu *remote; for (cpus_have_tags = cpumask_weight(&pool->cpus_have_tags); - cpus_have_tags * pool->percpu_max_size > pool->nr_tags / 2; - cpus_have_tags--) { + cpus_have_tags; cpus_have_tags--) { cpu = cpumask_next(cpu, &pool->cpus_have_tags); if (cpu >= nr_cpu_ids) { From f0276924fa35a3607920a58cf5d878212824b951 Mon Sep 17 00:00:00 2001 From: Shaohua Li Date: Tue, 31 Dec 2013 11:38:50 +0800 Subject: [PATCH 068/596] blk-mq: Don't reserve a tag for flush request Reserving a tag (request) for flush to avoid dead lock is a overkill. A tag is valuable resource. We can track the number of flush requests and disallow having too many pending flush requests allocated. With this patch, blk_mq_alloc_request_pinned() could do a busy nop (but not a dead loop) if too many pending requests are allocated and new flush request is allocated. But this should not be a problem, too many pending flush requests are very rare case. I verified this can fix the deadlock caused by too many pending flush requests. Signed-off-by: Shaohua Li Signed-off-by: Jens Axboe --- block/blk-flush.c | 8 +++++--- block/blk-mq.c | 46 +++++++++++++++++++++++++++--------------- include/linux/blk-mq.h | 3 +++ 3 files changed, 38 insertions(+), 19 deletions(-) diff --git a/block/blk-flush.c b/block/blk-flush.c index 9288aaf35c21..9143e85226c7 100644 --- a/block/blk-flush.c +++ b/block/blk-flush.c @@ -284,9 +284,8 @@ static void mq_flush_work(struct work_struct *work) q = container_of(work, struct request_queue, mq_flush_work); - /* We don't need set REQ_FLUSH_SEQ, it's for consistency */ rq = blk_mq_alloc_request(q, WRITE_FLUSH|REQ_FLUSH_SEQ, - __GFP_WAIT|GFP_ATOMIC, true); + __GFP_WAIT|GFP_ATOMIC, false); rq->cmd_type = REQ_TYPE_FS; rq->end_io = flush_end_io; @@ -408,8 +407,11 @@ void blk_insert_flush(struct request *rq) /* * @policy now records what operations need to be done. Adjust * REQ_FLUSH and FUA for the driver. + * We keep REQ_FLUSH for mq to track flush requests. For !FUA, + * we never dispatch the request directly. */ - rq->cmd_flags &= ~REQ_FLUSH; + if (rq->cmd_flags & REQ_FUA) + rq->cmd_flags &= ~REQ_FLUSH; if (!(fflags & REQ_FUA)) rq->cmd_flags &= ~REQ_FUA; diff --git a/block/blk-mq.c b/block/blk-mq.c index 57039fcd9c93..9072d0ab184f 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -194,9 +194,27 @@ static void blk_mq_rq_ctx_init(struct request_queue *q, struct blk_mq_ctx *ctx, } static struct request *__blk_mq_alloc_request(struct blk_mq_hw_ctx *hctx, - gfp_t gfp, bool reserved) + gfp_t gfp, bool reserved, + int rw) { - return blk_mq_alloc_rq(hctx, gfp, reserved); + struct request *req; + bool is_flush = false; + /* + * flush need allocate a request, leave at least one request for + * non-flush IO to avoid deadlock + */ + if ((rw & REQ_FLUSH) && !(rw & REQ_FLUSH_SEQ)) { + if (atomic_inc_return(&hctx->pending_flush) >= + hctx->queue_depth - hctx->reserved_tags - 1) { + atomic_dec(&hctx->pending_flush); + return NULL; + } + is_flush = true; + } + req = blk_mq_alloc_rq(hctx, gfp, reserved); + if (!req && is_flush) + atomic_dec(&hctx->pending_flush); + return req; } static struct request *blk_mq_alloc_request_pinned(struct request_queue *q, @@ -209,7 +227,7 @@ static struct request *blk_mq_alloc_request_pinned(struct request_queue *q, struct blk_mq_ctx *ctx = blk_mq_get_ctx(q); struct blk_mq_hw_ctx *hctx = q->mq_ops->map_queue(q, ctx->cpu); - rq = __blk_mq_alloc_request(hctx, gfp & ~__GFP_WAIT, reserved); + rq = __blk_mq_alloc_request(hctx, gfp & ~__GFP_WAIT, reserved, rw); if (rq) { blk_mq_rq_ctx_init(q, ctx, rq, rw); break; @@ -272,6 +290,9 @@ static void __blk_mq_free_request(struct blk_mq_hw_ctx *hctx, const int tag = rq->tag; struct request_queue *q = rq->q; + if ((rq->cmd_flags & REQ_FLUSH) && !(rq->cmd_flags & REQ_FLUSH_SEQ)) + atomic_dec(&hctx->pending_flush); + blk_mq_rq_init(hctx, rq); blk_mq_put_tag(hctx->tags, tag); @@ -900,14 +921,14 @@ static void blk_mq_make_request(struct request_queue *q, struct bio *bio) hctx = q->mq_ops->map_queue(q, ctx->cpu); trace_block_getrq(q, bio, rw); - rq = __blk_mq_alloc_request(hctx, GFP_ATOMIC, false); + rq = __blk_mq_alloc_request(hctx, GFP_ATOMIC, false, bio->bi_rw); if (likely(rq)) - blk_mq_rq_ctx_init(q, ctx, rq, rw); + blk_mq_rq_ctx_init(q, ctx, rq, bio->bi_rw); else { blk_mq_put_ctx(ctx); trace_block_sleeprq(q, bio, rw); - rq = blk_mq_alloc_request_pinned(q, rw, __GFP_WAIT|GFP_ATOMIC, - false); + rq = blk_mq_alloc_request_pinned(q, bio->bi_rw, + __GFP_WAIT|GFP_ATOMIC, false); ctx = rq->mq_ctx; hctx = q->mq_ops->map_queue(q, ctx->cpu); } @@ -1184,7 +1205,9 @@ static int blk_mq_init_hw_queues(struct request_queue *q, hctx->queue_num = i; hctx->flags = reg->flags; hctx->queue_depth = reg->queue_depth; + hctx->reserved_tags = reg->reserved_tags; hctx->cmd_size = reg->cmd_size; + atomic_set(&hctx->pending_flush, 0); blk_mq_init_cpu_notifier(&hctx->cpu_notifier, blk_mq_hctx_notify, hctx); @@ -1309,15 +1332,6 @@ struct request_queue *blk_mq_init_queue(struct blk_mq_reg *reg, reg->queue_depth = BLK_MQ_MAX_DEPTH; } - /* - * Set aside a tag for flush requests. It will only be used while - * another flush request is in progress but outside the driver. - * - * TODO: only allocate if flushes are supported - */ - reg->queue_depth++; - reg->reserved_tags++; - if (reg->queue_depth < (reg->reserved_tags + BLK_MQ_TAG_MIN)) return ERR_PTR(-EINVAL); diff --git a/include/linux/blk-mq.h b/include/linux/blk-mq.h index 161b23105b1e..1e8f16f65af4 100644 --- a/include/linux/blk-mq.h +++ b/include/linux/blk-mq.h @@ -36,12 +36,15 @@ struct blk_mq_hw_ctx { struct list_head page_list; struct blk_mq_tags *tags; + atomic_t pending_flush; + unsigned long queued; unsigned long run; #define BLK_MQ_MAX_DISPATCH_ORDER 10 unsigned long dispatched[BLK_MQ_MAX_DISPATCH_ORDER]; unsigned int queue_depth; + unsigned int reserved_tags; unsigned int numa_node; unsigned int cmd_size; /* per-request extra data */ From 556ee818c06f37b2e583af0363e6b16d0e0270de Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Wed, 29 Jan 2014 14:56:16 -0700 Subject: [PATCH 069/596] block: __elv_next_request() shouldn't call into the elevator if bypassing request_queue bypassing is used to suppress higher-level function of a request_queue so that they can be switched, reconfigured and shut down. A request_queue does the followings while bypassing. * bypasses elevator and io_cq association and queues requests directly to the FIFO dispatch queue. * bypasses block cgroup request_list lookup and always uses the root request_list. Once confirmed to be bypassing, specific elevator and block cgroup policy implementations can assume that nothing is in flight for them and perform various operations which would be dangerous otherwise. Such confirmation is acheived by short-circuiting all new requests directly to the dispatch queue and waiting for all the requests which were issued before to finish. Unfortunately, while the request allocating and draining sides were properly handled, we forgot to actually plug the request dispatch path. Even after bypassing mode is confirmed, if the attached driver tries to fetch a request and the dispatch queue is empty, __elv_next_request() would invoke the current elevator's elevator_dispatch_fn() callback. As all in-flight requests were drained, the elevator wouldn't contain any request but once bypass is confirmed we don't even know whether the elevator is even there. It might be in the process of being switched and half torn down. Frank Mayhar reports that this actually happened while switching elevators, leading to an oops. Let's fix it by making __elv_next_request() avoid invoking the elevator_dispatch_fn() callback if the queue is bypassing. It already avoids invoking the callback if the queue is dying. As a dying queue is guaranteed to be bypassing, we can simply replace blk_queue_dying() check with blk_queue_bypass(). Reported-by: Frank Mayhar References: http://lkml.kernel.org/g/1390319905.20232.38.camel@bobble.lax.corp.google.com Cc: stable@vger.kernel.org Tested-by: Frank Mayhar Signed-off-by: Jens Axboe --- block/blk.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/block/blk.h b/block/blk.h index c90e1d8f7a2b..d23b415b8a28 100644 --- a/block/blk.h +++ b/block/blk.h @@ -113,7 +113,7 @@ static inline struct request *__elv_next_request(struct request_queue *q) q->flush_queue_delayed = 1; return NULL; } - if (unlikely(blk_queue_dying(q)) || + if (unlikely(blk_queue_bypass(q)) || !q->elevator->type->ops.elevator_dispatch_fn(q, 0)) return NULL; } From f7b2e4032d52deff480e0c303fbd9180276f8dfc Mon Sep 17 00:00:00 2001 From: Sarah Sharp Date: Thu, 30 Jan 2014 13:27:49 -0800 Subject: [PATCH 070/596] Revert "xhci: replace xhci_read_64() with readq()" This reverts commit e8b373326d8efcaf9ec1da8b618556c89bd5ffc4. Many xHCI host controllers can only handle 32-bit addresses, and writing 64-bits at a time causes them to fail. Reading 64-bits at a time may also cause them to return 0xffffffff, so revert this commit as well. Signed-off-by: Sarah Sharp --- drivers/usb/host/xhci-dbg.c | 6 +++--- drivers/usb/host/xhci-mem.c | 6 +++--- drivers/usb/host/xhci-ring.c | 6 +++--- drivers/usb/host/xhci.c | 12 ++++++------ drivers/usb/host/xhci.h | 10 ++++++++-- 5 files changed, 23 insertions(+), 17 deletions(-) diff --git a/drivers/usb/host/xhci-dbg.c b/drivers/usb/host/xhci-dbg.c index b016d38199f2..eb009a457fb5 100644 --- a/drivers/usb/host/xhci-dbg.c +++ b/drivers/usb/host/xhci-dbg.c @@ -203,12 +203,12 @@ void xhci_print_ir_set(struct xhci_hcd *xhci, int set_num) addr, (unsigned int)temp); addr = &ir_set->erst_base; - temp_64 = readq(addr); + temp_64 = xhci_read_64(xhci, addr); xhci_dbg(xhci, " %p: ir_set.erst_base = @%08llx\n", addr, temp_64); addr = &ir_set->erst_dequeue; - temp_64 = readq(addr); + temp_64 = xhci_read_64(xhci, addr); xhci_dbg(xhci, " %p: ir_set.erst_dequeue = @%08llx\n", addr, temp_64); } @@ -412,7 +412,7 @@ void xhci_dbg_cmd_ptrs(struct xhci_hcd *xhci) { u64 val; - val = readq(&xhci->op_regs->cmd_ring); + val = xhci_read_64(xhci, &xhci->op_regs->cmd_ring); xhci_dbg(xhci, "// xHC command ring deq ptr low bits + flags = @%08x\n", lower_32_bits(val)); xhci_dbg(xhci, "// xHC command ring deq ptr high bits = @%08x\n", diff --git a/drivers/usb/host/xhci-mem.c b/drivers/usb/host/xhci-mem.c index 4b87026f8a5a..bce4391a0e7d 100644 --- a/drivers/usb/host/xhci-mem.c +++ b/drivers/usb/host/xhci-mem.c @@ -1958,7 +1958,7 @@ static void xhci_set_hc_event_deq(struct xhci_hcd *xhci) xhci_warn(xhci, "WARN something wrong with SW event ring " "dequeue ptr.\n"); /* Update HC event ring dequeue pointer */ - temp = readq(&xhci->ir_set->erst_dequeue); + temp = xhci_read_64(xhci, &xhci->ir_set->erst_dequeue); temp &= ERST_PTR_MASK; /* Don't clear the EHB bit (which is RW1C) because * there might be more events to service. @@ -2312,7 +2312,7 @@ int xhci_mem_init(struct xhci_hcd *xhci, gfp_t flags) (unsigned long long)xhci->cmd_ring->first_seg->dma); /* Set the address in the Command Ring Control register */ - val_64 = readq(&xhci->op_regs->cmd_ring); + val_64 = xhci_read_64(xhci, &xhci->op_regs->cmd_ring); val_64 = (val_64 & (u64) CMD_RING_RSVD_BITS) | (xhci->cmd_ring->first_seg->dma & (u64) ~CMD_RING_RSVD_BITS) | xhci->cmd_ring->cycle_state; @@ -2396,7 +2396,7 @@ int xhci_mem_init(struct xhci_hcd *xhci, gfp_t flags) xhci_dbg_trace(xhci, trace_xhci_dbg_init, "// Set ERST base address for ir_set 0 = 0x%llx", (unsigned long long)xhci->erst.erst_dma_addr); - val_64 = readq(&xhci->ir_set->erst_base); + val_64 = xhci_read_64(xhci, &xhci->ir_set->erst_base); val_64 &= ERST_PTR_MASK; val_64 |= (xhci->erst.erst_dma_addr & (u64) ~ERST_PTR_MASK); xhci_write_64(xhci, val_64, &xhci->ir_set->erst_base); diff --git a/drivers/usb/host/xhci-ring.c b/drivers/usb/host/xhci-ring.c index e19fae964f76..909b32a4412f 100644 --- a/drivers/usb/host/xhci-ring.c +++ b/drivers/usb/host/xhci-ring.c @@ -307,7 +307,7 @@ static int xhci_abort_cmd_ring(struct xhci_hcd *xhci) return 0; } - temp_64 = readq(&xhci->op_regs->cmd_ring); + temp_64 = xhci_read_64(xhci, &xhci->op_regs->cmd_ring); if (!(temp_64 & CMD_RING_RUNNING)) { xhci_dbg(xhci, "Command ring had been stopped\n"); return 0; @@ -2865,7 +2865,7 @@ irqreturn_t xhci_irq(struct usb_hcd *hcd) /* Clear the event handler busy flag (RW1C); * the event ring should be empty. */ - temp_64 = readq(&xhci->ir_set->erst_dequeue); + temp_64 = xhci_read_64(xhci, &xhci->ir_set->erst_dequeue); xhci_write_64(xhci, temp_64 | ERST_EHB, &xhci->ir_set->erst_dequeue); spin_unlock(&xhci->lock); @@ -2879,7 +2879,7 @@ irqreturn_t xhci_irq(struct usb_hcd *hcd) */ while (xhci_handle_event(xhci) > 0) {} - temp_64 = readq(&xhci->ir_set->erst_dequeue); + temp_64 = xhci_read_64(xhci, &xhci->ir_set->erst_dequeue); /* If necessary, update the HW's version of the event ring deq ptr. */ if (event_ring_deq != xhci->event_ring->dequeue) { deq = xhci_trb_virt_to_dma(xhci->event_ring->deq_seg, diff --git a/drivers/usb/host/xhci.c b/drivers/usb/host/xhci.c index ebbe52cafaa1..3712359d18ba 100644 --- a/drivers/usb/host/xhci.c +++ b/drivers/usb/host/xhci.c @@ -611,7 +611,7 @@ int xhci_run(struct usb_hcd *hcd) xhci_dbg(xhci, "Event ring:\n"); xhci_debug_ring(xhci, xhci->event_ring); xhci_dbg_ring_ptrs(xhci, xhci->event_ring); - temp_64 = readq(&xhci->ir_set->erst_dequeue); + temp_64 = xhci_read_64(xhci, &xhci->ir_set->erst_dequeue); temp_64 &= ~ERST_PTR_MASK; xhci_dbg_trace(xhci, trace_xhci_dbg_init, "ERST deq = 64'h%0lx", (long unsigned int) temp_64); @@ -756,11 +756,11 @@ static void xhci_save_registers(struct xhci_hcd *xhci) { xhci->s3.command = readl(&xhci->op_regs->command); xhci->s3.dev_nt = readl(&xhci->op_regs->dev_notification); - xhci->s3.dcbaa_ptr = readq(&xhci->op_regs->dcbaa_ptr); + xhci->s3.dcbaa_ptr = xhci_read_64(xhci, &xhci->op_regs->dcbaa_ptr); xhci->s3.config_reg = readl(&xhci->op_regs->config_reg); xhci->s3.erst_size = readl(&xhci->ir_set->erst_size); - xhci->s3.erst_base = readq(&xhci->ir_set->erst_base); - xhci->s3.erst_dequeue = readq(&xhci->ir_set->erst_dequeue); + xhci->s3.erst_base = xhci_read_64(xhci, &xhci->ir_set->erst_base); + xhci->s3.erst_dequeue = xhci_read_64(xhci, &xhci->ir_set->erst_dequeue); xhci->s3.irq_pending = readl(&xhci->ir_set->irq_pending); xhci->s3.irq_control = readl(&xhci->ir_set->irq_control); } @@ -783,7 +783,7 @@ static void xhci_set_cmd_ring_deq(struct xhci_hcd *xhci) u64 val_64; /* step 2: initialize command ring buffer */ - val_64 = readq(&xhci->op_regs->cmd_ring); + val_64 = xhci_read_64(xhci, &xhci->op_regs->cmd_ring); val_64 = (val_64 & (u64) CMD_RING_RSVD_BITS) | (xhci_trb_virt_to_dma(xhci->cmd_ring->deq_seg, xhci->cmd_ring->dequeue) & @@ -3842,7 +3842,7 @@ static int xhci_setup_device(struct usb_hcd *hcd, struct usb_device *udev, if (ret) { return ret; } - temp_64 = readq(&xhci->op_regs->dcbaa_ptr); + temp_64 = xhci_read_64(xhci, &xhci->op_regs->dcbaa_ptr); xhci_dbg_trace(xhci, trace_xhci_dbg_address, "Op regs DCBAA ptr = %#016llx", temp_64); xhci_dbg_trace(xhci, trace_xhci_dbg_address, diff --git a/drivers/usb/host/xhci.h b/drivers/usb/host/xhci.h index 60def15d129c..9154fd6cf24c 100644 --- a/drivers/usb/host/xhci.h +++ b/drivers/usb/host/xhci.h @@ -28,8 +28,6 @@ #include #include -#include - /* Code sharing between pci-quirks and xhci hcd */ #include "xhci-ext-caps.h" #include "pci-quirks.h" @@ -1614,6 +1612,14 @@ static inline struct usb_hcd *xhci_to_hcd(struct xhci_hcd *xhci) * xHCI implementations that do not support 64-bit address pointers will ignore * the high dword, and write order is irrelevant. */ +static inline u64 xhci_read_64(const struct xhci_hcd *xhci, + __le64 __iomem *regs) +{ + __u32 __iomem *ptr = (__u32 __iomem *) regs; + u64 val_lo = readl(ptr); + u64 val_hi = readl(ptr + 1); + return val_lo + (val_hi << 32); +} static inline void xhci_write_64(struct xhci_hcd *xhci, const u64 val, __le64 __iomem *regs) { From f8d56d8f892be43a2404356073e16401eb5a42e6 Mon Sep 17 00:00:00 2001 From: Stefan Roese Date: Wed, 29 Jan 2014 11:32:37 +0100 Subject: [PATCH 071/596] net: eth: cpsw: Correctly attach to GPIO bitbang MDIO driver When the GPIO bitbang MDIO driver is used instead of the Davinci MDIO driver we need to configure the phy_id string differently. Otherwise this string looks like this "gpio.6" instead of "gpio-0" and the PHY is not found when phy_connect() is called. Signed-off-by: Stefan Roese Cc: Lukas Stockmann Cc: Mugunthan V N Acked-by: Mugunthan V N Signed-off-by: David S. Miller --- drivers/net/ethernet/ti/cpsw.c | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c index bde63e3af96f..1d860ce914ed 100644 --- a/drivers/net/ethernet/ti/cpsw.c +++ b/drivers/net/ethernet/ti/cpsw.c @@ -1878,8 +1878,18 @@ static int cpsw_probe_dt(struct cpsw_platform_data *data, mdio_node = of_find_node_by_phandle(be32_to_cpup(parp)); phyid = be32_to_cpup(parp+1); mdio = of_find_device_by_node(mdio_node); - snprintf(slave_data->phy_id, sizeof(slave_data->phy_id), - PHY_ID_FMT, mdio->name, phyid); + + if (strncmp(mdio->name, "gpio", 4) == 0) { + /* GPIO bitbang MDIO driver attached */ + struct mii_bus *bus = dev_get_drvdata(&mdio->dev); + + snprintf(slave_data->phy_id, sizeof(slave_data->phy_id), + PHY_ID_FMT, bus->id, phyid); + } else { + /* davinci MDIO driver attached */ + snprintf(slave_data->phy_id, sizeof(slave_data->phy_id), + PHY_ID_FMT, mdio->name, phyid); + } mac_addr = of_get_mac_address(slave_node); if (mac_addr) From b5aaab12b2b4bc4acab7384c17a87f3406e5047d Mon Sep 17 00:00:00 2001 From: Or Gerlitz Date: Wed, 29 Jan 2014 18:08:59 +0200 Subject: [PATCH 072/596] net/ipv4: Use non-atomic allocation of udp offloads structure instance Since udp_add_offload() can be called from non-sleepable context e.g under this call tree from the vxlan driver use case: vxlan_socket_create() <-- holds the spinlock -> vxlan_notify_add_rx_port() -> udp_add_offload() <-- schedules we should allocate the udp_offloads structure in atomic manner. Fixes: b582ef0 ('net: Add GRO support for UDP encapsulating protocols') Reported-by: Dan Carpenter Signed-off-by: Or Gerlitz Signed-off-by: David S. Miller --- net/ipv4/udp_offload.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/ipv4/udp_offload.c b/net/ipv4/udp_offload.c index 25f5cee3a08a..2ffea6f31efc 100644 --- a/net/ipv4/udp_offload.c +++ b/net/ipv4/udp_offload.c @@ -101,7 +101,7 @@ static struct sk_buff *udp4_ufo_fragment(struct sk_buff *skb, int udp_add_offload(struct udp_offload *uo) { struct udp_offload_priv __rcu **head = &udp_offload_base; - struct udp_offload_priv *new_offload = kzalloc(sizeof(*new_offload), GFP_KERNEL); + struct udp_offload_priv *new_offload = kzalloc(sizeof(*new_offload), GFP_ATOMIC); if (!new_offload) return -ENOMEM; From 920a0fde5a32fd40b4d3c94ad72f7fc7039db8e3 Mon Sep 17 00:00:00 2001 From: Or Gerlitz Date: Wed, 29 Jan 2014 18:10:10 +0200 Subject: [PATCH 073/596] net/vxlan: Go over all candidate streams for GRO matching The loop in vxlan_gro_receive() over the current set of candidates for coalescing was wrongly aborted once a match was found. In rare cases, this can cause a false-positives matching in the next layer GRO checks. Signed-off-by: Or Gerlitz Signed-off-by: David S. Miller --- drivers/net/vxlan.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c index 026a313c2d2d..3d6cd1c52e95 100644 --- a/drivers/net/vxlan.c +++ b/drivers/net/vxlan.c @@ -596,10 +596,8 @@ static struct sk_buff **vxlan_gro_receive(struct sk_buff **head, struct sk_buff NAPI_GRO_CB(p)->same_flow = 0; continue; } - goto found; } -found: type = eh->h_proto; rcu_read_lock(); From 0ae89beb283a0db5980d1d4781c7d7be2f2810d6 Mon Sep 17 00:00:00 2001 From: Oliver Hartkopp Date: Thu, 30 Jan 2014 10:11:28 +0100 Subject: [PATCH 074/596] can: add destructor for self generated skbs Self generated skbuffs in net/can/bcm.c are setting a skb->sk reference but no explicit destructor which is enforced since Linux 3.11 with commit 376c7311bdb6 (net: add a temporary sanity check in skb_orphan()). This patch adds some helper functions to make sure that a destructor is properly defined when a sock reference is assigned to a CAN related skb. To create an unshared skb owned by the original sock a common helper function has been introduced to replace open coded functions to create CAN echo skbs. Signed-off-by: Oliver Hartkopp Tested-by: Andre Naujoks Reviewed-by: Eric Dumazet Signed-off-by: David S. Miller --- drivers/net/can/dev.c | 15 +++----------- drivers/net/can/janz-ican3.c | 18 ++++------------- drivers/net/can/vcan.c | 9 ++++----- include/linux/can/skb.h | 38 ++++++++++++++++++++++++++++++++++++ net/can/af_can.c | 3 ++- net/can/bcm.c | 4 ++-- 6 files changed, 53 insertions(+), 34 deletions(-) diff --git a/drivers/net/can/dev.c b/drivers/net/can/dev.c index 13a909822e25..fc59bc6f040b 100644 --- a/drivers/net/can/dev.c +++ b/drivers/net/can/dev.c @@ -323,19 +323,10 @@ void can_put_echo_skb(struct sk_buff *skb, struct net_device *dev, } if (!priv->echo_skb[idx]) { - struct sock *srcsk = skb->sk; - if (atomic_read(&skb->users) != 1) { - struct sk_buff *old_skb = skb; - - skb = skb_clone(old_skb, GFP_ATOMIC); - kfree_skb(old_skb); - if (!skb) - return; - } else - skb_orphan(skb); - - skb->sk = srcsk; + skb = can_create_echo_skb(skb); + if (!skb) + return; /* make settings for echo to reduce code in irq context */ skb->protocol = htons(ETH_P_CAN); diff --git a/drivers/net/can/janz-ican3.c b/drivers/net/can/janz-ican3.c index e24e6690d672..2124c6790687 100644 --- a/drivers/net/can/janz-ican3.c +++ b/drivers/net/can/janz-ican3.c @@ -18,6 +18,7 @@ #include #include #include +#include #include #include @@ -1133,20 +1134,9 @@ static void ican3_handle_message(struct ican3_dev *mod, struct ican3_msg *msg) */ static void ican3_put_echo_skb(struct ican3_dev *mod, struct sk_buff *skb) { - struct sock *srcsk = skb->sk; - - if (atomic_read(&skb->users) != 1) { - struct sk_buff *old_skb = skb; - - skb = skb_clone(old_skb, GFP_ATOMIC); - kfree_skb(old_skb); - if (!skb) - return; - } else { - skb_orphan(skb); - } - - skb->sk = srcsk; + skb = can_create_echo_skb(skb); + if (!skb) + return; /* save this skb for tx interrupt echo handling */ skb_queue_tail(&mod->echoq, skb); diff --git a/drivers/net/can/vcan.c b/drivers/net/can/vcan.c index 0a2a5ee79a17..4e94057ef5cf 100644 --- a/drivers/net/can/vcan.c +++ b/drivers/net/can/vcan.c @@ -46,6 +46,7 @@ #include #include #include +#include #include #include @@ -109,25 +110,23 @@ static netdev_tx_t vcan_tx(struct sk_buff *skb, struct net_device *dev) stats->rx_packets++; stats->rx_bytes += cfd->len; } - kfree_skb(skb); + consume_skb(skb); return NETDEV_TX_OK; } /* perform standard echo handling for CAN network interfaces */ if (loop) { - struct sock *srcsk = skb->sk; - skb = skb_share_check(skb, GFP_ATOMIC); + skb = can_create_echo_skb(skb); if (!skb) return NETDEV_TX_OK; /* receive with packet counting */ - skb->sk = srcsk; vcan_rx(skb, dev); } else { /* no looped packets => no counting */ - kfree_skb(skb); + consume_skb(skb); } return NETDEV_TX_OK; } diff --git a/include/linux/can/skb.h b/include/linux/can/skb.h index 2f0543f7510c..f9bbbb472663 100644 --- a/include/linux/can/skb.h +++ b/include/linux/can/skb.h @@ -11,7 +11,9 @@ #define CAN_SKB_H #include +#include #include +#include /* * The struct can_skb_priv is used to transport additional information along @@ -42,4 +44,40 @@ static inline void can_skb_reserve(struct sk_buff *skb) skb_reserve(skb, sizeof(struct can_skb_priv)); } +static inline void can_skb_destructor(struct sk_buff *skb) +{ + sock_put(skb->sk); +} + +static inline void can_skb_set_owner(struct sk_buff *skb, struct sock *sk) +{ + if (sk) { + sock_hold(sk); + skb->destructor = can_skb_destructor; + skb->sk = sk; + } +} + +/* + * returns an unshared skb owned by the original sock to be echo'ed back + */ +static inline struct sk_buff *can_create_echo_skb(struct sk_buff *skb) +{ + if (skb_shared(skb)) { + struct sk_buff *nskb = skb_clone(skb, GFP_ATOMIC); + + if (likely(nskb)) { + can_skb_set_owner(nskb, skb->sk); + consume_skb(skb); + return nskb; + } else { + kfree_skb(skb); + return NULL; + } + } + + /* we can assume to have an unshared skb with proper owner */ + return skb; +} + #endif /* CAN_SKB_H */ diff --git a/net/can/af_can.c b/net/can/af_can.c index d249874a366d..a27f8aad9e99 100644 --- a/net/can/af_can.c +++ b/net/can/af_can.c @@ -57,6 +57,7 @@ #include #include #include +#include #include #include #include @@ -290,7 +291,7 @@ int can_send(struct sk_buff *skb, int loop) return -ENOMEM; } - newskb->sk = skb->sk; + can_skb_set_owner(newskb, skb->sk); newskb->ip_summed = CHECKSUM_UNNECESSARY; newskb->pkt_type = PACKET_BROADCAST; } diff --git a/net/can/bcm.c b/net/can/bcm.c index 3fc737b214c7..dcb75c0e66c1 100644 --- a/net/can/bcm.c +++ b/net/can/bcm.c @@ -268,7 +268,7 @@ static void bcm_can_tx(struct bcm_op *op) /* send with loopback */ skb->dev = dev; - skb->sk = op->sk; + can_skb_set_owner(skb, op->sk); can_send(skb, 1); /* update statistics */ @@ -1223,7 +1223,7 @@ static int bcm_tx_send(struct msghdr *msg, int ifindex, struct sock *sk) can_skb_prv(skb)->ifindex = dev->ifindex; skb->dev = dev; - skb->sk = sk; + can_skb_set_owner(skb, sk); err = can_send(skb, 1); /* send with loopback */ dev_put(dev); From 39424e89d64661faa0a2e00c5ad1e6dbeebfa972 Mon Sep 17 00:00:00 2001 From: Prarit Bhargava Date: Tue, 28 Jan 2014 08:22:11 -0500 Subject: [PATCH 075/596] x86, cpu hotplug: Fix stack frame warning in check_irq_vectors_for_cpu_disable() Further discussion here: http://marc.info/?l=linux-kernel&m=139073901101034&w=2 kbuild, 0day kernel build service, outputs the warning: arch/x86/kernel/irq.c:333:1: warning: the frame size of 2056 bytes is larger than 2048 bytes [-Wframe-larger-than=] because check_irq_vectors_for_cpu_disable() allocates two cpumasks on the stack. Fix this by moving the two cpumasks to a global file context. Reported-by: Fengguang Wu Tested-by: David Rientjes Signed-off-by: Prarit Bhargava Link: http://lkml.kernel.org/r/1390915331-27375-1-git-send-email-prarit@redhat.com Cc: Andi Kleen Cc: Michel Lespinasse Cc: Seiji Aguchi Cc: Yang Zhang Cc: Paul Gortmaker Cc: Janet Morgan Cc: Tony Luck Cc: Ruiv Wang Cc: Gong Chen Cc: Yinghai Lu Signed-off-by: H. Peter Anvin --- arch/x86/kernel/irq.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/arch/x86/kernel/irq.c b/arch/x86/kernel/irq.c index dbb60878b744..d99f31d9a750 100644 --- a/arch/x86/kernel/irq.c +++ b/arch/x86/kernel/irq.c @@ -266,6 +266,14 @@ __visible void smp_trace_x86_platform_ipi(struct pt_regs *regs) EXPORT_SYMBOL_GPL(vector_used_by_percpu_irq); #ifdef CONFIG_HOTPLUG_CPU + +/* These two declarations are only used in check_irq_vectors_for_cpu_disable() + * below, which is protected by stop_machine(). Putting them on the stack + * results in a stack frame overflow. Dynamically allocating could result in a + * failure so declare these two cpumasks as global. + */ +static struct cpumask affinity_new, online_new; + /* * This cpu is going to be removed and its vectors migrated to the remaining * online cpus. Check to see if there are enough vectors in the remaining cpus. @@ -277,7 +285,6 @@ int check_irq_vectors_for_cpu_disable(void) unsigned int this_cpu, vector, this_count, count; struct irq_desc *desc; struct irq_data *data; - struct cpumask affinity_new, online_new; this_cpu = smp_processor_id(); cpumask_copy(&online_new, cpu_online_mask); From 8a85ac4bd8052e53f15fc0d646cc2b2b64438016 Mon Sep 17 00:00:00 2001 From: Venkat Venkatsubra Date: Thu, 30 Jan 2014 13:28:47 -0800 Subject: [PATCH 076/596] rds: The maintainer needs to be updated. Signed-off-by: Venkat Venkatsubra Signed-off-by: David S. Miller --- MAINTAINERS | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index a31a6e3e199f..fce05c4dc4e7 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -7175,7 +7175,7 @@ S: Maintained F: drivers/net/ethernet/rdc/r6040.c RDS - RELIABLE DATAGRAM SOCKETS -M: Venkat Venkatsubra +M: Chien Yen L: rds-devel@oss.oracle.com (moderated for non-subscribers) S: Supported F: net/rds/ From 2b6e0ca175fe4a20f21ba82b1e7ccc71029c4dd4 Mon Sep 17 00:00:00 2001 From: Michele Baldessari Date: Thu, 30 Jan 2014 10:51:04 +0000 Subject: [PATCH 077/596] e100: Fix "disabling already-disabled device" warning In https://bugzilla.redhat.com/show_bug.cgi?id=994438 and https://bugzilla.redhat.com/show_bug.cgi?id=970480 we received different reports of e100 throwing the following warning: [] ? pci_disable_device+0x85/0x90 [] warn_slowpath_fmt+0x33/0x40 [] pci_disable_device+0x85/0x90 [] __e100_shutdown+0x80/0x120 [e100] [] ? check_preempt_curr+0x65/0x90 [] e100_suspend+0x16/0x30 [e100] [] pci_legacy_suspend+0x2b/0xb0 [] ? wait_for_completion+0x1f/0xd0 [] ? pci_pm_poweroff+0xb0/0xb0 [] pci_pm_freeze+0x94/0xa0 [] dpm_run_callback+0x37/0x80 [] ? pm_wakeup_pending+0xc4/0x140 [] __device_suspend+0xb2/0x1f0 [] async_suspend+0x1f/0x90 [] async_run_entry_fn+0x35/0x140 [] ? wake_up_process+0x1f/0x40 [] process_one_work+0x115/0x370 [] ? start_worker+0x25/0x30 [] ? manage_workers.isra.27+0x1a5/0x250 [] worker_thread+0xfe/0x330 [] ? manage_workers.isra.27+0x250/0x250 [] kthread+0x94/0xa0 [] ret_from_kernel_thread+0x1b/0x28 [] ? insert_kthread_work+0x30/0x30 This patch removes pci_disable_device() from __e100_shutdown(). pci_clear_master() is enough. Signed-off-by: Michele Baldessari Tested-by: Mark Harig Signed-off-by: Aaron Brown Signed-off-by: David S. Miller --- drivers/net/ethernet/intel/e100.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/intel/e100.c b/drivers/net/ethernet/intel/e100.c index cbaba4442d4b..bf7a01ef9a57 100644 --- a/drivers/net/ethernet/intel/e100.c +++ b/drivers/net/ethernet/intel/e100.c @@ -3034,7 +3034,7 @@ static void __e100_shutdown(struct pci_dev *pdev, bool *enable_wake) *enable_wake = false; } - pci_disable_device(pdev); + pci_clear_master(pdev); } static int __e100_power_off(struct pci_dev *pdev, bool wake) From 54820f5802295993b78a373e404e7561039b399d Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Thu, 30 Jan 2014 14:51:19 +0100 Subject: [PATCH 078/596] regulator: s2mps11: Fix NULL pointer of_node value when using platform data When platform_data is used for regulator (of_node of sec-core MFD device is NULL) the config.of_node for regulator is not initialized. This NULL value of config.of_node is later stored during regulator_register(). Thus any call by regulator consumers to of_get_regulator() will fail on of_parse_phandle() returning NULL. In this case (using platform_data and parent's driver of_node is NULL) set the config.of_node to reg_node from platform_data. Signed-off-by: Krzysztof Kozlowski Signed-off-by: Mark Brown --- drivers/regulator/s2mps11.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/regulator/s2mps11.c b/drivers/regulator/s2mps11.c index d9e557990577..cd0b9e35a56d 100644 --- a/drivers/regulator/s2mps11.c +++ b/drivers/regulator/s2mps11.c @@ -441,6 +441,7 @@ static int s2mps11_pmic_probe(struct platform_device *pdev) for (i = 0; i < S2MPS11_REGULATOR_MAX; i++) { if (!reg_np) { config.init_data = pdata->regulators[i].initdata; + config.of_node = pdata->regulators[i].reg_node; } else { config.init_data = rdata[i].init_data; config.of_node = rdata[i].of_node; From 6a02652df511029127406cf8fa89cdf5e987f963 Mon Sep 17 00:00:00 2001 From: Francesco Fusco Date: Mon, 27 Jan 2014 14:39:13 +0100 Subject: [PATCH 079/596] perf tools: Fix include for non x86 architectures Commit 71ae8aac ("lib: introduce arch optimized hash library") added an include to for setting up an architecture specific fast hash. Since perf includes directly the non-uapi kernel header, it cannot find on non-x86 and thus prevents perf to be compiled on every architecture other than x86. The problem is the inclusion of in hash.h that results in the following error originating from util/evlist.c: fatal error: asm/hash.h: No such file or directory This commit simply adds an empty stub/file to fix the compile issue on non-x86 architectures. As perf does not use any of these new functions, it fixes the compilation and therefore seems to be the most appropriate solution to go with. Signed-off-by: Francesco Fusco Link: http://lkml.kernel.org/r/2cf8143aad65a6aa6fe30325ef8a65847141afa2.1390829373.git.ffusco@redhat.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/include/asm/hash.h | 6 ++++++ 1 file changed, 6 insertions(+) create mode 100644 tools/perf/util/include/asm/hash.h diff --git a/tools/perf/util/include/asm/hash.h b/tools/perf/util/include/asm/hash.h new file mode 100644 index 000000000000..d82b170bb216 --- /dev/null +++ b/tools/perf/util/include/asm/hash.h @@ -0,0 +1,6 @@ +#ifndef __ASM_GENERIC_HASH_H +#define __ASM_GENERIC_HASH_H + +/* Stub */ + +#endif /* __ASM_GENERIC_HASH_H */ From 9176753d1ed56951a6ee2a0f0a3f367904e35567 Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Wed, 29 Jan 2014 16:14:36 +0200 Subject: [PATCH 080/596] perf symbols: Fix symbol annotation for relocated kernel Kernel maps map memory addresses to file offsets. For symbol annotation, objdump needs the object VMA addresses. For an unrelocated kernel, that is the same as the memory address. The addresses passed to objdump for symbol annotation did not take into account kernel relocation. This patch fixes that. Reported-by: Linus Torvalds Signed-off-by: Adrian Hunter Tested-by: Jiri Olsa Cc: David Ahern Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Jiri Olsa Cc: Mike Galbraith Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/r/1391004884-10334-2-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/map.c | 5 +++-- tools/perf/util/map.h | 1 + tools/perf/util/symbol-elf.c | 2 ++ 3 files changed, 6 insertions(+), 2 deletions(-) diff --git a/tools/perf/util/map.c b/tools/perf/util/map.c index 3b97513f0e77..39cd2d0faff6 100644 --- a/tools/perf/util/map.c +++ b/tools/perf/util/map.c @@ -39,6 +39,7 @@ void map__init(struct map *map, enum map_type type, map->start = start; map->end = end; map->pgoff = pgoff; + map->reloc = 0; map->dso = dso; map->map_ip = map__map_ip; map->unmap_ip = map__unmap_ip; @@ -288,7 +289,7 @@ u64 map__rip_2objdump(struct map *map, u64 rip) if (map->dso->rel) return rip - map->pgoff; - return map->unmap_ip(map, rip); + return map->unmap_ip(map, rip) - map->reloc; } /** @@ -311,7 +312,7 @@ u64 map__objdump_2mem(struct map *map, u64 ip) if (map->dso->rel) return map->unmap_ip(map, ip + map->pgoff); - return ip; + return ip + map->reloc; } void map_groups__init(struct map_groups *mg) diff --git a/tools/perf/util/map.h b/tools/perf/util/map.h index 18068c6b71c1..257e513205ce 100644 --- a/tools/perf/util/map.h +++ b/tools/perf/util/map.h @@ -36,6 +36,7 @@ struct map { bool erange_warned; u32 priv; u64 pgoff; + u64 reloc; u32 maj, min; /* only valid for MMAP2 record */ u64 ino; /* only valid for MMAP2 record */ u64 ino_generation;/* only valid for MMAP2 record */ diff --git a/tools/perf/util/symbol-elf.c b/tools/perf/util/symbol-elf.c index 8f12f0f5101d..3e9f336740fa 100644 --- a/tools/perf/util/symbol-elf.c +++ b/tools/perf/util/symbol-elf.c @@ -751,6 +751,8 @@ int dso__load_sym(struct dso *dso, struct map *map, if (strcmp(elf_name, kmap->ref_reloc_sym->name)) continue; kmap->ref_reloc_sym->unrelocated_addr = sym.st_value; + map->reloc = kmap->ref_reloc_sym->addr - + kmap->ref_reloc_sym->unrelocated_addr; break; } } From 29b596b57426831fce92cd0ebb01c77627616fdf Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Wed, 29 Jan 2014 16:14:37 +0200 Subject: [PATCH 081/596] perf tools: Add kallsyms__get_function_start() Separate out the logic used to find the start address of the reference symbol used to track kernel relocation. kallsyms__get_function_start() is used in subsequent patches. Signed-off-by: Adrian Hunter Tested-by: Jiri Olsa Cc: David Ahern Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Jiri Olsa Cc: Mike Galbraith Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/r/1391004884-10334-3-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/event.c | 18 +++++++++++++++--- tools/perf/util/event.h | 3 +++ 2 files changed, 18 insertions(+), 3 deletions(-) diff --git a/tools/perf/util/event.c b/tools/perf/util/event.c index 1fc1c2f04772..17476df5c7b2 100644 --- a/tools/perf/util/event.c +++ b/tools/perf/util/event.c @@ -470,6 +470,17 @@ static int find_symbol_cb(void *arg, const char *name, char type, return 1; } +u64 kallsyms__get_function_start(const char *kallsyms_filename, + const char *symbol_name) +{ + struct process_symbol_args args = { .name = symbol_name, }; + + if (kallsyms__parse(kallsyms_filename, &args, find_symbol_cb) <= 0) + return 0; + + return args.start; +} + int perf_event__synthesize_kernel_mmap(struct perf_tool *tool, perf_event__handler_t process, struct machine *machine, @@ -480,13 +491,13 @@ int perf_event__synthesize_kernel_mmap(struct perf_tool *tool, char path[PATH_MAX]; char name_buff[PATH_MAX]; struct map *map; + u64 start; int err; /* * We should get this from /sys/kernel/sections/.text, but till that is * available use this, and after it is use this as a fallback for older * kernels. */ - struct process_symbol_args args = { .name = symbol_name, }; union perf_event *event = zalloc((sizeof(event->mmap) + machine->id_hdr_size)); if (event == NULL) { @@ -513,7 +524,8 @@ int perf_event__synthesize_kernel_mmap(struct perf_tool *tool, } } - if (kallsyms__parse(filename, &args, find_symbol_cb) <= 0) { + start = kallsyms__get_function_start(filename, symbol_name); + if (!start) { free(event); return -ENOENT; } @@ -525,7 +537,7 @@ int perf_event__synthesize_kernel_mmap(struct perf_tool *tool, event->mmap.header.type = PERF_RECORD_MMAP; event->mmap.header.size = (sizeof(event->mmap) - (sizeof(event->mmap.filename) - size) + machine->id_hdr_size); - event->mmap.pgoff = args.start; + event->mmap.pgoff = start; event->mmap.start = map->start; event->mmap.len = map->end - event->mmap.start; event->mmap.pid = machine->pid; diff --git a/tools/perf/util/event.h b/tools/perf/util/event.h index faf6e219be21..66a0c0392863 100644 --- a/tools/perf/util/event.h +++ b/tools/perf/util/event.h @@ -279,4 +279,7 @@ size_t perf_event__fprintf_mmap2(union perf_event *event, FILE *fp); size_t perf_event__fprintf_task(union perf_event *event, FILE *fp); size_t perf_event__fprintf(union perf_event *event, FILE *fp); +u64 kallsyms__get_function_start(const char *kallsyms_filename, + const char *symbol_name); + #endif /* __PERF_RECORD_H */ From 15a0a8706c32bd38bff9ebf7c6ef24f32d1ea921 Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Wed, 29 Jan 2014 16:14:38 +0200 Subject: [PATCH 082/596] perf machine: Add machine__get_kallsyms_filename() Separate out the logic used to make the kallsyms full path name for a machine. It will be reused in a subsequent patch. Signed-off-by: Adrian Hunter Tested-by: Jiri Olsa Cc: David Ahern Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Jiri Olsa Cc: Mike Galbraith Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/r/1391004884-10334-4-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/machine.c | 19 +++++++++++-------- 1 file changed, 11 insertions(+), 8 deletions(-) diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c index ded74590b92f..290c2e6d4001 100644 --- a/tools/perf/util/machine.c +++ b/tools/perf/util/machine.c @@ -496,19 +496,22 @@ static int symbol__in_kernel(void *arg, const char *name, return 1; } +static void machine__get_kallsyms_filename(struct machine *machine, char *buf, + size_t bufsz) +{ + if (machine__is_default_guest(machine)) + scnprintf(buf, bufsz, "%s", symbol_conf.default_guest_kallsyms); + else + scnprintf(buf, bufsz, "%s/proc/kallsyms", machine->root_dir); +} + /* Figure out the start address of kernel map from /proc/kallsyms */ static u64 machine__get_kernel_start_addr(struct machine *machine) { - const char *filename; - char path[PATH_MAX]; + char filename[PATH_MAX]; struct process_args args; - if (machine__is_default_guest(machine)) - filename = (char *)symbol_conf.default_guest_kallsyms; - else { - sprintf(path, "%s/proc/kallsyms", machine->root_dir); - filename = path; - } + machine__get_kallsyms_filename(machine, filename, PATH_MAX); if (symbol__restricted_filename(filename, "/proc/kallsyms")) return 0; From 5512cf24bed2de56f1ef44b6cc9a0a9b15499cea Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Wed, 29 Jan 2014 16:14:39 +0200 Subject: [PATCH 083/596] perf machine: Set up ref_reloc_sym in machine__create_kernel_maps() The ref_reloc_sym is always needed for the kernel map in order to check for relocation. Consequently set it up when the kernel map is created. Otherwise it was only being set up by 'perf record'. Signed-off-by: Adrian Hunter Tested-by: Jiri Olsa Cc: David Ahern Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Jiri Olsa Cc: Mike Galbraith Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/r/1391004884-10334-5-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/machine.c | 23 +++++++++++++++++++++++ tools/perf/util/machine.h | 2 ++ 2 files changed, 25 insertions(+) diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c index 290c2e6d4001..c872991e0f65 100644 --- a/tools/perf/util/machine.c +++ b/tools/perf/util/machine.c @@ -832,9 +832,25 @@ static int machine__create_modules(struct machine *machine) return 0; } +const char *ref_reloc_sym_names[] = {"_text", "_stext", NULL}; + int machine__create_kernel_maps(struct machine *machine) { struct dso *kernel = machine__get_kernel(machine); + char filename[PATH_MAX]; + const char *name; + u64 addr = 0; + int i; + + machine__get_kallsyms_filename(machine, filename, PATH_MAX); + + for (i = 0; (name = ref_reloc_sym_names[i]) != NULL; i++) { + addr = kallsyms__get_function_start(filename, name); + if (addr) + break; + } + if (!addr) + return -1; if (kernel == NULL || __machine__create_kernel_maps(machine, kernel) < 0) @@ -853,6 +869,13 @@ int machine__create_kernel_maps(struct machine *machine) * Now that we have all the maps created, just set the ->end of them: */ map_groups__fixup_end(&machine->kmaps); + + if (maps__set_kallsyms_ref_reloc_sym(machine->vmlinux_maps, name, + addr)) { + machine__destroy_kernel_maps(machine); + return -1; + } + return 0; } diff --git a/tools/perf/util/machine.h b/tools/perf/util/machine.h index 477133015440..f77e91e483dc 100644 --- a/tools/perf/util/machine.h +++ b/tools/perf/util/machine.h @@ -18,6 +18,8 @@ union perf_event; #define HOST_KERNEL_ID (-1) #define DEFAULT_GUEST_KERNEL_ID (0) +extern const char *ref_reloc_sym_names[]; + struct machine { struct rb_node rb_node; pid_t pid; From 0ae617bedde062003fd70e566e9a2601e273ea0e Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Wed, 29 Jan 2014 16:14:40 +0200 Subject: [PATCH 084/596] perf record: Get ref_reloc_sym from kernel map Now that ref_reloc_sym is set up when the kernel map is created, 'perf record' does not need to pass the symbol names to perf_event__synthesize_kernel_mmap() which can read the values needed from ref_reloc_sym directly. Signed-off-by: Adrian Hunter Tested-by: Jiri Olsa Cc: David Ahern Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Jiri Olsa Cc: Mike Galbraith Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/r/1391004884-10334-6-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/builtin-record.c | 10 ++-------- tools/perf/util/event.c | 26 ++++++-------------------- tools/perf/util/event.h | 3 +-- 3 files changed, 9 insertions(+), 30 deletions(-) diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c index 3c394bf16fa8..af47531b82ec 100644 --- a/tools/perf/builtin-record.c +++ b/tools/perf/builtin-record.c @@ -287,10 +287,7 @@ static void perf_event__synthesize_guest_os(struct machine *machine, void *data) * have no _text sometimes. */ err = perf_event__synthesize_kernel_mmap(tool, process_synthesized_event, - machine, "_text"); - if (err < 0) - err = perf_event__synthesize_kernel_mmap(tool, process_synthesized_event, - machine, "_stext"); + machine); if (err < 0) pr_err("Couldn't record guest kernel [%d]'s reference" " relocation symbol.\n", machine->pid); @@ -457,10 +454,7 @@ static int __cmd_record(struct record *rec, int argc, const char **argv) } err = perf_event__synthesize_kernel_mmap(tool, process_synthesized_event, - machine, "_text"); - if (err < 0) - err = perf_event__synthesize_kernel_mmap(tool, process_synthesized_event, - machine, "_stext"); + machine); if (err < 0) pr_err("Couldn't record kernel reference relocation symbol\n" "Symbol resolution may be skewed if relocation was used (e.g. kexec).\n" diff --git a/tools/perf/util/event.c b/tools/perf/util/event.c index 17476df5c7b2..b0f3ca850e9e 100644 --- a/tools/perf/util/event.c +++ b/tools/perf/util/event.c @@ -483,15 +483,13 @@ u64 kallsyms__get_function_start(const char *kallsyms_filename, int perf_event__synthesize_kernel_mmap(struct perf_tool *tool, perf_event__handler_t process, - struct machine *machine, - const char *symbol_name) + struct machine *machine) { size_t size; - const char *filename, *mmap_name; - char path[PATH_MAX]; + const char *mmap_name; char name_buff[PATH_MAX]; struct map *map; - u64 start; + struct kmap *kmap; int err; /* * We should get this from /sys/kernel/sections/.text, but till that is @@ -513,31 +511,19 @@ int perf_event__synthesize_kernel_mmap(struct perf_tool *tool, * see kernel/perf_event.c __perf_event_mmap */ event->header.misc = PERF_RECORD_MISC_KERNEL; - filename = "/proc/kallsyms"; } else { event->header.misc = PERF_RECORD_MISC_GUEST_KERNEL; - if (machine__is_default_guest(machine)) - filename = (char *) symbol_conf.default_guest_kallsyms; - else { - sprintf(path, "%s/proc/kallsyms", machine->root_dir); - filename = path; - } - } - - start = kallsyms__get_function_start(filename, symbol_name); - if (!start) { - free(event); - return -ENOENT; } map = machine->vmlinux_maps[MAP__FUNCTION]; + kmap = map__kmap(map); size = snprintf(event->mmap.filename, sizeof(event->mmap.filename), - "%s%s", mmap_name, symbol_name) + 1; + "%s%s", mmap_name, kmap->ref_reloc_sym->name) + 1; size = PERF_ALIGN(size, sizeof(u64)); event->mmap.header.type = PERF_RECORD_MMAP; event->mmap.header.size = (sizeof(event->mmap) - (sizeof(event->mmap.filename) - size) + machine->id_hdr_size); - event->mmap.pgoff = start; + event->mmap.pgoff = kmap->ref_reloc_sym->addr; event->mmap.start = map->start; event->mmap.len = map->end - event->mmap.start; event->mmap.pid = machine->pid; diff --git a/tools/perf/util/event.h b/tools/perf/util/event.h index 66a0c0392863..851fa06f4a42 100644 --- a/tools/perf/util/event.h +++ b/tools/perf/util/event.h @@ -214,8 +214,7 @@ int perf_event__synthesize_threads(struct perf_tool *tool, struct machine *machine, bool mmap_data); int perf_event__synthesize_kernel_mmap(struct perf_tool *tool, perf_event__handler_t process, - struct machine *machine, - const char *symbol_name); + struct machine *machine); int perf_event__synthesize_modules(struct perf_tool *tool, perf_event__handler_t process, From a00d28cb72d3629c6481fe21ba6c6b4f96caed49 Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Wed, 29 Jan 2014 16:14:41 +0200 Subject: [PATCH 085/596] perf symbols: Prevent the use of kcore if the kernel has moved Use of kcore is predicated upon it matching the recorded data. If the kernel has been relocated at boot time (i.e. since the data was recorded) then do not use kcore. Note that it is possible to make a copy of kcore at the time the data is recorded using 'perf buildid-cache'. Then the perf tools will use the copy because it does match the data. Signed-off-by: Adrian Hunter Tested-by: Jiri Olsa Cc: David Ahern Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Jiri Olsa Cc: Mike Galbraith Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/r/1391004884-10334-7-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/symbol.c | 25 +++++++++++++++++++++---- 1 file changed, 21 insertions(+), 4 deletions(-) diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c index 39ce9adbaaf0..4ac1f871ec27 100644 --- a/tools/perf/util/symbol.c +++ b/tools/perf/util/symbol.c @@ -976,6 +976,23 @@ static int validate_kcore_modules(const char *kallsyms_filename, return 0; } +static int validate_kcore_addresses(const char *kallsyms_filename, + struct map *map) +{ + struct kmap *kmap = map__kmap(map); + + if (kmap->ref_reloc_sym && kmap->ref_reloc_sym->name) { + u64 start; + + start = kallsyms__get_function_start(kallsyms_filename, + kmap->ref_reloc_sym->name); + if (start != kmap->ref_reloc_sym->addr) + return -EINVAL; + } + + return validate_kcore_modules(kallsyms_filename, map); +} + struct kcore_mapfn_data { struct dso *dso; enum map_type type; @@ -1019,8 +1036,8 @@ static int dso__load_kcore(struct dso *dso, struct map *map, kallsyms_filename)) return -EINVAL; - /* All modules must be present at their original addresses */ - if (validate_kcore_modules(kallsyms_filename, map)) + /* Modules and kernel must be present at their original addresses */ + if (validate_kcore_addresses(kallsyms_filename, map)) return -EINVAL; md.dso = dso; @@ -1424,7 +1441,7 @@ static int find_matching_kcore(struct map *map, char *dir, size_t dir_sz) continue; scnprintf(kallsyms_filename, sizeof(kallsyms_filename), "%s/%s/kallsyms", dir, dent->d_name); - if (!validate_kcore_modules(kallsyms_filename, map)) { + if (!validate_kcore_addresses(kallsyms_filename, map)) { strlcpy(dir, kallsyms_filename, dir_sz); ret = 0; break; @@ -1479,7 +1496,7 @@ static char *dso__find_kallsyms(struct dso *dso, struct map *map) if (fd != -1) { close(fd); /* If module maps match go with /proc/kallsyms */ - if (!validate_kcore_modules("/proc/kallsyms", map)) + if (!validate_kcore_addresses("/proc/kallsyms", map)) goto proc_kallsyms; } From c080f72753def150993144d755379941f8b14683 Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Wed, 29 Jan 2014 16:14:42 +0200 Subject: [PATCH 086/596] perf tests: No need to set up ref_reloc_sym Now that ref_reloc_sym is set up by machine__create_kernel_maps(), the "vmlinux symtab matches kallsyms" test does have to do it. Signed-off-by: Adrian Hunter Tested-by: Jiri Olsa Cc: David Ahern Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Jiri Olsa Cc: Mike Galbraith Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/r/1391004884-10334-8-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/tests/vmlinux-kallsyms.c | 10 ---------- 1 file changed, 10 deletions(-) diff --git a/tools/perf/tests/vmlinux-kallsyms.c b/tools/perf/tests/vmlinux-kallsyms.c index 2bd13edcbc17..3d9088003a5b 100644 --- a/tools/perf/tests/vmlinux-kallsyms.c +++ b/tools/perf/tests/vmlinux-kallsyms.c @@ -26,7 +26,6 @@ int test__vmlinux_matches_kallsyms(void) struct map *kallsyms_map, *vmlinux_map; struct machine kallsyms, vmlinux; enum map_type type = MAP__FUNCTION; - struct ref_reloc_sym ref_reloc_sym = { .name = "_stext", }; u64 mem_start, mem_end; /* @@ -70,14 +69,6 @@ int test__vmlinux_matches_kallsyms(void) */ kallsyms_map = machine__kernel_map(&kallsyms, type); - sym = map__find_symbol_by_name(kallsyms_map, ref_reloc_sym.name, NULL); - if (sym == NULL) { - pr_debug("dso__find_symbol_by_name "); - goto out; - } - - ref_reloc_sym.addr = UM(sym->start); - /* * Step 5: * @@ -89,7 +80,6 @@ int test__vmlinux_matches_kallsyms(void) } vmlinux_map = machine__kernel_map(&vmlinux, type); - map__kmap(vmlinux_map)->ref_reloc_sym = &ref_reloc_sym; /* * Step 6: From d9b62aba87a82939c73f451a166c7a21342350d6 Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Wed, 29 Jan 2014 16:14:43 +0200 Subject: [PATCH 087/596] perf tools: Adjust kallsyms for relocated kernel If the kernel is relocated at boot time, kallsyms will not match data recorded previously. That does not matter for modules because they are corrected anyway. It also does not matter if vmlinux is being used for symbols. But if perf tools has only kallsyms then the symbols will not match. Fix by applying the delta gained by comparing the old and current addresses of the relocation reference symbol. Signed-off-by: Adrian Hunter Tested-by: Jiri Olsa Cc: David Ahern Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Jiri Olsa Cc: Mike Galbraith Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/r/1391004884-10334-9-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/symbol.c | 40 ++++++++++++++++++++++++++++++++++++++-- 1 file changed, 38 insertions(+), 2 deletions(-) diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c index 4ac1f871ec27..a9d758a3b371 100644 --- a/tools/perf/util/symbol.c +++ b/tools/perf/util/symbol.c @@ -627,7 +627,7 @@ static int dso__split_kallsyms_for_kcore(struct dso *dso, struct map *map, * kernel range is broken in several maps, named [kernel].N, as we don't have * the original ELF section names vmlinux have. */ -static int dso__split_kallsyms(struct dso *dso, struct map *map, +static int dso__split_kallsyms(struct dso *dso, struct map *map, u64 delta, symbol_filter_t filter) { struct map_groups *kmaps = map__kmap(map)->kmaps; @@ -692,6 +692,12 @@ static int dso__split_kallsyms(struct dso *dso, struct map *map, char dso_name[PATH_MAX]; struct dso *ndso; + if (delta) { + /* Kernel was relocated at boot time */ + pos->start -= delta; + pos->end -= delta; + } + if (count == 0) { curr_map = map; goto filter_symbol; @@ -721,6 +727,10 @@ static int dso__split_kallsyms(struct dso *dso, struct map *map, curr_map->map_ip = curr_map->unmap_ip = identity__map_ip; map_groups__insert(kmaps, curr_map); ++kernel_range; + } else if (delta) { + /* Kernel was relocated at boot time */ + pos->start -= delta; + pos->end -= delta; } filter_symbol: if (filter && filter(curr_map, pos)) { @@ -1130,15 +1140,41 @@ static int dso__load_kcore(struct dso *dso, struct map *map, return -EINVAL; } +/* + * If the kernel is relocated at boot time, kallsyms won't match. Compute the + * delta based on the relocation reference symbol. + */ +static int kallsyms__delta(struct map *map, const char *filename, u64 *delta) +{ + struct kmap *kmap = map__kmap(map); + u64 addr; + + if (!kmap->ref_reloc_sym || !kmap->ref_reloc_sym->name) + return 0; + + addr = kallsyms__get_function_start(filename, + kmap->ref_reloc_sym->name); + if (!addr) + return -1; + + *delta = addr - kmap->ref_reloc_sym->addr; + return 0; +} + int dso__load_kallsyms(struct dso *dso, const char *filename, struct map *map, symbol_filter_t filter) { + u64 delta = 0; + if (symbol__restricted_filename(filename, "/proc/kallsyms")) return -1; if (dso__load_all_kallsyms(dso, filename, map) < 0) return -1; + if (kallsyms__delta(map, filename, &delta)) + return -1; + symbols__fixup_duplicate(&dso->symbols[map->type]); symbols__fixup_end(&dso->symbols[map->type]); @@ -1150,7 +1186,7 @@ int dso__load_kallsyms(struct dso *dso, const char *filename, if (!dso__load_kcore(dso, map, filename)) return dso__split_kallsyms_for_kcore(dso, map, filter); else - return dso__split_kallsyms(dso, map, filter); + return dso__split_kallsyms(dso, map, delta, filter); } static int dso__load_perf_map(struct dso *dso, struct map *map, From d3b70220292c40d3b499797fd2f33f608fc35edb Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Wed, 29 Jan 2014 16:14:44 +0200 Subject: [PATCH 088/596] perf buildid-cache: Check relocation when checking for existing kcore perf buildid-cache does not make another copy of kcore if the buildid and modules match an existing copy. That does not take into account the possibility that the kernel has been relocated. Extend the check to check if the reference relocation symbol matches too, otherwise do make a copy. Signed-off-by: Adrian Hunter Tested-by: Jiri Olsa Cc: David Ahern Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Jiri Olsa Cc: Mike Galbraith Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/r/1391004884-10334-10-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/builtin-buildid-cache.c | 33 ++++++++++++++++++++++++++---- 1 file changed, 29 insertions(+), 4 deletions(-) diff --git a/tools/perf/builtin-buildid-cache.c b/tools/perf/builtin-buildid-cache.c index cfede86161d8..b22dbb16f877 100644 --- a/tools/perf/builtin-buildid-cache.c +++ b/tools/perf/builtin-buildid-cache.c @@ -63,11 +63,35 @@ static int build_id_cache__kcore_dir(char *dir, size_t sz) return 0; } +static bool same_kallsyms_reloc(const char *from_dir, char *to_dir) +{ + char from[PATH_MAX]; + char to[PATH_MAX]; + const char *name; + u64 addr1 = 0, addr2 = 0; + int i; + + scnprintf(from, sizeof(from), "%s/kallsyms", from_dir); + scnprintf(to, sizeof(to), "%s/kallsyms", to_dir); + + for (i = 0; (name = ref_reloc_sym_names[i]) != NULL; i++) { + addr1 = kallsyms__get_function_start(from, name); + if (addr1) + break; + } + + if (name) + addr2 = kallsyms__get_function_start(to, name); + + return addr1 == addr2; +} + static int build_id_cache__kcore_existing(const char *from_dir, char *to_dir, size_t to_dir_sz) { char from[PATH_MAX]; char to[PATH_MAX]; + char to_subdir[PATH_MAX]; struct dirent *dent; int ret = -1; DIR *d; @@ -86,10 +110,11 @@ static int build_id_cache__kcore_existing(const char *from_dir, char *to_dir, continue; scnprintf(to, sizeof(to), "%s/%s/modules", to_dir, dent->d_name); - if (!compare_proc_modules(from, to)) { - scnprintf(to, sizeof(to), "%s/%s", to_dir, - dent->d_name); - strlcpy(to_dir, to, to_dir_sz); + scnprintf(to_subdir, sizeof(to_subdir), "%s/%s", + to_dir, dent->d_name); + if (!compare_proc_modules(from, to) && + same_kallsyms_reloc(from_dir, to_subdir)) { + strlcpy(to_dir, to_subdir, to_dir_sz); ret = 0; break; } From 718360c59f34b80d9878429300c1c688f7c2031d Mon Sep 17 00:00:00 2001 From: Noah Massey Date: Thu, 30 Jan 2014 21:31:12 -0500 Subject: [PATCH 089/596] nfs: fix setting of ACLs on file creation. nfs3_get_acl() tries to skip posix equivalent ACLs, but misinterprets the return value of posix_acl_equiv_mode(). Fix it. This is a regression introduced by "nfs: use generic posix ACL infrastructure for v3 Posix ACLs" CC: Christoph Hellwig CC: linux-nfs@vger.kernel.org CC: linux-fsdevel@vger.kernel.org Signed-off-by: Trond Myklebust --- fs/nfs/nfs3acl.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/nfs/nfs3acl.c b/fs/nfs/nfs3acl.c index 9a5ca03fa539..0851f852568d 100644 --- a/fs/nfs/nfs3acl.c +++ b/fs/nfs/nfs3acl.c @@ -80,7 +80,7 @@ struct posix_acl *nfs3_get_acl(struct inode *inode, int type) } if (res.acl_access != NULL) { - if (posix_acl_equiv_mode(res.acl_access, NULL) || + if ((posix_acl_equiv_mode(res.acl_access, NULL) == 0) || res.acl_access->a_count == 0) { posix_acl_release(res.acl_access); res.acl_access = NULL; From ac323d8d807060f7c95a685a9fe861e7b6300993 Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Thu, 30 Jan 2014 14:32:45 +0100 Subject: [PATCH 090/596] power: max17040: Fix NULL pointer dereference when there is no platform_data Fix NULL pointer dereference of "chip->pdata" if platform_data was not supplied to the driver. The driver during probe stored the pointer to the platform_data: chip->pdata = client->dev.platform_data; Later it was dereferenced in max17040_get_online() and max17040_get_status(). If platform_data was not supplied, the NULL pointer exception would happen: [ 6.626094] Unable to handle kernel of a at virtual address 00000000 [ 6.628557] pgd = c0004000 [ 6.632868] [00000000] *pgd=66262564 [ 6.634636] Unable to handle kernel paging request at virtual address e6262000 [ 6.642014] pgd = de468000 [ 6.644700] [e6262000] *pgd=00000000 [ 6.648265] Internal error: Oops: 5 [#1] PREEMPT SMP ARM [ 6.653552] Modules linked in: [ 6.656598] CPU: 0 PID: 31 Comm: kworker/0:1 Not tainted 3.10.14-02717-gc58b4b4 #505 [ 6.664334] Workqueue: events max17040_work [ 6.668488] task: dfa11b80 ti: df9f6000 task.ti: df9f6000 [ 6.673873] PC is at show_pte+0x80/0xb8 [ 6.677687] LR is at show_pte+0x3c/0xb8 [ 6.681503] pc : [] lr : [] psr: 600f0113 [ 6.681503] sp : df9f7d58 ip : 600f0113 fp : 00000009 [ 6.692965] r10: 00000000 r9 : 00000000 r8 : dfa11b80 [ 6.698171] r7 : df9f7ea0 r6 : e6262000 r5 : 00000000 r4 : 00000000 [ 6.704680] r3 : 00000000 r2 : e6262000 r1 : 600f0193 r0 : c05b3750 [ 6.711194] Flags: nZCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment kernel [ 6.718485] Control: 10c53c7d Table: 5e46806a DAC: 00000015 [ 6.724218] Process kworker/0:1 (pid: 31, stack limit = 0xdf9f6238) [ 6.730465] Stack: (0xdf9f7d58 to 0xdf9f8000) [ 6.914325] [] (show_pte+0x80/0xb8) from [] (__do_kernel_fault.part.9+0x44/0x74) [ 6.923425] [] (__do_kernel_fault.part.9+0x44/0x74) from [] (do_page_fault+0x2c4/0x360) [ 6.933144] [] (do_page_fault+0x2c4/0x360) from [] (do_DataAbort+0x34/0x9c) [ 6.941825] [] (do_DataAbort+0x34/0x9c) from [] (__dabt_svc+0x38/0x60) [ 6.950058] Exception stack(0xdf9f7ea0 to 0xdf9f7ee8) [ 6.955099] 7ea0: df0c1790 00000000 00000002 00000000 df0c1794 df0c1790 df0c1790 00000042 [ 6.963271] 7ec0: df0c1794 00000001 00000000 00000009 00000000 df9f7ee8 c0306268 c0306270 [ 6.971419] 7ee0: a00f0113 ffffffff [ 6.974902] [] (__dabt_svc+0x38/0x60) from [] (max17040_work+0x8c/0x144) [ 6.983317] [] (max17040_work+0x8c/0x144) from [] (process_one_work+0x138/0x440) [ 6.992429] [] (process_one_work+0x138/0x440) from [] (worker_thread+0x134/0x3b8) [ 7.001628] [] (worker_thread+0x134/0x3b8) from [] (kthread+0xa4/0xb0) [ 7.009875] [] (kthread+0xa4/0xb0) from [] (ret_from_fork+0x14/0x2c) [ 7.017943] Code: e1a03005 e2422480 e0826104 e59f002c (e7922104) [ 7.024017] ---[ end trace 73bc7006b9cc5c79 ]--- Signed-off-by: Krzysztof Kozlowski Fixes: c6f4a42de60b981dd210de01cd3e575835e3158e Cc: --- drivers/power/max17040_battery.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/power/max17040_battery.c b/drivers/power/max17040_battery.c index c7ff6d67f158..0fbac861080d 100644 --- a/drivers/power/max17040_battery.c +++ b/drivers/power/max17040_battery.c @@ -148,7 +148,7 @@ static void max17040_get_online(struct i2c_client *client) { struct max17040_chip *chip = i2c_get_clientdata(client); - if (chip->pdata->battery_online) + if (chip->pdata && chip->pdata->battery_online) chip->online = chip->pdata->battery_online(); else chip->online = 1; @@ -158,7 +158,8 @@ static void max17040_get_status(struct i2c_client *client) { struct max17040_chip *chip = i2c_get_clientdata(client); - if (!chip->pdata->charger_online || !chip->pdata->charger_enable) { + if (!chip->pdata || !chip->pdata->charger_online + || !chip->pdata->charger_enable) { chip->status = POWER_SUPPLY_STATUS_UNKNOWN; return; } From 17ead6c85c3d0ef57a14d1373f1f1cee2ce60ea8 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sat, 1 Feb 2014 14:53:23 -0500 Subject: [PATCH 091/596] NFSv4: Fix memory corruption in nfs4_proc_open_confirm nfs41_wake_and_assign_slot() relies on the task->tk_msg.rpc_argp and task->tk_msg.rpc_resp always pointing to the session sequence arguments. nfs4_proc_open_confirm tries to pull a fast one by reusing the open sequence structure, thus causing corruption of the NFSv4 slot table. Cc: stable@vger.kernel.org # 3.12+ Signed-off-by: Trond Myklebust --- fs/nfs/nfs4proc.c | 8 ++++---- include/linux/nfs_xdr.h | 2 ++ 2 files changed, 6 insertions(+), 4 deletions(-) diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 42da6af77587..2da6a698b8f7 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -1620,15 +1620,15 @@ static void nfs4_open_confirm_prepare(struct rpc_task *task, void *calldata) { struct nfs4_opendata *data = calldata; - nfs40_setup_sequence(data->o_arg.server, &data->o_arg.seq_args, - &data->o_res.seq_res, task); + nfs40_setup_sequence(data->o_arg.server, &data->c_arg.seq_args, + &data->c_res.seq_res, task); } static void nfs4_open_confirm_done(struct rpc_task *task, void *calldata) { struct nfs4_opendata *data = calldata; - nfs40_sequence_done(task, &data->o_res.seq_res); + nfs40_sequence_done(task, &data->c_res.seq_res); data->rpc_status = task->tk_status; if (data->rpc_status == 0) { @@ -1686,7 +1686,7 @@ static int _nfs4_proc_open_confirm(struct nfs4_opendata *data) }; int status; - nfs4_init_sequence(&data->o_arg.seq_args, &data->o_res.seq_res, 1); + nfs4_init_sequence(&data->c_arg.seq_args, &data->c_res.seq_res, 1); kref_get(&data->kref); data->rpc_done = 0; data->rpc_status = 0; diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index 3ccfcecf8999..b2fb167b2e6d 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -379,12 +379,14 @@ struct nfs_openres { * Arguments to the open_confirm call. */ struct nfs_open_confirmargs { + struct nfs4_sequence_args seq_args; const struct nfs_fh * fh; nfs4_stateid * stateid; struct nfs_seqid * seqid; }; struct nfs_open_confirmres { + struct nfs4_sequence_res seq_res; nfs4_stateid stateid; struct nfs_seqid * seqid; }; From 20b9a9024540a775395d5d1f41eec0ec6ec41f9b Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sat, 1 Feb 2014 13:47:06 -0500 Subject: [PATCH 092/596] NFSv4.1: nfs4_destroy_session must call rpc_destroy_waitqueue There may still be timers active on the session waitqueues. Make sure that we kill them before freeing the memory. Cc: stable@vger.kernel.org # 3.12+ Signed-off-by: Trond Myklebust --- fs/nfs/nfs4client.c | 2 +- fs/nfs/nfs4session.c | 25 ++++++++++++++++++++----- fs/nfs/nfs4session.h | 2 +- 3 files changed, 22 insertions(+), 7 deletions(-) diff --git a/fs/nfs/nfs4client.c b/fs/nfs/nfs4client.c index dbb3e1f30c68..860ad26a5590 100644 --- a/fs/nfs/nfs4client.c +++ b/fs/nfs/nfs4client.c @@ -170,7 +170,7 @@ void nfs41_shutdown_client(struct nfs_client *clp) void nfs40_shutdown_client(struct nfs_client *clp) { if (clp->cl_slot_tbl) { - nfs4_release_slot_table(clp->cl_slot_tbl); + nfs4_shutdown_slot_table(clp->cl_slot_tbl); kfree(clp->cl_slot_tbl); } } diff --git a/fs/nfs/nfs4session.c b/fs/nfs/nfs4session.c index cf883c7ae053..e799dc3c3b1d 100644 --- a/fs/nfs/nfs4session.c +++ b/fs/nfs/nfs4session.c @@ -231,14 +231,23 @@ static int nfs4_realloc_slot_table(struct nfs4_slot_table *tbl, return ret; } +/* + * nfs4_release_slot_table - release all slot table entries + */ +static void nfs4_release_slot_table(struct nfs4_slot_table *tbl) +{ + nfs4_shrink_slot_table(tbl, 0); +} + /** - * nfs4_release_slot_table - release resources attached to a slot table + * nfs4_shutdown_slot_table - release resources attached to a slot table * @tbl: slot table to shut down * */ -void nfs4_release_slot_table(struct nfs4_slot_table *tbl) +void nfs4_shutdown_slot_table(struct nfs4_slot_table *tbl) { - nfs4_shrink_slot_table(tbl, 0); + nfs4_release_slot_table(tbl); + rpc_destroy_wait_queue(&tbl->slot_tbl_waitq); } /** @@ -422,7 +431,7 @@ void nfs41_update_target_slotid(struct nfs4_slot_table *tbl, spin_unlock(&tbl->slot_tbl_lock); } -static void nfs4_destroy_session_slot_tables(struct nfs4_session *session) +static void nfs4_release_session_slot_tables(struct nfs4_session *session) { nfs4_release_slot_table(&session->fc_slot_table); nfs4_release_slot_table(&session->bc_slot_table); @@ -450,7 +459,7 @@ int nfs4_setup_session_slot_tables(struct nfs4_session *ses) if (status && tbl->slots == NULL) /* Fore and back channel share a connection so get * both slot tables or neither */ - nfs4_destroy_session_slot_tables(ses); + nfs4_release_session_slot_tables(ses); return status; } @@ -470,6 +479,12 @@ struct nfs4_session *nfs4_alloc_session(struct nfs_client *clp) return session; } +static void nfs4_destroy_session_slot_tables(struct nfs4_session *session) +{ + nfs4_shutdown_slot_table(&session->fc_slot_table); + nfs4_shutdown_slot_table(&session->bc_slot_table); +} + void nfs4_destroy_session(struct nfs4_session *session) { struct rpc_xprt *xprt; diff --git a/fs/nfs/nfs4session.h b/fs/nfs/nfs4session.h index 232306100651..b34ada9bc6a2 100644 --- a/fs/nfs/nfs4session.h +++ b/fs/nfs/nfs4session.h @@ -74,7 +74,7 @@ enum nfs4_session_state { extern int nfs4_setup_slot_table(struct nfs4_slot_table *tbl, unsigned int max_reqs, const char *queue); -extern void nfs4_release_slot_table(struct nfs4_slot_table *tbl); +extern void nfs4_shutdown_slot_table(struct nfs4_slot_table *tbl); extern struct nfs4_slot *nfs4_alloc_slot(struct nfs4_slot_table *tbl); extern void nfs4_free_slot(struct nfs4_slot_table *tbl, struct nfs4_slot *slot); extern void nfs4_slot_tbl_drain_complete(struct nfs4_slot_table *tbl); From fba9110c47855050b38a5a19bbcd3728b5f21bfe Mon Sep 17 00:00:00 2001 From: Max Filippov Date: Fri, 31 Jan 2014 09:41:04 +0400 Subject: [PATCH 093/596] net: ethoc: implement basic ethtool operations The following methods are implemented: - get link state (standard implementation); - get timestamping info (standard implementation). Signed-off-by: Max Filippov Reviewed-by: Florian Fainelli Reviewed-by: Ben Hutchings Signed-off-by: David S. Miller --- drivers/net/ethernet/ethoc.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/net/ethernet/ethoc.c b/drivers/net/ethernet/ethoc.c index 4de8cfd149cf..0623c20131bb 100644 --- a/drivers/net/ethernet/ethoc.c +++ b/drivers/net/ethernet/ethoc.c @@ -890,6 +890,11 @@ static netdev_tx_t ethoc_start_xmit(struct sk_buff *skb, struct net_device *dev) return NETDEV_TX_OK; } +const struct ethtool_ops ethoc_ethtool_ops = { + .get_link = ethtool_op_get_link, + .get_ts_info = ethtool_op_get_ts_info, +}; + static const struct net_device_ops ethoc_netdev_ops = { .ndo_open = ethoc_open, .ndo_stop = ethoc_stop, @@ -1111,6 +1116,7 @@ static int ethoc_probe(struct platform_device *pdev) netdev->netdev_ops = ðoc_netdev_ops; netdev->watchdog_timeo = ETHOC_TIMEOUT; netdev->features |= 0; + netdev->ethtool_ops = ðoc_ethtool_ops; /* setup NAPI */ netif_napi_add(netdev, &priv->napi, ethoc_poll, 64); From 01cd7d578e01d342b1c06f0dc41c00b445415753 Mon Sep 17 00:00:00 2001 From: Max Filippov Date: Fri, 31 Jan 2014 09:41:05 +0400 Subject: [PATCH 094/596] net: ethoc: implement ethtool get/set settings Signed-off-by: Max Filippov Reviewed-by: Ben Hutchings Signed-off-by: David S. Miller --- drivers/net/ethernet/ethoc.c | 24 ++++++++++++++++++++++++ 1 file changed, 24 insertions(+) diff --git a/drivers/net/ethernet/ethoc.c b/drivers/net/ethernet/ethoc.c index 0623c20131bb..779d3c386ea7 100644 --- a/drivers/net/ethernet/ethoc.c +++ b/drivers/net/ethernet/ethoc.c @@ -890,7 +890,31 @@ static netdev_tx_t ethoc_start_xmit(struct sk_buff *skb, struct net_device *dev) return NETDEV_TX_OK; } +static int ethoc_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) +{ + struct ethoc *priv = netdev_priv(dev); + struct phy_device *phydev = priv->phy; + + if (!phydev) + return -EOPNOTSUPP; + + return phy_ethtool_gset(phydev, cmd); +} + +static int ethoc_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) +{ + struct ethoc *priv = netdev_priv(dev); + struct phy_device *phydev = priv->phy; + + if (!phydev) + return -EOPNOTSUPP; + + return phy_ethtool_sset(phydev, cmd); +} + const struct ethtool_ops ethoc_ethtool_ops = { + .get_settings = ethoc_get_settings, + .set_settings = ethoc_set_settings, .get_link = ethtool_op_get_link, .get_ts_info = ethtool_op_get_ts_info, }; From 1112909f76650e1d4c6de44b80c2d044c7c11e59 Mon Sep 17 00:00:00 2001 From: Max Filippov Date: Fri, 31 Jan 2014 09:41:06 +0400 Subject: [PATCH 095/596] net: ethoc: implement ethtool get registers Signed-off-by: Max Filippov Reviewed-by: Florian Fainelli Reviewed-by: Ben Hutchings Signed-off-by: David S. Miller --- drivers/net/ethernet/ethoc.c | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+) diff --git a/drivers/net/ethernet/ethoc.c b/drivers/net/ethernet/ethoc.c index 779d3c386ea7..5da32a71e8ac 100644 --- a/drivers/net/ethernet/ethoc.c +++ b/drivers/net/ethernet/ethoc.c @@ -51,6 +51,7 @@ MODULE_PARM_DESC(buffer_size, "DMA buffer allocation size"); #define ETH_HASH0 0x48 #define ETH_HASH1 0x4c #define ETH_TXCTRL 0x50 +#define ETH_END 0x54 /* mode register */ #define MODER_RXEN (1 << 0) /* receive enable */ @@ -912,9 +913,28 @@ static int ethoc_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) return phy_ethtool_sset(phydev, cmd); } +static int ethoc_get_regs_len(struct net_device *netdev) +{ + return ETH_END; +} + +static void ethoc_get_regs(struct net_device *dev, struct ethtool_regs *regs, + void *p) +{ + struct ethoc *priv = netdev_priv(dev); + u32 *regs_buff = p; + unsigned i; + + regs->version = 0; + for (i = 0; i < ETH_END / sizeof(u32); ++i) + regs_buff[i] = ethoc_read(priv, i * sizeof(u32)); +} + const struct ethtool_ops ethoc_ethtool_ops = { .get_settings = ethoc_get_settings, .set_settings = ethoc_set_settings, + .get_regs_len = ethoc_get_regs_len, + .get_regs = ethoc_get_regs, .get_link = ethtool_op_get_link, .get_ts_info = ethtool_op_get_ts_info, }; From bee7bacd18612564b85dc0b3e54063316c5bb32d Mon Sep 17 00:00:00 2001 From: Max Filippov Date: Fri, 31 Jan 2014 09:41:07 +0400 Subject: [PATCH 096/596] net: ethoc: implement ethtool get/set ring parameters TX and RX rings share memory and descriptors. Maximal number of descriptors reported is one less than the total available nuber of descriptors. For the set operation the requested number of TX descriptors is rounded down to the nearest power of two (driver logic requirement). Signed-off-by: Max Filippov Signed-off-by: David S. Miller --- drivers/net/ethernet/ethoc.c | 51 ++++++++++++++++++++++++++++++++++++ 1 file changed, 51 insertions(+) diff --git a/drivers/net/ethernet/ethoc.c b/drivers/net/ethernet/ethoc.c index 5da32a71e8ac..f9c1cf536298 100644 --- a/drivers/net/ethernet/ethoc.c +++ b/drivers/net/ethernet/ethoc.c @@ -180,6 +180,7 @@ MODULE_PARM_DESC(buffer_size, "DMA buffer allocation size"); * @membase: pointer to buffer memory region * @dma_alloc: dma allocated buffer size * @io_region_size: I/O memory region size + * @num_bd: number of buffer descriptors * @num_tx: number of send buffers * @cur_tx: last send buffer written * @dty_tx: last buffer actually sent @@ -200,6 +201,7 @@ struct ethoc { int dma_alloc; resource_size_t io_region_size; + unsigned int num_bd; unsigned int num_tx; unsigned int cur_tx; unsigned int dty_tx; @@ -930,12 +932,60 @@ static void ethoc_get_regs(struct net_device *dev, struct ethtool_regs *regs, regs_buff[i] = ethoc_read(priv, i * sizeof(u32)); } +static void ethoc_get_ringparam(struct net_device *dev, + struct ethtool_ringparam *ring) +{ + struct ethoc *priv = netdev_priv(dev); + + ring->rx_max_pending = priv->num_bd - 1; + ring->rx_mini_max_pending = 0; + ring->rx_jumbo_max_pending = 0; + ring->tx_max_pending = priv->num_bd - 1; + + ring->rx_pending = priv->num_rx; + ring->rx_mini_pending = 0; + ring->rx_jumbo_pending = 0; + ring->tx_pending = priv->num_tx; +} + +static int ethoc_set_ringparam(struct net_device *dev, + struct ethtool_ringparam *ring) +{ + struct ethoc *priv = netdev_priv(dev); + + if (ring->tx_pending < 1 || ring->rx_pending < 1 || + ring->tx_pending + ring->rx_pending > priv->num_bd) + return -EINVAL; + if (ring->rx_mini_pending || ring->rx_jumbo_pending) + return -EINVAL; + + if (netif_running(dev)) { + netif_tx_disable(dev); + ethoc_disable_rx_and_tx(priv); + ethoc_disable_irq(priv, INT_MASK_TX | INT_MASK_RX); + synchronize_irq(dev->irq); + } + + priv->num_tx = rounddown_pow_of_two(ring->tx_pending); + priv->num_rx = ring->rx_pending; + ethoc_init_ring(priv, dev->mem_start); + + if (netif_running(dev)) { + ethoc_enable_irq(priv, INT_MASK_TX | INT_MASK_RX); + ethoc_enable_rx_and_tx(priv); + netif_wake_queue(dev); + } + return 0; +} + const struct ethtool_ops ethoc_ethtool_ops = { .get_settings = ethoc_get_settings, .set_settings = ethoc_set_settings, .get_regs_len = ethoc_get_regs_len, .get_regs = ethoc_get_regs, .get_link = ethtool_op_get_link, + .get_ringparam = ethoc_get_ringparam, + .set_ringparam = ethoc_set_ringparam, .get_ts_info = ethtool_op_get_ts_info, }; @@ -1065,6 +1115,7 @@ static int ethoc_probe(struct platform_device *pdev) ret = -ENODEV; goto error; } + priv->num_bd = num_bd; /* num_tx must be a power of two */ priv->num_tx = rounddown_pow_of_two(num_bd >> 1); priv->num_rx = num_bd - priv->num_tx; From 4fe46b9a4d0b5eef96867e6d5134159e5a65d2a5 Mon Sep 17 00:00:00 2001 From: Daniel Baluta Date: Fri, 31 Jan 2014 09:50:12 +0200 Subject: [PATCH 097/596] vxlan: remove extra newline after function definition Signed-off-by: Daniel Baluta Signed-off-by: David S. Miller --- drivers/net/vxlan.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c index 3d6cd1c52e95..b0f705c2378f 100644 --- a/drivers/net/vxlan.c +++ b/drivers/net/vxlan.c @@ -469,7 +469,6 @@ static inline struct hlist_head *vxlan_fdb_head(struct vxlan_dev *vxlan, /* Look up Ethernet address in forwarding table */ static struct vxlan_fdb *__vxlan_find_mac(struct vxlan_dev *vxlan, const u8 *mac) - { struct hlist_head *head = vxlan_fdb_head(vxlan, mac); struct vxlan_fdb *f; From 85fc73a2cdf10cf42bc36fb3bca3896b2095a1c2 Mon Sep 17 00:00:00 2001 From: Petr Tesarik Date: Sat, 1 Feb 2014 13:30:19 +0100 Subject: [PATCH 098/596] x86: Fix the initialization of physnode_map With DISCONTIGMEM, the mapping between a pfn and its owning node is initialized using data provided by the BIOS. However, the initialization may fail if the extents are not aligned to section boundary (64M). The symptom of this bug is an early boot failure in pfn_to_page(), as it tries to access NODE_DATA(__nid) using index from an unitialized element of the physnode_map[] array. While the bug is always present, it is more likely to be hit in kdump kernels on large machines, because: 1. The memory map for a kdump kernel is specified as exactmap, and exactmap is more likely to be unaligned. 2. Large reservations are more likely to span across a 64M boundary. [ hpa: fixed incorrect use of "pfn" instead of "start" ] Signed-off-by: Petr Tesarik Link: http://lkml.kernel.org/r/20140201133019.32e56f86@hananiah.suse.cz Acked-by: David Rientjes Signed-off-by: H. Peter Anvin --- arch/x86/mm/numa_32.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/x86/mm/numa_32.c b/arch/x86/mm/numa_32.c index 0342d27ca798..47b6436e41c2 100644 --- a/arch/x86/mm/numa_32.c +++ b/arch/x86/mm/numa_32.c @@ -52,6 +52,8 @@ void memory_present(int nid, unsigned long start, unsigned long end) nid, start, end); printk(KERN_DEBUG " Setting physnode_map array to node %d for pfns:\n", nid); printk(KERN_DEBUG " "); + start = round_down(start, PAGES_PER_SECTION); + end = round_up(end, PAGES_PER_SECTION); for (pfn = start; pfn < end; pfn += PAGES_PER_SECTION) { physnode_map[pfn / PAGES_PER_SECTION] = nid; printk(KERN_CONT "%lx ", pfn); From 8e2a866ef214af4e104ec8d593e3269d8fe66d19 Mon Sep 17 00:00:00 2001 From: Emmanuel Grumbach Date: Tue, 28 Jan 2014 12:27:31 +0200 Subject: [PATCH 099/596] iwlwifi: mvm: BT Coex - disable BT when TXing probe request in scan Not doing so will let BT kill our probe requests leading to failures in scan. Cc: [3.10+] Reviewed-by: Johannes Berg Signed-off-by: Emmanuel Grumbach --- drivers/net/wireless/iwlwifi/mvm/scan.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/iwlwifi/mvm/scan.c b/drivers/net/wireless/iwlwifi/mvm/scan.c index c35f35cc4391..742afc429c94 100644 --- a/drivers/net/wireless/iwlwifi/mvm/scan.c +++ b/drivers/net/wireless/iwlwifi/mvm/scan.c @@ -344,7 +344,8 @@ int iwl_mvm_scan_request(struct iwl_mvm *mvm, iwl_mvm_scan_fill_ssids(cmd, req, basic_ssid ? 1 : 0); - cmd->tx_cmd.tx_flags = cpu_to_le32(TX_CMD_FLG_SEQ_CTL); + cmd->tx_cmd.tx_flags = cpu_to_le32(TX_CMD_FLG_SEQ_CTL | + TX_CMD_FLG_BT_DIS); cmd->tx_cmd.sta_id = mvm->aux_sta.sta_id; cmd->tx_cmd.life_time = cpu_to_le32(TX_CMD_LIFE_TIME_INFINITE); cmd->tx_cmd.rate_n_flags = From c512865446e6dd5b6e91e81187e75b734ad7cfc7 Mon Sep 17 00:00:00 2001 From: Emmanuel Grumbach Date: Thu, 5 Dec 2013 22:42:55 +0200 Subject: [PATCH 100/596] iwlwifi: mvm: don't allow A band if SKU forbids it The driver wasn't reading the NVM properly. While this didn't lead to any issue until now, it seems that there is an old version of the NVM in the wild. In this version, the A band channels appear to be valid but the SKU capabilities (another field of the NVM) says that A band isn't supported at all. With this specific version of the NVM, the driver would think that A band is supported while the HW / firmware don't. This leads to asserts. Cc: [3.10+] Reviewed-by: Johannes Berg Signed-off-by: Emmanuel Grumbach --- drivers/net/wireless/iwlwifi/iwl-nvm-parse.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/net/wireless/iwlwifi/iwl-nvm-parse.c b/drivers/net/wireless/iwlwifi/iwl-nvm-parse.c index f06f4cbe1317..725e954d8475 100644 --- a/drivers/net/wireless/iwlwifi/iwl-nvm-parse.c +++ b/drivers/net/wireless/iwlwifi/iwl-nvm-parse.c @@ -182,6 +182,11 @@ static int iwl_init_channel_map(struct device *dev, const struct iwl_cfg *cfg, for (ch_idx = 0; ch_idx < IWL_NUM_CHANNELS; ch_idx++) { ch_flags = __le16_to_cpup(nvm_ch_flags + ch_idx); + + if (ch_idx >= NUM_2GHZ_CHANNELS && + !data->sku_cap_band_52GHz_enable) + ch_flags &= ~NVM_CHANNEL_VALID; + if (!(ch_flags & NVM_CHANNEL_VALID)) { IWL_DEBUG_EEPROM(dev, "Ch. %d Flags %x [%sGHz] - No traffic\n", From 7288ca07b638db485abec5752bd6b1faed1c33ef Mon Sep 17 00:00:00 2001 From: Jean-Francois Moine Date: Sat, 25 Jan 2014 18:14:44 +0100 Subject: [PATCH 101/596] drm/i2c: tda998x: fix bad value in the AIF The AIF has an uninitialized byte. This patch clears the whole buffer before filling it. Tested-by: Russell King Acked-by: Russell King Signed-off-by: Jean-Francois Moine Signed-off-by: Russell King --- drivers/gpu/drm/i2c/tda998x_drv.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i2c/tda998x_drv.c b/drivers/gpu/drm/i2c/tda998x_drv.c index 400b0c4a10fb..e62082654e01 100644 --- a/drivers/gpu/drm/i2c/tda998x_drv.c +++ b/drivers/gpu/drm/i2c/tda998x_drv.c @@ -528,10 +528,10 @@ tda998x_write_aif(struct drm_encoder *encoder, struct tda998x_encoder_params *p) { uint8_t buf[PB(5) + 1]; + memset(buf, 0, sizeof(buf)); buf[HB(0)] = 0x84; buf[HB(1)] = 0x01; buf[HB(2)] = 10; - buf[PB(0)] = 0; buf[PB(1)] = p->audio_frame[1] & 0x07; /* CC */ buf[PB(2)] = p->audio_frame[2] & 0x1c; /* SF */ buf[PB(4)] = p->audio_frame[4]; From 6ae668cc19e8b18df28cd67b3448d9abd79284a4 Mon Sep 17 00:00:00 2001 From: Jean-Francois Moine Date: Sat, 25 Jan 2014 18:14:43 +0100 Subject: [PATCH 102/596] drm/i2c: tda998x: check the CEC device creation This patch checks if the CEC device is well created at intialization time. Tested-by: Russell King Acked-by: Russell King Signed-off-by: Jean-Francois Moine Signed-off-by: Russell King --- drivers/gpu/drm/i2c/tda998x_drv.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/gpu/drm/i2c/tda998x_drv.c b/drivers/gpu/drm/i2c/tda998x_drv.c index e62082654e01..c37180abcc4b 100644 --- a/drivers/gpu/drm/i2c/tda998x_drv.c +++ b/drivers/gpu/drm/i2c/tda998x_drv.c @@ -1144,6 +1144,8 @@ tda998x_encoder_init(struct i2c_client *client, priv->current_page = 0; priv->cec = i2c_new_dummy(client->adapter, 0x34); + if (!priv->cec) + return -ENODEV; priv->dpms = DRM_MODE_DPMS_OFF; encoder_slave->slave_priv = priv; From fc275a74eb816c12d4fc226344e734872ed0b2f9 Mon Sep 17 00:00:00 2001 From: Jean-Francois Moine Date: Sat, 25 Jan 2014 18:14:42 +0100 Subject: [PATCH 103/596] drm/i2c: tda998x: free the CEC device on encoder_destroy The cec i2c device is created in tda998x_encoder_init() when the DRM driver starts. This patch frees it when the DRM driver is unloaded. Tested-by: Russell King Acked-by: Russell King Signed-off-by: Jean-Francois Moine Signed-off-by: Russell King --- drivers/gpu/drm/i2c/tda998x_drv.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/gpu/drm/i2c/tda998x_drv.c b/drivers/gpu/drm/i2c/tda998x_drv.c index c37180abcc4b..de0572ec597e 100644 --- a/drivers/gpu/drm/i2c/tda998x_drv.c +++ b/drivers/gpu/drm/i2c/tda998x_drv.c @@ -1094,6 +1094,8 @@ tda998x_encoder_destroy(struct drm_encoder *encoder) { struct tda998x_priv *priv = to_tda998x_priv(encoder); drm_i2c_encoder_destroy(encoder); + if (priv->cec) + i2c_unregister_device(priv->cec); kfree(priv); } From 2eb4c7b1e7f275fe833aabe0a251b8e3f767fb08 Mon Sep 17 00:00:00 2001 From: Jean-Francois Moine Date: Sat, 25 Jan 2014 18:14:45 +0100 Subject: [PATCH 104/596] drm/i2c: tda998x: force the page register at startup time This patch forces the page register to be set on the first I/O operation. Tested-by: Russell King Acked-by: Russell King Signed-off-by: Jean-Francois Moine Signed-off-by: Russell King --- drivers/gpu/drm/i2c/tda998x_drv.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i2c/tda998x_drv.c b/drivers/gpu/drm/i2c/tda998x_drv.c index de0572ec597e..e871a4246379 100644 --- a/drivers/gpu/drm/i2c/tda998x_drv.c +++ b/drivers/gpu/drm/i2c/tda998x_drv.c @@ -1144,7 +1144,7 @@ tda998x_encoder_init(struct i2c_client *client, priv->vip_cntrl_1 = VIP_CNTRL_1_SWAP_C(0) | VIP_CNTRL_1_SWAP_D(1); priv->vip_cntrl_2 = VIP_CNTRL_2_SWAP_E(4) | VIP_CNTRL_2_SWAP_F(5); - priv->current_page = 0; + priv->current_page = 0xff; priv->cec = i2c_new_dummy(client->adapter, 0x34); if (!priv->cec) return -ENODEV; From 3ae471f73a1d581e078b5b06d08d7b82833a093f Mon Sep 17 00:00:00 2001 From: Jean-Francois Moine Date: Sat, 25 Jan 2014 18:14:36 +0100 Subject: [PATCH 105/596] drm/i2c: tda998x: set the PLL division factor in range 0..3 The predivider division factor of the register PLL_SERIAL_2 is in the range 0..3, the value 0 being used for a division by 1. Tested-by: Russell King Acked-by: Russell King Signed-off-by: Jean-Francois Moine Signed-off-by: Russell King --- drivers/gpu/drm/i2c/tda998x_drv.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i2c/tda998x_drv.c b/drivers/gpu/drm/i2c/tda998x_drv.c index e871a4246379..a65250b62f33 100644 --- a/drivers/gpu/drm/i2c/tda998x_drv.c +++ b/drivers/gpu/drm/i2c/tda998x_drv.c @@ -208,7 +208,7 @@ struct tda998x_priv { # define PLL_SERIAL_1_SRL_IZ(x) (((x) & 3) << 1) # define PLL_SERIAL_1_SRL_MAN_IZ (1 << 6) #define REG_PLL_SERIAL_2 REG(0x02, 0x01) /* read/write */ -# define PLL_SERIAL_2_SRL_NOSC(x) (((x) & 3) << 0) +# define PLL_SERIAL_2_SRL_NOSC(x) ((x) << 0) # define PLL_SERIAL_2_SRL_PR(x) (((x) & 0xf) << 4) #define REG_PLL_SERIAL_3 REG(0x02, 0x02) /* read/write */ # define PLL_SERIAL_3_SRL_CCIR (1 << 0) @@ -824,6 +824,11 @@ tda998x_encoder_mode_set(struct drm_encoder *encoder, } div = 148500 / mode->clock; + if (div != 0) { + div--; + if (div > 3) + div = 3; + } /* mute the audio FIFO: */ reg_set(encoder, REG_AIP_CNTRL_0, AIP_CNTRL_0_RST_FIFO); From 2e9a3fc3a360ac180f5b4c3c4416a0d0dec60dd8 Mon Sep 17 00:00:00 2001 From: Jean-Francois Moine Date: Sat, 25 Jan 2014 18:14:37 +0100 Subject: [PATCH 106/596] drm/i2c: tda998x: fix the ENABLE_SPACE register This patch fixes the ENABLE_SPACE register, the value of which was inverted. Tested-by: Russell King Acked-by: Russell King Signed-off-by: Jean-Francois Moine Signed-off-by: Russell King --- drivers/gpu/drm/i2c/tda998x_drv.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i2c/tda998x_drv.c b/drivers/gpu/drm/i2c/tda998x_drv.c index a65250b62f33..fa18cf374470 100644 --- a/drivers/gpu/drm/i2c/tda998x_drv.c +++ b/drivers/gpu/drm/i2c/tda998x_drv.c @@ -918,7 +918,7 @@ tda998x_encoder_mode_set(struct drm_encoder *encoder, if (priv->rev == TDA19988) { /* let incoming pixels fill the active space (if any) */ - reg_write(encoder, REG_ENABLE_SPACE, 0x01); + reg_write(encoder, REG_ENABLE_SPACE, 0x00); } /* must be last register set: */ From 9ac27090f61ea6735a62b0a98c7669c833bcdc09 Mon Sep 17 00:00:00 2001 From: Keith Busch Date: Fri, 31 Jan 2014 16:53:39 -0700 Subject: [PATCH 107/596] NVMe: Namespace use after free on surprise removal An nvme block device may have open references when the device is removed. New commands may still be sent on the removed device, so we need to ref count the opens, return errors for new commands, and not free the namespace and nvme_dev until all references are closed. Signed-off-by: Keith Busch Signed-off-by: Matthew Wilcox --- drivers/block/nvme-core.c | 55 ++++++++++++++++++++++++++++++--------- 1 file changed, 42 insertions(+), 13 deletions(-) diff --git a/drivers/block/nvme-core.c b/drivers/block/nvme-core.c index 23728099e36f..cd39390710a0 100644 --- a/drivers/block/nvme-core.c +++ b/drivers/block/nvme-core.c @@ -1716,10 +1716,31 @@ static int nvme_compat_ioctl(struct block_device *bdev, fmode_t mode, #define nvme_compat_ioctl NULL #endif +static int nvme_open(struct block_device *bdev, fmode_t mode) +{ + struct nvme_ns *ns = bdev->bd_disk->private_data; + struct nvme_dev *dev = ns->dev; + + kref_get(&dev->kref); + return 0; +} + +static void nvme_free_dev(struct kref *kref); + +static void nvme_release(struct gendisk *disk, fmode_t mode) +{ + struct nvme_ns *ns = disk->private_data; + struct nvme_dev *dev = ns->dev; + + kref_put(&dev->kref, nvme_free_dev); +} + static const struct block_device_operations nvme_fops = { .owner = THIS_MODULE, .ioctl = nvme_ioctl, .compat_ioctl = nvme_compat_ioctl, + .open = nvme_open, + .release = nvme_release, }; static void nvme_resubmit_bios(struct nvme_queue *nvmeq) @@ -1849,13 +1870,6 @@ static struct nvme_ns *nvme_alloc_ns(struct nvme_dev *dev, unsigned nsid, return NULL; } -static void nvme_ns_free(struct nvme_ns *ns) -{ - put_disk(ns->disk); - blk_cleanup_queue(ns->queue); - kfree(ns); -} - static int set_queue_count(struct nvme_dev *dev, int count) { int status; @@ -2287,12 +2301,13 @@ static void nvme_dev_shutdown(struct nvme_dev *dev) static void nvme_dev_remove(struct nvme_dev *dev) { - struct nvme_ns *ns, *next; + struct nvme_ns *ns; - list_for_each_entry_safe(ns, next, &dev->namespaces, list) { - list_del(&ns->list); - del_gendisk(ns->disk); - nvme_ns_free(ns); + list_for_each_entry(ns, &dev->namespaces, list) { + if (ns->disk->flags & GENHD_FL_UP) + del_gendisk(ns->disk); + if (!blk_queue_dying(ns->queue)) + blk_cleanup_queue(ns->queue); } } @@ -2349,9 +2364,22 @@ static void nvme_release_instance(struct nvme_dev *dev) spin_unlock(&dev_list_lock); } +static void nvme_free_namespaces(struct nvme_dev *dev) +{ + struct nvme_ns *ns, *next; + + list_for_each_entry_safe(ns, next, &dev->namespaces, list) { + list_del(&ns->list); + put_disk(ns->disk); + kfree(ns); + } +} + static void nvme_free_dev(struct kref *kref) { struct nvme_dev *dev = container_of(kref, struct nvme_dev, kref); + + nvme_free_namespaces(dev); kfree(dev->queues); kfree(dev->entry); kfree(dev); @@ -2525,6 +2553,7 @@ static int nvme_probe(struct pci_dev *pdev, const struct pci_device_id *id) goto release_pools; } + kref_init(&dev->kref); result = nvme_dev_add(dev); if (result) goto shutdown; @@ -2540,11 +2569,11 @@ static int nvme_probe(struct pci_dev *pdev, const struct pci_device_id *id) goto remove; dev->initialized = 1; - kref_init(&dev->kref); return 0; remove: nvme_dev_remove(dev); + nvme_free_namespaces(dev); shutdown: nvme_dev_shutdown(dev); release_pools: From 57789d69c5e7cdfdad6c0f1c54ab17240d800ac2 Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Sat, 1 Feb 2014 16:35:42 -0600 Subject: [PATCH 108/596] ARM: hisi: fix kconfig warning on HAVE_ARM_TWD Commit "ARM: hisi: don't select SMP" introduced a kconfig warning: warning: (ARCH_HI3xxx) selects HAVE_ARM_TWD which has unmet direct dependencies (SMP) Fix HAVE_ARM_TWD to depend on SMP. Signed-off-by: Rob Herring Signed-off-by: Olof Johansson --- arch/arm/mach-hisi/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/mach-hisi/Kconfig b/arch/arm/mach-hisi/Kconfig index 8f4649b301b2..1abae5f6a418 100644 --- a/arch/arm/mach-hisi/Kconfig +++ b/arch/arm/mach-hisi/Kconfig @@ -8,7 +8,7 @@ config ARCH_HI3xxx select CLKSRC_OF select GENERIC_CLOCKEVENTS select HAVE_ARM_SCU - select HAVE_ARM_TWD + select HAVE_ARM_TWD if SMP select HAVE_SMP select PINCTRL select PINCTRL_SINGLE From e803d988662ff365c2ed3c2d10a0aad811c1f492 Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Sat, 1 Feb 2014 16:35:43 -0600 Subject: [PATCH 109/596] ARM: moxart: move DMA_OF selection to driver Move the DMA_OF selection to the DMA driver to fix kconfig warning: warning: (ARCH_MOXART) selects DMA_OF which has unmet direct dependencies (DMADEVICES && OF) Signed-off-by: Rob Herring Signed-off-by: Olof Johansson --- arch/arm/mach-moxart/Kconfig | 1 - drivers/dma/Kconfig | 1 + 2 files changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/mach-moxart/Kconfig b/arch/arm/mach-moxart/Kconfig index ba470d64493b..3795ae28a613 100644 --- a/arch/arm/mach-moxart/Kconfig +++ b/arch/arm/mach-moxart/Kconfig @@ -2,7 +2,6 @@ config ARCH_MOXART bool "MOXA ART SoC" if ARCH_MULTI_V4T select CPU_FA526 select ARM_DMA_MEM_BUFFERABLE - select DMA_OF select USE_OF select CLKSRC_OF select CLKSRC_MMIO diff --git a/drivers/dma/Kconfig b/drivers/dma/Kconfig index 9bed1a2a67a1..605b016bcea4 100644 --- a/drivers/dma/Kconfig +++ b/drivers/dma/Kconfig @@ -346,6 +346,7 @@ config MOXART_DMA tristate "MOXART DMA support" depends on ARCH_MOXART select DMA_ENGINE + select DMA_OF select DMA_VIRTUAL_CHANNELS help Enable support for the MOXA ART SoC DMA controller. From 8b7dfa7dc701a27ac8ea4c4dfc0dbc8dce4825de Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Sat, 1 Feb 2014 16:35:44 -0600 Subject: [PATCH 110/596] ARM: fix HAVE_ARM_TWD selection for OMAP and shmobile The selection of HAVE_ARM_TWD for OMAP and shmobile depend on LOCAL_TIMER which no longer exists. They should depend on SMP instead. Cc: Tony Lindgren Cc: Simon Horman Cc: Magnus Damm Signed-off-by: Rob Herring Signed-off-by: Olof Johansson --- arch/arm/mach-omap2/Kconfig | 2 +- arch/arm/mach-shmobile/Kconfig | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/arm/mach-omap2/Kconfig b/arch/arm/mach-omap2/Kconfig index 653b489479e0..e2ce4f8366a7 100644 --- a/arch/arm/mach-omap2/Kconfig +++ b/arch/arm/mach-omap2/Kconfig @@ -54,7 +54,7 @@ config SOC_OMAP5 select ARM_GIC select CPU_V7 select HAVE_ARM_SCU if SMP - select HAVE_ARM_TWD if LOCAL_TIMERS + select HAVE_ARM_TWD if SMP select HAVE_SMP select HAVE_ARM_ARCH_TIMER select ARM_ERRATA_798181 if SMP diff --git a/arch/arm/mach-shmobile/Kconfig b/arch/arm/mach-shmobile/Kconfig index 338640631e08..05fa505df585 100644 --- a/arch/arm/mach-shmobile/Kconfig +++ b/arch/arm/mach-shmobile/Kconfig @@ -8,7 +8,7 @@ config ARCH_SHMOBILE_MULTI select CPU_V7 select GENERIC_CLOCKEVENTS select HAVE_ARM_SCU if SMP - select HAVE_ARM_TWD if LOCAL_TIMERS + select HAVE_ARM_TWD if SMP select HAVE_SMP select ARM_GIC select MIGHT_HAVE_CACHE_L2X0 From daa436e67cd2dcac7cbb505bb4425fdfdafaa5a7 Mon Sep 17 00:00:00 2001 From: Guenter Roeck Date: Thu, 30 Jan 2014 19:51:14 -0800 Subject: [PATCH 111/596] hwmon: (pmbus) Support per-page exponent in linear mode Some chips use different exponents for sensors on different pages or rails. Detect and store exponent per page to support this situation. This fixes a problem with wrong voltages seen on UCD90120. Reported-by: Soren Brinkmann Tested-by: Soren Brinkmann Signed-off-by: Guenter Roeck --- drivers/hwmon/pmbus/pmbus_core.c | 68 +++++++++++++++++--------------- 1 file changed, 36 insertions(+), 32 deletions(-) diff --git a/drivers/hwmon/pmbus/pmbus_core.c b/drivers/hwmon/pmbus/pmbus_core.c index 3cbf66e9d861..291d11fe93e7 100644 --- a/drivers/hwmon/pmbus/pmbus_core.c +++ b/drivers/hwmon/pmbus/pmbus_core.c @@ -90,7 +90,8 @@ struct pmbus_data { u32 flags; /* from platform data */ - int exponent; /* linear mode: exponent for output voltages */ + int exponent[PMBUS_PAGES]; + /* linear mode: exponent for output voltages */ const struct pmbus_driver_info *info; @@ -410,7 +411,7 @@ static long pmbus_reg2data_linear(struct pmbus_data *data, long val; if (sensor->class == PSC_VOLTAGE_OUT) { /* LINEAR16 */ - exponent = data->exponent; + exponent = data->exponent[sensor->page]; mantissa = (u16) sensor->data; } else { /* LINEAR11 */ exponent = ((s16)sensor->data) >> 11; @@ -516,7 +517,7 @@ static long pmbus_reg2data(struct pmbus_data *data, struct pmbus_sensor *sensor) #define MIN_MANTISSA (511 * 1000) static u16 pmbus_data2reg_linear(struct pmbus_data *data, - enum pmbus_sensor_classes class, long val) + struct pmbus_sensor *sensor, long val) { s16 exponent = 0, mantissa; bool negative = false; @@ -525,7 +526,7 @@ static u16 pmbus_data2reg_linear(struct pmbus_data *data, if (val == 0) return 0; - if (class == PSC_VOLTAGE_OUT) { + if (sensor->class == PSC_VOLTAGE_OUT) { /* LINEAR16 does not support negative voltages */ if (val < 0) return 0; @@ -534,10 +535,10 @@ static u16 pmbus_data2reg_linear(struct pmbus_data *data, * For a static exponents, we don't have a choice * but to adjust the value to it. */ - if (data->exponent < 0) - val <<= -data->exponent; + if (data->exponent[sensor->page] < 0) + val <<= -data->exponent[sensor->page]; else - val >>= data->exponent; + val >>= data->exponent[sensor->page]; val = DIV_ROUND_CLOSEST(val, 1000); return val & 0xffff; } @@ -548,14 +549,14 @@ static u16 pmbus_data2reg_linear(struct pmbus_data *data, } /* Power is in uW. Convert to mW before converting. */ - if (class == PSC_POWER) + if (sensor->class == PSC_POWER) val = DIV_ROUND_CLOSEST(val, 1000L); /* * For simplicity, convert fan data to milli-units * before calculating the exponent. */ - if (class == PSC_FAN) + if (sensor->class == PSC_FAN) val = val * 1000; /* Reduce large mantissa until it fits into 10 bit */ @@ -585,22 +586,22 @@ static u16 pmbus_data2reg_linear(struct pmbus_data *data, } static u16 pmbus_data2reg_direct(struct pmbus_data *data, - enum pmbus_sensor_classes class, long val) + struct pmbus_sensor *sensor, long val) { long m, b, R; - m = data->info->m[class]; - b = data->info->b[class]; - R = data->info->R[class]; + m = data->info->m[sensor->class]; + b = data->info->b[sensor->class]; + R = data->info->R[sensor->class]; /* Power is in uW. Adjust R and b. */ - if (class == PSC_POWER) { + if (sensor->class == PSC_POWER) { R -= 3; b *= 1000; } /* Calculate Y = (m * X + b) * 10^R */ - if (class != PSC_FAN) { + if (sensor->class != PSC_FAN) { R -= 3; /* Adjust R and b for data in milli-units */ b *= 1000; } @@ -619,7 +620,7 @@ static u16 pmbus_data2reg_direct(struct pmbus_data *data, } static u16 pmbus_data2reg_vid(struct pmbus_data *data, - enum pmbus_sensor_classes class, long val) + struct pmbus_sensor *sensor, long val) { val = clamp_val(val, 500, 1600); @@ -627,20 +628,20 @@ static u16 pmbus_data2reg_vid(struct pmbus_data *data, } static u16 pmbus_data2reg(struct pmbus_data *data, - enum pmbus_sensor_classes class, long val) + struct pmbus_sensor *sensor, long val) { u16 regval; - switch (data->info->format[class]) { + switch (data->info->format[sensor->class]) { case direct: - regval = pmbus_data2reg_direct(data, class, val); + regval = pmbus_data2reg_direct(data, sensor, val); break; case vid: - regval = pmbus_data2reg_vid(data, class, val); + regval = pmbus_data2reg_vid(data, sensor, val); break; case linear: default: - regval = pmbus_data2reg_linear(data, class, val); + regval = pmbus_data2reg_linear(data, sensor, val); break; } return regval; @@ -746,7 +747,7 @@ static ssize_t pmbus_set_sensor(struct device *dev, return -EINVAL; mutex_lock(&data->update_lock); - regval = pmbus_data2reg(data, sensor->class, val); + regval = pmbus_data2reg(data, sensor, val); ret = _pmbus_write_word_data(client, sensor->page, sensor->reg, regval); if (ret < 0) rv = ret; @@ -1643,12 +1644,13 @@ static int pmbus_find_attributes(struct i2c_client *client, * This function is called for all chips. */ static int pmbus_identify_common(struct i2c_client *client, - struct pmbus_data *data) + struct pmbus_data *data, int page) { int vout_mode = -1; - if (pmbus_check_byte_register(client, 0, PMBUS_VOUT_MODE)) - vout_mode = _pmbus_read_byte_data(client, 0, PMBUS_VOUT_MODE); + if (pmbus_check_byte_register(client, page, PMBUS_VOUT_MODE)) + vout_mode = _pmbus_read_byte_data(client, page, + PMBUS_VOUT_MODE); if (vout_mode >= 0 && vout_mode != 0xff) { /* * Not all chips support the VOUT_MODE command, @@ -1659,7 +1661,7 @@ static int pmbus_identify_common(struct i2c_client *client, if (data->info->format[PSC_VOLTAGE_OUT] != linear) return -ENODEV; - data->exponent = ((s8)(vout_mode << 3)) >> 3; + data->exponent[page] = ((s8)(vout_mode << 3)) >> 3; break; case 1: /* VID mode */ if (data->info->format[PSC_VOLTAGE_OUT] != vid) @@ -1674,7 +1676,7 @@ static int pmbus_identify_common(struct i2c_client *client, } } - pmbus_clear_fault_page(client, 0); + pmbus_clear_fault_page(client, page); return 0; } @@ -1682,7 +1684,7 @@ static int pmbus_init_common(struct i2c_client *client, struct pmbus_data *data, struct pmbus_driver_info *info) { struct device *dev = &client->dev; - int ret; + int page, ret; /* * Some PMBus chips don't support PMBUS_STATUS_BYTE, so try @@ -1715,10 +1717,12 @@ static int pmbus_init_common(struct i2c_client *client, struct pmbus_data *data, return -ENODEV; } - ret = pmbus_identify_common(client, data); - if (ret < 0) { - dev_err(dev, "Failed to identify chip capabilities\n"); - return ret; + for (page = 0; page < info->pages; page++) { + ret = pmbus_identify_common(client, data, page); + if (ret < 0) { + dev_err(dev, "Failed to identify chip capabilities\n"); + return ret; + } } return 0; } From b0dcfd87323ea86501e93d0fa2a98d2fd3579bcf Mon Sep 17 00:00:00 2001 From: Nicolas Ferre Date: Tue, 21 Jan 2014 16:55:18 +0100 Subject: [PATCH 112/596] pinctrl: at91: use locked variant of irq_set_handler When setting the gpio irq type, use the __irq_set_handler_locked() variant instead of the irq_set_handler() to prevent false spinlock recursion warning. Signed-off-by: Nicolas Ferre Cc: stable # v3.12 Signed-off-by: Linus Walleij --- drivers/pinctrl/pinctrl-at91.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/pinctrl/pinctrl-at91.c b/drivers/pinctrl/pinctrl-at91.c index 38c6f8b9790e..d990e33d8aa7 100644 --- a/drivers/pinctrl/pinctrl-at91.c +++ b/drivers/pinctrl/pinctrl-at91.c @@ -1286,22 +1286,22 @@ static int alt_gpio_irq_type(struct irq_data *d, unsigned type) switch (type) { case IRQ_TYPE_EDGE_RISING: - irq_set_handler(d->irq, handle_simple_irq); + __irq_set_handler_locked(d->irq, handle_simple_irq); writel_relaxed(mask, pio + PIO_ESR); writel_relaxed(mask, pio + PIO_REHLSR); break; case IRQ_TYPE_EDGE_FALLING: - irq_set_handler(d->irq, handle_simple_irq); + __irq_set_handler_locked(d->irq, handle_simple_irq); writel_relaxed(mask, pio + PIO_ESR); writel_relaxed(mask, pio + PIO_FELLSR); break; case IRQ_TYPE_LEVEL_LOW: - irq_set_handler(d->irq, handle_level_irq); + __irq_set_handler_locked(d->irq, handle_level_irq); writel_relaxed(mask, pio + PIO_LSR); writel_relaxed(mask, pio + PIO_FELLSR); break; case IRQ_TYPE_LEVEL_HIGH: - irq_set_handler(d->irq, handle_level_irq); + __irq_set_handler_locked(d->irq, handle_level_irq); writel_relaxed(mask, pio + PIO_LSR); writel_relaxed(mask, pio + PIO_REHLSR); break; @@ -1310,7 +1310,7 @@ static int alt_gpio_irq_type(struct irq_data *d, unsigned type) * disable additional interrupt modes: * fall back to default behavior */ - irq_set_handler(d->irq, handle_simple_irq); + __irq_set_handler_locked(d->irq, handle_simple_irq); writel_relaxed(mask, pio + PIO_AIMDR); return 0; case IRQ_TYPE_NONE: From 795779df22afc8bdee4e9fbe5c18c47e44974d75 Mon Sep 17 00:00:00 2001 From: Chris Ruehl Date: Wed, 22 Jan 2014 11:14:51 +0800 Subject: [PATCH 113/596] pinctrl: imx27: fix wrong offset to ICONFB The offset to ICONFB was incorrect, this patch set the correct value 0x14. dev_dbg in function imx1_write_2bit print the wrong address and had been moved after address calculation. Cc: stable@vger.kernel.org Signed-off-by: Chris Ruehl Reviewed-by: Markus Pargmann Signed-off-by: Linus Walleij --- drivers/pinctrl/pinctrl-imx1-core.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/pinctrl/pinctrl-imx1-core.c b/drivers/pinctrl/pinctrl-imx1-core.c index 17aecde1b51d..9e7893fe96e0 100644 --- a/drivers/pinctrl/pinctrl-imx1-core.c +++ b/drivers/pinctrl/pinctrl-imx1-core.c @@ -45,7 +45,7 @@ struct imx1_pinctrl { #define MX1_DDIR 0x00 #define MX1_OCR 0x04 #define MX1_ICONFA 0x0c -#define MX1_ICONFB 0x10 +#define MX1_ICONFB 0x14 #define MX1_GIUS 0x20 #define MX1_GPR 0x38 #define MX1_PUEN 0x40 @@ -97,13 +97,13 @@ static void imx1_write_2bit(struct imx1_pinctrl *ipctl, unsigned int pin_id, u32 old_val; u32 new_val; - dev_dbg(ipctl->dev, "write: register 0x%p offset %d value 0x%x\n", - reg, offset, value); - /* Use the next register if the pin's port pin number is >=16 */ if (pin_id % 32 >= 16) reg += 0x04; + dev_dbg(ipctl->dev, "write: register 0x%p offset %d value 0x%x\n", + reg, offset, value); + /* Get current state of pins */ old_val = readl(reg); old_val &= mask; From f17248ed868767567298e1cdf06faf8159a81f7c Mon Sep 17 00:00:00 2001 From: Tony Prisk Date: Thu, 23 Jan 2014 21:57:33 +1300 Subject: [PATCH 114/596] pinctrl: vt8500: Change devicetree data parsing Due to an assumption in the VT8500 pinctrl driver, the value passed from devicetree for 'wm,pull' was not explicitly translated before being passed to pinconf. Since v3.10, changes to 'enum pin_config_param', PIN_CONFIG_BIAS_PULL_(UP/DOWN) no longer map 1-to-1 with the expected values in devicetree. This patch adds a small translation between the devicetree values (0..2) and the enum pin_config_param equivalent values. Cc: # v3.10+ Signed-off-by: Tony Prisk Signed-off-by: Linus Walleij --- drivers/pinctrl/vt8500/pinctrl-wmt.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-) diff --git a/drivers/pinctrl/vt8500/pinctrl-wmt.c b/drivers/pinctrl/vt8500/pinctrl-wmt.c index b28d1af9c232..9802b67040cc 100644 --- a/drivers/pinctrl/vt8500/pinctrl-wmt.c +++ b/drivers/pinctrl/vt8500/pinctrl-wmt.c @@ -276,7 +276,20 @@ static int wmt_pctl_dt_node_to_map_pull(struct wmt_pinctrl_data *data, if (!configs) return -ENOMEM; - configs[0] = pull; + switch (pull) { + case 0: + configs[0] = PIN_CONFIG_BIAS_DISABLE; + break; + case 1: + configs[0] = PIN_CONFIG_BIAS_PULL_DOWN; + break; + case 2: + configs[0] = PIN_CONFIG_BIAS_PULL_UP; + break; + default: + configs[0] = PIN_CONFIG_BIAS_DISABLE; + dev_err(data->dev, "invalid pull state %d - disabling\n", pull); + } map->type = PIN_MAP_TYPE_CONFIGS_PIN; map->data.configs.group_or_pin = data->groups[group]; From e3365d0974ed64157f5b5a576c611057dc40a595 Mon Sep 17 00:00:00 2001 From: Chris Ruehl Date: Wed, 22 Jan 2014 11:14:52 +0800 Subject: [PATCH 115/596] pinctrl: imx27: fix offset calculation in imx_read_2bit The offset for the 2bit register calculate wrong, this patch fixes the problem. The debugfs printout for oconf, iconfa, iconfb now shows the real values. Cc: stable@vger.kernel.org Signed-off-by: Chris Ruehl Reviewed-by: Markus Pargmann Signed-off-by: Linus Walleij --- drivers/pinctrl/pinctrl-imx1-core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/pinctrl/pinctrl-imx1-core.c b/drivers/pinctrl/pinctrl-imx1-core.c index 9e7893fe96e0..815384b377b5 100644 --- a/drivers/pinctrl/pinctrl-imx1-core.c +++ b/drivers/pinctrl/pinctrl-imx1-core.c @@ -139,7 +139,7 @@ static int imx1_read_2bit(struct imx1_pinctrl *ipctl, unsigned int pin_id, u32 reg_offset) { void __iomem *reg = imx1_mem(ipctl, pin_id) + reg_offset; - int offset = pin_id % 16; + int offset = (pin_id % 16) * 2; /* Use the next register if the pin's port pin number is >=16 */ if (pin_id % 32 >= 16) From fa74d0d3e30cf079e94db327ea6d4726cd7d5871 Mon Sep 17 00:00:00 2001 From: Qipan Li Date: Mon, 27 Jan 2014 14:01:29 +0800 Subject: [PATCH 116/596] pinctrl: sirf: correct the pin index of ac97_pins group according to datasheet and ac97_muxmask assignment, ac97_pins should be corrected. Signed-off-by: Qipan Li Signed-off-by: Barry Song Signed-off-by: Linus Walleij --- drivers/pinctrl/sirf/pinctrl-prima2.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/pinctrl/sirf/pinctrl-prima2.c b/drivers/pinctrl/sirf/pinctrl-prima2.c index 37b42651d76a..dde0285544d6 100644 --- a/drivers/pinctrl/sirf/pinctrl-prima2.c +++ b/drivers/pinctrl/sirf/pinctrl-prima2.c @@ -413,7 +413,7 @@ static const struct sirfsoc_padmux ac97_padmux = { .funcval = 0, }; -static const unsigned ac97_pins[] = { 33, 34, 35, 36 }; +static const unsigned ac97_pins[] = { 43, 44, 45, 46 }; static const struct sirfsoc_muxmask spi1_muxmask[] = { { From cffcc92e966911e3ea9320fac943f545a353f2e0 Mon Sep 17 00:00:00 2001 From: Baruch Siach Date: Thu, 23 Jan 2014 08:50:38 +0200 Subject: [PATCH 117/596] gpio: xtensa: fix build when XCHAL_HAVE_CP is 0 In xtensa coprocessors may exist without coprocessor context, i.e. they cannot be disabled/enabled. In this case the RSR_CPENABLE/WSR_CPENABLE are undefined, thus breaking the build. Fix the build by adding dummy versions of enable_cp/disable_cp in this case. Reported-by: Fengguang Wu Signed-off-by: Baruch Siach Signed-off-by: Linus Walleij --- drivers/gpio/gpio-xtensa.c | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) diff --git a/drivers/gpio/gpio-xtensa.c b/drivers/gpio/gpio-xtensa.c index 1d136eceda62..7081304d6797 100644 --- a/drivers/gpio/gpio-xtensa.c +++ b/drivers/gpio/gpio-xtensa.c @@ -40,6 +40,8 @@ #error GPIO32 option is not enabled for your xtensa core variant #endif +#if XCHAL_HAVE_CP + static inline unsigned long enable_cp(unsigned long *cpenable) { unsigned long flags; @@ -57,6 +59,20 @@ static inline void disable_cp(unsigned long flags, unsigned long cpenable) local_irq_restore(flags); } +#else + +static inline unsigned long enable_cp(unsigned long *cpenable) +{ + *cpenable = 0; /* avoid uninitialized value warning */ + return 0; +} + +static inline void disable_cp(unsigned long flags, unsigned long cpenable) +{ +} + +#endif /* XCHAL_HAVE_CP */ + static int xtensa_impwire_get_direction(struct gpio_chip *gc, unsigned offset) { return 1; /* input only */ From 4fa71c1550a857ff1dbfe9e99acc1f4cfec5f0d0 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Mon, 3 Feb 2014 09:37:59 +0100 Subject: [PATCH 118/596] ALSA: usb-audio: Add missing kconfig dependecy The commit 44dcbbb1cd61 introduced the usage of bitreverse helpers but forgot to add the dependency. This patch adds the selection for CONFIG_BITREVERSE. Fixes: 44dcbbb1cd61 ('ALSA: snd-usb: add support for bit-reversed byte formats') Reported-by: Fengguang Wu Cc: Signed-off-by: Takashi Iwai --- sound/usb/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/sound/usb/Kconfig b/sound/usb/Kconfig index de9408b83f75..e05a86b7c0da 100644 --- a/sound/usb/Kconfig +++ b/sound/usb/Kconfig @@ -14,6 +14,7 @@ config SND_USB_AUDIO select SND_HWDEP select SND_RAWMIDI select SND_PCM + select BITREVERSE help Say Y here to include support for USB audio and USB MIDI devices. From e85fc9805591a17ca8af50023ee8e2b61d9a123b Mon Sep 17 00:00:00 2001 From: Konrad Rzeszutek Wilk Date: Mon, 3 Feb 2014 06:43:59 -0500 Subject: [PATCH 119/596] Revert "xen/grant-table: Avoid m2p_override during mapping" This reverts commit 08ece5bb2312b4510b161a6ef6682f37f4eac8a1. As it breaks ARM builds and needs more attention on the ARM side. Acked-by: David Vrabel Signed-off-by: Konrad Rzeszutek Wilk --- arch/x86/include/asm/xen/page.h | 5 +- arch/x86/xen/p2m.c | 17 +++++- drivers/block/xen-blkback/blkback.c | 15 +++-- drivers/xen/gntdev.c | 13 ++--- drivers/xen/grant-table.c | 89 +++++------------------------ include/xen/grant_table.h | 8 +-- 6 files changed, 46 insertions(+), 101 deletions(-) diff --git a/arch/x86/include/asm/xen/page.h b/arch/x86/include/asm/xen/page.h index 787e1bb5aafc..3e276eb23d1b 100644 --- a/arch/x86/include/asm/xen/page.h +++ b/arch/x86/include/asm/xen/page.h @@ -52,8 +52,7 @@ extern unsigned long set_phys_range_identity(unsigned long pfn_s, extern int m2p_add_override(unsigned long mfn, struct page *page, struct gnttab_map_grant_ref *kmap_op); extern int m2p_remove_override(struct page *page, - struct gnttab_map_grant_ref *kmap_op, - unsigned long mfn); + struct gnttab_map_grant_ref *kmap_op); extern struct page *m2p_find_override(unsigned long mfn); extern unsigned long m2p_find_override_pfn(unsigned long mfn, unsigned long pfn); @@ -122,7 +121,7 @@ static inline unsigned long mfn_to_pfn(unsigned long mfn) pfn = m2p_find_override_pfn(mfn, ~0); } - /* + /* * pfn is ~0 if there are no entries in the m2p for mfn or if the * entry doesn't map back to the mfn and m2p_override doesn't have a * valid entry for it. diff --git a/arch/x86/xen/p2m.c b/arch/x86/xen/p2m.c index 8009acbe41e4..696c694986d0 100644 --- a/arch/x86/xen/p2m.c +++ b/arch/x86/xen/p2m.c @@ -899,6 +899,13 @@ int m2p_add_override(unsigned long mfn, struct page *page, "m2p_add_override: pfn %lx not mapped", pfn)) return -EINVAL; } + WARN_ON(PagePrivate(page)); + SetPagePrivate(page); + set_page_private(page, mfn); + page->index = pfn_to_mfn(pfn); + + if (unlikely(!set_phys_to_machine(pfn, FOREIGN_FRAME(mfn)))) + return -ENOMEM; if (kmap_op != NULL) { if (!PageHighMem(page)) { @@ -937,16 +944,19 @@ int m2p_add_override(unsigned long mfn, struct page *page, } EXPORT_SYMBOL_GPL(m2p_add_override); int m2p_remove_override(struct page *page, - struct gnttab_map_grant_ref *kmap_op, - unsigned long mfn) + struct gnttab_map_grant_ref *kmap_op) { unsigned long flags; + unsigned long mfn; unsigned long pfn; unsigned long uninitialized_var(address); unsigned level; pte_t *ptep = NULL; pfn = page_to_pfn(page); + mfn = get_phys_to_machine(pfn); + if (mfn == INVALID_P2M_ENTRY || !(mfn & FOREIGN_FRAME_BIT)) + return -EINVAL; if (!PageHighMem(page)) { address = (unsigned long)__va(pfn << PAGE_SHIFT); @@ -960,7 +970,10 @@ int m2p_remove_override(struct page *page, spin_lock_irqsave(&m2p_override_lock, flags); list_del(&page->lru); spin_unlock_irqrestore(&m2p_override_lock, flags); + WARN_ON(!PagePrivate(page)); + ClearPagePrivate(page); + set_phys_to_machine(pfn, page->index); if (kmap_op != NULL) { if (!PageHighMem(page)) { struct multicall_space mcs; diff --git a/drivers/block/xen-blkback/blkback.c b/drivers/block/xen-blkback/blkback.c index 875025f299b6..6620b73d0490 100644 --- a/drivers/block/xen-blkback/blkback.c +++ b/drivers/block/xen-blkback/blkback.c @@ -285,7 +285,8 @@ static void free_persistent_gnts(struct xen_blkif *blkif, struct rb_root *root, if (++segs_to_unmap == BLKIF_MAX_SEGMENTS_PER_REQUEST || !rb_next(&persistent_gnt->node)) { - ret = gnttab_unmap_refs(unmap, pages, segs_to_unmap); + ret = gnttab_unmap_refs(unmap, NULL, pages, + segs_to_unmap); BUG_ON(ret); put_free_pages(blkif, pages, segs_to_unmap); segs_to_unmap = 0; @@ -320,7 +321,8 @@ static void unmap_purged_grants(struct work_struct *work) pages[segs_to_unmap] = persistent_gnt->page; if (++segs_to_unmap == BLKIF_MAX_SEGMENTS_PER_REQUEST) { - ret = gnttab_unmap_refs(unmap, pages, segs_to_unmap); + ret = gnttab_unmap_refs(unmap, NULL, pages, + segs_to_unmap); BUG_ON(ret); put_free_pages(blkif, pages, segs_to_unmap); segs_to_unmap = 0; @@ -328,7 +330,7 @@ static void unmap_purged_grants(struct work_struct *work) kfree(persistent_gnt); } if (segs_to_unmap > 0) { - ret = gnttab_unmap_refs(unmap, pages, segs_to_unmap); + ret = gnttab_unmap_refs(unmap, NULL, pages, segs_to_unmap); BUG_ON(ret); put_free_pages(blkif, pages, segs_to_unmap); } @@ -668,14 +670,15 @@ static void xen_blkbk_unmap(struct xen_blkif *blkif, GNTMAP_host_map, pages[i]->handle); pages[i]->handle = BLKBACK_INVALID_HANDLE; if (++invcount == BLKIF_MAX_SEGMENTS_PER_REQUEST) { - ret = gnttab_unmap_refs(unmap, unmap_pages, invcount); + ret = gnttab_unmap_refs(unmap, NULL, unmap_pages, + invcount); BUG_ON(ret); put_free_pages(blkif, unmap_pages, invcount); invcount = 0; } } if (invcount) { - ret = gnttab_unmap_refs(unmap, unmap_pages, invcount); + ret = gnttab_unmap_refs(unmap, NULL, unmap_pages, invcount); BUG_ON(ret); put_free_pages(blkif, unmap_pages, invcount); } @@ -737,7 +740,7 @@ static int xen_blkbk_map(struct xen_blkif *blkif, } if (segs_to_map) { - ret = gnttab_map_refs(map, pages_to_gnt, segs_to_map); + ret = gnttab_map_refs(map, NULL, pages_to_gnt, segs_to_map); BUG_ON(ret); } diff --git a/drivers/xen/gntdev.c b/drivers/xen/gntdev.c index 34a2704fbc88..073b4a19a8b0 100644 --- a/drivers/xen/gntdev.c +++ b/drivers/xen/gntdev.c @@ -284,10 +284,8 @@ static int map_grant_pages(struct grant_map *map) } pr_debug("map %d+%d\n", map->index, map->count); - err = gnttab_map_refs_userspace(map->map_ops, - use_ptemod ? map->kmap_ops : NULL, - map->pages, - map->count); + err = gnttab_map_refs(map->map_ops, use_ptemod ? map->kmap_ops : NULL, + map->pages, map->count); if (err) return err; @@ -317,10 +315,9 @@ static int __unmap_grant_pages(struct grant_map *map, int offset, int pages) } } - err = gnttab_unmap_refs_userspace(map->unmap_ops + offset, - use_ptemod ? map->kmap_ops + offset : NULL, - map->pages + offset, - pages); + err = gnttab_unmap_refs(map->unmap_ops + offset, + use_ptemod ? map->kmap_ops + offset : NULL, map->pages + offset, + pages); if (err) return err; diff --git a/drivers/xen/grant-table.c b/drivers/xen/grant-table.c index 8ee13e2e45e2..b84e3ab839aa 100644 --- a/drivers/xen/grant-table.c +++ b/drivers/xen/grant-table.c @@ -928,17 +928,15 @@ void gnttab_batch_copy(struct gnttab_copy *batch, unsigned count) } EXPORT_SYMBOL_GPL(gnttab_batch_copy); -int __gnttab_map_refs(struct gnttab_map_grant_ref *map_ops, +int gnttab_map_refs(struct gnttab_map_grant_ref *map_ops, struct gnttab_map_grant_ref *kmap_ops, - struct page **pages, unsigned int count, - bool m2p_override) + struct page **pages, unsigned int count) { int i, ret; bool lazy = false; pte_t *pte; - unsigned long mfn, pfn; + unsigned long mfn; - BUG_ON(kmap_ops && !m2p_override); ret = HYPERVISOR_grant_table_op(GNTTABOP_map_grant_ref, map_ops, count); if (ret) return ret; @@ -957,12 +955,10 @@ int __gnttab_map_refs(struct gnttab_map_grant_ref *map_ops, set_phys_to_machine(map_ops[i].host_addr >> PAGE_SHIFT, map_ops[i].dev_bus_addr >> PAGE_SHIFT); } - return 0; + return ret; } - if (m2p_override && - !in_interrupt() && - paravirt_get_lazy_mode() == PARAVIRT_LAZY_NONE) { + if (!in_interrupt() && paravirt_get_lazy_mode() == PARAVIRT_LAZY_NONE) { arch_enter_lazy_mmu_mode(); lazy = true; } @@ -979,20 +975,8 @@ int __gnttab_map_refs(struct gnttab_map_grant_ref *map_ops, } else { mfn = PFN_DOWN(map_ops[i].dev_bus_addr); } - pfn = page_to_pfn(pages[i]); - - WARN_ON(PagePrivate(pages[i])); - SetPagePrivate(pages[i]); - set_page_private(pages[i], mfn); - - pages[i]->index = pfn_to_mfn(pfn); - if (unlikely(!set_phys_to_machine(pfn, FOREIGN_FRAME(mfn)))) { - ret = -ENOMEM; - goto out; - } - if (m2p_override) - ret = m2p_add_override(mfn, pages[i], kmap_ops ? - &kmap_ops[i] : NULL); + ret = m2p_add_override(mfn, pages[i], kmap_ops ? + &kmap_ops[i] : NULL); if (ret) goto out; } @@ -1003,32 +987,15 @@ int __gnttab_map_refs(struct gnttab_map_grant_ref *map_ops, return ret; } - -int gnttab_map_refs(struct gnttab_map_grant_ref *map_ops, - struct page **pages, unsigned int count) -{ - return __gnttab_map_refs(map_ops, NULL, pages, count, false); -} EXPORT_SYMBOL_GPL(gnttab_map_refs); -int gnttab_map_refs_userspace(struct gnttab_map_grant_ref *map_ops, - struct gnttab_map_grant_ref *kmap_ops, - struct page **pages, unsigned int count) -{ - return __gnttab_map_refs(map_ops, kmap_ops, pages, count, true); -} -EXPORT_SYMBOL_GPL(gnttab_map_refs_userspace); - -int __gnttab_unmap_refs(struct gnttab_unmap_grant_ref *unmap_ops, +int gnttab_unmap_refs(struct gnttab_unmap_grant_ref *unmap_ops, struct gnttab_map_grant_ref *kmap_ops, - struct page **pages, unsigned int count, - bool m2p_override) + struct page **pages, unsigned int count) { int i, ret; bool lazy = false; - unsigned long pfn, mfn; - BUG_ON(kmap_ops && !m2p_override); ret = HYPERVISOR_grant_table_op(GNTTABOP_unmap_grant_ref, unmap_ops, count); if (ret) return ret; @@ -1039,33 +1006,17 @@ int __gnttab_unmap_refs(struct gnttab_unmap_grant_ref *unmap_ops, set_phys_to_machine(unmap_ops[i].host_addr >> PAGE_SHIFT, INVALID_P2M_ENTRY); } - return 0; + return ret; } - if (m2p_override && - !in_interrupt() && - paravirt_get_lazy_mode() == PARAVIRT_LAZY_NONE) { + if (!in_interrupt() && paravirt_get_lazy_mode() == PARAVIRT_LAZY_NONE) { arch_enter_lazy_mmu_mode(); lazy = true; } for (i = 0; i < count; i++) { - pfn = page_to_pfn(pages[i]); - mfn = get_phys_to_machine(pfn); - if (mfn == INVALID_P2M_ENTRY || !(mfn & FOREIGN_FRAME_BIT)) { - ret = -EINVAL; - goto out; - } - - set_page_private(pages[i], INVALID_P2M_ENTRY); - WARN_ON(!PagePrivate(pages[i])); - ClearPagePrivate(pages[i]); - set_phys_to_machine(pfn, pages[i]->index); - if (m2p_override) - ret = m2p_remove_override(pages[i], - kmap_ops ? - &kmap_ops[i] : NULL, - mfn); + ret = m2p_remove_override(pages[i], kmap_ops ? + &kmap_ops[i] : NULL); if (ret) goto out; } @@ -1076,22 +1027,8 @@ int __gnttab_unmap_refs(struct gnttab_unmap_grant_ref *unmap_ops, return ret; } - -int gnttab_unmap_refs(struct gnttab_unmap_grant_ref *map_ops, - struct page **pages, unsigned int count) -{ - return __gnttab_unmap_refs(map_ops, NULL, pages, count, false); -} EXPORT_SYMBOL_GPL(gnttab_unmap_refs); -int gnttab_unmap_refs_userspace(struct gnttab_unmap_grant_ref *map_ops, - struct gnttab_map_grant_ref *kmap_ops, - struct page **pages, unsigned int count) -{ - return __gnttab_unmap_refs(map_ops, kmap_ops, pages, count, true); -} -EXPORT_SYMBOL_GPL(gnttab_unmap_refs_userspace); - static unsigned nr_status_frames(unsigned nr_grant_frames) { BUG_ON(grefs_per_grant_frame == 0); diff --git a/include/xen/grant_table.h b/include/xen/grant_table.h index 7ad033dbc845..a5af2a26d94f 100644 --- a/include/xen/grant_table.h +++ b/include/xen/grant_table.h @@ -191,15 +191,11 @@ void gnttab_free_auto_xlat_frames(void); #define gnttab_map_vaddr(map) ((void *)(map.host_virt_addr)) int gnttab_map_refs(struct gnttab_map_grant_ref *map_ops, + struct gnttab_map_grant_ref *kmap_ops, struct page **pages, unsigned int count); -int gnttab_map_refs_userspace(struct gnttab_map_grant_ref *map_ops, - struct gnttab_map_grant_ref *kmap_ops, - struct page **pages, unsigned int count); int gnttab_unmap_refs(struct gnttab_unmap_grant_ref *unmap_ops, + struct gnttab_map_grant_ref *kunmap_ops, struct page **pages, unsigned int count); -int gnttab_unmap_refs_userspace(struct gnttab_unmap_grant_ref *unmap_ops, - struct gnttab_map_grant_ref *kunmap_ops, - struct page **pages, unsigned int count); /* Perform a batch of grant map/copy operations. Retry every batch slot * for which the hypervisor returns GNTST_eagain. This is typically due From 375f5fb2e4609b0577778db90e2fc40a2d709cac Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Emilio=20L=C3=B3pez?= Date: Wed, 29 Jan 2014 22:36:38 -0300 Subject: [PATCH 120/596] of: add vendor prefix for Allwinner Technology MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit We have been using the "allwinner" prefix for everything so far; let's document it here. Signed-off-by: Emilio López Signed-off-by: Rob Herring --- Documentation/devicetree/bindings/vendor-prefixes.txt | 1 + 1 file changed, 1 insertion(+) diff --git a/Documentation/devicetree/bindings/vendor-prefixes.txt b/Documentation/devicetree/bindings/vendor-prefixes.txt index 3f900cd51bf0..e1d1a225a8d1 100644 --- a/Documentation/devicetree/bindings/vendor-prefixes.txt +++ b/Documentation/devicetree/bindings/vendor-prefixes.txt @@ -8,6 +8,7 @@ ad Avionic Design GmbH adi Analog Devices, Inc. aeroflexgaisler Aeroflex Gaisler AB ak Asahi Kasei Corp. +allwinner Allwinner Technology Co., Ltd. altr Altera Corp. amcc Applied Micro Circuits Corporation (APM, formally AMCC) amstaos AMS-Taos Inc. From d5f93b8ad01e23d4351cc66c855d09cc5c5cf42c Mon Sep 17 00:00:00 2001 From: Kumar Gala Date: Wed, 18 Dec 2013 17:41:16 -0600 Subject: [PATCH 121/596] of: Update qcom vendor prefix description Update vendor description to be the full name of the corporate entity that produces the SoCs associated with this prefix. Acked-by: Olof Johansson Signed-off-by: Kumar Gala Signed-off-by: Rob Herring --- Documentation/devicetree/bindings/vendor-prefixes.txt | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Documentation/devicetree/bindings/vendor-prefixes.txt b/Documentation/devicetree/bindings/vendor-prefixes.txt index e1d1a225a8d1..6f2a022def4f 100644 --- a/Documentation/devicetree/bindings/vendor-prefixes.txt +++ b/Documentation/devicetree/bindings/vendor-prefixes.txt @@ -65,7 +65,7 @@ phytec PHYTEC Messtechnik GmbH picochip Picochip Ltd powervr PowerVR (deprecated, use img) qca Qualcomm Atheros, Inc. -qcom Qualcomm, Inc. +qcom Qualcomm Technologies, Inc ralink Mediatek/Ralink Technology Corp. ramtron Ramtron International realtek Realtek Semiconductor Corp. From b2d960e4061986c6aab2269103d61e8408dcc71d Mon Sep 17 00:00:00 2001 From: Marek Belisko Date: Sat, 25 Jan 2014 22:38:44 +0100 Subject: [PATCH 122/596] of: add vendor prefix for Honeywell Signed-off-by: Marek Belisko Signed-off-by: Rob Herring --- Documentation/devicetree/bindings/vendor-prefixes.txt | 1 + 1 file changed, 1 insertion(+) diff --git a/Documentation/devicetree/bindings/vendor-prefixes.txt b/Documentation/devicetree/bindings/vendor-prefixes.txt index 6f2a022def4f..ba43b521218f 100644 --- a/Documentation/devicetree/bindings/vendor-prefixes.txt +++ b/Documentation/devicetree/bindings/vendor-prefixes.txt @@ -41,6 +41,7 @@ gmt Global Mixed-mode Technology, Inc. gumstix Gumstix, Inc. haoyu Haoyu Microelectronic Co. Ltd. hisilicon Hisilicon Limited. +honeywell Honeywell hp Hewlett Packard ibm International Business Machines (IBM) idt Integrated Device Technologies, Inc. From 8101c8dbf6243ba517aab58d69bf1bc37d8b7b9c Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Wed, 29 Jan 2014 16:05:30 -0500 Subject: [PATCH 123/596] Btrfs: disable snapshot aware defrag for now It's just broken and it's taking a lot of effort to fix it, so for now just disable it so people can defrag in peace. Thanks, Cc: stable@vger.kernel.org Signed-off-by: Josef Bacik Signed-off-by: Chris Mason --- fs/btrfs/inode.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index fb74a536add3..1af34d0c744b 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -2629,7 +2629,7 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent) EXTENT_DEFRAG, 1, cached_state); if (ret) { u64 last_snapshot = btrfs_root_last_snapshot(&root->root_item); - if (last_snapshot >= BTRFS_I(inode)->generation) + if (0 && last_snapshot >= BTRFS_I(inode)->generation) /* the inode is shared */ new = record_old_file_extents(inode, ordered_extent); From 0b947aff1599afbbd2ec07ada87b05af0f94cf10 Mon Sep 17 00:00:00 2001 From: Filipe David Borba Manana Date: Wed, 29 Jan 2014 21:06:04 +0000 Subject: [PATCH 124/596] Btrfs: use btrfs_crc32c everywhere instead of libcrc32c After the commit titled "Btrfs: fix btrfs boot when compiled as built-in", LIBCRC32C requirement was removed from btrfs' Kconfig. This made it not possible to build a kernel with btrfs enabled (either as module or built-in) if libcrc32c is not enabled as well. So just replace all uses of libcrc32c with the equivalent function in btrfs hash.h - btrfs_crc32c. Signed-off-by: Filipe David Borba Manana Signed-off-by: Chris Mason --- fs/btrfs/check-integrity.c | 4 ++-- fs/btrfs/disk-io.c | 4 ++-- fs/btrfs/send.c | 4 ++-- 3 files changed, 6 insertions(+), 6 deletions(-) diff --git a/fs/btrfs/check-integrity.c b/fs/btrfs/check-integrity.c index 160fb509d720..39bfd56a1f26 100644 --- a/fs/btrfs/check-integrity.c +++ b/fs/btrfs/check-integrity.c @@ -92,11 +92,11 @@ #include #include #include -#include #include #include #include "ctree.h" #include "disk-io.h" +#include "hash.h" #include "transaction.h" #include "extent_io.h" #include "volumes.h" @@ -1823,7 +1823,7 @@ static int btrfsic_test_for_metadata(struct btrfsic_state *state, size_t sublen = i ? PAGE_CACHE_SIZE : (PAGE_CACHE_SIZE - BTRFS_CSUM_SIZE); - crc = crc32c(crc, data, sublen); + crc = btrfs_crc32c(crc, data, sublen); } btrfs_csum_final(crc, csum); if (memcmp(csum, h->csum, state->csum_size)) diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 7619147da382..3903bd3f8d2b 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -26,7 +26,6 @@ #include #include #include -#include #include #include #include @@ -35,6 +34,7 @@ #include #include "ctree.h" #include "disk-io.h" +#include "hash.h" #include "transaction.h" #include "btrfs_inode.h" #include "volumes.h" @@ -244,7 +244,7 @@ static struct extent_map *btree_get_extent(struct inode *inode, u32 btrfs_csum_data(char *data, u32 seed, size_t len) { - return crc32c(seed, data, len); + return btrfs_crc32c(seed, data, len); } void btrfs_csum_final(u32 crc, char *result) diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c index 730dce395858..cf9107a64204 100644 --- a/fs/btrfs/send.c +++ b/fs/btrfs/send.c @@ -24,12 +24,12 @@ #include #include #include -#include #include #include #include "send.h" #include "backref.h" +#include "hash.h" #include "locking.h" #include "disk-io.h" #include "btrfs_inode.h" @@ -620,7 +620,7 @@ static int send_cmd(struct send_ctx *sctx) hdr->len = cpu_to_le32(sctx->send_size - sizeof(*hdr)); hdr->crc = 0; - crc = crc32c(0, (unsigned char *)sctx->send_buf, sctx->send_size); + crc = btrfs_crc32c(0, (unsigned char *)sctx->send_buf, sctx->send_size); hdr->crc = cpu_to_le32(crc); ret = write_buf(sctx->send_filp, sctx->send_buf, sctx->send_size, From 60efa5eb2e886852a0d5f9e1ffa7c896a1099da8 Mon Sep 17 00:00:00 2001 From: Filipe David Borba Manana Date: Sat, 1 Feb 2014 21:27:56 +0000 Subject: [PATCH 125/596] Btrfs: use late_initcall instead of module_init It seems that when init_btrfs_fs() is called, crc32c/crc32c-intel might not always be already initialized, which results in the call to crypto_alloc_shash() returning -ENOENT, as experienced by Ahmet who reported this. Therefore make sure init_btrfs_fs() is called after crc32c is initialized (which is at initialization level 6, module_init), by using late_initcall (which is at initialization level 7) instead of module_init for btrfs. Reported-and-Tested-by: Ahmet Inan Signed-off-by: Filipe David Borba Manana Signed-off-by: Chris Mason --- fs/btrfs/super.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c index c02f63356895..97cc24198554 100644 --- a/fs/btrfs/super.c +++ b/fs/btrfs/super.c @@ -1996,7 +1996,7 @@ static void __exit exit_btrfs_fs(void) btrfs_hash_exit(); } -module_init(init_btrfs_fs) +late_initcall(init_btrfs_fs); module_exit(exit_btrfs_fs) MODULE_LICENSE("GPL"); From c172ec5c8dc8c09dd5958f4ae542fa321bb15a92 Mon Sep 17 00:00:00 2001 From: Ilya Dryomov Date: Fri, 31 Jan 2014 17:49:22 +0200 Subject: [PATCH 126/596] libceph: fix error handling in ceph_osdc_init() msgpool_op_reply message pool isn't destroyed if workqueue construction fails. Fix it. Signed-off-by: Ilya Dryomov Reviewed-by: Sage Weil --- net/ceph/osd_client.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/net/ceph/osd_client.c b/net/ceph/osd_client.c index 010ff3bd58ad..166d4c7ba065 100644 --- a/net/ceph/osd_client.c +++ b/net/ceph/osd_client.c @@ -2504,9 +2504,12 @@ int ceph_osdc_init(struct ceph_osd_client *osdc, struct ceph_client *client) err = -ENOMEM; osdc->notify_wq = create_singlethread_workqueue("ceph-watch-notify"); if (!osdc->notify_wq) - goto out_msgpool; + goto out_msgpool_reply; + return 0; +out_msgpool_reply: + ceph_msgpool_destroy(&osdc->msgpool_op_reply); out_msgpool: ceph_msgpool_destroy(&osdc->msgpool_op); out_mempool: From d4c42fb493e018e9240810bb6dc5334ae0505145 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sun, 2 Feb 2014 14:41:42 -0500 Subject: [PATCH 127/596] NFSv3: Remove unused function nfs3_proc_set_default_acl Cc: Christoph Hellwig Signed-off-by: Trond Myklebust --- fs/nfs/nfs3acl.c | 19 ------------------- 1 file changed, 19 deletions(-) diff --git a/fs/nfs/nfs3acl.c b/fs/nfs/nfs3acl.c index 0851f852568d..9271a6bb9a41 100644 --- a/fs/nfs/nfs3acl.c +++ b/fs/nfs/nfs3acl.c @@ -233,25 +233,6 @@ int nfs3_set_acl(struct inode *inode, struct posix_acl *acl, int type) return PTR_ERR(alloc); } -int nfs3_proc_set_default_acl(struct inode *dir, struct inode *inode, - umode_t mode) -{ - struct posix_acl *default_acl, *acl; - int error; - - error = posix_acl_create(dir, &mode, &default_acl, &acl); - if (error) - return (error == -EOPNOTSUPP) ? 0 : error; - - error = nfs3_proc_setacls(inode, acl, default_acl); - - if (acl) - posix_acl_release(acl); - if (default_acl) - posix_acl_release(default_acl); - return error; -} - const struct xattr_handler *nfs3_xattr_handlers[] = { &posix_acl_access_xattr_handler, &posix_acl_default_xattr_handler, From 8f493b9cfcd8941c6b27d6ce8e3b4a78c094b3c1 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sun, 2 Feb 2014 14:36:42 -0500 Subject: [PATCH 128/596] NFSv3: Fix return value of nfs3_proc_setacls nfs3_proc_setacls is used internally by the NFSv3 create operations to set the acl after the file has been created. If the operation fails because the server doesn't support acls, then it must return '0', not -EOPNOTSUPP. Reported-by: Russell King Link: http://lkml.kernel.org/r/20140201010328.GI15937@n2100.arm.linux.org.uk Cc: Christoph Hellwig Tested-by: Takashi Iwai Signed-off-by: Trond Myklebust --- fs/nfs/nfs3acl.c | 13 +++++++++++-- 1 file changed, 11 insertions(+), 2 deletions(-) diff --git a/fs/nfs/nfs3acl.c b/fs/nfs/nfs3acl.c index 9271a6bb9a41..871d6eda8dba 100644 --- a/fs/nfs/nfs3acl.c +++ b/fs/nfs/nfs3acl.c @@ -113,7 +113,7 @@ struct posix_acl *nfs3_get_acl(struct inode *inode, int type) return ERR_PTR(status); } -int nfs3_proc_setacls(struct inode *inode, struct posix_acl *acl, +static int __nfs3_proc_setacls(struct inode *inode, struct posix_acl *acl, struct posix_acl *dfacl) { struct nfs_server *server = NFS_SERVER(inode); @@ -198,6 +198,15 @@ int nfs3_proc_setacls(struct inode *inode, struct posix_acl *acl, return status; } +int nfs3_proc_setacls(struct inode *inode, struct posix_acl *acl, + struct posix_acl *dfacl) +{ + int ret; + ret = __nfs3_proc_setacls(inode, acl, dfacl); + return (ret == -EOPNOTSUPP) ? 0 : ret; + +} + int nfs3_set_acl(struct inode *inode, struct posix_acl *acl, int type) { struct posix_acl *alloc = NULL, *dfacl = NULL; @@ -225,7 +234,7 @@ int nfs3_set_acl(struct inode *inode, struct posix_acl *acl, int type) if (IS_ERR(alloc)) goto fail; } - status = nfs3_proc_setacls(inode, acl, dfacl); + status = __nfs3_proc_setacls(inode, acl, dfacl); posix_acl_release(alloc); return status; From afca50132cfa7bfc5646676d8c67dc18454f38f8 Mon Sep 17 00:00:00 2001 From: Mukesh Rathor Date: Wed, 29 Jan 2014 16:15:18 -0800 Subject: [PATCH 129/596] xen/pvh: set CR4 flags for APs During bootup in the 'probe_page_size_mask' these CR4 flags are set in there. But for AP processors they are not set as we do not use 'secondary_startup_64' which the baremetal kernels uses. Instead do it in this function which we use in Xen PVH during our startup for AP processors. As such fix it up to make sure we have that flag set. Signed-off-by: Mukesh Rathor Signed-off-by: Konrad Rzeszutek Wilk --- arch/x86/xen/enlighten.c | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c index a4d7b647867f..201d09a7c46b 100644 --- a/arch/x86/xen/enlighten.c +++ b/arch/x86/xen/enlighten.c @@ -1473,6 +1473,18 @@ static void xen_pvh_set_cr_flags(int cpu) * X86_CR0_TS, X86_CR0_PE, X86_CR0_ET are set by Xen for HVM guests * (which PVH shared codepaths), while X86_CR0_PG is for PVH. */ write_cr0(read_cr0() | X86_CR0_MP | X86_CR0_NE | X86_CR0_WP | X86_CR0_AM); + + if (!cpu) + return; + /* + * For BSP, PSE PGE are set in probe_page_size_mask(), for APs + * set them here. For all, OSFXSR OSXMMEXCPT are set in fpu_init. + */ + if (cpu_has_pse) + set_in_cr4(X86_CR4_PSE); + + if (cpu_has_pge) + set_in_cr4(X86_CR4_PGE); } /* From b045d37bd68c20ca88123c2b363cac5e3dae815f Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 3 Feb 2014 12:52:14 -0800 Subject: [PATCH 130/596] ip_tunnel: fix panic in ip_tunnel_xmit() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Setting rt variable to NULL at the beginning of ip_tunnel_xmit() missed possible use of this variable as a scratch value. Also fixes a possible dst leak in tunnel_dst_check() : If we had to call tunnel_dst_reset(), we forgot to release the reference on dst. Merges tunnel_dst_get()/tunnel_dst_check() into a single tunnel_rtable_get() function for clarity. Many thanks to Tommi for his report and tests. Fixes: 7d442fab0a67 ("ipv4: Cache dst in tunnels") Reported-by: Tommi Rantala Signed-off-by: Eric Dumazet Tested-by: Tommi Rantala Cc: Tom Herbert Cc: Maciej Żenczykowski Signed-off-by: David S. Miller --- net/ipv4/ip_tunnel.c | 29 +++++++++++------------------ 1 file changed, 11 insertions(+), 18 deletions(-) diff --git a/net/ipv4/ip_tunnel.c b/net/ipv4/ip_tunnel.c index bd28f386bd02..50228be5c17b 100644 --- a/net/ipv4/ip_tunnel.c +++ b/net/ipv4/ip_tunnel.c @@ -101,28 +101,22 @@ static void tunnel_dst_reset_all(struct ip_tunnel *t) __tunnel_dst_set(per_cpu_ptr(t->dst_cache, i), NULL); } -static struct dst_entry *tunnel_dst_get(struct ip_tunnel *t) +static struct rtable *tunnel_rtable_get(struct ip_tunnel *t, u32 cookie) { struct dst_entry *dst; rcu_read_lock(); dst = rcu_dereference(this_cpu_ptr(t->dst_cache)->dst); - if (dst) + if (dst) { + if (dst->obsolete && dst->ops->check(dst, cookie) == NULL) { + rcu_read_unlock(); + tunnel_dst_reset(t); + return NULL; + } dst_hold(dst); - rcu_read_unlock(); - return dst; -} - -static struct dst_entry *tunnel_dst_check(struct ip_tunnel *t, u32 cookie) -{ - struct dst_entry *dst = tunnel_dst_get(t); - - if (dst && dst->obsolete && dst->ops->check(dst, cookie) == NULL) { - tunnel_dst_reset(t); - return NULL; } - - return dst; + rcu_read_unlock(); + return (struct rtable *)dst; } /* Often modified stats are per cpu, other are shared (netdev->stats) */ @@ -584,7 +578,7 @@ void ip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev, struct flowi4 fl4; u8 tos, ttl; __be16 df; - struct rtable *rt = NULL; /* Route to the other host */ + struct rtable *rt; /* Route to the other host */ unsigned int max_headroom; /* The extra header space needed */ __be32 dst; int err; @@ -657,8 +651,7 @@ void ip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev, init_tunnel_flow(&fl4, protocol, dst, tnl_params->saddr, tunnel->parms.o_key, RT_TOS(tos), tunnel->parms.link); - if (connected) - rt = (struct rtable *)tunnel_dst_check(tunnel, 0); + rt = connected ? tunnel_rtable_get(tunnel, 0) : NULL; if (!rt) { rt = ip_route_output_key(tunnel->net, &fl4); From 2d7c1b77dd59387070aab355532dd157f888325c Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Mon, 3 Feb 2014 02:22:07 +0100 Subject: [PATCH 131/596] ACPI / hotplug / PCI: Remove entries from bus->devices in reverse order According to the changelog of commit 29ed1f29b68a (PCI: pciehp: Fix null pointer deref when hot-removing SR-IOV device) it is unsafe to walk the bus->devices list of a PCI bus and remove devices from it in direct order, because that may lead to NULL pointer dereferences related to virtual functions. For this reason, change all of the bus->devices list walks in acpiphp_glue.c during which devices may be removed to be carried out in reverse order. Signed-off-by: Rafael J. Wysocki Tested-by: Mika Westerberg --- drivers/pci/hotplug/acpiphp_glue.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/pci/hotplug/acpiphp_glue.c b/drivers/pci/hotplug/acpiphp_glue.c index cd929aed3613..6a4b4b734fbd 100644 --- a/drivers/pci/hotplug/acpiphp_glue.c +++ b/drivers/pci/hotplug/acpiphp_glue.c @@ -742,7 +742,7 @@ static void trim_stale_devices(struct pci_dev *dev) /* The device is a bridge. so check the bus below it. */ pm_runtime_get_sync(&dev->dev); - list_for_each_entry_safe(child, tmp, &bus->devices, bus_list) + list_for_each_entry_safe_reverse(child, tmp, &bus->devices, bus_list) trim_stale_devices(child); pm_runtime_put(&dev->dev); @@ -773,8 +773,8 @@ static void acpiphp_check_bridge(struct acpiphp_bridge *bridge) ; /* do nothing */ } else if (get_slot_status(slot) == ACPI_STA_ALL) { /* remove stale devices if any */ - list_for_each_entry_safe(dev, tmp, &bus->devices, - bus_list) + list_for_each_entry_safe_reverse(dev, tmp, + &bus->devices, bus_list) if (PCI_SLOT(dev->devfn) == slot->device) trim_stale_devices(dev); @@ -805,7 +805,7 @@ static void acpiphp_sanitize_bus(struct pci_bus *bus) int i; unsigned long type_mask = IORESOURCE_IO | IORESOURCE_MEM; - list_for_each_entry_safe(dev, tmp, &bus->devices, bus_list) { + list_for_each_entry_safe_reverse(dev, tmp, &bus->devices, bus_list) { for (i=0; iresource[i]; if ((res->flags & type_mask) && !res->start && From f41b32613138ae05329a0f0e7170223b775d6b24 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Mon, 3 Feb 2014 02:22:17 +0100 Subject: [PATCH 132/596] ACPI / hotplug / PCI: Move PCI rescan-remove locking to hotplug_event() Commit 9217a984671e (ACPI / hotplug / PCI: Use global PCI rescan-remove locking) modified ACPIPHP to protect its PCI device removal and addition code paths from races against sysfs-driven rescan and remove operations with the help of PCI rescan-remove locking. However, it overlooked the fact that hotplug_event_work() is not the only caller of hotplug_event() which may also be called by dock_hotplug_event() and that code path is missing the PCI rescan-remove locking. This means that, although the PCI rescan-remove lock is held as appropriate during the handling of events originating from handle_hotplug_event(), the ACPIPHP's operations resulting from dock events may still suffer the race conditions that commit 9217a984671e was supposed to eliminate. To address that problem, move the PCI rescan-remove locking from hotplug_event_work() to hotplug_event() so that it is used regardless of the way that function is invoked. Revamps: 9217a984671e (ACPI / hotplug / PCI: Use global PCI rescan-remove locking) Signed-off-by: Rafael J. Wysocki Tested-by: Mika Westerberg --- drivers/pci/hotplug/acpiphp_glue.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/pci/hotplug/acpiphp_glue.c b/drivers/pci/hotplug/acpiphp_glue.c index 6a4b4b734fbd..6e5bd79af810 100644 --- a/drivers/pci/hotplug/acpiphp_glue.c +++ b/drivers/pci/hotplug/acpiphp_glue.c @@ -852,6 +852,7 @@ static void hotplug_event(acpi_handle handle, u32 type, void *data) mutex_unlock(&acpiphp_context_lock); + pci_lock_rescan_remove(); acpi_get_name(handle, ACPI_FULL_PATHNAME, &buffer); switch (type) { @@ -905,6 +906,7 @@ static void hotplug_event(acpi_handle handle, u32 type, void *data) break; } + pci_unlock_rescan_remove(); if (bridge) put_bridge(bridge); } @@ -915,11 +917,9 @@ static void hotplug_event_work(void *data, u32 type) acpi_handle handle = context->handle; acpi_scan_lock_acquire(); - pci_lock_rescan_remove(); hotplug_event(handle, type, context); - pci_unlock_rescan_remove(); acpi_scan_lock_release(); acpi_evaluate_hotplug_ost(handle, type, ACPI_OST_SC_SUCCESS, NULL); put_bridge(context->func.parent); From d42f5da2340083301dd2c48ff2d75f6ce4b30767 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Mon, 3 Feb 2014 02:22:27 +0100 Subject: [PATCH 133/596] ACPI / hotplug / PCI: Scan root bus under the PCI rescan-remove lock Since acpiphp_check_bridge() called by acpiphp_check_host_bridge() does things that require PCI rescan-remove locking around it, make acpiphp_check_host_bridge() use that locking. Signed-off-by: Rafael J. Wysocki Tested-by: Mika Westerberg --- drivers/pci/hotplug/acpiphp_glue.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/pci/hotplug/acpiphp_glue.c b/drivers/pci/hotplug/acpiphp_glue.c index 6e5bd79af810..931d0b44eace 100644 --- a/drivers/pci/hotplug/acpiphp_glue.c +++ b/drivers/pci/hotplug/acpiphp_glue.c @@ -829,7 +829,11 @@ void acpiphp_check_host_bridge(acpi_handle handle) bridge = acpiphp_handle_to_bridge(handle); if (bridge) { + pci_lock_rescan_remove(); + acpiphp_check_bridge(bridge); + + pci_unlock_rescan_remove(); put_bridge(bridge); } } From 1b360f44d009059e446532f29c1a889951e72667 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Mon, 3 Feb 2014 22:30:06 +0100 Subject: [PATCH 134/596] ACPI / hotplug / PCI: Fix bridge removal race in handle_hotplug_event() If a PCI bridge with an ACPIPHP context attached is removed via sysfs, the code path executed as a result is the following: pci_stop_and_remove_bus_device_locked pci_remove_bus pcibios_remove_bus acpi_pci_remove_bus acpiphp_remove_slots cleanup_bridge put_bridge free_bridge acpiphp_put_context (for each child, under context lock) kfree (child context) Now, if a hotplug notify is dispatched for one of the bridge's children and the timing is such that handle_hotplug_event() for that notify is executed while free_bridge() above is running, the get_bridge(context->func.parent) in handle_hotplug_event() will not really help, because it is too late to prevent the bridge from going away and the child's context may be freed before hotplug_event_work() scheduled from handle_hotplug_event() dereferences the pointer to it passed via the data argument. That will cause a kernel crash to happpen in hotplug_event_work(). To prevent that from happening, make handle_hotplug_event() check the is_going_away flag of the function's parent bridge (under acpiphp_context_lock) and bail out if it's set. Also, make cleanup_bridge() set the bridge's is_going_away flag under acpiphp_context_lock so that it cannot be changed between the check and the subsequent get_bridge(context->func.parent) in handle_hotplug_event(). Then, in the above scenario, handle_hotplug_event() will notice that context->func.parent->is_going_away is already set and it will exit immediately preventing the crash from happening. Signed-off-by: Rafael J. Wysocki --- drivers/pci/hotplug/acpiphp_glue.c | 18 ++++++++++++++---- 1 file changed, 14 insertions(+), 4 deletions(-) diff --git a/drivers/pci/hotplug/acpiphp_glue.c b/drivers/pci/hotplug/acpiphp_glue.c index 931d0b44eace..91eceaf3131b 100644 --- a/drivers/pci/hotplug/acpiphp_glue.c +++ b/drivers/pci/hotplug/acpiphp_glue.c @@ -441,7 +441,9 @@ static void cleanup_bridge(struct acpiphp_bridge *bridge) list_del(&bridge->list); mutex_unlock(&bridge_mutex); + mutex_lock(&acpiphp_context_lock); bridge->is_going_away = true; + mutex_unlock(&acpiphp_context_lock); } /** @@ -941,6 +943,7 @@ static void handle_hotplug_event(acpi_handle handle, u32 type, void *data) { struct acpiphp_context *context; u32 ost_code = ACPI_OST_SC_SUCCESS; + acpi_status status; switch (type) { case ACPI_NOTIFY_BUS_CHECK: @@ -976,13 +979,20 @@ static void handle_hotplug_event(acpi_handle handle, u32 type, void *data) mutex_lock(&acpiphp_context_lock); context = acpiphp_get_context(handle); - if (context && !WARN_ON(context->handle != handle)) { - get_bridge(context->func.parent); - acpiphp_put_context(context); - acpi_hotplug_execute(hotplug_event_work, context, type); + if (!context || WARN_ON(context->handle != handle) + || context->func.parent->is_going_away) + goto err_out; + + get_bridge(context->func.parent); + acpiphp_put_context(context); + status = acpi_hotplug_execute(hotplug_event_work, context, type); + if (ACPI_SUCCESS(status)) { mutex_unlock(&acpiphp_context_lock); return; } + put_bridge(context->func.parent); + + err_out: mutex_unlock(&acpiphp_context_lock); ost_code = ACPI_OST_SC_NON_SPECIFIC_FAILURE; From af9d8adc6b832003bbe3d83fde665ae6b4f072eb Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Mon, 3 Feb 2014 22:30:15 +0100 Subject: [PATCH 135/596] ACPI / hotplug / PCI: Fix bridge removal race vs dock events If a PCI bridge with an ACPIPHP context attached is removed via sysfs, the code path executed as a result is the following: pci_stop_and_remove_bus_device_locked pci_remove_bus pcibios_remove_bus acpi_pci_remove_bus acpiphp_remove_slots cleanup_bridge unregister_hotplug_dock_device (drops dock references to the bridge) put_bridge free_bridge acpiphp_put_context (for each child, under context lock) kfree (context) Now, if a dock event affecting one of the bridge's child devices occurs (roughly at the same time), it will lead to the following code path: acpi_dock_deferred_cb dock_notify handle_eject_request hot_remove_dock_devices dock_hotplug_event hotplug_event (dereferences context) That may lead to a kernel crash in hotplug_event() if it is executed after the last kfree() in the bridge removal code path. To prevent that from happening, add a wrapper around hotplug_event() called dock_event() and point the .handler pointer in acpiphp_dock_ops to it. Make that wrapper retrieve the device's ACPIPHP context using acpiphp_get_context() (instead of taking it from the data argument) under acpiphp_context_lock and check if the parent bridge's is_going_away flag is set. If that flag is set, it will return immediately and if it is not set it will grab a reference to the device's parent bridge before executing hotplug_event(). Then, in the above scenario, the reference to the parent bridge held by dock_event() will prevent free_bridge() from being executed for it until hotplug_event() returns. Signed-off-by: Rafael J. Wysocki --- drivers/pci/hotplug/acpiphp_glue.c | 21 ++++++++++++++++++++- 1 file changed, 20 insertions(+), 1 deletion(-) diff --git a/drivers/pci/hotplug/acpiphp_glue.c b/drivers/pci/hotplug/acpiphp_glue.c index 91eceaf3131b..e2a783fdb98f 100644 --- a/drivers/pci/hotplug/acpiphp_glue.c +++ b/drivers/pci/hotplug/acpiphp_glue.c @@ -210,10 +210,29 @@ static void post_dock_fixups(acpi_handle not_used, u32 event, void *data) } } +static void dock_event(acpi_handle handle, u32 type, void *data) +{ + struct acpiphp_context *context; + + mutex_lock(&acpiphp_context_lock); + context = acpiphp_get_context(handle); + if (!context || WARN_ON(context->handle != handle) + || context->func.parent->is_going_away) { + mutex_unlock(&acpiphp_context_lock); + return; + } + get_bridge(context->func.parent); + acpiphp_put_context(context); + mutex_unlock(&acpiphp_context_lock); + + hotplug_event(handle, type, data); + + put_bridge(context->func.parent); +} static const struct acpi_dock_ops acpiphp_dock_ops = { .fixup = post_dock_fixups, - .handler = hotplug_event, + .handler = dock_event, }; /* Check whether the PCI device is managed by native PCIe hotplug driver */ From 789b663ae3d427ea9c50505339a13276e7228c9d Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Fri, 31 Jan 2014 14:25:19 -0500 Subject: [PATCH 136/596] fs: get_acl() must be allowed to return EOPNOTSUPP posix_acl_xattr_get requires get_acl() to return EOPNOTSUPP if the filesystem cannot support acls. This is needed for NFS, which can't know whether or not the server supports acls until it tries to get/set one. This patch converts posix_acl_chmod and posix_acl_create to deal with EOPNOTSUPP return values from get_acl(). Reported-by: Russell King Link: http://lkml.kernel.org/r/20140130140834.GW15937@n2100.arm.linux.org.uk Cc: Al Viro viro@zeniv.linux.org.uk> Reviewed-by: Christoph Hellwig Tested-by: Takashi Iwai Signed-off-by: Trond Myklebust --- fs/posix_acl.c | 18 ++++++++++++------ 1 file changed, 12 insertions(+), 6 deletions(-) diff --git a/fs/posix_acl.c b/fs/posix_acl.c index 38bae5a0ea25..11c54fd51e16 100644 --- a/fs/posix_acl.c +++ b/fs/posix_acl.c @@ -521,8 +521,11 @@ posix_acl_chmod(struct inode *inode, umode_t mode) return -EOPNOTSUPP; acl = get_acl(inode, ACL_TYPE_ACCESS); - if (IS_ERR_OR_NULL(acl)) + if (IS_ERR_OR_NULL(acl)) { + if (acl == ERR_PTR(-EOPNOTSUPP)) + return 0; return PTR_ERR(acl); + } ret = __posix_acl_chmod(&acl, GFP_KERNEL, mode); if (ret) @@ -544,14 +547,15 @@ posix_acl_create(struct inode *dir, umode_t *mode, goto no_acl; p = get_acl(dir, ACL_TYPE_DEFAULT); - if (IS_ERR(p)) + if (IS_ERR(p)) { + if (p == ERR_PTR(-EOPNOTSUPP)) + goto apply_umask; return PTR_ERR(p); - - if (!p) { - *mode &= ~current_umask(); - goto no_acl; } + if (!p) + goto apply_umask; + *acl = posix_acl_clone(p, GFP_NOFS); if (!*acl) return -ENOMEM; @@ -575,6 +579,8 @@ posix_acl_create(struct inode *dir, umode_t *mode, } return 0; +apply_umask: + *mode &= ~current_umask(); no_acl: *default_acl = NULL; *acl = NULL; From 4528eb19b00d9ccd65ded6f8201eec704267edd8 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Tue, 4 Feb 2014 07:39:06 +0100 Subject: [PATCH 137/596] ALSA: hda - Fix silent output on Toshiba Satellite L40 Toshiba Satellite L40 with AD1986A codec requires the EAPD of NID 0x1b to be constantly on, otherwise the output doesn't work. Unlike most of other AD1986A machines, EAPD is correctly implemented in HD-audio manner (that is, bit set = amp on), so we need to clear the inv_eapd flag in the fixup, too. Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=67481 Cc: [v3.11+] Signed-off-by: Takashi Iwai --- sound/pci/hda/patch_analog.c | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+) diff --git a/sound/pci/hda/patch_analog.c b/sound/pci/hda/patch_analog.c index 7a426ed491f2..50b2427f19ca 100644 --- a/sound/pci/hda/patch_analog.c +++ b/sound/pci/hda/patch_analog.c @@ -244,6 +244,19 @@ static void ad_fixup_inv_jack_detect(struct hda_codec *codec, } } +/* Toshiba Satellite L40 implements EAPD in a standard way unlike others */ +static void ad1986a_fixup_eapd(struct hda_codec *codec, + const struct hda_fixup *fix, int action) +{ + struct ad198x_spec *spec = codec->spec; + + if (action == HDA_FIXUP_ACT_PRE_PROBE) { + codec->inv_eapd = 0; + spec->gen.keep_eapd_on = 1; + spec->eapd_nid = 0x1b; + } +} + enum { AD1986A_FIXUP_INV_JACK_DETECT, AD1986A_FIXUP_ULTRA, @@ -251,6 +264,7 @@ enum { AD1986A_FIXUP_3STACK, AD1986A_FIXUP_LAPTOP, AD1986A_FIXUP_LAPTOP_IMIC, + AD1986A_FIXUP_EAPD, }; static const struct hda_fixup ad1986a_fixups[] = { @@ -311,6 +325,10 @@ static const struct hda_fixup ad1986a_fixups[] = { .chained_before = 1, .chain_id = AD1986A_FIXUP_LAPTOP, }, + [AD1986A_FIXUP_EAPD] = { + .type = HDA_FIXUP_FUNC, + .v.func = ad1986a_fixup_eapd, + }, }; static const struct snd_pci_quirk ad1986a_fixup_tbl[] = { @@ -318,6 +336,7 @@ static const struct snd_pci_quirk ad1986a_fixup_tbl[] = { SND_PCI_QUIRK_MASK(0x1043, 0xff00, 0x8100, "ASUS P5", AD1986A_FIXUP_3STACK), SND_PCI_QUIRK_MASK(0x1043, 0xff00, 0x8200, "ASUS M2", AD1986A_FIXUP_3STACK), SND_PCI_QUIRK(0x10de, 0xcb84, "ASUS A8N-VM", AD1986A_FIXUP_3STACK), + SND_PCI_QUIRK(0x1179, 0xff40, "Toshiba Satellite L40", AD1986A_FIXUP_EAPD), SND_PCI_QUIRK(0x144d, 0xc01e, "FSC V2060", AD1986A_FIXUP_LAPTOP), SND_PCI_QUIRK_MASK(0x144d, 0xff00, 0xc000, "Samsung", AD1986A_FIXUP_SAMSUNG), SND_PCI_QUIRK(0x144d, 0xc027, "Samsung Q1", AD1986A_FIXUP_ULTRA), From 7e8f15c5aa9b8021ee933336f3e055f279482051 Mon Sep 17 00:00:00 2001 From: Andrzej Hajda Date: Thu, 9 Jan 2014 09:55:43 -0300 Subject: [PATCH 138/596] [media] s5k5baf: allow to handle arbitrary long i2c sequences Using variable length array in s5k5baf_write_arr_seq caused an implicit assumption that i2c sequences should be short. The patch rewrites the function so it can handle sequences of any length and does not use variable length array. Reported-by: Dan Carpenter Signed-off-by: Andrzej Hajda Signed-off-by: Sylwester Nawrocki Signed-off-by: Mauro Carvalho Chehab --- drivers/media/i2c/s5k5baf.c | 32 ++++++++++++++++++++------------ 1 file changed, 20 insertions(+), 12 deletions(-) diff --git a/drivers/media/i2c/s5k5baf.c b/drivers/media/i2c/s5k5baf.c index 4b8381111cbd..77e10e0fd8d6 100644 --- a/drivers/media/i2c/s5k5baf.c +++ b/drivers/media/i2c/s5k5baf.c @@ -478,25 +478,33 @@ static void s5k5baf_write_arr_seq(struct s5k5baf *state, u16 addr, u16 count, const u16 *seq) { struct i2c_client *c = v4l2_get_subdevdata(&state->sd); - __be16 buf[count + 1]; - int ret, n; + __be16 buf[65]; s5k5baf_i2c_write(state, REG_CMDWR_ADDR, addr); if (state->error) return; - buf[0] = __constant_cpu_to_be16(REG_CMD_BUF); - for (n = 1; n <= count; ++n) - buf[n] = cpu_to_be16(*seq++); - - n *= 2; - ret = i2c_master_send(c, (char *)buf, n); v4l2_dbg(3, debug, c, "i2c_write_seq(count=%d): %*ph\n", count, - min(2 * count, 64), seq - count); + min(2 * count, 64), seq); - if (ret != n) { - v4l2_err(c, "i2c_write_seq: error during transfer (%d)\n", ret); - state->error = ret; + buf[0] = __constant_cpu_to_be16(REG_CMD_BUF); + + while (count > 0) { + int n = min_t(int, count, ARRAY_SIZE(buf) - 1); + int ret, i; + + for (i = 1; i <= n; ++i) + buf[i] = cpu_to_be16(*seq++); + + i *= 2; + ret = i2c_master_send(c, (char *)buf, i); + if (ret != i) { + v4l2_err(c, "i2c_write_seq: error during transfer (%d)\n", ret); + state->error = ret; + break; + } + + count -= n; } } From a62cffefc9a327a5bc45ba9318ef601b525f4bd1 Mon Sep 17 00:00:00 2001 From: Jacek Anaszewski Date: Thu, 16 Jan 2014 08:26:33 -0300 Subject: [PATCH 139/596] [media] s5p-jpeg: Fix wrong NV12 format parameters NV12 format entries in the sjpeg_formats array had wrong colplanes, depth and v_align values. Signed-off-by: Jacek Anaszewski Signed-off-by: Kyungmin Park Signed-off-by: Sylwester Nawrocki Signed-off-by: Mauro Carvalho Chehab --- drivers/media/platform/s5p-jpeg/jpeg-core.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/media/platform/s5p-jpeg/jpeg-core.c b/drivers/media/platform/s5p-jpeg/jpeg-core.c index a1c78c870b68..7d68d0b9966a 100644 --- a/drivers/media/platform/s5p-jpeg/jpeg-core.c +++ b/drivers/media/platform/s5p-jpeg/jpeg-core.c @@ -175,7 +175,7 @@ static struct s5p_jpeg_fmt sjpeg_formats[] = { { .name = "YUV 4:2:0 planar, Y/CbCr", .fourcc = V4L2_PIX_FMT_NV12, - .depth = 16, + .depth = 12, .colplanes = 2, .h_align = 1, .v_align = 1, @@ -188,10 +188,10 @@ static struct s5p_jpeg_fmt sjpeg_formats[] = { { .name = "YUV 4:2:0 planar, Y/CbCr", .fourcc = V4L2_PIX_FMT_NV12, - .depth = 16, - .colplanes = 4, + .depth = 12, + .colplanes = 2, .h_align = 4, - .v_align = 1, + .v_align = 4, .flags = SJPEG_FMT_FLAG_ENC_OUTPUT | SJPEG_FMT_FLAG_DEC_CAPTURE | SJPEG_FMT_FLAG_S5P | From a27a19d61535ce61a1e2f9274f99316c7532d5ef Mon Sep 17 00:00:00 2001 From: Sylwester Nawrocki Date: Tue, 7 Jan 2014 19:07:52 -0300 Subject: [PATCH 140/596] [media] exynos4-is: Fix error paths in probe() for !pm_runtime_enabled() Ensure clk_disable() is called on error paths only when clk_enable() was previously called. This fixes following build warning: .../media-git/drivers/media/platform/exynos4-is/fimc-lite.c: In function 'fimc_lite_probe': .../media-git/drivers/media/platform/exynos4-is/fimc-lite.c:1583:1: warning: label 'err_sd' defined but not used [-Wunused-label] Reported-by: Hans Verkuil Signed-off-by: Sylwester Nawrocki Signed-off-by: Mauro Carvalho Chehab --- drivers/media/platform/exynos4-is/fimc-core.c | 3 ++- drivers/media/platform/exynos4-is/fimc-lite.c | 5 +++-- 2 files changed, 5 insertions(+), 3 deletions(-) diff --git a/drivers/media/platform/exynos4-is/fimc-core.c b/drivers/media/platform/exynos4-is/fimc-core.c index a7dfd07e8389..dcbea59c8c69 100644 --- a/drivers/media/platform/exynos4-is/fimc-core.c +++ b/drivers/media/platform/exynos4-is/fimc-core.c @@ -1027,7 +1027,8 @@ static int fimc_probe(struct platform_device *pdev) return 0; err_gclk: - clk_disable(fimc->clock[CLK_GATE]); + if (!pm_runtime_enabled(dev)) + clk_disable(fimc->clock[CLK_GATE]); err_sd: fimc_unregister_capture_subdev(fimc); err_sclk: diff --git a/drivers/media/platform/exynos4-is/fimc-lite.c b/drivers/media/platform/exynos4-is/fimc-lite.c index 1234734bccf4..5213ff03d28a 100644 --- a/drivers/media/platform/exynos4-is/fimc-lite.c +++ b/drivers/media/platform/exynos4-is/fimc-lite.c @@ -1563,7 +1563,7 @@ static int fimc_lite_probe(struct platform_device *pdev) if (!pm_runtime_enabled(dev)) { ret = clk_enable(fimc->clock); if (ret < 0) - goto err_clk_put; + goto err_sd; } fimc->alloc_ctx = vb2_dma_contig_init_ctx(dev); @@ -1579,7 +1579,8 @@ static int fimc_lite_probe(struct platform_device *pdev) return 0; err_clk_dis: - clk_disable(fimc->clock); + if (!pm_runtime_enabled(dev)) + clk_disable(fimc->clock); err_sd: fimc_lite_unregister_capture_subdev(fimc); err_clk_put: From d003a30dd7401f38c876bb97e3e51f4fd9fed15e Mon Sep 17 00:00:00 2001 From: Sylwester Nawrocki Date: Tue, 7 Jan 2014 19:08:48 -0300 Subject: [PATCH 141/596] [media] exynos4-is: Compile in fimc runtime PM callbacks conditionally MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Enclose the runtime PM helpers in #ifdef CONFIG_PM_RUNTIME/#endif to avoid following compile warning when CONFIG_PM_RUNTIME is disabled: CC drivers/media/platform/exynos4-is/fimc-core.o drivers/media/platform/exynos4-is/fimc-core.c:1040:12: warning: ‘fimc_runtime_resume’ defined but not used drivers/media/platform/exynos4-is/fimc-core.c:1057:12: warning: ‘fimc_runtime_suspend’ defined but not used Signed-off-by: Sylwester Nawrocki Reviewed-by: Jingoo Han Signed-off-by: Mauro Carvalho Chehab --- drivers/media/platform/exynos4-is/fimc-core.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/media/platform/exynos4-is/fimc-core.c b/drivers/media/platform/exynos4-is/fimc-core.c index dcbea59c8c69..da2fc86cc524 100644 --- a/drivers/media/platform/exynos4-is/fimc-core.c +++ b/drivers/media/platform/exynos4-is/fimc-core.c @@ -1037,6 +1037,7 @@ static int fimc_probe(struct platform_device *pdev) return ret; } +#ifdef CONFIG_PM_RUNTIME static int fimc_runtime_resume(struct device *dev) { struct fimc_dev *fimc = dev_get_drvdata(dev); @@ -1069,6 +1070,7 @@ static int fimc_runtime_suspend(struct device *dev) dbg("fimc%d: state: 0x%lx", fimc->id, fimc->state); return ret; } +#endif #ifdef CONFIG_PM_SLEEP static int fimc_resume(struct device *dev) From 656e62dc84a38fa847d9501636acb098a32d6f89 Mon Sep 17 00:00:00 2001 From: Sylwester Nawrocki Date: Thu, 16 Jan 2014 11:49:28 -0300 Subject: [PATCH 142/596] [media] exynos4-is: Compile in fimc-lite runtime PM callbacks conditionally MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Enclose the runtime PM helpers in #ifdef CONFIG_PM_RUNTIME/#endif to avoid following compile warning when CONFIG_PM_RUNTIME is disabled: CC drivers/media/platform/exynos4-is/fimc-lite.o drivers/media/platform/exynos4-is/fimc-lite.c:1591:12: warning: ‘fimc_lite_runtime_resume’ defined but not used [-Wunused-function] drivers/media/platform/exynos4-is/fimc-lite.c:1599:12: warning: ‘fimc_lite_runtime_suspend’ defined but not used [-Wunused-function] Signed-off-by: Sylwester Nawrocki Signed-off-by: Mauro Carvalho Chehab --- drivers/media/platform/exynos4-is/fimc-lite.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/media/platform/exynos4-is/fimc-lite.c b/drivers/media/platform/exynos4-is/fimc-lite.c index 5213ff03d28a..779ec3cd259d 100644 --- a/drivers/media/platform/exynos4-is/fimc-lite.c +++ b/drivers/media/platform/exynos4-is/fimc-lite.c @@ -1588,6 +1588,7 @@ static int fimc_lite_probe(struct platform_device *pdev) return ret; } +#ifdef CONFIG_PM_RUNTIME static int fimc_lite_runtime_resume(struct device *dev) { struct fimc_lite *fimc = dev_get_drvdata(dev); @@ -1603,6 +1604,7 @@ static int fimc_lite_runtime_suspend(struct device *dev) clk_disable(fimc->clock); return 0; } +#endif #ifdef CONFIG_PM_SLEEP static int fimc_lite_resume(struct device *dev) From 38121b6ef314e556d3b59acc0c41a4d6f7e597fa Mon Sep 17 00:00:00 2001 From: Levente Kurusa Date: Thu, 19 Dec 2013 12:06:49 -0300 Subject: [PATCH 143/596] [media] media: bt8xx: add missing put_device call This is required so that we give up the last reference to the device. Remove the kfree() because the put_device() call will actually call release_sub_device which in turn kfrees the device. Signed-off-by: Levente Kurusa Signed-off-by: Hans Verkuil Signed-off-by: Mauro Carvalho Chehab --- drivers/media/pci/bt8xx/bttv-gpio.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/media/pci/bt8xx/bttv-gpio.c b/drivers/media/pci/bt8xx/bttv-gpio.c index 922e8233fd0b..3f364b7062b9 100644 --- a/drivers/media/pci/bt8xx/bttv-gpio.c +++ b/drivers/media/pci/bt8xx/bttv-gpio.c @@ -98,7 +98,7 @@ int bttv_sub_add_device(struct bttv_core *core, char *name) err = device_register(&sub->dev); if (0 != err) { - kfree(sub); + put_device(&sub->dev); return err; } pr_info("%d: add subdevice \"%s\"\n", core->nr, dev_name(&sub->dev)); From 50c88544d225baadd6de1c4365d4ed16cc942a37 Mon Sep 17 00:00:00 2001 From: Alexey Khoroshilov Date: Fri, 20 Dec 2013 16:17:32 -0300 Subject: [PATCH 144/596] [media] go7007-loader: fix usb_dev leak There is usb_get_dev() in go7007_loader_probe(), but there is no usb_put_dev() anywhere. Found by Linux Driver Verification project (linuxtesting.org). Signed-off-by: Alexey Khoroshilov Signed-off-by: Hans Verkuil Signed-off-by: Mauro Carvalho Chehab --- drivers/staging/media/go7007/go7007-loader.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/staging/media/go7007/go7007-loader.c b/drivers/staging/media/go7007/go7007-loader.c index 10bb41c2fb6d..eecb1f2a5574 100644 --- a/drivers/staging/media/go7007/go7007-loader.c +++ b/drivers/staging/media/go7007/go7007-loader.c @@ -59,7 +59,7 @@ static int go7007_loader_probe(struct usb_interface *interface, if (usbdev->descriptor.bNumConfigurations != 1) { dev_err(&interface->dev, "can't handle multiple config\n"); - return -ENODEV; + goto failed2; } vendor = le16_to_cpu(usbdev->descriptor.idVendor); @@ -108,6 +108,7 @@ static int go7007_loader_probe(struct usb_interface *interface, return 0; failed2: + usb_put_dev(usbdev); dev_err(&interface->dev, "probe failed\n"); return -ENODEV; } @@ -115,6 +116,7 @@ static int go7007_loader_probe(struct usb_interface *interface, static void go7007_loader_disconnect(struct usb_interface *interface) { dev_info(&interface->dev, "disconnect\n"); + usb_put_dev(interface_to_usbdev(interface)); usb_set_intfdata(interface, NULL); } From cca36e2eecec2b8fc869a50ffd3bd0adeed92b8b Mon Sep 17 00:00:00 2001 From: Hans Verkuil Date: Fri, 3 Jan 2014 08:10:49 -0300 Subject: [PATCH 145/596] [media] Revert "[media] videobuf_vm_{open,close} race fixes" This reverts commit a242f426108c284049a69710f871cc9f11b13e61. That commit actually caused deadlocks, rather then fixing them. If ext_lock is set to NULL (otherwise videobuf_queue_lock doesn't do anything), then you get this deadlock: The driver's mmap function calls videobuf_mmap_mapper which calls videobuf_queue_lock on q. videobuf_mmap_mapper calls __videobuf_mmap_mapper, __videobuf_mmap_mapper calls videobuf_vm_open and videobuf_vm_open calls videobuf_queue_lock on q (introduced by above patch): deadlocked. This affects drivers using dma-contig and dma-vmalloc. Only dma-sg is not affected since it doesn't call videobuf_vm_open from __videobuf_mmap_mapper. Most drivers these days have a non-NULL ext_lock. Those that still use NULL there are all fairly obscure drivers, which is why this hasn't been seen earlier. Since everything worked perfectly fine for many years I prefer to just revert this patch rather than trying to fix it. videobuf is quite fragile and I rather not touch it too much. Work is (slowly) progressing to move everything over to vb2 or at the very least use non-NULL ext_lock in videobuf. Signed-off-by: Hans Verkuil Cc: # for v3.11 and up Cc: Al Viro Reported-by: Pete Eberlein Signed-off-by: Mauro Carvalho Chehab --- drivers/media/v4l2-core/videobuf-dma-contig.c | 12 +++++------- drivers/media/v4l2-core/videobuf-dma-sg.c | 10 ++++------ drivers/media/v4l2-core/videobuf-vmalloc.c | 10 ++++------ 3 files changed, 13 insertions(+), 19 deletions(-) diff --git a/drivers/media/v4l2-core/videobuf-dma-contig.c b/drivers/media/v4l2-core/videobuf-dma-contig.c index 65411adcd0ea..7e6b209b7002 100644 --- a/drivers/media/v4l2-core/videobuf-dma-contig.c +++ b/drivers/media/v4l2-core/videobuf-dma-contig.c @@ -66,14 +66,11 @@ static void __videobuf_dc_free(struct device *dev, static void videobuf_vm_open(struct vm_area_struct *vma) { struct videobuf_mapping *map = vma->vm_private_data; - struct videobuf_queue *q = map->q; - dev_dbg(q->dev, "vm_open %p [count=%u,vma=%08lx-%08lx]\n", + dev_dbg(map->q->dev, "vm_open %p [count=%u,vma=%08lx-%08lx]\n", map, map->count, vma->vm_start, vma->vm_end); - videobuf_queue_lock(q); map->count++; - videobuf_queue_unlock(q); } static void videobuf_vm_close(struct vm_area_struct *vma) @@ -85,11 +82,12 @@ static void videobuf_vm_close(struct vm_area_struct *vma) dev_dbg(q->dev, "vm_close %p [count=%u,vma=%08lx-%08lx]\n", map, map->count, vma->vm_start, vma->vm_end); - videobuf_queue_lock(q); - if (!--map->count) { + map->count--; + if (0 == map->count) { struct videobuf_dma_contig_memory *mem; dev_dbg(q->dev, "munmap %p q=%p\n", map, q); + videobuf_queue_lock(q); /* We need first to cancel streams, before unmapping */ if (q->streaming) @@ -128,8 +126,8 @@ static void videobuf_vm_close(struct vm_area_struct *vma) kfree(map); + videobuf_queue_unlock(q); } - videobuf_queue_unlock(q); } static const struct vm_operations_struct videobuf_vm_ops = { diff --git a/drivers/media/v4l2-core/videobuf-dma-sg.c b/drivers/media/v4l2-core/videobuf-dma-sg.c index 9db674ccdc68..828e7c10bd70 100644 --- a/drivers/media/v4l2-core/videobuf-dma-sg.c +++ b/drivers/media/v4l2-core/videobuf-dma-sg.c @@ -338,14 +338,11 @@ EXPORT_SYMBOL_GPL(videobuf_dma_free); static void videobuf_vm_open(struct vm_area_struct *vma) { struct videobuf_mapping *map = vma->vm_private_data; - struct videobuf_queue *q = map->q; dprintk(2, "vm_open %p [count=%d,vma=%08lx-%08lx]\n", map, map->count, vma->vm_start, vma->vm_end); - videobuf_queue_lock(q); map->count++; - videobuf_queue_unlock(q); } static void videobuf_vm_close(struct vm_area_struct *vma) @@ -358,9 +355,10 @@ static void videobuf_vm_close(struct vm_area_struct *vma) dprintk(2, "vm_close %p [count=%d,vma=%08lx-%08lx]\n", map, map->count, vma->vm_start, vma->vm_end); - videobuf_queue_lock(q); - if (!--map->count) { + map->count--; + if (0 == map->count) { dprintk(1, "munmap %p q=%p\n", map, q); + videobuf_queue_lock(q); for (i = 0; i < VIDEO_MAX_FRAME; i++) { if (NULL == q->bufs[i]) continue; @@ -376,9 +374,9 @@ static void videobuf_vm_close(struct vm_area_struct *vma) q->bufs[i]->baddr = 0; q->ops->buf_release(q, q->bufs[i]); } + videobuf_queue_unlock(q); kfree(map); } - videobuf_queue_unlock(q); return; } diff --git a/drivers/media/v4l2-core/videobuf-vmalloc.c b/drivers/media/v4l2-core/videobuf-vmalloc.c index 1365c651c177..2ff7fcc77b11 100644 --- a/drivers/media/v4l2-core/videobuf-vmalloc.c +++ b/drivers/media/v4l2-core/videobuf-vmalloc.c @@ -54,14 +54,11 @@ MODULE_LICENSE("GPL"); static void videobuf_vm_open(struct vm_area_struct *vma) { struct videobuf_mapping *map = vma->vm_private_data; - struct videobuf_queue *q = map->q; dprintk(2, "vm_open %p [count=%u,vma=%08lx-%08lx]\n", map, map->count, vma->vm_start, vma->vm_end); - videobuf_queue_lock(q); map->count++; - videobuf_queue_unlock(q); } static void videobuf_vm_close(struct vm_area_struct *vma) @@ -73,11 +70,12 @@ static void videobuf_vm_close(struct vm_area_struct *vma) dprintk(2, "vm_close %p [count=%u,vma=%08lx-%08lx]\n", map, map->count, vma->vm_start, vma->vm_end); - videobuf_queue_lock(q); - if (!--map->count) { + map->count--; + if (0 == map->count) { struct videobuf_vmalloc_memory *mem; dprintk(1, "munmap %p q=%p\n", map, q); + videobuf_queue_lock(q); /* We need first to cancel streams, before unmapping */ if (q->streaming) @@ -116,8 +114,8 @@ static void videobuf_vm_close(struct vm_area_struct *vma) kfree(map); + videobuf_queue_unlock(q); } - videobuf_queue_unlock(q); return; } From 548df7831adc34eca3ccefa29f768cef84315d6e Mon Sep 17 00:00:00 2001 From: Ricardo Ribalda Date: Wed, 8 Jan 2014 05:01:33 -0300 Subject: [PATCH 146/596] [media] vb2: Check if there are buffers before streamon This patch adds a test preventing streamon() if there is no buffer ready. Without this patch, a user could call streamon() before preparing any buffer. This leads to a situation where if he calls close() before calling streamoff() the device is kept streaming. Signed-off-by: Ricardo Ribalda Delgado Reviewed-by: Marek Szyprowski Signed-off-by: Hans Verkuil Signed-off-by: Mauro Carvalho Chehab --- drivers/media/v4l2-core/videobuf2-core.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/media/v4l2-core/videobuf2-core.c b/drivers/media/v4l2-core/videobuf2-core.c index 5a5fb7f09b7b..a127925c9d61 100644 --- a/drivers/media/v4l2-core/videobuf2-core.c +++ b/drivers/media/v4l2-core/videobuf2-core.c @@ -1776,6 +1776,11 @@ static int vb2_internal_streamon(struct vb2_queue *q, enum v4l2_buf_type type) return 0; } + if (!q->num_buffers) { + dprintk(1, "streamon: no buffers have been allocated\n"); + return -EINVAL; + } + /* * If any buffers were queued before streamon, * we can now pass them to driver for processing. From 08e10972661db78d0e0a2d4a4c28fc3bf93617ba Mon Sep 17 00:00:00 2001 From: Michael Krufky Date: Wed, 29 Jan 2014 23:10:11 -0300 Subject: [PATCH 147/596] [media] update Michael Krufky's email address I am no longer available at the kernellabs.com or m1k.net email addresses. Update each instance of my email to my linuxtv.org account. Signed-off-by: Michael Krufky Signed-off-by: Mauro Carvalho Chehab --- Documentation/dvb/contributors.txt | 2 +- drivers/media/dvb-frontends/nxt200x.c | 2 +- drivers/media/pci/bt8xx/bttv-cards.c | 2 +- drivers/media/pci/saa7134/saa7134-cards.c | 2 +- drivers/media/usb/dvb-usb-v2/mxl111sf-demod.c | 4 ++-- drivers/media/usb/dvb-usb-v2/mxl111sf-demod.h | 2 +- drivers/media/usb/dvb-usb-v2/mxl111sf-gpio.c | 2 +- drivers/media/usb/dvb-usb-v2/mxl111sf-gpio.h | 2 +- drivers/media/usb/dvb-usb-v2/mxl111sf-i2c.c | 2 +- drivers/media/usb/dvb-usb-v2/mxl111sf-i2c.h | 2 +- drivers/media/usb/dvb-usb-v2/mxl111sf-phy.c | 2 +- drivers/media/usb/dvb-usb-v2/mxl111sf-phy.h | 2 +- drivers/media/usb/dvb-usb-v2/mxl111sf-reg.h | 2 +- drivers/media/usb/dvb-usb-v2/mxl111sf-tuner.c | 4 ++-- drivers/media/usb/dvb-usb-v2/mxl111sf-tuner.h | 2 +- drivers/media/usb/dvb-usb-v2/mxl111sf.c | 4 ++-- drivers/media/usb/dvb-usb-v2/mxl111sf.h | 2 +- 17 files changed, 20 insertions(+), 20 deletions(-) diff --git a/Documentation/dvb/contributors.txt b/Documentation/dvb/contributors.txt index 47c30098dab6..731a009723c7 100644 --- a/Documentation/dvb/contributors.txt +++ b/Documentation/dvb/contributors.txt @@ -78,7 +78,7 @@ Peter Beutner Wilson Michaels for the lgdt330x frontend driver, and various bugfixes -Michael Krufky +Michael Krufky for maintaining v4l/dvb inter-tree dependencies Taylor Jacob diff --git a/drivers/media/dvb-frontends/nxt200x.c b/drivers/media/dvb-frontends/nxt200x.c index 4bf057544607..8a8e1ecb762d 100644 --- a/drivers/media/dvb-frontends/nxt200x.c +++ b/drivers/media/dvb-frontends/nxt200x.c @@ -2,7 +2,7 @@ * Support for NXT2002 and NXT2004 - VSB/QAM * * Copyright (C) 2005 Kirk Lapray - * Copyright (C) 2006 Michael Krufky + * Copyright (C) 2006-2014 Michael Krufky * based on nxt2002 by Taylor Jacob * and nxt2004 by Jean-Francois Thibert * diff --git a/drivers/media/pci/bt8xx/bttv-cards.c b/drivers/media/pci/bt8xx/bttv-cards.c index d85cb0ace4dc..6662b495b22c 100644 --- a/drivers/media/pci/bt8xx/bttv-cards.c +++ b/drivers/media/pci/bt8xx/bttv-cards.c @@ -2426,7 +2426,7 @@ struct tvcard bttv_tvcards[] = { }, /* ---- card 0x87---------------------------------- */ [BTTV_BOARD_DVICO_FUSIONHDTV_5_LITE] = { - /* Michael Krufky */ + /* Michael Krufky */ .name = "DViCO FusionHDTV 5 Lite", .tuner_type = TUNER_LG_TDVS_H06XF, /* TDVS-H064F */ .tuner_addr = ADDR_UNSET, diff --git a/drivers/media/pci/saa7134/saa7134-cards.c b/drivers/media/pci/saa7134/saa7134-cards.c index d45e7f6ff332..c9b2350e92c8 100644 --- a/drivers/media/pci/saa7134/saa7134-cards.c +++ b/drivers/media/pci/saa7134/saa7134-cards.c @@ -2590,7 +2590,7 @@ struct saa7134_board saa7134_boards[] = { }}, }, [SAA7134_BOARD_AVERMEDIA_AVERTVHD_A180] = { - /* Michael Krufky + /* Michael Krufky * Uses Alps Electric TDHU2, containing NXT2004 ATSC Decoder * AFAIK, there is no analog demod, thus, * no support for analog television. diff --git a/drivers/media/usb/dvb-usb-v2/mxl111sf-demod.c b/drivers/media/usb/dvb-usb-v2/mxl111sf-demod.c index d83df4bb72d3..0a98d04c53e4 100644 --- a/drivers/media/usb/dvb-usb-v2/mxl111sf-demod.c +++ b/drivers/media/usb/dvb-usb-v2/mxl111sf-demod.c @@ -1,7 +1,7 @@ /* * mxl111sf-demod.c - driver for the MaxLinear MXL111SF DVB-T demodulator * - * Copyright (C) 2010 Michael Krufky + * Copyright (C) 2010-2014 Michael Krufky * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by @@ -601,7 +601,7 @@ struct dvb_frontend *mxl111sf_demod_attach(struct mxl111sf_state *mxl_state, EXPORT_SYMBOL_GPL(mxl111sf_demod_attach); MODULE_DESCRIPTION("MaxLinear MxL111SF DVB-T demodulator driver"); -MODULE_AUTHOR("Michael Krufky "); +MODULE_AUTHOR("Michael Krufky "); MODULE_LICENSE("GPL"); MODULE_VERSION("0.1"); diff --git a/drivers/media/usb/dvb-usb-v2/mxl111sf-demod.h b/drivers/media/usb/dvb-usb-v2/mxl111sf-demod.h index 3f3f8bfd190b..2d4530f5be54 100644 --- a/drivers/media/usb/dvb-usb-v2/mxl111sf-demod.h +++ b/drivers/media/usb/dvb-usb-v2/mxl111sf-demod.h @@ -1,7 +1,7 @@ /* * mxl111sf-demod.h - driver for the MaxLinear MXL111SF DVB-T demodulator * - * Copyright (C) 2010 Michael Krufky + * Copyright (C) 2010-2014 Michael Krufky * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by diff --git a/drivers/media/usb/dvb-usb-v2/mxl111sf-gpio.c b/drivers/media/usb/dvb-usb-v2/mxl111sf-gpio.c index e4121cb8f5ef..a619410adde4 100644 --- a/drivers/media/usb/dvb-usb-v2/mxl111sf-gpio.c +++ b/drivers/media/usb/dvb-usb-v2/mxl111sf-gpio.c @@ -1,7 +1,7 @@ /* * mxl111sf-gpio.c - driver for the MaxLinear MXL111SF * - * Copyright (C) 2010 Michael Krufky + * Copyright (C) 2010-2014 Michael Krufky * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by diff --git a/drivers/media/usb/dvb-usb-v2/mxl111sf-gpio.h b/drivers/media/usb/dvb-usb-v2/mxl111sf-gpio.h index 0220f54299a5..b85a5772d771 100644 --- a/drivers/media/usb/dvb-usb-v2/mxl111sf-gpio.h +++ b/drivers/media/usb/dvb-usb-v2/mxl111sf-gpio.h @@ -1,7 +1,7 @@ /* * mxl111sf-gpio.h - driver for the MaxLinear MXL111SF * - * Copyright (C) 2010 Michael Krufky + * Copyright (C) 2010-2014 Michael Krufky * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by diff --git a/drivers/media/usb/dvb-usb-v2/mxl111sf-i2c.c b/drivers/media/usb/dvb-usb-v2/mxl111sf-i2c.c index 34434557ef65..a101d06eb143 100644 --- a/drivers/media/usb/dvb-usb-v2/mxl111sf-i2c.c +++ b/drivers/media/usb/dvb-usb-v2/mxl111sf-i2c.c @@ -1,7 +1,7 @@ /* * mxl111sf-i2c.c - driver for the MaxLinear MXL111SF * - * Copyright (C) 2010 Michael Krufky + * Copyright (C) 2010-2014 Michael Krufky * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by diff --git a/drivers/media/usb/dvb-usb-v2/mxl111sf-i2c.h b/drivers/media/usb/dvb-usb-v2/mxl111sf-i2c.h index a57a45ffb9e4..465762145ad2 100644 --- a/drivers/media/usb/dvb-usb-v2/mxl111sf-i2c.h +++ b/drivers/media/usb/dvb-usb-v2/mxl111sf-i2c.h @@ -1,7 +1,7 @@ /* * mxl111sf-i2c.h - driver for the MaxLinear MXL111SF * - * Copyright (C) 2010 Michael Krufky + * Copyright (C) 2010-2014 Michael Krufky * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by diff --git a/drivers/media/usb/dvb-usb-v2/mxl111sf-phy.c b/drivers/media/usb/dvb-usb-v2/mxl111sf-phy.c index b741b3a7a325..f6b348024bec 100644 --- a/drivers/media/usb/dvb-usb-v2/mxl111sf-phy.c +++ b/drivers/media/usb/dvb-usb-v2/mxl111sf-phy.c @@ -1,7 +1,7 @@ /* * mxl111sf-phy.c - driver for the MaxLinear MXL111SF * - * Copyright (C) 2010 Michael Krufky + * Copyright (C) 2010-2014 Michael Krufky * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by diff --git a/drivers/media/usb/dvb-usb-v2/mxl111sf-phy.h b/drivers/media/usb/dvb-usb-v2/mxl111sf-phy.h index f0756071d347..0643738de7de 100644 --- a/drivers/media/usb/dvb-usb-v2/mxl111sf-phy.h +++ b/drivers/media/usb/dvb-usb-v2/mxl111sf-phy.h @@ -1,7 +1,7 @@ /* * mxl111sf-phy.h - driver for the MaxLinear MXL111SF * - * Copyright (C) 2010 Michael Krufky + * Copyright (C) 2010-2014 Michael Krufky * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by diff --git a/drivers/media/usb/dvb-usb-v2/mxl111sf-reg.h b/drivers/media/usb/dvb-usb-v2/mxl111sf-reg.h index 17831b0fb9db..89bf115e927e 100644 --- a/drivers/media/usb/dvb-usb-v2/mxl111sf-reg.h +++ b/drivers/media/usb/dvb-usb-v2/mxl111sf-reg.h @@ -1,7 +1,7 @@ /* * mxl111sf-reg.h - driver for the MaxLinear MXL111SF * - * Copyright (C) 2010 Michael Krufky + * Copyright (C) 2010-2014 Michael Krufky * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by diff --git a/drivers/media/usb/dvb-usb-v2/mxl111sf-tuner.c b/drivers/media/usb/dvb-usb-v2/mxl111sf-tuner.c index 879c529640f7..a8d2c7053674 100644 --- a/drivers/media/usb/dvb-usb-v2/mxl111sf-tuner.c +++ b/drivers/media/usb/dvb-usb-v2/mxl111sf-tuner.c @@ -1,7 +1,7 @@ /* * mxl111sf-tuner.c - driver for the MaxLinear MXL111SF CMOS tuner * - * Copyright (C) 2010 Michael Krufky + * Copyright (C) 2010-2014 Michael Krufky * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by @@ -512,7 +512,7 @@ struct dvb_frontend *mxl111sf_tuner_attach(struct dvb_frontend *fe, EXPORT_SYMBOL_GPL(mxl111sf_tuner_attach); MODULE_DESCRIPTION("MaxLinear MxL111SF CMOS tuner driver"); -MODULE_AUTHOR("Michael Krufky "); +MODULE_AUTHOR("Michael Krufky "); MODULE_LICENSE("GPL"); MODULE_VERSION("0.1"); diff --git a/drivers/media/usb/dvb-usb-v2/mxl111sf-tuner.h b/drivers/media/usb/dvb-usb-v2/mxl111sf-tuner.h index 90f583e5d6a6..5fc0121736ff 100644 --- a/drivers/media/usb/dvb-usb-v2/mxl111sf-tuner.h +++ b/drivers/media/usb/dvb-usb-v2/mxl111sf-tuner.h @@ -1,7 +1,7 @@ /* * mxl111sf-tuner.h - driver for the MaxLinear MXL111SF CMOS tuner * - * Copyright (C) 2010 Michael Krufky + * Copyright (C) 2010-2014 Michael Krufky * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by diff --git a/drivers/media/usb/dvb-usb-v2/mxl111sf.c b/drivers/media/usb/dvb-usb-v2/mxl111sf.c index 08240e498451..8ceff4290bf1 100644 --- a/drivers/media/usb/dvb-usb-v2/mxl111sf.c +++ b/drivers/media/usb/dvb-usb-v2/mxl111sf.c @@ -1,5 +1,5 @@ /* - * Copyright (C) 2010 Michael Krufky (mkrufky@kernellabs.com) + * Copyright (C) 2010-2014 Michael Krufky (mkrufky@linuxtv.org) * * This program is free software; you can redistribute it and/or modify it * under the terms of the GNU General Public License as published by the Free @@ -1421,7 +1421,7 @@ static struct usb_driver mxl111sf_usb_driver = { module_usb_driver(mxl111sf_usb_driver); -MODULE_AUTHOR("Michael Krufky "); +MODULE_AUTHOR("Michael Krufky "); MODULE_DESCRIPTION("Driver for MaxLinear MxL111SF"); MODULE_VERSION("1.0"); MODULE_LICENSE("GPL"); diff --git a/drivers/media/usb/dvb-usb-v2/mxl111sf.h b/drivers/media/usb/dvb-usb-v2/mxl111sf.h index 9816de86e48c..8516c011b7cc 100644 --- a/drivers/media/usb/dvb-usb-v2/mxl111sf.h +++ b/drivers/media/usb/dvb-usb-v2/mxl111sf.h @@ -1,5 +1,5 @@ /* - * Copyright (C) 2010 Michael Krufky (mkrufky@kernellabs.com) + * Copyright (C) 2010-2014 Michael Krufky (mkrufky@linuxtv.org) * * This program is free software; you can redistribute it and/or modify it * under the terms of the GNU General Public License as published by the Free From 62fd0d30e1fbd12d5a07ff722b2726bb8fa630c7 Mon Sep 17 00:00:00 2001 From: Andi Shyti Date: Thu, 30 Jan 2014 00:05:01 -0300 Subject: [PATCH 148/596] [media] cx24117: remove dead code in always 'false' if statement At this point of the execution in the function cx24117_attach() demod cannot be '0'. In that case the function returns earlier with an error value ('NULL'). Remove the if statement. This error has been reported by scan.coverity.com Signed-off-by: Andi Shyti Signed-off-by: Michael Krufky Signed-off-by: Mauro Carvalho Chehab --- drivers/media/dvb-frontends/cx24117.c | 8 -------- 1 file changed, 8 deletions(-) diff --git a/drivers/media/dvb-frontends/cx24117.c b/drivers/media/dvb-frontends/cx24117.c index 68f768a5422d..782dcfe6ae69 100644 --- a/drivers/media/dvb-frontends/cx24117.c +++ b/drivers/media/dvb-frontends/cx24117.c @@ -1200,12 +1200,6 @@ struct dvb_frontend *cx24117_attach(const struct cx24117_config *config, state->demod = demod - 1; state->priv = priv; - /* test i2c bus for ack */ - if (demod == 0) { - if (cx24117_readreg(state, 0x00) < 0) - goto error3; - } - dev_info(&state->priv->i2c->dev, "%s: Attaching frontend %d\n", KBUILD_MODNAME, state->demod); @@ -1216,8 +1210,6 @@ struct dvb_frontend *cx24117_attach(const struct cx24117_config *config, state->frontend.demodulator_priv = state; return &state->frontend; -error3: - kfree(state); error2: cx24117_release_priv(priv); error1: From a33dd5171d141c378df498aba3fa3c9a573cacb6 Mon Sep 17 00:00:00 2001 From: Andi Shyti Date: Thu, 30 Jan 2014 00:06:41 -0300 Subject: [PATCH 149/596] [media] cx24117: use a valid dev pointer for dev_err printout Don't use '&state->priv->i2c->dev' reference to device because state is still 'NULL'. Use '&i2c->dev' instead. This bug has been reported by scan.coverity.com Signed-off-by: Andi Shyti Signed-off-by: Michael Krufky Cc: vger@stable.kernel.org Signed-off-by: Mauro Carvalho Chehab --- drivers/media/dvb-frontends/cx24117.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/media/dvb-frontends/cx24117.c b/drivers/media/dvb-frontends/cx24117.c index 782dcfe6ae69..a6c3c9e2e897 100644 --- a/drivers/media/dvb-frontends/cx24117.c +++ b/drivers/media/dvb-frontends/cx24117.c @@ -1176,7 +1176,7 @@ struct dvb_frontend *cx24117_attach(const struct cx24117_config *config, switch (demod) { case 0: - dev_err(&state->priv->i2c->dev, + dev_err(&i2c->dev, "%s: Error attaching frontend %d\n", KBUILD_MODNAME, demod); goto error1; From 866e8d8a9dc1ebb4f9e67197e264ac2df81f7d4b Mon Sep 17 00:00:00 2001 From: Dave Jones Date: Thu, 30 Jan 2014 00:11:33 -0300 Subject: [PATCH 150/596] [media] mxl111sf: Fix unintentional garbage stack read mxl111sf_read_reg takes an address of a variable to write to as an argument. drivers/media/usb/dvb-usb-v2/mxl111sf-gpio.c:mxl111sf_config_pin_mux_modes passes several uninitialized stack variables to this routine, expecting them to be filled in. In the event that something unexpected happens when reading from the chip, we end up doing a pr_debug of the value passed in, revealing whatever garbage happened to be on the stack. Change the pr_debug to match what happens in the 'success' case, where we assign buf[1] to *data. Spotted with Coverity (Bugs 731910 through 731917) Signed-off-by: Dave Jones Signed-off-by: Michael Krufky Cc: stable@vger.kernel.org Signed-off-by: Mauro Carvalho Chehab --- drivers/media/usb/dvb-usb-v2/mxl111sf.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/media/usb/dvb-usb-v2/mxl111sf.c b/drivers/media/usb/dvb-usb-v2/mxl111sf.c index 8ceff4290bf1..c7304fa8ab73 100644 --- a/drivers/media/usb/dvb-usb-v2/mxl111sf.c +++ b/drivers/media/usb/dvb-usb-v2/mxl111sf.c @@ -105,7 +105,7 @@ int mxl111sf_read_reg(struct mxl111sf_state *state, u8 addr, u8 *data) ret = -EINVAL; } - pr_debug("R: (0x%02x, 0x%02x)\n", addr, *data); + pr_debug("R: (0x%02x, 0x%02x)\n", addr, buf[1]); fail: return ret; } From 13e1b87c986100169b0695aeb26970943665eda9 Mon Sep 17 00:00:00 2001 From: Dave Jones Date: Thu, 30 Jan 2014 00:17:09 -0300 Subject: [PATCH 151/596] [media] mxl111sf: Fix compile when CONFIG_DVB_USB_MXL111SF is unset MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Fix the following build error: drivers/media/usb/dvb-usb-v2/ mxl111sf-tuner.h:72:9: error: expected ‘;’, ‘,’ or ‘)’ before ‘struct’ struct mxl111sf_tuner_config *cfg) Signed-off-by: Dave Jones Signed-off-by: Michael Krufky Cc: stable@vger.kernel.org Signed-off-by: Mauro Carvalho Chehab --- drivers/media/usb/dvb-usb-v2/mxl111sf-tuner.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/media/usb/dvb-usb-v2/mxl111sf-tuner.h b/drivers/media/usb/dvb-usb-v2/mxl111sf-tuner.h index 5fc0121736ff..2046db22519e 100644 --- a/drivers/media/usb/dvb-usb-v2/mxl111sf-tuner.h +++ b/drivers/media/usb/dvb-usb-v2/mxl111sf-tuner.h @@ -68,7 +68,7 @@ struct dvb_frontend *mxl111sf_tuner_attach(struct dvb_frontend *fe, #else static inline struct dvb_frontend *mxl111sf_tuner_attach(struct dvb_frontend *fe, - struct mxl111sf_state *mxl_state + struct mxl111sf_state *mxl_state, struct mxl111sf_tuner_config *cfg) { printk(KERN_WARNING "%s: driver disabled by Kconfig\n", __func__); From f2e4c5e004691dfe37d0e4b363296f28abdb9bc7 Mon Sep 17 00:00:00 2001 From: Antti Palosaari Date: Thu, 16 Jan 2014 08:59:30 -0300 Subject: [PATCH 152/596] [media] af9035: add ID [2040:f900] Hauppauge WinTV-MiniStick 2 Add USB ID [2040:f900] for Hauppauge WinTV-MiniStick 2. Device is build upon IT9135 chipset. Tested-by: Stefan Becker Signed-off-by: Antti Palosaari Cc: stable@vger.kernel.org Signed-off-by: Mauro Carvalho Chehab --- drivers/media/usb/dvb-usb-v2/af9035.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/media/usb/dvb-usb-v2/af9035.c b/drivers/media/usb/dvb-usb-v2/af9035.c index 8f9b2cea88f0..8ede8ea762e6 100644 --- a/drivers/media/usb/dvb-usb-v2/af9035.c +++ b/drivers/media/usb/dvb-usb-v2/af9035.c @@ -1539,6 +1539,8 @@ static const struct usb_device_id af9035_id_table[] = { &af9035_props, "TerraTec Cinergy T Stick Dual RC (rev. 2)", NULL) }, { DVB_USB_DEVICE(USB_VID_LEADTEK, 0x6a05, &af9035_props, "Leadtek WinFast DTV Dongle Dual", NULL) }, + { DVB_USB_DEVICE(USB_VID_HAUPPAUGE, 0xf900, + &af9035_props, "Hauppauge WinTV-MiniStick 2", NULL) }, { } }; MODULE_DEVICE_TABLE(usb, af9035_id_table); From 1ba6c90161ac058f580e6ceb5ccc14dcd86365d1 Mon Sep 17 00:00:00 2001 From: Masanari Iida Date: Fri, 17 Jan 2014 13:38:00 -0300 Subject: [PATCH 153/596] [media] hdpvr: Fix memory leak in debug cppcheck reported memory leak in device_authorizatio() within hdpvr-core.c. When the debug option is specified and the code jump to "unlock:" label, print_buf was not freed. Confirm the module succesfully compiled without error. Signed-off-by: Masanari Iida Signed-off-by: Hans Verkuil Signed-off-by: Mauro Carvalho Chehab --- drivers/media/usb/hdpvr/hdpvr-core.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/media/usb/hdpvr/hdpvr-core.c b/drivers/media/usb/hdpvr/hdpvr-core.c index 2f0c89cbac76..c5638964c3f2 100644 --- a/drivers/media/usb/hdpvr/hdpvr-core.c +++ b/drivers/media/usb/hdpvr/hdpvr-core.c @@ -198,7 +198,6 @@ static int device_authorization(struct hdpvr_device *dev) hex_dump_to_buffer(response, 8, 16, 1, print_buf, 5*buf_size+1, 0); v4l2_dbg(MSG_INFO, hdpvr_debug, &dev->v4l2_dev, " response: %s\n", print_buf); - kfree(print_buf); #endif msleep(100); @@ -214,6 +213,9 @@ static int device_authorization(struct hdpvr_device *dev) retval = ret != 8; unlock: mutex_unlock(&dev->usbc_mutex); +#ifdef HDPVR_DEBUG + kfree(print_buf); +#endif return retval; } From 257cc4b5c57060a65e9c805fb1c225688ebbca80 Mon Sep 17 00:00:00 2001 From: Martin Bugge Date: Thu, 23 Jan 2014 06:40:00 -0300 Subject: [PATCH 154/596] [media] v4l2-dv-timings: fix GTF calculation Round off image width to nearest 8 (GTF_CELL_GRAN) A source sending a GTF (Generalized Timing Formula) format have no means of signalling image width. The assumed aspect ratio may result in an odd image width but according to the standard image width should be in multiple of 8. Cc: Mats Randgaard Signed-off-by: Martin Bugge Reviewed-by: Hans Verkuil Signed-off-by: Hans Verkuil Signed-off-by: Mauro Carvalho Chehab --- drivers/media/v4l2-core/v4l2-dv-timings.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/media/v4l2-core/v4l2-dv-timings.c b/drivers/media/v4l2-core/v4l2-dv-timings.c index ee52b9f4a944..f7902fe8a526 100644 --- a/drivers/media/v4l2-core/v4l2-dv-timings.c +++ b/drivers/media/v4l2-core/v4l2-dv-timings.c @@ -515,6 +515,7 @@ bool v4l2_detect_gtf(unsigned frame_height, aspect.denominator = 9; } image_width = ((image_height * aspect.numerator) / aspect.denominator); + image_width = (image_width + GTF_CELL_GRAN/2) & ~(GTF_CELL_GRAN - 1); /* Horizontal */ if (default_gtf) From 57f0547fbc1e925f5e58c76f311a6632c3f37740 Mon Sep 17 00:00:00 2001 From: Martin Bugge Date: Wed, 29 Jan 2014 06:50:20 -0300 Subject: [PATCH 155/596] [media] adv7842: Composite free-run platfrom-data fix Incorrectly setting of free-run for Composite. Copy/paste regression fix. Signed-off-by: Martin Bugge Signed-off-by: Hans Verkuil Signed-off-by: Mauro Carvalho Chehab --- drivers/media/i2c/adv7842.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/media/i2c/adv7842.c b/drivers/media/i2c/adv7842.c index 1effc21e1cdd..9bbd6656fb8f 100644 --- a/drivers/media/i2c/adv7842.c +++ b/drivers/media/i2c/adv7842.c @@ -2554,7 +2554,7 @@ static int adv7842_core_init(struct v4l2_subdev *sd) sdp_write_and_or(sd, 0xdd, 0xf0, pdata->sdp_free_run_force | (pdata->sdp_free_run_cbar_en << 1) | (pdata->sdp_free_run_man_col_en << 2) | - (pdata->sdp_free_run_force << 3)); + (pdata->sdp_free_run_auto << 3)); /* TODO from platform data */ cp_write(sd, 0x69, 0x14); /* Enable CP CSC */ From 2a971354e74f3837d14b9c8d7f7983b0c9c330e4 Mon Sep 17 00:00:00 2001 From: Michal Kubecek Date: Thu, 30 Jan 2014 08:50:20 +0100 Subject: [PATCH 156/596] ipvs: fix AF assignment in ip_vs_conn_new() If a fwmark is passed to ip_vs_conn_new(), it is passed in vaddr, not daddr. Therefore we should set AF to AF_UNSPEC in vaddr assignment (like we do in ip_vs_ct_in_get()), otherwise we may copy only first 4 bytes of an IPv6 address into cp->daddr. Signed-off-by: Bogdano Arendartchuk Signed-off-by: Michal Kubecek Acked-by: Julian Anastasov Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_conn.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/net/netfilter/ipvs/ip_vs_conn.c b/net/netfilter/ipvs/ip_vs_conn.c index 59a1a85bcb3e..a8eb0a89326a 100644 --- a/net/netfilter/ipvs/ip_vs_conn.c +++ b/net/netfilter/ipvs/ip_vs_conn.c @@ -871,11 +871,11 @@ ip_vs_conn_new(const struct ip_vs_conn_param *p, cp->protocol = p->protocol; ip_vs_addr_set(p->af, &cp->caddr, p->caddr); cp->cport = p->cport; - ip_vs_addr_set(p->af, &cp->vaddr, p->vaddr); - cp->vport = p->vport; - /* proto should only be IPPROTO_IP if d_addr is a fwmark */ + /* proto should only be IPPROTO_IP if p->vaddr is a fwmark */ ip_vs_addr_set(p->protocol == IPPROTO_IP ? AF_UNSPEC : p->af, - &cp->daddr, daddr); + &cp->vaddr, p->vaddr); + cp->vport = p->vport; + ip_vs_addr_set(p->af, &cp->daddr, daddr); cp->dport = dport; cp->flags = flags; cp->fwmark = fwmark; From 4294ad1c52caad3239404cea9e0cbfeb435e3a25 Mon Sep 17 00:00:00 2001 From: Manuel Lauss Date: Wed, 29 Jan 2014 19:11:46 +0100 Subject: [PATCH 157/596] MIPS: Alchemy: Fix DB1100 GPIO registration With CONFIG_GPIOLIB=y gpios need to be requested before they can be modified. Request the SD carddetect pins, and drop the SPI direction setup, as the driver does that for us anyway. This gets rid of a lot of WARN_ON()s triggered by GPIO core, and restores functionality of the touschreen controller. Signed-off-by: Manuel Lauss Cc: Linux-MIPS Patchwork: https://patchwork.linux-mips.org/patch/6497/ Signed-off-by: Ralf Baechle --- arch/mips/alchemy/devboards/db1000.c | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) diff --git a/arch/mips/alchemy/devboards/db1000.c b/arch/mips/alchemy/devboards/db1000.c index 11f3ad20321c..5483906e0f86 100644 --- a/arch/mips/alchemy/devboards/db1000.c +++ b/arch/mips/alchemy/devboards/db1000.c @@ -534,13 +534,10 @@ static int __init db1000_dev_init(void) s0 = AU1100_GPIO1_INT; s1 = AU1100_GPIO4_INT; + gpio_request(19, "sd0_cd"); + gpio_request(20, "sd1_cd"); gpio_direction_input(19); /* sd0 cd# */ gpio_direction_input(20); /* sd1 cd# */ - gpio_direction_input(21); /* touch pendown# */ - gpio_direction_input(207); /* SPI MISO */ - gpio_direction_output(208, 0); /* SPI MOSI */ - gpio_direction_output(209, 1); /* SPI SCK */ - gpio_direction_output(210, 1); /* SPI CS# */ /* spi_gpio on SSI0 pins */ pfc = __raw_readl((void __iomem *)SYS_PINFUNC); From 6776254b1c28fece9535d42baf2db88756121fe7 Mon Sep 17 00:00:00 2001 From: James Hogan Date: Tue, 4 Feb 2014 12:29:01 +0000 Subject: [PATCH 158/596] MIPS: Wire up sched_setattr/sched_getattr syscalls Wire up for MIPS the new sched_setattr and sched_getattr system calls added in commit d50dde5a10f3 (sched: Add new scheduler syscalls to support an extended scheduling parameters ABI) merged in v3.14-rc1. Signed-off-by: James Hogan Reviewed-by: Markos Chandras Cc: Ralf Baechle Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/6502/ Signed-off-by: Ralf Baechle --- arch/mips/include/uapi/asm/unistd.h | 18 ++++++++++++------ arch/mips/kernel/scall32-o32.S | 2 ++ arch/mips/kernel/scall64-64.S | 2 ++ arch/mips/kernel/scall64-n32.S | 2 ++ arch/mips/kernel/scall64-o32.S | 2 ++ 5 files changed, 20 insertions(+), 6 deletions(-) diff --git a/arch/mips/include/uapi/asm/unistd.h b/arch/mips/include/uapi/asm/unistd.h index 1dee279f9665..d6e154a9e6a5 100644 --- a/arch/mips/include/uapi/asm/unistd.h +++ b/arch/mips/include/uapi/asm/unistd.h @@ -369,16 +369,18 @@ #define __NR_process_vm_writev (__NR_Linux + 346) #define __NR_kcmp (__NR_Linux + 347) #define __NR_finit_module (__NR_Linux + 348) +#define __NR_sched_setattr (__NR_Linux + 349) +#define __NR_sched_getattr (__NR_Linux + 350) /* * Offset of the last Linux o32 flavoured syscall */ -#define __NR_Linux_syscalls 348 +#define __NR_Linux_syscalls 350 #endif /* _MIPS_SIM == _MIPS_SIM_ABI32 */ #define __NR_O32_Linux 4000 -#define __NR_O32_Linux_syscalls 348 +#define __NR_O32_Linux_syscalls 350 #if _MIPS_SIM == _MIPS_SIM_ABI64 @@ -695,16 +697,18 @@ #define __NR_kcmp (__NR_Linux + 306) #define __NR_finit_module (__NR_Linux + 307) #define __NR_getdents64 (__NR_Linux + 308) +#define __NR_sched_setattr (__NR_Linux + 309) +#define __NR_sched_getattr (__NR_Linux + 310) /* * Offset of the last Linux 64-bit flavoured syscall */ -#define __NR_Linux_syscalls 308 +#define __NR_Linux_syscalls 310 #endif /* _MIPS_SIM == _MIPS_SIM_ABI64 */ #define __NR_64_Linux 5000 -#define __NR_64_Linux_syscalls 308 +#define __NR_64_Linux_syscalls 310 #if _MIPS_SIM == _MIPS_SIM_NABI32 @@ -1025,15 +1029,17 @@ #define __NR_process_vm_writev (__NR_Linux + 310) #define __NR_kcmp (__NR_Linux + 311) #define __NR_finit_module (__NR_Linux + 312) +#define __NR_sched_setattr (__NR_Linux + 313) +#define __NR_sched_getattr (__NR_Linux + 314) /* * Offset of the last N32 flavoured syscall */ -#define __NR_Linux_syscalls 312 +#define __NR_Linux_syscalls 314 #endif /* _MIPS_SIM == _MIPS_SIM_NABI32 */ #define __NR_N32_Linux 6000 -#define __NR_N32_Linux_syscalls 312 +#define __NR_N32_Linux_syscalls 314 #endif /* _UAPI_ASM_UNISTD_H */ diff --git a/arch/mips/kernel/scall32-o32.S b/arch/mips/kernel/scall32-o32.S index e8e541b40d86..a5b14f48e1af 100644 --- a/arch/mips/kernel/scall32-o32.S +++ b/arch/mips/kernel/scall32-o32.S @@ -563,3 +563,5 @@ EXPORT(sys_call_table) PTR sys_process_vm_writev PTR sys_kcmp PTR sys_finit_module + PTR sys_sched_setattr + PTR sys_sched_getattr /* 4350 */ diff --git a/arch/mips/kernel/scall64-64.S b/arch/mips/kernel/scall64-64.S index 57e3742fec59..b56e254beb15 100644 --- a/arch/mips/kernel/scall64-64.S +++ b/arch/mips/kernel/scall64-64.S @@ -425,4 +425,6 @@ EXPORT(sys_call_table) PTR sys_kcmp PTR sys_finit_module PTR sys_getdents64 + PTR sys_sched_setattr + PTR sys_sched_getattr /* 5310 */ .size sys_call_table,.-sys_call_table diff --git a/arch/mips/kernel/scall64-n32.S b/arch/mips/kernel/scall64-n32.S index 2f48f5934399..f7e5b72cf481 100644 --- a/arch/mips/kernel/scall64-n32.S +++ b/arch/mips/kernel/scall64-n32.S @@ -418,4 +418,6 @@ EXPORT(sysn32_call_table) PTR compat_sys_process_vm_writev /* 6310 */ PTR sys_kcmp PTR sys_finit_module + PTR sys_sched_setattr + PTR sys_sched_getattr .size sysn32_call_table,.-sysn32_call_table diff --git a/arch/mips/kernel/scall64-o32.S b/arch/mips/kernel/scall64-o32.S index f1acdb429f4f..6788727d91af 100644 --- a/arch/mips/kernel/scall64-o32.S +++ b/arch/mips/kernel/scall64-o32.S @@ -541,4 +541,6 @@ EXPORT(sys32_call_table) PTR compat_sys_process_vm_writev PTR sys_kcmp PTR sys_finit_module + PTR sys_sched_setattr + PTR sys_sched_getattr /* 4350 */ .size sys32_call_table,.-sys32_call_table From d7736ff5be31edaa4fe5ab62810c64529a24b149 Mon Sep 17 00:00:00 2001 From: Michael Holzheu Date: Thu, 30 Jan 2014 16:14:02 +0100 Subject: [PATCH 159/596] s390/dump: Fix dump memory detection Dumps created by kdump or zfcpdump can contain invalid memory holes when dumping z/VM systems that have memory pressure. For example: # zgetdump -i /proc/vmcore. Memory map: 0000000000000000 - 0000000000bfffff (12 MB) 0000000000e00000 - 00000000014fffff (7 MB) 000000000bd00000 - 00000000f3bfffff (3711 MB) The memory detection function find_memory_chunks() issues tprot to find valid memory chunks. In case of CMM it can happen that pages are marked as unstable via set_page_unstable() in arch_free_page(). If z/VM has released that pages, tprot returns -EFAULT and indicates a memory hole. So fix this and switch off CMM in case of kdump or zfcpdump. Cc: Signed-off-by: Michael Holzheu Signed-off-by: Martin Schwidefsky --- arch/s390/mm/page-states.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/arch/s390/mm/page-states.c b/arch/s390/mm/page-states.c index a90d45e9dfb0..27c50f4d90cb 100644 --- a/arch/s390/mm/page-states.c +++ b/arch/s390/mm/page-states.c @@ -12,6 +12,8 @@ #include #include #include +#include +#include #define ESSA_SET_STABLE 1 #define ESSA_SET_UNUSED 2 @@ -41,6 +43,14 @@ void __init cmma_init(void) if (!cmma_flag) return; + /* + * Disable CMM for dump, otherwise the tprot based memory + * detection can fail because of unstable pages. + */ + if (OLDMEM_BASE || ipl_info.type == IPL_TYPE_FCP_DUMP) { + cmma_flag = 0; + return; + } asm volatile( " .insn rrf,0xb9ab0000,%1,%1,0,0\n" "0: la %0,0\n" From 40507403485fcb56b83d6ddfc954e9b08305054c Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Tue, 4 Feb 2014 14:41:26 +0000 Subject: [PATCH 160/596] arm64: vdso: prevent ld from aligning PT_LOAD segments to 64k Whilst the text segment for our VDSO is marked as PT_LOAD in the ELF headers, it is mapped by the kernel and not actually subject to demand-paging. ld doesn't realise this, and emits a p_align field of 64k (the maximum supported page size), which conflicts with the load address picked by the kernel on 4k systems, which will be 4k aligned. This causes GDB to fail with "Failed to read a valid object file image from memory" when attempting to load the VDSO. This patch passes the -n option to ld, which prevents it from aligning PT_LOAD segments to the maximum page size. Cc: Reported-by: Kyle McMartin Acked-by: Kyle McMartin Signed-off-by: Will Deacon Signed-off-by: Catalin Marinas --- arch/arm64/kernel/vdso/Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm64/kernel/vdso/Makefile b/arch/arm64/kernel/vdso/Makefile index d8064af42e62..6d20b7d162d8 100644 --- a/arch/arm64/kernel/vdso/Makefile +++ b/arch/arm64/kernel/vdso/Makefile @@ -48,7 +48,7 @@ $(obj-vdso): %.o: %.S # Actual build commands quiet_cmd_vdsold = VDSOL $@ - cmd_vdsold = $(CC) $(c_flags) -Wl,-T $^ -o $@ + cmd_vdsold = $(CC) $(c_flags) -Wl,-n -Wl,-T $^ -o $@ quiet_cmd_vdsoas = VDSOA $@ cmd_vdsoas = $(CC) $(a_flags) -c -o $@ $< From e290c3434b70b2eb200f9a54d992b7c4fd938879 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Tue, 21 Jan 2014 16:10:09 +0100 Subject: [PATCH 161/596] spi: rspi: Document support for Renesas QSPI in Kconfig As of commit 5ce0ba88650f2606244a761d92e2b725f4ab3583 ("spi: rcar: add Renesas QSPI support on RSPI") the rspi driver handles Renesas QSPI, too, but this was not reflected in the Kconfig help text. Signed-off-by: Geert Uytterhoeven Signed-off-by: Mark Brown --- drivers/spi/Kconfig | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/spi/Kconfig b/drivers/spi/Kconfig index ba9310bc9acb..581ee2a8856b 100644 --- a/drivers/spi/Kconfig +++ b/drivers/spi/Kconfig @@ -376,10 +376,10 @@ config SPI_PXA2XX_PCI def_tristate SPI_PXA2XX && PCI config SPI_RSPI - tristate "Renesas RSPI controller" + tristate "Renesas RSPI/QSPI controller" depends on (SUPERH && SH_DMAE_BASE) || ARCH_SHMOBILE help - SPI driver for Renesas RSPI blocks. + SPI driver for Renesas RSPI and QSPI blocks. config SPI_S3C24XX tristate "Samsung S3C24XX series SPI" From 89258a97531124cd1e212f7a8ed44fccf8f9a891 Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Tue, 4 Feb 2014 19:57:38 +0100 Subject: [PATCH 162/596] MAINTAINERS: Update drm/i915 git repo Moved to a common location so that Jani also can push to it, to avoid moving it every time I go on vacation. Please update autobuilders and everything else pointing at the drm-intel.git repo, the old one won't be updated any more. Cc: Dave Airlie Cc: Jani Nikula Signed-off-by: Daniel Vetter --- MAINTAINERS | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index 31a046213274..e3aaf277cf58 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2832,7 +2832,7 @@ M: Jani Nikula L: intel-gfx@lists.freedesktop.org L: dri-devel@lists.freedesktop.org Q: http://patchwork.freedesktop.org/project/intel-gfx/ -T: git git://people.freedesktop.org/~danvet/drm-intel +T: git git://anongit.freedesktop.org/drm-intel S: Supported F: drivers/gpu/drm/i915/ F: include/drm/i915* From 48b8f6315a6a230f859fcae8ef4a0ccf4f48e802 Mon Sep 17 00:00:00 2001 From: Jani Nikula Date: Fri, 31 Jan 2014 15:49:07 +0200 Subject: [PATCH 163/596] drm: add DRM_INFO_ONCE() to print a one-time DRM_INFO() message Just like DRM_INFO(), but only do it once. Signed-off-by: Jani Nikula Reviewed-by: Chris Wilson Acked-by: Dave Airlie Signed-off-by: Daniel Vetter --- include/drm/drmP.h | 3 +++ 1 file changed, 3 insertions(+) diff --git a/include/drm/drmP.h b/include/drm/drmP.h index 04086c5be930..04a7f31301f8 100644 --- a/include/drm/drmP.h +++ b/include/drm/drmP.h @@ -199,6 +199,9 @@ int drm_err(const char *func, const char *format, ...); #define DRM_INFO(fmt, ...) \ printk(KERN_INFO "[" DRM_NAME "] " fmt, ##__VA_ARGS__) +#define DRM_INFO_ONCE(fmt, ...) \ + printk_once(KERN_INFO "[" DRM_NAME "] " fmt, ##__VA_ARGS__) + /** * Debug output. * From bdde5c6a258a702bdfa7d1f4ae804a7bc405e788 Mon Sep 17 00:00:00 2001 From: Jani Nikula Date: Fri, 31 Jan 2014 15:49:08 +0200 Subject: [PATCH 164/596] drm/i915: demote opregion excessive timeout WARN_ONCE to DRM_INFO_ONCE The WARN_ONCE is a bit too verbose, make it a DRM_INFO_ONCE. While at it, add a #define for MAX_DSLP and make the message a bit more informative. v2: use DRM_INFO_ONCE, add MAX_DSLP, pimp the message. Suggested-by: Chris Wilson Signed-off-by: Jani Nikula Reviewed-by: Chris Wilson Signed-off-by: Daniel Vetter --- drivers/gpu/drm/i915/intel_opregion.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/i915/intel_opregion.c b/drivers/gpu/drm/i915/intel_opregion.c index 37e9a96777ef..68459605bd12 100644 --- a/drivers/gpu/drm/i915/intel_opregion.c +++ b/drivers/gpu/drm/i915/intel_opregion.c @@ -227,6 +227,8 @@ struct opregion_asle { #define ACPI_DIGITAL_OUTPUT (3<<8) #define ACPI_LVDS_OUTPUT (4<<8) +#define MAX_DSLP 1500 + #ifdef CONFIG_ACPI static int swsci(struct drm_device *dev, u32 function, u32 parm, u32 *parm_out) { @@ -261,10 +263,11 @@ static int swsci(struct drm_device *dev, u32 function, u32 parm, u32 *parm_out) /* The spec says 2ms should be the default, but it's too small * for some machines. */ dslp = 50; - } else if (dslp > 500) { + } else if (dslp > MAX_DSLP) { /* Hey bios, trust must be earned. */ - WARN_ONCE(1, "excessive driver sleep timeout (DSPL) %u\n", dslp); - dslp = 500; + DRM_INFO_ONCE("ACPI BIOS requests an excessive sleep of %u ms, " + "using %u ms instead\n", dslp, MAX_DSLP); + dslp = MAX_DSLP; } /* The spec tells us to do this, but we are the only user... */ From 12b13835a0a8bfabea68741e1ab4d4a4cb77d037 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Tue, 4 Feb 2014 12:20:01 -0800 Subject: [PATCH 165/596] kbuild: don't enable DEBUG_INFO when building for COMPILE_TEST It really isn't very interesting to have DEBUG_INFO when doing compile coverage stuff (you wouldn't want to run the result anyway, that's kind of the whole point of COMPILE_TEST), and it currently makes the build take longer and use much more disk space for "all{yes,mod}config". There's somewhat active discussion about this still, and we might end up with some new config option for things like this (Andi points out that the silly X86_DECODER_SELFTEST option also slows down the normal coverage tests hugely), but I'm starting the ball rolling with this simple one-liner. DEBUG_INFO isn't that noticeable if you have tons of memory and a good IO subsystem, but it hurts you a lot if you don't - for very little upside for the common use. Signed-off-by: Linus Torvalds --- lib/Kconfig.debug | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index dbf94a7d25a8..a48abeac753f 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -119,7 +119,7 @@ menu "Compile-time checks and compiler options" config DEBUG_INFO bool "Compile the kernel with debug info" - depends on DEBUG_KERNEL + depends on DEBUG_KERNEL && !COMPILE_TEST help If you say Y here the resulting kernel image will include debugging info resulting in a larger kernel image. From 2fa4cb905605c863bf570027233af7afd8149ae4 Mon Sep 17 00:00:00 2001 From: Stanislaw Gruszka Date: Tue, 28 Jan 2014 09:14:48 +0100 Subject: [PATCH 166/596] ath9k_htc: make ->sta_rc_update atomic for most calls sta_rc_update() callback must be atomic, hence we can not take mutexes or do other operations, which can sleep in ath9k_htc_sta_rc_update(). I think we can just return from ath9k_htc_sta_rc_update(), if it is called without IEEE80211_RC_SUPP_RATES_CHANGED bit. That will help with scheduling while atomic bug for most cases (except mesh and IBSS modes). For mesh and IBSS I do not see other solution like creating additional workqueue, because sending firmware command require us to sleep, but this can be done in additional patch. Patch partially fixes bug: https://bugzilla.redhat.com/show_bug.cgi?id=990955 Cc: stable@vger.kernel.org Signed-off-by: Stanislaw Gruszka Signed-off-by: John W. Linville --- drivers/net/wireless/ath/ath9k/htc_drv_main.c | 25 ++++++++++--------- 1 file changed, 13 insertions(+), 12 deletions(-) diff --git a/drivers/net/wireless/ath/ath9k/htc_drv_main.c b/drivers/net/wireless/ath/ath9k/htc_drv_main.c index 608d739d1378..a57af9b96a39 100644 --- a/drivers/net/wireless/ath/ath9k/htc_drv_main.c +++ b/drivers/net/wireless/ath/ath9k/htc_drv_main.c @@ -1315,21 +1315,22 @@ static void ath9k_htc_sta_rc_update(struct ieee80211_hw *hw, struct ath_common *common = ath9k_hw_common(priv->ah); struct ath9k_htc_target_rate trate; + if (!(changed & IEEE80211_RC_SUPP_RATES_CHANGED)) + return; + mutex_lock(&priv->mutex); ath9k_htc_ps_wakeup(priv); - if (changed & IEEE80211_RC_SUPP_RATES_CHANGED) { - memset(&trate, 0, sizeof(struct ath9k_htc_target_rate)); - ath9k_htc_setup_rate(priv, sta, &trate); - if (!ath9k_htc_send_rate_cmd(priv, &trate)) - ath_dbg(common, CONFIG, - "Supported rates for sta: %pM updated, rate caps: 0x%X\n", - sta->addr, be32_to_cpu(trate.capflags)); - else - ath_dbg(common, CONFIG, - "Unable to update supported rates for sta: %pM\n", - sta->addr); - } + memset(&trate, 0, sizeof(struct ath9k_htc_target_rate)); + ath9k_htc_setup_rate(priv, sta, &trate); + if (!ath9k_htc_send_rate_cmd(priv, &trate)) + ath_dbg(common, CONFIG, + "Supported rates for sta: %pM updated, rate caps: 0x%X\n", + sta->addr, be32_to_cpu(trate.capflags)); + else + ath_dbg(common, CONFIG, + "Unable to update supported rates for sta: %pM\n", + sta->addr); ath9k_htc_ps_restore(priv); mutex_unlock(&priv->mutex); From 58e33a219298c5ba0f88ce9a7d0f448eb3ae884a Mon Sep 17 00:00:00 2001 From: Stanislaw Gruszka Date: Wed, 29 Jan 2014 17:42:37 +0100 Subject: [PATCH 167/596] rt2800: disable PS by default on USB We have disabled it currently on other buses. PS can cause some issues, not necessarily with our driver but on AP, that are not easy to debug. Since behaviour differs on rt2800usb and rt2800pci, user usually blame for malfunction rt2800usb driver, whereas issue is on AP side. Signed-off-by: Stanislaw Gruszka Acked-by: Gertjan van Wingerde Signed-off-by: John W. Linville --- drivers/net/wireless/rt2x00/rt2800lib.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/drivers/net/wireless/rt2x00/rt2800lib.c b/drivers/net/wireless/rt2x00/rt2800lib.c index b8f5b06006c4..7f8b5d156c8c 100644 --- a/drivers/net/wireless/rt2x00/rt2800lib.c +++ b/drivers/net/wireless/rt2x00/rt2800lib.c @@ -7458,10 +7458,9 @@ static int rt2800_probe_hw_mode(struct rt2x00_dev *rt2x00dev) u32 reg; /* - * Disable powersaving as default on PCI devices. + * Disable powersaving as default. */ - if (rt2x00_is_pci(rt2x00dev) || rt2x00_is_soc(rt2x00dev)) - rt2x00dev->hw->wiphy->flags &= ~WIPHY_FLAG_PS_ON_BY_DEFAULT; + rt2x00dev->hw->wiphy->flags &= ~WIPHY_FLAG_PS_ON_BY_DEFAULT; /* * Initialize all hw fields. From 8b0df00fe650197ac47c42ba71b2588334a529be Mon Sep 17 00:00:00 2001 From: Stanislaw Gruszka Date: Wed, 29 Jan 2014 17:54:17 +0100 Subject: [PATCH 168/596] rt2500: disable PS by default It is know that PS cause issues on that old devices, disable it by default. Signed-off-by: Stanislaw Gruszka Acked-by: Gertjan van Wingerde Signed-off-by: John W. Linville --- drivers/net/wireless/rt2x00/rt2500pci.c | 5 +++++ drivers/net/wireless/rt2x00/rt2500usb.c | 5 +++++ 2 files changed, 10 insertions(+) diff --git a/drivers/net/wireless/rt2x00/rt2500pci.c b/drivers/net/wireless/rt2x00/rt2500pci.c index abc5f56f29fe..2f1cd929c6f6 100644 --- a/drivers/net/wireless/rt2x00/rt2500pci.c +++ b/drivers/net/wireless/rt2x00/rt2500pci.c @@ -1876,6 +1876,11 @@ static int rt2500pci_probe_hw_mode(struct rt2x00_dev *rt2x00dev) rt2x00_eeprom_addr(rt2x00dev, EEPROM_MAC_ADDR_0)); + /* + * Disable powersaving as default. + */ + rt2x00dev->hw->wiphy->flags &= ~WIPHY_FLAG_PS_ON_BY_DEFAULT; + /* * Initialize hw_mode information. */ diff --git a/drivers/net/wireless/rt2x00/rt2500usb.c b/drivers/net/wireless/rt2x00/rt2500usb.c index 9f16824cd1bc..d849d590de25 100644 --- a/drivers/net/wireless/rt2x00/rt2500usb.c +++ b/drivers/net/wireless/rt2x00/rt2500usb.c @@ -1706,6 +1706,11 @@ static int rt2500usb_probe_hw_mode(struct rt2x00_dev *rt2x00dev) IEEE80211_HW_SUPPORTS_PS | IEEE80211_HW_PS_NULLFUNC_STACK; + /* + * Disable powersaving as default. + */ + rt2x00dev->hw->wiphy->flags &= ~WIPHY_FLAG_PS_ON_BY_DEFAULT; + SET_IEEE80211_DEV(rt2x00dev->hw, rt2x00dev->dev); SET_IEEE80211_PERM_ADDR(rt2x00dev->hw, rt2x00_eeprom_addr(rt2x00dev, From 6bca610d97b6139a1d7598b8009da9d339daa50f Mon Sep 17 00:00:00 2001 From: Oleksij Rempel Date: Thu, 30 Jan 2014 09:14:53 +0100 Subject: [PATCH 169/596] ath9k_htc: Do not support PowerSave by default It is a copy/paste of patch provided by Sujith for ath9k. "Even though we make sure PowerSave is not enabled by default by disabling the flag, WIPHY_FLAG_PS_ON_BY_DEFAULT on init, PS could be enabled by userspace based on various factors like battery usage etc. Since PS in ath9k is just broken and has been untested for years, remove support for it, but allow a user to explicitly enable it using a module parameter." Cc: stable@vger.kernel.org Signed-off-by: Oleksij Rempel Signed-off-by: John W. Linville --- drivers/net/wireless/ath/ath9k/htc_drv_init.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/ath/ath9k/htc_drv_init.c b/drivers/net/wireless/ath/ath9k/htc_drv_init.c index f4e1de20d99c..c57d6b859c04 100644 --- a/drivers/net/wireless/ath/ath9k/htc_drv_init.c +++ b/drivers/net/wireless/ath/ath9k/htc_drv_init.c @@ -34,6 +34,10 @@ static int ath9k_htc_btcoex_enable; module_param_named(btcoex_enable, ath9k_htc_btcoex_enable, int, 0444); MODULE_PARM_DESC(btcoex_enable, "Enable wifi-BT coexistence"); +static int ath9k_ps_enable; +module_param_named(ps_enable, ath9k_ps_enable, int, 0444); +MODULE_PARM_DESC(ps_enable, "Enable WLAN PowerSave"); + #define CHAN2G(_freq, _idx) { \ .center_freq = (_freq), \ .hw_value = (_idx), \ @@ -725,12 +729,14 @@ static void ath9k_set_hw_capab(struct ath9k_htc_priv *priv, IEEE80211_HW_SPECTRUM_MGMT | IEEE80211_HW_HAS_RATE_CONTROL | IEEE80211_HW_RX_INCLUDES_FCS | - IEEE80211_HW_SUPPORTS_PS | IEEE80211_HW_PS_NULLFUNC_STACK | IEEE80211_HW_REPORTS_TX_ACK_STATUS | IEEE80211_HW_MFP_CAPABLE | IEEE80211_HW_HOST_BROADCAST_PS_BUFFERING; + if (ath9k_ps_enable) + hw->flags |= IEEE80211_HW_SUPPORTS_PS; + hw->wiphy->interface_modes = BIT(NL80211_IFTYPE_STATION) | BIT(NL80211_IFTYPE_ADHOC) | From 4fcfc7443d072582b5047b8b391d711590e5645c Mon Sep 17 00:00:00 2001 From: Oleksij Rempel Date: Sun, 2 Feb 2014 10:55:18 +0100 Subject: [PATCH 170/596] ar5523: fix usb id for Gigaset. Raw id and FW id should be switched. Tested-by: Oleksij Rempel Signed-off-by: Oleksij Rempel Cc: stable@vger.kernel.org Signed-off-by: John W. Linville --- drivers/net/wireless/ath/ar5523/ar5523.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/wireless/ath/ar5523/ar5523.c b/drivers/net/wireless/ath/ar5523/ar5523.c index 8aa20df55e50..507d9a9ee69a 100644 --- a/drivers/net/wireless/ath/ar5523/ar5523.c +++ b/drivers/net/wireless/ath/ar5523/ar5523.c @@ -1764,7 +1764,7 @@ static struct usb_device_id ar5523_id_table[] = { AR5523_DEVICE_UG(0x07d1, 0x3a07), /* D-Link / WUA-2340 rev A1 */ AR5523_DEVICE_UG(0x1690, 0x0712), /* Gigaset / AR5523 */ AR5523_DEVICE_UG(0x1690, 0x0710), /* Gigaset / SMCWUSBTG */ - AR5523_DEVICE_UG(0x129b, 0x160c), /* Gigaset / USB stick 108 + AR5523_DEVICE_UG(0x129b, 0x160b), /* Gigaset / USB stick 108 (CyberTAN Technology) */ AR5523_DEVICE_UG(0x16ab, 0x7801), /* Globalsun / AR5523_1 */ AR5523_DEVICE_UX(0x16ab, 0x7811), /* Globalsun / AR5523_2 */ From a243de48558397f438e299178cac29f6da8fc0ce Mon Sep 17 00:00:00 2001 From: Stanislaw Gruszka Date: Mon, 3 Feb 2014 11:45:51 +0100 Subject: [PATCH 171/596] ath9k_htc: avoid scheduling while atomic on sta_rc_update mac80211 ->sta_rc_update() callback must be atomic. Since we have to take mutex and do other operations that can sleep when sending fimrware commands to device, the only option to satisfy atomicity requirement of ->sta_rc_update(), that I can see, is introduce work_struct and defer uploading new rates to that work. Tested-by: Oleksij Rempel Signed-off-by: Stanislaw Gruszka Signed-off-by: John W. Linville --- drivers/net/wireless/ath/ath9k/htc.h | 2 + drivers/net/wireless/ath/ath9k/htc_drv_main.c | 104 ++++++++++-------- 2 files changed, 62 insertions(+), 44 deletions(-) diff --git a/drivers/net/wireless/ath/ath9k/htc.h b/drivers/net/wireless/ath/ath9k/htc.h index 58da3468d1f0..99a203174f45 100644 --- a/drivers/net/wireless/ath/ath9k/htc.h +++ b/drivers/net/wireless/ath/ath9k/htc.h @@ -262,6 +262,8 @@ enum tid_aggr_state { struct ath9k_htc_sta { u8 index; enum tid_aggr_state tid_state[ATH9K_HTC_MAX_TID]; + struct work_struct rc_update_work; + struct ath9k_htc_priv *htc_priv; }; #define ATH9K_HTC_RXBUF 256 diff --git a/drivers/net/wireless/ath/ath9k/htc_drv_main.c b/drivers/net/wireless/ath/ath9k/htc_drv_main.c index a57af9b96a39..c9254a61ca52 100644 --- a/drivers/net/wireless/ath/ath9k/htc_drv_main.c +++ b/drivers/net/wireless/ath/ath9k/htc_drv_main.c @@ -1270,54 +1270,16 @@ static void ath9k_htc_configure_filter(struct ieee80211_hw *hw, mutex_unlock(&priv->mutex); } -static int ath9k_htc_sta_add(struct ieee80211_hw *hw, - struct ieee80211_vif *vif, - struct ieee80211_sta *sta) +static void ath9k_htc_sta_rc_update_work(struct work_struct *work) { - struct ath9k_htc_priv *priv = hw->priv; - int ret; - - mutex_lock(&priv->mutex); - ath9k_htc_ps_wakeup(priv); - ret = ath9k_htc_add_station(priv, vif, sta); - if (!ret) - ath9k_htc_init_rate(priv, sta); - ath9k_htc_ps_restore(priv); - mutex_unlock(&priv->mutex); - - return ret; -} - -static int ath9k_htc_sta_remove(struct ieee80211_hw *hw, - struct ieee80211_vif *vif, - struct ieee80211_sta *sta) -{ - struct ath9k_htc_priv *priv = hw->priv; - struct ath9k_htc_sta *ista; - int ret; - - mutex_lock(&priv->mutex); - ath9k_htc_ps_wakeup(priv); - ista = (struct ath9k_htc_sta *) sta->drv_priv; - htc_sta_drain(priv->htc, ista->index); - ret = ath9k_htc_remove_station(priv, vif, sta); - ath9k_htc_ps_restore(priv); - mutex_unlock(&priv->mutex); - - return ret; -} - -static void ath9k_htc_sta_rc_update(struct ieee80211_hw *hw, - struct ieee80211_vif *vif, - struct ieee80211_sta *sta, u32 changed) -{ - struct ath9k_htc_priv *priv = hw->priv; + struct ath9k_htc_sta *ista = + container_of(work, struct ath9k_htc_sta, rc_update_work); + struct ieee80211_sta *sta = + container_of((void *)ista, struct ieee80211_sta, drv_priv); + struct ath9k_htc_priv *priv = ista->htc_priv; struct ath_common *common = ath9k_hw_common(priv->ah); struct ath9k_htc_target_rate trate; - if (!(changed & IEEE80211_RC_SUPP_RATES_CHANGED)) - return; - mutex_lock(&priv->mutex); ath9k_htc_ps_wakeup(priv); @@ -1336,6 +1298,60 @@ static void ath9k_htc_sta_rc_update(struct ieee80211_hw *hw, mutex_unlock(&priv->mutex); } +static int ath9k_htc_sta_add(struct ieee80211_hw *hw, + struct ieee80211_vif *vif, + struct ieee80211_sta *sta) +{ + struct ath9k_htc_priv *priv = hw->priv; + struct ath9k_htc_sta *ista = (struct ath9k_htc_sta *) sta->drv_priv; + int ret; + + mutex_lock(&priv->mutex); + ath9k_htc_ps_wakeup(priv); + ret = ath9k_htc_add_station(priv, vif, sta); + if (!ret) { + INIT_WORK(&ista->rc_update_work, ath9k_htc_sta_rc_update_work); + ista->htc_priv = priv; + ath9k_htc_init_rate(priv, sta); + } + ath9k_htc_ps_restore(priv); + mutex_unlock(&priv->mutex); + + return ret; +} + +static int ath9k_htc_sta_remove(struct ieee80211_hw *hw, + struct ieee80211_vif *vif, + struct ieee80211_sta *sta) +{ + struct ath9k_htc_priv *priv = hw->priv; + struct ath9k_htc_sta *ista = (struct ath9k_htc_sta *) sta->drv_priv; + int ret; + + cancel_work_sync(&ista->rc_update_work); + + mutex_lock(&priv->mutex); + ath9k_htc_ps_wakeup(priv); + htc_sta_drain(priv->htc, ista->index); + ret = ath9k_htc_remove_station(priv, vif, sta); + ath9k_htc_ps_restore(priv); + mutex_unlock(&priv->mutex); + + return ret; +} + +static void ath9k_htc_sta_rc_update(struct ieee80211_hw *hw, + struct ieee80211_vif *vif, + struct ieee80211_sta *sta, u32 changed) +{ + struct ath9k_htc_sta *ista = (struct ath9k_htc_sta *) sta->drv_priv; + + if (!(changed & IEEE80211_RC_SUPP_RATES_CHANGED)) + return; + + schedule_work(&ista->rc_update_work); +} + static int ath9k_htc_conf_tx(struct ieee80211_hw *hw, struct ieee80211_vif *vif, u16 queue, const struct ieee80211_tx_queue_params *params) From 3683a07b29d2bddebb903f1400860c77d8e423f3 Mon Sep 17 00:00:00 2001 From: Sujith Manoharan Date: Tue, 4 Feb 2014 08:37:52 +0530 Subject: [PATCH 172/596] ath9k: Fix build error on ARM Use mdelay instead of udelay to fix this error: ERROR: "__bad_udelay" [drivers/net/wireless/ath/ath9k/ath9k_hw.ko] undefined! make[1]: *** [__modpost] Error 1 make: *** [modules] Error 2 Reported-by: Josh Boyer Signed-off-by: Sujith Manoharan Signed-off-by: John W. Linville --- drivers/net/wireless/ath/ath9k/hw.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/drivers/net/wireless/ath/ath9k/hw.c b/drivers/net/wireless/ath/ath9k/hw.c index fbf43c05713f..11eab9f01fd8 100644 --- a/drivers/net/wireless/ath/ath9k/hw.c +++ b/drivers/net/wireless/ath/ath9k/hw.c @@ -1316,7 +1316,7 @@ static bool ath9k_hw_set_reset(struct ath_hw *ah, int type) if (AR_SREV_9300_20_OR_LATER(ah)) udelay(50); else if (AR_SREV_9100(ah)) - udelay(10000); + mdelay(10); else udelay(100); @@ -2051,9 +2051,8 @@ static bool ath9k_hw_set_power_awake(struct ath_hw *ah) REG_SET_BIT(ah, AR_RTC_FORCE_WAKE, AR_RTC_FORCE_WAKE_EN); - if (AR_SREV_9100(ah)) - udelay(10000); + mdelay(10); else udelay(50); From 8298383c2cd5a6d0639f1bb1781fba181bd20154 Mon Sep 17 00:00:00 2001 From: Sujith Manoharan Date: Tue, 4 Feb 2014 08:37:53 +0530 Subject: [PATCH 173/596] ath9k: Do not support PowerSave by default Even though we make sure PowerSave is not enabled by default by disabling the flag, WIPHY_FLAG_PS_ON_BY_DEFAULT on init, PS could be enabled by userspace based on various factors like battery usage etc. Since PS in ath9k is just broken and has been untested for years, remove support for it, but allow a user to explicitly enable it using a module parameter. Cc: stable@vger.kernel.org Signed-off-by: Sujith Manoharan Signed-off-by: John W. Linville --- drivers/net/wireless/ath/ath9k/init.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/ath/ath9k/init.c b/drivers/net/wireless/ath/ath9k/init.c index c36de303c8f3..1fc2e5a26b52 100644 --- a/drivers/net/wireless/ath/ath9k/init.c +++ b/drivers/net/wireless/ath/ath9k/init.c @@ -57,6 +57,10 @@ static int ath9k_bt_ant_diversity; module_param_named(bt_ant_diversity, ath9k_bt_ant_diversity, int, 0444); MODULE_PARM_DESC(bt_ant_diversity, "Enable WLAN/BT RX antenna diversity"); +static int ath9k_ps_enable; +module_param_named(ps_enable, ath9k_ps_enable, int, 0444); +MODULE_PARM_DESC(ps_enable, "Enable WLAN PowerSave"); + bool is_ath9k_unloaded; /* We use the hw_value as an index into our private channel structure */ @@ -903,13 +907,15 @@ static void ath9k_set_hw_capab(struct ath_softc *sc, struct ieee80211_hw *hw) hw->flags = IEEE80211_HW_RX_INCLUDES_FCS | IEEE80211_HW_HOST_BROADCAST_PS_BUFFERING | IEEE80211_HW_SIGNAL_DBM | - IEEE80211_HW_SUPPORTS_PS | IEEE80211_HW_PS_NULLFUNC_STACK | IEEE80211_HW_SPECTRUM_MGMT | IEEE80211_HW_REPORTS_TX_ACK_STATUS | IEEE80211_HW_SUPPORTS_RC_TABLE | IEEE80211_HW_SUPPORTS_HT_CCK_RATES; + if (ath9k_ps_enable) + hw->flags |= IEEE80211_HW_SUPPORTS_PS; + if (sc->sc_ah->caps.hw_caps & ATH9K_HW_CAP_HT) { hw->flags |= IEEE80211_HW_AMPDU_AGGREGATION; From 4cfe9a8d58292dc8e32a8093a95187d47507d394 Mon Sep 17 00:00:00 2001 From: Sujith Manoharan Date: Tue, 4 Feb 2014 08:37:54 +0530 Subject: [PATCH 174/596] ath9k: Fix TX power calculation The commit, "ath9k_hw: Fix incorrect Tx control power in AR9003 template" fixed the incorrect values in the eeprom templates, but if boards have already been calibrated with incorrect values, they would still be using the wrong TX power. Fix this by assigning a default value in such cases. Cc: Rajkumar Manoharan Signed-off-by: Sujith Manoharan Signed-off-by: John W. Linville --- drivers/net/wireless/ath/ath9k/ar9003_eeprom.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/net/wireless/ath/ath9k/ar9003_eeprom.c b/drivers/net/wireless/ath/ath9k/ar9003_eeprom.c index 25243cbc07f0..b8daff78b9d1 100644 --- a/drivers/net/wireless/ath/ath9k/ar9003_eeprom.c +++ b/drivers/net/wireless/ath/ath9k/ar9003_eeprom.c @@ -5065,6 +5065,10 @@ static u16 ar9003_hw_get_max_edge_power(struct ar9300_eeprom *eep, break; } } + + if (is2GHz && !twiceMaxEdgePower) + twiceMaxEdgePower = 60; + return twiceMaxEdgePower; } From f7db1588d6028c97c098bb6445eaabc56a25fed8 Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Mon, 20 Jan 2014 23:22:07 +0800 Subject: [PATCH 175/596] spi: nuc900: Set SPI_LSB_FIRST for master->mode_bits if hw->pdata->lsb is true Otherwise, spi_setup() fails with unsupported mode bits message. Signed-off-by: Axel Lin Signed-off-by: Mark Brown Cc: stable@vger.kernel.org --- drivers/spi/spi-nuc900.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/spi/spi-nuc900.c b/drivers/spi/spi-nuc900.c index 50406306bc20..bae97ffec4b9 100644 --- a/drivers/spi/spi-nuc900.c +++ b/drivers/spi/spi-nuc900.c @@ -361,6 +361,8 @@ static int nuc900_spi_probe(struct platform_device *pdev) init_completion(&hw->done); master->mode_bits = SPI_CPOL | SPI_CPHA | SPI_CS_HIGH; + if (hw->pdata->lsb) + master->mode_bits |= SPI_LSB_FIRST; master->num_chipselect = hw->pdata->num_cs; master->bus_num = hw->pdata->bus_num; hw->bitbang.master = hw->master; From fcb6a15c2e7e76d493e6f91ea889ab40e1c643a4 Mon Sep 17 00:00:00 2001 From: Dirk Brandewie Date: Mon, 3 Feb 2014 08:55:31 -0800 Subject: [PATCH 176/596] intel_pstate: Take core C0 time into account for core busy calculation Take non-idle time into account when calculating core busy time. This ensures that intel_pstate will notice a decrease in load. References: https://bugzilla.kernel.org/show_bug.cgi?id=66581 Cc: 3.10+ # 3.10+ Signed-off-by: Dirk Brandewie Signed-off-by: Rafael J. Wysocki --- drivers/cpufreq/intel_pstate.c | 21 +++++++++++++++++---- 1 file changed, 17 insertions(+), 4 deletions(-) diff --git a/drivers/cpufreq/intel_pstate.c b/drivers/cpufreq/intel_pstate.c index 7e257b233602..79606f473f48 100644 --- a/drivers/cpufreq/intel_pstate.c +++ b/drivers/cpufreq/intel_pstate.c @@ -57,6 +57,7 @@ struct sample { int32_t core_pct_busy; u64 aperf; u64 mperf; + unsigned long long tsc; int freq; }; @@ -96,6 +97,7 @@ struct cpudata { u64 prev_aperf; u64 prev_mperf; + unsigned long long prev_tsc; int sample_ptr; struct sample samples[SAMPLE_COUNT]; }; @@ -548,30 +550,41 @@ static inline void intel_pstate_calc_busy(struct cpudata *cpu, struct sample *sample) { u64 core_pct; - core_pct = div64_u64(int_tofp(sample->aperf * 100), - sample->mperf); - sample->freq = fp_toint(cpu->pstate.max_pstate * core_pct * 1000); + u64 c0_pct; - sample->core_pct_busy = core_pct; + core_pct = div64_u64(sample->aperf * 100, sample->mperf); + + c0_pct = div64_u64(sample->mperf * 100, sample->tsc); + sample->freq = fp_toint( + mul_fp(int_tofp(cpu->pstate.max_pstate), + int_tofp(core_pct * 1000))); + + sample->core_pct_busy = mul_fp(int_tofp(core_pct), + div_fp(int_tofp(c0_pct + 1), int_tofp(100))); } static inline void intel_pstate_sample(struct cpudata *cpu) { u64 aperf, mperf; + unsigned long long tsc; rdmsrl(MSR_IA32_APERF, aperf); rdmsrl(MSR_IA32_MPERF, mperf); + tsc = native_read_tsc(); cpu->sample_ptr = (cpu->sample_ptr + 1) % SAMPLE_COUNT; cpu->samples[cpu->sample_ptr].aperf = aperf; cpu->samples[cpu->sample_ptr].mperf = mperf; + cpu->samples[cpu->sample_ptr].tsc = tsc; cpu->samples[cpu->sample_ptr].aperf -= cpu->prev_aperf; cpu->samples[cpu->sample_ptr].mperf -= cpu->prev_mperf; + cpu->samples[cpu->sample_ptr].tsc -= cpu->prev_tsc; intel_pstate_calc_busy(cpu, &cpu->samples[cpu->sample_ptr]); cpu->prev_aperf = aperf; cpu->prev_mperf = mperf; + cpu->prev_tsc = tsc; } static inline void intel_pstate_set_sample_time(struct cpudata *cpu) From 861e0f5bae1c4f4026b58ca10fa3b0fc02f1540c Mon Sep 17 00:00:00 2001 From: Andre Heider Date: Tue, 4 Feb 2014 21:27:44 +0100 Subject: [PATCH 177/596] usb: dwc2: bail out early when booting with "nousb" Add usb_disabled() check to prevent kernel oops when booting with "nousb" in the cmdline: Unable to handle kernel NULL pointer dereference at virtual address 00000030 ... PC is at bus_add_device+0xe0/0x18c LR is at device_add_groups+0x1c/0x20 ... [] (bus_add_device) from [] (device_add+0x41c/0x538) [] (device_add) from [] (usb_new_device+0x270/0x35c) [] (usb_new_device) from [] (usb_add_hcd+0x4fc/0x760) [] (usb_add_hcd) from [] (dwc2_hcd_init+0x434/0x510) [] (dwc2_hcd_init) from [] (dwc2_driver_probe+0x130/0x170) [] (dwc2_driver_probe) from [] (platform_drv_probe+0x28/0x58) Signed-off-by: Andre Heider Acked-by: Paul Zimmerman Signed-off-by: Greg Kroah-Hartman --- drivers/usb/dwc2/platform.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/usb/dwc2/platform.c b/drivers/usb/dwc2/platform.c index d01d0d3f2cf0..eaba547ce26b 100644 --- a/drivers/usb/dwc2/platform.c +++ b/drivers/usb/dwc2/platform.c @@ -124,6 +124,9 @@ static int dwc2_driver_probe(struct platform_device *dev) int retval; int irq; + if (usb_disabled()) + return -ENODEV; + match = of_match_device(dwc2_of_match_table, &dev->dev); if (match && match->data) { params = match->data; From a23666c49938bba08df3654fc66cd051587f7673 Mon Sep 17 00:00:00 2001 From: Paul Zimmerman Date: Tue, 4 Feb 2014 11:42:15 -0800 Subject: [PATCH 178/596] usb: dwc2: fix role switch breakage Commit beb7e592bc "staging: dwc2: add check on dwc2_core_reset return" broke the B -> A role switching on OTG-enabled platforms. This commit fixes it. Reported-by: Dinh Nguyen Tested-by: Dinh Nguyen Signed-off-by: Paul Zimmerman Signed-off-by: Greg Kroah-Hartman --- drivers/usb/dwc2/core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/usb/dwc2/core.c b/drivers/usb/dwc2/core.c index 8565d87f94b4..1d129884cc39 100644 --- a/drivers/usb/dwc2/core.c +++ b/drivers/usb/dwc2/core.c @@ -216,7 +216,7 @@ static int dwc2_hs_phy_init(struct dwc2_hsotg *hsotg, bool select_phy) int retval = 0; if (!select_phy) - return -ENODEV; + return 0; usbcfg = readl(hsotg->regs + GUSBCFG); From 51b1b6491752ac066ee8d32cc66042fcc955fef6 Mon Sep 17 00:00:00 2001 From: Paul Zimmerman Date: Tue, 4 Feb 2014 11:21:24 -0800 Subject: [PATCH 179/596] usb: dwc2: fix memory corruption in dwc2 driver The move from the staging tree to the main tree exposed a longstanding memory corruption bug in the dwc2 driver. The reordering of the driver initialization caused the dwc2 driver to corrupt the initialization data of the sdhci driver on the Raspberry Pi platform, which made the bug show up. The error is in calling to_usb_device(hsotg->dev), since ->dev is not a member of struct usb_device. The easiest fix is to just remove the offending code, since it is not really needed. Thanks to Stephen Warren for tracking down the cause of this. Reported-by: Andre Heider Tested-by: Stephen Warren Signed-off-by: Paul Zimmerman Signed-off-by: Greg Kroah-Hartman --- drivers/usb/dwc2/hcd.c | 11 ----------- 1 file changed, 11 deletions(-) diff --git a/drivers/usb/dwc2/hcd.c b/drivers/usb/dwc2/hcd.c index f59484d43b35..4d918ed8d343 100644 --- a/drivers/usb/dwc2/hcd.c +++ b/drivers/usb/dwc2/hcd.c @@ -2565,25 +2565,14 @@ static void _dwc2_hcd_endpoint_reset(struct usb_hcd *hcd, struct usb_host_endpoint *ep) { struct dwc2_hsotg *hsotg = dwc2_hcd_to_hsotg(hcd); - int is_control = usb_endpoint_xfer_control(&ep->desc); - int is_out = usb_endpoint_dir_out(&ep->desc); - int epnum = usb_endpoint_num(&ep->desc); - struct usb_device *udev; unsigned long flags; dev_dbg(hsotg->dev, "DWC OTG HCD EP RESET: bEndpointAddress=0x%02x\n", ep->desc.bEndpointAddress); - udev = to_usb_device(hsotg->dev); - spin_lock_irqsave(&hsotg->lock, flags); - - usb_settoggle(udev, epnum, is_out, 0); - if (is_control) - usb_settoggle(udev, epnum, !is_out, 0); dwc2_hcd_endpoint_reset(hsotg, ep); - spin_unlock_irqrestore(&hsotg->lock, flags); } From 67847baee056892dc35efb9c3fd05ae7f075588c Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Bj=C3=B8rn=20Mork?= Date: Tue, 14 Jan 2014 18:56:54 +0100 Subject: [PATCH 180/596] usb: ftdi_sio: add Mindstorms EV3 console adapter MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Cc: stable Signed-off-by: Bjørn Mork Signed-off-by: Greg Kroah-Hartman --- drivers/usb/serial/ftdi_sio.c | 1 + drivers/usb/serial/ftdi_sio_ids.h | 1 + 2 files changed, 2 insertions(+) diff --git a/drivers/usb/serial/ftdi_sio.c b/drivers/usb/serial/ftdi_sio.c index ce0d7b0db012..0f3248a017b3 100644 --- a/drivers/usb/serial/ftdi_sio.c +++ b/drivers/usb/serial/ftdi_sio.c @@ -152,6 +152,7 @@ static const struct usb_device_id id_table_combined[] = { { USB_DEVICE(FTDI_VID, FTDI_CANUSB_PID) }, { USB_DEVICE(FTDI_VID, FTDI_CANDAPTER_PID) }, { USB_DEVICE(FTDI_VID, FTDI_NXTCAM_PID) }, + { USB_DEVICE(FTDI_VID, FTDI_EV3CON_PID) }, { USB_DEVICE(FTDI_VID, FTDI_SCS_DEVICE_0_PID) }, { USB_DEVICE(FTDI_VID, FTDI_SCS_DEVICE_1_PID) }, { USB_DEVICE(FTDI_VID, FTDI_SCS_DEVICE_2_PID) }, diff --git a/drivers/usb/serial/ftdi_sio_ids.h b/drivers/usb/serial/ftdi_sio_ids.h index a7019d1e3058..eb0302135348 100644 --- a/drivers/usb/serial/ftdi_sio_ids.h +++ b/drivers/usb/serial/ftdi_sio_ids.h @@ -50,6 +50,7 @@ #define TI_XDS100V2_PID 0xa6d0 #define FTDI_NXTCAM_PID 0xABB8 /* NXTCam for Mindstorms NXT */ +#define FTDI_EV3CON_PID 0xABB9 /* Mindstorms EV3 Console Adapter */ /* US Interface Navigator (http://www.usinterface.com/) */ #define FTDI_USINT_CAT_PID 0xb810 /* Navigator CAT and 2nd PTT lines */ From e9d123a50b30f2b7b487d46df8820c3ae1d589ff Mon Sep 17 00:00:00 2001 From: Josh Boyer Date: Tue, 28 Jan 2014 09:53:50 -0500 Subject: [PATCH 181/596] usb: phy: move some error messages to debug the PHY layer is supposed to be optional, considering some PHY have no control bus for SW to poke around. After commit 1ae5799 (usb: hcd: Initialize USB phy if needed) any HCD which didn't provide a PHY driver would emit annoying error messages. In this patch we're decreasing those messages to debugging only and we also add a PHY prefix or use dev_dbg so we know where they're coming from. Reported-by: Josh Boyer Signed-off-by: Felipe Balbi Signed-off-by: Josh Boyer Signed-off-by: Greg Kroah-Hartman --- drivers/usb/phy/phy.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) diff --git a/drivers/usb/phy/phy.c b/drivers/usb/phy/phy.c index e6f61e4361df..8afa813d690b 100644 --- a/drivers/usb/phy/phy.c +++ b/drivers/usb/phy/phy.c @@ -130,7 +130,7 @@ struct usb_phy *usb_get_phy(enum usb_phy_type type) phy = __usb_find_phy(&phy_list, type); if (IS_ERR(phy) || !try_module_get(phy->dev->driver->owner)) { - pr_err("unable to find transceiver of type %s\n", + pr_debug("PHY: unable to find transceiver of type %s\n", usb_phy_type_string(type)); goto err0; } @@ -228,7 +228,7 @@ struct usb_phy *usb_get_phy_dev(struct device *dev, u8 index) phy = __usb_find_phy_dev(dev, &phy_bind_list, index); if (IS_ERR(phy) || !try_module_get(phy->dev->driver->owner)) { - pr_err("unable to find transceiver\n"); + dev_dbg(dev, "unable to find transceiver\n"); goto err0; } @@ -424,10 +424,8 @@ int usb_bind_phy(const char *dev_name, u8 index, unsigned long flags; phy_bind = kzalloc(sizeof(*phy_bind), GFP_KERNEL); - if (!phy_bind) { - pr_err("phy_bind(): No memory for phy_bind"); + if (!phy_bind) return -ENOMEM; - } phy_bind->dev_name = dev_name; phy_bind->phy_dev_name = phy_dev_name; From a9c143c82608bee2a36410caa56d82cd86bdc7fa Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Thu, 30 Jan 2014 10:20:29 -0500 Subject: [PATCH 182/596] usb-storage: restrict bcdDevice range for Super Top in Cypress ATACB The Cypress ATACB unusual-devs entry for the Super Top SATA bridge causes problems. Although it was originally reported only for bcdDevice = 0x160, its range was much larger. This resulted in a bug report for bcdDevice 0x220, so the range was capped at 0x219. Now Milan reports errors with bcdDevice 0x150. Therefore this patch restricts the range to just 0x160. Signed-off-by: Alan Stern Reported-and-tested-by: Milan Svoboda CC: Signed-off-by: Greg Kroah-Hartman --- drivers/usb/storage/unusual_cypress.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/usb/storage/unusual_cypress.h b/drivers/usb/storage/unusual_cypress.h index 65a6a75066a8..82e8ed0324e3 100644 --- a/drivers/usb/storage/unusual_cypress.h +++ b/drivers/usb/storage/unusual_cypress.h @@ -31,7 +31,7 @@ UNUSUAL_DEV( 0x04b4, 0x6831, 0x0000, 0x9999, "Cypress ISD-300LP", USB_SC_CYP_ATACB, USB_PR_DEVICE, NULL, 0), -UNUSUAL_DEV( 0x14cd, 0x6116, 0x0000, 0x0219, +UNUSUAL_DEV( 0x14cd, 0x6116, 0x0160, 0x0160, "Super Top", "USB 2.0 SATA BRIDGE", USB_SC_CYP_ATACB, USB_PR_DEVICE, NULL, 0), From c5637e5119c43452a00e27c274356b072263ecbb Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Tue, 21 Jan 2014 10:38:45 -0500 Subject: [PATCH 183/596] usb-storage: add unusual-devs entry for BlackBerry 9000 This patch adds an unusual-devs entry for the BlackBerry 9000. This fixes Bugzilla #22442. Signed-off-by: Alan Stern Reported-by: Moritz Moeller-Herrmann Tested-by: Moritz Moeller-Herrmann Cc: stable Signed-off-by: Greg Kroah-Hartman --- drivers/usb/storage/unusual_devs.h | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/usb/storage/unusual_devs.h b/drivers/usb/storage/unusual_devs.h index ad06255c2ade..adbeb255616a 100644 --- a/drivers/usb/storage/unusual_devs.h +++ b/drivers/usb/storage/unusual_devs.h @@ -1455,6 +1455,13 @@ UNUSUAL_DEV( 0x0f88, 0x042e, 0x0100, 0x0100, USB_SC_DEVICE, USB_PR_DEVICE, NULL, US_FL_FIX_CAPACITY ), +/* Reported by Moritz Moeller-Herrmann */ +UNUSUAL_DEV( 0x0fca, 0x8004, 0x0201, 0x0201, + "Research In Motion", + "BlackBerry Bold 9000", + USB_SC_DEVICE, USB_PR_DEVICE, NULL, + US_FL_MAX_SECTORS_64 ), + /* Reported by Michael Stattmann */ UNUSUAL_DEV( 0x0fce, 0xd008, 0x0000, 0x0000, "Sony Ericsson", From 2240c365108adbc4100a55654a5707e8e877a401 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Krist=C3=B3f=20Ralovich?= Date: Fri, 24 Jan 2014 12:18:35 +0100 Subject: [PATCH 184/596] USB: simple: add Dynastream ANT USB-m Stick device support MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Add support for ANT USB-m Stick from Dynastream Innovations, by listing USB pid [34366.944805] usb 6-1: New USB device found, idVendor=0fcf, idProduct=1009 [34366.944817] usb 6-1: New USB device strings: Mfr=1, Product=2, SerialNumber=3 [34366.944824] usb 6-1: Product: ANT USB-m Stick [34366.944831] usb 6-1: Manufacturer: Dynastream Innovations Device reported (https://code.google.com/p/antpm/issues/detail?id=5) to work through: $ modprobe usbserial vendor=0x0fcf product=0x1009 Signed-off-by: Kristóf Ralovich Signed-off-by: Johan Hovold Cc: stable Signed-off-by: Greg Kroah-Hartman --- drivers/usb/serial/usb-serial-simple.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/usb/serial/usb-serial-simple.c b/drivers/usb/serial/usb-serial-simple.c index f112b079ddfc..fb79775447b0 100644 --- a/drivers/usb/serial/usb-serial-simple.c +++ b/drivers/usb/serial/usb-serial-simple.c @@ -71,7 +71,8 @@ DEVICE(hp4x, HP4X_IDS); /* Suunto ANT+ USB Driver */ #define SUUNTO_IDS() \ - { USB_DEVICE(0x0fcf, 0x1008) } + { USB_DEVICE(0x0fcf, 0x1008) }, \ + { USB_DEVICE(0x0fcf, 0x1009) } /* Dynastream ANT USB-m Stick */ DEVICE(suunto, SUUNTO_IDS); /* Siemens USB/MPI adapter */ From 823d12c95c666fa7ab7dad208d735f6bc6afabdc Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Thu, 30 Jan 2014 10:43:22 -0500 Subject: [PATCH 185/596] usb-storage: enable multi-LUN scanning when needed People sometimes create their own custom-configured kernels and forget to enable CONFIG_SCSI_MULTI_LUN. This causes problems when they plug in a USB storage device (such as a card reader) with more than one LUN. Fortunately, we can tell fairly easily when a storage device claims to have more than one LUN. When that happens, this patch asks the SCSI layer to probe all the LUNs automatically, regardless of the config setting. The patch also updates the Kconfig help text for usb-storage, explaining that CONFIG_SCSI_MULTI_LUN may be necessary. Signed-off-by: Alan Stern Reported-by: Thomas Raschbacher CC: Matthew Dharm CC: James Bottomley Cc: stable Signed-off-by: Greg Kroah-Hartman --- drivers/usb/storage/Kconfig | 4 +++- drivers/usb/storage/scsiglue.c | 6 ++++++ 2 files changed, 9 insertions(+), 1 deletion(-) diff --git a/drivers/usb/storage/Kconfig b/drivers/usb/storage/Kconfig index 8470e1b114f2..1dd0604d1911 100644 --- a/drivers/usb/storage/Kconfig +++ b/drivers/usb/storage/Kconfig @@ -18,7 +18,9 @@ config USB_STORAGE This option depends on 'SCSI' support being enabled, but you probably also need 'SCSI device support: SCSI disk support' - (BLK_DEV_SD) for most USB storage devices. + (BLK_DEV_SD) for most USB storage devices. Some devices also + will require 'Probe all LUNs on each SCSI device' + (SCSI_MULTI_LUN). To compile this driver as a module, choose M here: the module will be called usb-storage. diff --git a/drivers/usb/storage/scsiglue.c b/drivers/usb/storage/scsiglue.c index 18509e6c21ab..9d38ddc8da49 100644 --- a/drivers/usb/storage/scsiglue.c +++ b/drivers/usb/storage/scsiglue.c @@ -78,6 +78,8 @@ static const char* host_info(struct Scsi_Host *host) static int slave_alloc (struct scsi_device *sdev) { + struct us_data *us = host_to_us(sdev->host); + /* * Set the INQUIRY transfer length to 36. We don't use any of * the extra data and many devices choke if asked for more or @@ -102,6 +104,10 @@ static int slave_alloc (struct scsi_device *sdev) */ blk_queue_update_dma_alignment(sdev->request_queue, (512 - 1)); + /* Tell the SCSI layer if we know there is more than one LUN */ + if (us->protocol == USB_PR_BULK && us->max_lun > 0) + sdev->sdev_bflags |= BLIST_FORCELUN; + return 0; } From 7b320cb1ed2dbd2c5f2a778197baf76fd6bf545a Mon Sep 17 00:00:00 2001 From: Stanislaw Gruszka Date: Tue, 4 Feb 2014 09:07:09 +0100 Subject: [PATCH 186/596] pinctrl: protect pinctrl_list add We have few fedora bug reports about list corruption on pinctrl, for example: https://bugzilla.redhat.com/show_bug.cgi?id=1051918 Most likely corruption happen due lack of protection of pinctrl_list when adding new nodes to it. Patch corrects that. Fixes: 42fed7ba44e ("pinctrl: move subsystem mutex to pinctrl_dev struct") Cc: stable@vger.kernel.org Signed-off-by: Stanislaw Gruszka Acked-by: Stephen Warren Signed-off-by: Linus Walleij --- drivers/pinctrl/core.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/pinctrl/core.c b/drivers/pinctrl/core.c index 5ee61a470016..cab020a58bf5 100644 --- a/drivers/pinctrl/core.c +++ b/drivers/pinctrl/core.c @@ -851,7 +851,9 @@ static struct pinctrl *create_pinctrl(struct device *dev) kref_init(&p->users); /* Add the pinctrl handle to the global list */ + mutex_lock(&pinctrl_list_mutex); list_add_tail(&p->node, &pinctrl_list); + mutex_unlock(&pinctrl_list_mutex); return p; } From e18ac62fa4b3f16234bab0d5a6627c57dbae9e7e Mon Sep 17 00:00:00 2001 From: Mika Westerberg Date: Mon, 3 Feb 2014 12:59:16 +0200 Subject: [PATCH 187/596] ACPI / video: Add HP EliteBook Revolve 810 to the blacklist On HP EliteBook Revolve 810 the ACPI backlight device doesn't work as expected. For example when resuming from system sleep, it seems to lose backlight settings. Forcing Intel driver fixes the problem so add this machine the ACPI video detect blacklist. Signed-off-by: Mika Westerberg Signed-off-by: Rafael J. Wysocki --- drivers/acpi/video_detect.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/acpi/video_detect.c b/drivers/acpi/video_detect.c index f0447d3daf2c..a697b77b8865 100644 --- a/drivers/acpi/video_detect.c +++ b/drivers/acpi/video_detect.c @@ -170,6 +170,14 @@ static struct dmi_system_id video_detect_dmi_table[] = { }, { .callback = video_detect_force_vendor, + .ident = "HP EliteBook Revolve 810", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Hewlett-Packard"), + DMI_MATCH(DMI_PRODUCT_NAME, "HP EliteBook Revolve 810 G1"), + }, + }, + { + .callback = video_detect_force_vendor, .ident = "Lenovo Yoga 13", .matches = { DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"), From eb5ed9a3221a30edc7b48b87af550dbcc0e82c80 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Thu, 30 Jan 2014 14:56:56 +0300 Subject: [PATCH 188/596] ACPI / utils: remove a pointless NULL check "element" can't be NULL because it is the address of a struct member. Signed-off-by: Dan Carpenter Signed-off-by: Rafael J. Wysocki --- drivers/acpi/utils.c | 4 ---- 1 file changed, 4 deletions(-) diff --git a/drivers/acpi/utils.c b/drivers/acpi/utils.c index 0347a37eb438..85e3b612bdc0 100644 --- a/drivers/acpi/utils.c +++ b/drivers/acpi/utils.c @@ -99,10 +99,6 @@ acpi_extract_package(union acpi_object *package, union acpi_object *element = &(package->package.elements[i]); - if (!element) { - return AE_BAD_DATA; - } - switch (element->type) { case ACPI_TYPE_INTEGER: From 085ca1175cdccc8e400f6d06cf64e209b46021a2 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Thu, 30 Jan 2014 14:55:15 +0300 Subject: [PATCH 189/596] ACPI / proc: remove unneeded NULL check We already verified that "ldev" was non-NULL earlier and also we dereference again without checking a three lines later. Signed-off-by: Dan Carpenter Signed-off-by: Rafael J. Wysocki --- drivers/acpi/proc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/acpi/proc.c b/drivers/acpi/proc.c index 50fe34ffe932..75c28eae8860 100644 --- a/drivers/acpi/proc.c +++ b/drivers/acpi/proc.c @@ -60,7 +60,7 @@ acpi_system_wakeup_device_seq_show(struct seq_file *seq, void *offset) seq_printf(seq, "%c%-8s %s:%s\n", dev->wakeup.flags.run_wake ? '*' : ' ', (device_may_wakeup(&dev->dev) || - (ldev && device_may_wakeup(ldev))) ? + device_may_wakeup(ldev)) ? "enabled" : "disabled", ldev->bus ? ldev->bus->name : "no-bus", dev_name(ldev)); From 47a08c85f70ebdbaaf42b444af57dd6f83e04485 Mon Sep 17 00:00:00 2001 From: "Luis G.F" Date: Tue, 21 Jan 2014 15:40:43 +0100 Subject: [PATCH 190/596] ACPI / battery: Fix incorrect sscanf() string in acpi_battery_init_alarm() Fix incorrect sscanf() string in acpi_battery_init_alarm(). Change from %ld to %lu, because 'x' is unsigned long. Signed-off-by: Luis G.F Signed-off-by: Rafael J. Wysocki --- drivers/acpi/battery.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/acpi/battery.c b/drivers/acpi/battery.c index 470e7542bf31..018a42883706 100644 --- a/drivers/acpi/battery.c +++ b/drivers/acpi/battery.c @@ -549,7 +549,7 @@ static ssize_t acpi_battery_alarm_store(struct device *dev, { unsigned long x; struct acpi_battery *battery = to_acpi_battery(dev_get_drvdata(dev)); - if (sscanf(buf, "%ld\n", &x) == 1) + if (sscanf(buf, "%lu\n", &x) == 1) battery->alarm = x/1000; if (acpi_battery_present(battery)) acpi_battery_set_alarm(battery); From 1877db75589a895bbdc4c4c3f23558e57b521141 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Wed, 5 Feb 2014 12:17:01 +1100 Subject: [PATCH 191/596] md/raid1: restore ability for check and repair to fix read errors. commit 30bc9b53878a9921b02e3b5bc4283ac1c6de102a md/raid1: fix bio handling problems in process_checks() Move the bio_reset() to a point before where BIO_UPTODATE is checked, so that check now always report that the bio is uptodate, even if it is not. This causes process_check() to sometimes treat read-errors as successful matches so the good data isn't written out. This patch preserves the flag until it is needed. Bug was introduced in 3.11, but backported to 3.10-stable (as it fixed an even worse bug). So suitable for any -stable since 3.10. Reported-and-tested-by: Michael Tokarev Cc: stable@vger.kernel.org (3.10+) Fixed: 30bc9b53878a9921b02e3b5bc4283ac1c6de102a Signed-off-by: NeilBrown --- drivers/md/raid1.c | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c index fd3a2a14b587..4a6ca1cb2e78 100644 --- a/drivers/md/raid1.c +++ b/drivers/md/raid1.c @@ -1953,11 +1953,15 @@ static int process_checks(struct r1bio *r1_bio) for (i = 0; i < conf->raid_disks * 2; i++) { int j; int size; + int uptodate; struct bio *b = r1_bio->bios[i]; if (b->bi_end_io != end_sync_read) continue; - /* fixup the bio for reuse */ + /* fixup the bio for reuse, but preserve BIO_UPTODATE */ + uptodate = test_bit(BIO_UPTODATE, &b->bi_flags); bio_reset(b); + if (!uptodate) + clear_bit(BIO_UPTODATE, &b->bi_flags); b->bi_vcnt = vcnt; b->bi_iter.bi_size = r1_bio->sectors << 9; b->bi_iter.bi_sector = r1_bio->sector + @@ -1990,11 +1994,14 @@ static int process_checks(struct r1bio *r1_bio) int j; struct bio *pbio = r1_bio->bios[primary]; struct bio *sbio = r1_bio->bios[i]; + int uptodate = test_bit(BIO_UPTODATE, &sbio->bi_flags); if (sbio->bi_end_io != end_sync_read) continue; + /* Now we can 'fixup' the BIO_UPTODATE flag */ + set_bit(BIO_UPTODATE, &sbio->bi_flags); - if (test_bit(BIO_UPTODATE, &sbio->bi_flags)) { + if (uptodate) { for (j = vcnt; j-- ; ) { struct page *p, *s; p = pbio->bi_io_vec[j].bv_page; @@ -2009,7 +2016,7 @@ static int process_checks(struct r1bio *r1_bio) if (j >= 0) atomic64_add(r1_bio->sectors, &mddev->resync_mismatches); if (j < 0 || (test_bit(MD_RECOVERY_CHECK, &mddev->recovery) - && test_bit(BIO_UPTODATE, &sbio->bi_flags))) { + && uptodate)) { /* No need to write to this device. */ sbio->bi_end_io = NULL; rdev_dec_pending(conf->mirrors[i].rdev, mddev); From 00503b6f702eaf23e7257d6287da72805d7d014c Mon Sep 17 00:00:00 2001 From: dingtianhong Date: Sat, 25 Jan 2014 13:00:29 +0800 Subject: [PATCH 192/596] bonding: fail_over_mac should only affect AB mode at enslave and removal processing According to bonding.txt, the fail_over_ma should only affect active-backup mode, but I found that the fail_over_mac could be set to active or follow in all modes, this will cause new slave could not be set to bond's MAC address at enslave processing and restore its own MAC address at removal processing. The correct way to fix the problem is that we should not add restrictions when setting options, just need to modify the bond enslave and removal processing to check the mode in addition to fail_over_mac when setting a slave's MAC during enslavement. The change active slave processing already only calls the fail_over_mac function when in active-backup mode. Thanks for Jay's suggestion. The patch also modify the pr_warning() to pr_warn(). Cc: Jay Vosburgh Cc: Veaceslav Falico Cc: Andy Gospodarek Signed-off-by: Ding Tianhong Signed-off-by: David S. Miller --- drivers/net/bonding/bond_main.c | 22 +++++++++++++++------- 1 file changed, 15 insertions(+), 7 deletions(-) diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 4c08018d7333..8e6dd13b3607 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -1270,9 +1270,13 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) if (slave_ops->ndo_set_mac_address == NULL) { if (!bond_has_slaves(bond)) { - pr_warning("%s: Warning: The first slave device specified does not support setting the MAC address. Setting fail_over_mac to active.", - bond_dev->name); - bond->params.fail_over_mac = BOND_FOM_ACTIVE; + pr_warn("%s: Warning: The first slave device specified does not support setting the MAC address.\n", + bond_dev->name); + if (bond->params.mode == BOND_MODE_ACTIVEBACKUP) { + bond->params.fail_over_mac = BOND_FOM_ACTIVE; + pr_warn("%s: Setting fail_over_mac to active for active-backup mode.\n", + bond_dev->name); + } } else if (bond->params.fail_over_mac != BOND_FOM_ACTIVE) { pr_err("%s: Error: The slave device specified does not support setting the MAC address, but fail_over_mac is not set to active.\n", bond_dev->name); @@ -1315,7 +1319,8 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) */ memcpy(new_slave->perm_hwaddr, slave_dev->dev_addr, ETH_ALEN); - if (!bond->params.fail_over_mac) { + if (!bond->params.fail_over_mac || + bond->params.mode != BOND_MODE_ACTIVEBACKUP) { /* * Set slave to master's mac address. The application already * set the master's mac address to that of the first slave @@ -1579,7 +1584,8 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) dev_close(slave_dev); err_restore_mac: - if (!bond->params.fail_over_mac) { + if (!bond->params.fail_over_mac || + bond->params.mode != BOND_MODE_ACTIVEBACKUP) { /* XXX TODO - fom follow mode needs to change master's * MAC if this slave's MAC is in use by the bond, or at * least print a warning. @@ -1672,7 +1678,8 @@ static int __bond_release_one(struct net_device *bond_dev, bond->current_arp_slave = NULL; - if (!all && !bond->params.fail_over_mac) { + if (!all && (!bond->params.fail_over_mac || + bond->params.mode != BOND_MODE_ACTIVEBACKUP)) { if (ether_addr_equal_64bits(bond_dev->dev_addr, slave->perm_hwaddr) && bond_has_slaves(bond)) pr_warn("%s: Warning: the permanent HWaddr of %s - %pM - is still in use by %s. Set the HWaddr of %s to a different address to avoid conflicts.\n", @@ -1769,7 +1776,8 @@ static int __bond_release_one(struct net_device *bond_dev, /* close slave before restoring its mac address */ dev_close(slave_dev); - if (bond->params.fail_over_mac != BOND_FOM_ACTIVE) { + if (bond->params.fail_over_mac != BOND_FOM_ACTIVE || + bond->params.mode != BOND_MODE_ACTIVEBACKUP) { /* restore original ("permanent") mac address */ memcpy(addr.sa_data, slave->perm_hwaddr, ETH_ALEN); addr.sa_family = slave_dev->type; From cc689aaa7abf33b2ccb02482e5e17885ea8903d1 Mon Sep 17 00:00:00 2001 From: dingtianhong Date: Sat, 25 Jan 2014 13:00:57 +0800 Subject: [PATCH 193/596] bonding: fail_over_mac should only affect AB mode in bond_set_mac_address() The fail_over_mac could be set to active or follow in any time for all modes, so if the fail_over_mac is not none and the current mode is not active-backup, the bond_set_mac_address() could not change the master and slave's MAC address. In bond_set_mac_address(), the fail_over_mac should only affect AB mode, so modify to check the mode in addition to fail_over_mac when setting bond's MAC address. Cc: Jay Vosburgh Cc: Veaceslav Falico Cc: Andy Gospodarek Signed-off-by: Ding Tianhong Signed-off-by: David S. Miller --- drivers/net/bonding/bond_main.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 8e6dd13b3607..35f7e90aef72 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -3439,7 +3439,8 @@ static int bond_set_mac_address(struct net_device *bond_dev, void *addr) /* If fail_over_mac is enabled, do nothing and return success. * Returning an error causes ifenslave to fail. */ - if (bond->params.fail_over_mac) + if (bond->params.fail_over_mac && + bond->params.mode == BOND_MODE_ACTIVEBACKUP) return 0; if (!is_valid_ether_addr(sa->sa_data)) From a664a4f7aa4f01ca8728e3ec43618327416fc8ff Mon Sep 17 00:00:00 2001 From: Shlomo Pongratz Date: Sun, 2 Feb 2014 15:42:10 +0200 Subject: [PATCH 194/596] net/ipv4: Use proper RCU APIs for writer-side in udp_offload.c RCU writer side should use rcu_dereference_protected() and not rcu_dereference(), fix that. This also removes the "suspicious RCU usage" warning seen when running with CONFIG_PROVE_RCU. Also, don't use rcu_assign_pointer/rcu_dereference for pointers which are invisible beyond the udp offload code. Fixes: b582ef0 ('net: Add GRO support for UDP encapsulating protocols') Reported-by: Eric Dumazet Cc: Eric Dumazet Signed-off-by: Or Gerlitz Signed-off-by: Shlomo Pongratz Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- net/ipv4/udp_offload.c | 15 ++++++++------- 1 file changed, 8 insertions(+), 7 deletions(-) diff --git a/net/ipv4/udp_offload.c b/net/ipv4/udp_offload.c index 2ffea6f31efc..88b4023ecfcf 100644 --- a/net/ipv4/udp_offload.c +++ b/net/ipv4/udp_offload.c @@ -17,6 +17,8 @@ static DEFINE_SPINLOCK(udp_offload_lock); static struct udp_offload_priv __rcu *udp_offload_base __read_mostly; +#define udp_deref_protected(X) rcu_dereference_protected(X, lockdep_is_held(&udp_offload_lock)) + struct udp_offload_priv { struct udp_offload *offload; struct rcu_head rcu; @@ -100,7 +102,6 @@ static struct sk_buff *udp4_ufo_fragment(struct sk_buff *skb, int udp_add_offload(struct udp_offload *uo) { - struct udp_offload_priv __rcu **head = &udp_offload_base; struct udp_offload_priv *new_offload = kzalloc(sizeof(*new_offload), GFP_ATOMIC); if (!new_offload) @@ -109,8 +110,8 @@ int udp_add_offload(struct udp_offload *uo) new_offload->offload = uo; spin_lock(&udp_offload_lock); - rcu_assign_pointer(new_offload->next, rcu_dereference(*head)); - rcu_assign_pointer(*head, new_offload); + new_offload->next = udp_offload_base; + rcu_assign_pointer(udp_offload_base, new_offload); spin_unlock(&udp_offload_lock); return 0; @@ -130,12 +131,12 @@ void udp_del_offload(struct udp_offload *uo) spin_lock(&udp_offload_lock); - uo_priv = rcu_dereference(*head); + uo_priv = udp_deref_protected(*head); for (; uo_priv != NULL; - uo_priv = rcu_dereference(*head)) { - + uo_priv = udp_deref_protected(*head)) { if (uo_priv->offload == uo) { - rcu_assign_pointer(*head, rcu_dereference(uo_priv->next)); + rcu_assign_pointer(*head, + udp_deref_protected(uo_priv->next)); goto unlock; } head = &uo_priv->next; From e2e2f51dd0254fa0002bcd1c5fda180348163f09 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Stefan=20S=C3=B8rensen?= Date: Mon, 3 Feb 2014 15:36:35 +0100 Subject: [PATCH 195/596] net:phy:dp83640: Declare that TX timestamping possible MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Set the SKBTX_IN_PROGRESS bit in tx_flags dp83640_txtstamp when doing tx timestamps as per Documentation/networking/timestamping.txt. Signed-off-by: Stefan Sørensen Acked-by: Richard Cochran Signed-off-by: David S. Miller --- drivers/net/phy/dp83640.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/phy/dp83640.c b/drivers/net/phy/dp83640.c index 547725fa8671..3f882eea6e1d 100644 --- a/drivers/net/phy/dp83640.c +++ b/drivers/net/phy/dp83640.c @@ -1281,6 +1281,7 @@ static void dp83640_txtstamp(struct phy_device *phydev, } /* fall through */ case HWTSTAMP_TX_ON: + skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS; skb_queue_tail(&dp83640->tx_queue, skb); schedule_work(&dp83640->ts_work); break; From 80671bd2a91b81b8e06634e8309ce94df688c42c Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Stefan=20S=C3=B8rensen?= Date: Mon, 3 Feb 2014 15:36:50 +0100 Subject: [PATCH 196/596] net:phy:dp83640: Do not hardcode timestamping event edge MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Currently the external timestamping code is hardcoded to use the rising edge even though the hardware has configurable event edge detection. This patch changes the code to use falling edge detection if PTP_FALLING_EDGE is set in the user supplied flags. Signed-off-by: Stefan Sørensen Acked-by: Richard Cochran Signed-off-by: David S. Miller --- drivers/net/phy/dp83640.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/net/phy/dp83640.c b/drivers/net/phy/dp83640.c index 3f882eea6e1d..e347c8a218fc 100644 --- a/drivers/net/phy/dp83640.c +++ b/drivers/net/phy/dp83640.c @@ -437,7 +437,10 @@ static int ptp_dp83640_enable(struct ptp_clock_info *ptp, if (on) { gpio_num = gpio_tab[EXTTS0_GPIO + index]; evnt |= (gpio_num & EVNT_GPIO_MASK) << EVNT_GPIO_SHIFT; - evnt |= EVNT_RISE; + if (rq->extts.flags & PTP_FALLING_EDGE) + evnt |= EVNT_FALL; + else + evnt |= EVNT_RISE; } ext_write(0, phydev, PAGE5, PTP_EVNT, evnt); return 0; From 62ad968402466c4777a6634c86010e86521bc745 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Stefan=20S=C3=B8rensen?= Date: Mon, 3 Feb 2014 15:36:58 +0100 Subject: [PATCH 197/596] net:phy:dp83640: Initialize PTP clocks at device init. MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The trigger and events functionality can be useful even if packet timestamping is not used, but the required PTP clock is only enabled when packet timestamping is started. This patch moves the clock enable to when the interface is configured. Signed-off-by: Stefan Sørensen Acked-by: Richard Cochran Signed-off-by: David S. Miller --- drivers/net/phy/dp83640.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) diff --git a/drivers/net/phy/dp83640.c b/drivers/net/phy/dp83640.c index e347c8a218fc..9414fa272160 100644 --- a/drivers/net/phy/dp83640.c +++ b/drivers/net/phy/dp83640.c @@ -1061,6 +1061,13 @@ static void dp83640_remove(struct phy_device *phydev) kfree(dp83640); } +static int dp83640_config_init(struct phy_device *phydev) +{ + enable_status_frames(phydev, true); + ext_write(0, phydev, PAGE4, PTP_CTL, PTP_ENABLE); + return 0; +} + static int dp83640_ack_interrupt(struct phy_device *phydev) { int err = phy_read(phydev, MII_DP83640_MISR); @@ -1198,11 +1205,6 @@ static int dp83640_hwtstamp(struct phy_device *phydev, struct ifreq *ifr) mutex_lock(&dp83640->clock->extreg_lock); - if (dp83640->hwts_tx_en || dp83640->hwts_rx_en) { - enable_status_frames(phydev, true); - ext_write(0, phydev, PAGE4, PTP_CTL, PTP_ENABLE); - } - ext_write(0, phydev, PAGE5, PTP_TXCFG0, txcfg0); ext_write(0, phydev, PAGE5, PTP_RXCFG0, rxcfg0); @@ -1334,6 +1336,7 @@ static struct phy_driver dp83640_driver = { .flags = PHY_HAS_INTERRUPT, .probe = dp83640_probe, .remove = dp83640_remove, + .config_init = dp83640_config_init, .config_aneg = genphy_config_aneg, .read_status = genphy_read_status, .ack_interrupt = dp83640_ack_interrupt, From 5273e3a5ca94fbeb8e07d31203069220d5e682aa Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Mon, 3 Feb 2014 12:35:46 -0800 Subject: [PATCH 198/596] net: phy: ensure Gigabit features are masked off if requested When a Gigabit PHY device is connected to a 10/100Mbits capable Ethernet MAC, the driver will restrict the phydev->supported modes to mask off Gigabit. If the Gigabit PHY comes out of reset with the Gigabit features set by default in MII_CTRL1000, it will keep advertising these feature, so by the time we call genphy_config_advert(), the condition on phydev->supported having the Gigabit features on is false, and we do not update MII_CTRL1000 with updated values, and we keep advertising Gigabit features, eventually configuring the PHY for Gigabit whilst the Ethernet MAC does not support that. This patches fixes the problem by ensuring that the Gigabit feature bits are always cleared in MII_CTRL1000, if the PHY happens to be a Gigabit PHY, and then, if Gigabit features are supported, setting those and updating MII_CTRL1000 accordingly. Reported-by: Max Filippov Signed-off-by: Florian Fainelli Tested-by: Max Filippov Signed-off-by: David S. Miller --- drivers/net/phy/phy_device.c | 38 +++++++++++++++++++++++------------- 1 file changed, 24 insertions(+), 14 deletions(-) diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c index 4b03e63639b7..82514e72b3d8 100644 --- a/drivers/net/phy/phy_device.c +++ b/drivers/net/phy/phy_device.c @@ -719,7 +719,7 @@ int phy_resume(struct phy_device *phydev) static int genphy_config_advert(struct phy_device *phydev) { u32 advertise; - int oldadv, adv; + int oldadv, adv, bmsr; int err, changed = 0; /* Only allow advertising what this PHY supports */ @@ -744,26 +744,36 @@ static int genphy_config_advert(struct phy_device *phydev) changed = 1; } + bmsr = phy_read(phydev, MII_BMSR); + if (bmsr < 0) + return bmsr; + + /* Per 802.3-2008, Section 22.2.4.2.16 Extended status all + * 1000Mbits/sec capable PHYs shall have the BMSR_ESTATEN bit set to a + * logical 1. + */ + if (!(bmsr & BMSR_ESTATEN)) + return changed; + /* Configure gigabit if it's supported */ + adv = phy_read(phydev, MII_CTRL1000); + if (adv < 0) + return adv; + + oldadv = adv; + adv &= ~(ADVERTISE_1000FULL | ADVERTISE_1000HALF); + if (phydev->supported & (SUPPORTED_1000baseT_Half | SUPPORTED_1000baseT_Full)) { - adv = phy_read(phydev, MII_CTRL1000); - if (adv < 0) - return adv; - - oldadv = adv; - adv &= ~(ADVERTISE_1000FULL | ADVERTISE_1000HALF); adv |= ethtool_adv_to_mii_ctrl1000_t(advertise); - - if (adv != oldadv) { - err = phy_write(phydev, MII_CTRL1000, adv); - - if (err < 0) - return err; + if (adv != oldadv) changed = 1; - } } + err = phy_write(phydev, MII_CTRL1000, adv); + if (err < 0) + return err; + return changed; } From 445a48cc9df90775d6479ef4c73d3a90184af4ed Mon Sep 17 00:00:00 2001 From: Max Filippov Date: Tue, 4 Feb 2014 03:33:09 +0400 Subject: [PATCH 199/596] net: ethoc: don't advertise gigabit speed on attached PHY OpenCores 10/100 Mbps MAC does not support speeds above 100 Mbps, but does not disable advertisement when PHY supports them. This results in non-functioning network when the MAC is connected to a gigabit PHY connected to a gigabit switch. The fix is to disable gigabit speed advertisement on attached PHY unconditionally. Signed-off-by: Max Filippov Signed-off-by: David S. Miller --- drivers/net/ethernet/ethoc.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/net/ethernet/ethoc.c b/drivers/net/ethernet/ethoc.c index f9c1cf536298..0a8533c0c01a 100644 --- a/drivers/net/ethernet/ethoc.c +++ b/drivers/net/ethernet/ethoc.c @@ -691,6 +691,11 @@ static int ethoc_mdio_probe(struct net_device *dev) } priv->phy = phy; + phy->advertising &= ~(ADVERTISED_1000baseT_Full | + ADVERTISED_1000baseT_Half); + phy->supported &= ~(SUPPORTED_1000baseT_Full | + SUPPORTED_1000baseT_Half); + return 0; } From a13aff0641a92dc0b95136e32526e2ce81ffc4ef Mon Sep 17 00:00:00 2001 From: Max Filippov Date: Tue, 4 Feb 2014 03:33:10 +0400 Subject: [PATCH 200/596] net: ethoc: set up MII management bus clock MII management bus clock is derived from the MAC clock by dividing it by MIIMODER register CLKDIV field value. This value may need to be set up in case it is undefined or its default value is too high (and communication with PHY is too slow) or too low (and communication with PHY is impossible). The value of CLKDIV is not specified directly, but is derived from the MAC clock for the default MII management bus frequency of 2.5MHz. The MAC clock may be specified in the platform data, or in the 'clocks' device tree attribute. Signed-off-by: Max Filippov Signed-off-by: David S. Miller --- drivers/net/ethernet/ethoc.c | 32 ++++++++++++++++++++++++++++++-- include/net/ethoc.h | 1 + 2 files changed, 31 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/ethoc.c b/drivers/net/ethernet/ethoc.c index 0a8533c0c01a..55e0fa03dc90 100644 --- a/drivers/net/ethernet/ethoc.c +++ b/drivers/net/ethernet/ethoc.c @@ -13,6 +13,7 @@ #include #include +#include #include #include #include @@ -219,6 +220,7 @@ struct ethoc { struct phy_device *phy; struct mii_bus *mdio; + struct clk *clk; s8 phy_id; }; @@ -1021,6 +1023,8 @@ static int ethoc_probe(struct platform_device *pdev) int num_bd; int ret = 0; bool random_mac = false; + struct ethoc_platform_data *pdata = dev_get_platdata(&pdev->dev); + u32 eth_clkfreq = pdata ? pdata->eth_clkfreq : 0; /* allocate networking device */ netdev = alloc_etherdev(sizeof(struct ethoc)); @@ -1135,8 +1139,7 @@ static int ethoc_probe(struct platform_device *pdev) } /* Allow the platform setup code to pass in a MAC address. */ - if (dev_get_platdata(&pdev->dev)) { - struct ethoc_platform_data *pdata = dev_get_platdata(&pdev->dev); + if (pdata) { memcpy(netdev->dev_addr, pdata->hwaddr, IFHWADDRLEN); priv->phy_id = pdata->phy_id; } else { @@ -1174,6 +1177,27 @@ static int ethoc_probe(struct platform_device *pdev) if (random_mac) netdev->addr_assign_type = NET_ADDR_RANDOM; + /* Allow the platform setup code to adjust MII management bus clock. */ + if (!eth_clkfreq) { + struct clk *clk = devm_clk_get(&pdev->dev, NULL); + + if (!IS_ERR(clk)) { + priv->clk = clk; + clk_prepare_enable(clk); + eth_clkfreq = clk_get_rate(clk); + } + } + if (eth_clkfreq) { + u32 clkdiv = MIIMODER_CLKDIV(eth_clkfreq / 2500000 + 1); + + if (!clkdiv) + clkdiv = 2; + dev_dbg(&pdev->dev, "setting MII clkdiv to %u\n", clkdiv); + ethoc_write(priv, MIIMODER, + (ethoc_read(priv, MIIMODER) & MIIMODER_NOPRE) | + clkdiv); + } + /* register MII bus */ priv->mdio = mdiobus_alloc(); if (!priv->mdio) { @@ -1239,6 +1263,8 @@ static int ethoc_probe(struct platform_device *pdev) kfree(priv->mdio->irq); mdiobus_free(priv->mdio); free: + if (priv->clk) + clk_disable_unprepare(priv->clk); free_netdev(netdev); out: return ret; @@ -1263,6 +1289,8 @@ static int ethoc_remove(struct platform_device *pdev) kfree(priv->mdio->irq); mdiobus_free(priv->mdio); } + if (priv->clk) + clk_disable_unprepare(priv->clk); unregister_netdev(netdev); free_netdev(netdev); } diff --git a/include/net/ethoc.h b/include/net/ethoc.h index 96f3789b27bc..2a2d6bb34eb8 100644 --- a/include/net/ethoc.h +++ b/include/net/ethoc.h @@ -16,6 +16,7 @@ struct ethoc_platform_data { u8 hwaddr[IFHWADDRLEN]; s8 phy_id; + u32 eth_clkfreq; }; #endif /* !LINUX_NET_ETHOC_H */ From 61950e82b384cbe581f7e9cb668f77dc1d8cef35 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Stefan=20S=C3=B8rensen?= Date: Tue, 4 Feb 2014 08:46:36 +0100 Subject: [PATCH 201/596] ptp: Allow selecting trigger/event index in testptp MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Currently the trigger/event is hardcoded to 0, this patch adds a new command line argument -i to select an arbitrary trigger/ event. Signed-off-by: Stefan Sørensen Acked-by: Richard Cochran Signed-off-by: David S. Miller --- Documentation/ptp/testptp.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/Documentation/ptp/testptp.c b/Documentation/ptp/testptp.c index a74d0a84d329..4aba0436da65 100644 --- a/Documentation/ptp/testptp.c +++ b/Documentation/ptp/testptp.c @@ -117,6 +117,7 @@ static void usage(char *progname) " -f val adjust the ptp clock frequency by 'val' ppb\n" " -g get the ptp clock time\n" " -h prints this message\n" + " -i val index for event/trigger\n" " -k val measure the time offset between system and phc clock\n" " for 'val' times (Maximum 25)\n" " -p val enable output with a period of 'val' nanoseconds\n" @@ -154,6 +155,7 @@ int main(int argc, char *argv[]) int capabilities = 0; int extts = 0; int gettime = 0; + int index = 0; int oneshot = 0; int pct_offset = 0; int n_samples = 0; @@ -167,7 +169,7 @@ int main(int argc, char *argv[]) progname = strrchr(argv[0], '/'); progname = progname ? 1+progname : argv[0]; - while (EOF != (c = getopt(argc, argv, "a:A:cd:e:f:ghk:p:P:sSt:v"))) { + while (EOF != (c = getopt(argc, argv, "a:A:cd:e:f:ghi:k:p:P:sSt:v"))) { switch (c) { case 'a': oneshot = atoi(optarg); @@ -190,6 +192,9 @@ int main(int argc, char *argv[]) case 'g': gettime = 1; break; + case 'i': + index = atoi(optarg); + break; case 'k': pct_offset = 1; n_samples = atoi(optarg); @@ -301,7 +306,7 @@ int main(int argc, char *argv[]) if (extts) { memset(&extts_request, 0, sizeof(extts_request)); - extts_request.index = 0; + extts_request.index = index; extts_request.flags = PTP_ENABLE_FEATURE; if (ioctl(fd, PTP_EXTTS_REQUEST, &extts_request)) { perror("PTP_EXTTS_REQUEST"); @@ -375,7 +380,7 @@ int main(int argc, char *argv[]) return -1; } memset(&perout_request, 0, sizeof(perout_request)); - perout_request.index = 0; + perout_request.index = index; perout_request.start.sec = ts.tv_sec + 2; perout_request.start.nsec = 0; perout_request.period.sec = 0; From 6049f2530cf2cb48a6fe8735309cc0b97aa7f700 Mon Sep 17 00:00:00 2001 From: Fernando Luis Vazquez Cao Date: Tue, 4 Feb 2014 19:35:02 +0900 Subject: [PATCH 202/596] rtnetlink: fix oops in rtnl_link_get_slave_info_data_size We should check whether rtnetlink link operations are defined before calling get_slave_size(). Without this, the following oops can occur when adding a tap device to OVS. [ 87.839553] BUG: unable to handle kernel NULL pointer dereference at 00000000000000a8 [ 87.839595] IP: [] if_nlmsg_size+0xf0/0x220 [...] [ 87.840651] Call Trace: [ 87.840664] [] ? rtmsg_ifinfo+0x2b/0x100 [ 87.840688] [] ? __netdev_adjacent_dev_insert+0x150/0x1a0 [ 87.840718] [] ? rtnetlink_event+0x30/0x40 [ 87.840742] [] ? notifier_call_chain+0x44/0x70 [ 87.840768] [] ? __netdev_upper_dev_link+0x3c6/0x3f0 [ 87.840798] [] ? netdev_create+0xcc/0x160 [openvswitch] [ 87.840828] [] ? ovs_vport_add+0x4a/0xd0 [openvswitch] [ 87.840857] [] ? new_vport+0x9/0x50 [openvswitch] [ 87.840884] [] ? ovs_vport_cmd_new+0x11e/0x210 [openvswitch] [ 87.840915] [] ? genl_family_rcv_msg+0x19a/0x360 [ 87.840941] [] ? genl_family_rcv_msg+0x360/0x360 [ 87.840967] [] ? genl_rcv_msg+0x79/0xc0 [ 87.840991] [] ? __kmalloc_reserve.isra.25+0x29/0x80 [ 87.841018] [] ? netlink_rcv_skb+0xa9/0xc0 [ 87.841042] [] ? genl_rcv+0x1f/0x30 [ 87.841064] [] ? netlink_unicast+0xe8/0x1e0 [ 87.841088] [] ? netlink_sendmsg+0x31a/0x750 [ 87.841113] [] ? sock_sendmsg+0x86/0xc0 [ 87.841136] [] ? __netdev_update_features+0x4d/0x200 [ 87.841163] [] ? ethtool_get_value+0x2e/0x50 [ 87.841188] [] ? ___sys_sendmsg+0x359/0x370 [ 87.841212] [] ? dev_ioctl+0x1a6/0x5c0 [ 87.841236] [] ? autoremove_wake_function+0x30/0x30 [ 87.841264] [] ? sock_do_ioctl+0x3d/0x50 [ 87.841288] [] ? sock_ioctl+0x1e8/0x2c0 [ 87.841312] [] ? do_vfs_ioctl+0x2cf/0x4b0 [ 87.841335] [] ? __sys_sendmsg+0x39/0x70 [ 87.841362] [] ? system_call_fastpath+0x16/0x1b [ 87.841386] Code: c0 74 10 48 89 ef ff d0 83 c0 07 83 e0 fc 48 98 49 01 c7 48 89 ef e8 d0 d6 fe ff 48 85 c0 0f 84 df 00 00 00 48 8b 90 08 07 00 00 <48> 8b 8a a8 00 00 00 31 d2 48 85 c9 74 0c 48 89 ee 48 89 c7 ff [ 87.841529] RIP [] if_nlmsg_size+0xf0/0x220 [ 87.841555] RSP [ 87.841569] CR2: 00000000000000a8 [ 87.851442] ---[ end trace e42ab217691b4fc2 ]--- Signed-off-by: Fernando Luis Vazquez Cao Acked-by: Jiri Pirko Signed-off-by: David S. Miller --- net/core/rtnetlink.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 393b1bc9a618..048dc8d183aa 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -374,7 +374,7 @@ static size_t rtnl_link_get_slave_info_data_size(const struct net_device *dev) if (!master_dev) return 0; ops = master_dev->rtnl_link_ops; - if (!ops->get_slave_size) + if (!ops || !ops->get_slave_size) return 0; /* IFLA_INFO_SLAVE_DATA + nested data */ return nla_total_size(sizeof(struct nlattr)) + From ec22b4aa993abbd18f5bbbcb20a1c56be3b1d38b Mon Sep 17 00:00:00 2001 From: Dave Airlie Date: Wed, 5 Feb 2014 14:13:56 +1000 Subject: [PATCH 203/596] drm/mgag200: fix typo causing bw limits to be ignored on some chips mode->mdev otherwise the bw limits never kick in. Reported in RHEL testing. Cc: stable@vger.kernel.org Signed-off-by: Dave Airlie --- drivers/gpu/drm/mgag200/mgag200_mode.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/mgag200/mgag200_mode.c b/drivers/gpu/drm/mgag200/mgag200_mode.c index b8583f275e80..968374776db9 100644 --- a/drivers/gpu/drm/mgag200/mgag200_mode.c +++ b/drivers/gpu/drm/mgag200/mgag200_mode.c @@ -1519,11 +1519,11 @@ static int mga_vga_mode_valid(struct drm_connector *connector, (mga_vga_calculate_mode_bandwidth(mode, bpp) > (32700 * 1024))) { return MODE_BANDWIDTH; - } else if (mode->type == G200_EH && + } else if (mdev->type == G200_EH && (mga_vga_calculate_mode_bandwidth(mode, bpp) > (37500 * 1024))) { return MODE_BANDWIDTH; - } else if (mode->type == G200_ER && + } else if (mdev->type == G200_ER && (mga_vga_calculate_mode_bandwidth(mode, bpp) > (55000 * 1024))) { return MODE_BANDWIDTH; From fbd3a77d813f211060f86cc7a2f8416caf0e03b1 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Bj=C3=B8rn=20Mork?= Date: Tue, 4 Feb 2014 13:04:33 +0100 Subject: [PATCH 204/596] net: qmi_wwan: add Netgear Aircard 340U MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This device was mentioned in an OpenWRT forum. Seems to have a "standard" Sierra Wireless ifnumber to function layout: 0: qcdm 2: nmea 3: modem 8: qmi 9: storage Signed-off-by: Bjørn Mork Signed-off-by: David S. Miller --- drivers/net/usb/qmi_wwan.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/usb/qmi_wwan.c b/drivers/net/usb/qmi_wwan.c index 23bdd5b9274d..ef4a45aea5aa 100644 --- a/drivers/net/usb/qmi_wwan.c +++ b/drivers/net/usb/qmi_wwan.c @@ -723,6 +723,7 @@ static const struct usb_device_id products[] = { {QMI_FIXED_INTF(0x1199, 0x68a2, 8)}, /* Sierra Wireless MC7710 in QMI mode */ {QMI_FIXED_INTF(0x1199, 0x68a2, 19)}, /* Sierra Wireless MC7710 in QMI mode */ {QMI_FIXED_INTF(0x1199, 0x901c, 8)}, /* Sierra Wireless EM7700 */ + {QMI_FIXED_INTF(0x1199, 0x9051, 8)}, /* Netgear AirCard 340U */ {QMI_FIXED_INTF(0x1bbb, 0x011e, 4)}, /* Telekom Speedstick LTE II (Alcatel One Touch L100V LTE) */ {QMI_FIXED_INTF(0x2357, 0x0201, 4)}, /* TP-LINK HSUPA Modem MA180 */ {QMI_FIXED_INTF(0x2357, 0x9000, 4)}, /* TP-LINK MA260 */ From e8c37aff2a392c31fefa62ec045058a0877c45a1 Mon Sep 17 00:00:00 2001 From: Dmitry Kravkov Date: Tue, 4 Feb 2014 17:43:03 +0200 Subject: [PATCH 205/596] bnx2x: fix L2-GRE TCP issues When configuring GRE tunnel using OVS, tcp stream is distributed over all RSS queues which may cause TCP reordering. It happens since OVS uses L2GRE protocol when kernel gre uses IPGRE. Patch defaults gre tunnel to L2GRE which allows proper RSS for L2GRE packets and (implicitly) disables RSS for IPGRE traffic. Signed-off-by: Dmitry Kravkov Signed-off-by: Ariel Elior Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h index 17d1689aec6b..bfc58d488bb5 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h @@ -936,7 +936,7 @@ static inline int bnx2x_func_start(struct bnx2x *bp) else /* CHIP_IS_E1X */ start_params->network_cos_mode = FW_WRR; - start_params->gre_tunnel_mode = IPGRE_TUNNEL; + start_params->gre_tunnel_mode = L2GRE_TUNNEL; start_params->gre_tunnel_rss = GRE_INNER_HEADERS_RSS; return bnx2x_func_state_change(bp, &func_params); From bce3ea81d5b2a33ed0e275d58a45f10ce95cbd50 Mon Sep 17 00:00:00 2001 From: David Vrabel Date: Tue, 4 Feb 2014 18:50:26 +0000 Subject: [PATCH 206/596] xen-netfront: handle backend CLOSED without CLOSING Backend drivers shouldn't transistion to CLOSED unless the frontend is CLOSED. If a backend does transition to CLOSED too soon then the frontend may not see the CLOSING state and will not properly shutdown. So, treat an unexpected backend CLOSED state the same as CLOSING. Signed-off-by: David Vrabel Acked-by: Konrad Rzeszutek Wilk Signed-off-by: David S. Miller --- drivers/net/xen-netfront.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/net/xen-netfront.c b/drivers/net/xen-netfront.c index ff04d4f95baa..f9daa9e183f2 100644 --- a/drivers/net/xen-netfront.c +++ b/drivers/net/xen-netfront.c @@ -1832,7 +1832,6 @@ static void netback_changed(struct xenbus_device *dev, case XenbusStateReconfiguring: case XenbusStateReconfigured: case XenbusStateUnknown: - case XenbusStateClosed: break; case XenbusStateInitWait: @@ -1847,6 +1846,10 @@ static void netback_changed(struct xenbus_device *dev, netdev_notify_peers(netdev); break; + case XenbusStateClosed: + if (dev->state == XenbusStateClosed) + break; + /* Missed the backend's CLOSING state -- fallthrough */ case XenbusStateClosing: xenbus_frontend_closed(dev); break; From 04c2facad8fee66c981a51852806d8923336f362 Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Tue, 4 Feb 2014 18:33:11 +0100 Subject: [PATCH 207/596] drivers: phy: Make NULL a valid phy reference The common clock framework considers NULL a valid clock reference. This makes handling optional clocks simple, in that if the optional clock is not available, a NULL reference can be used in the place of a real clock, simplifying the clock consumer. Extend this concept to the phy consumer API. A NULL can be passed to the release calls, the phy_init() and phy_exit() calls, and phy_power_on() and phy_power_off() and a NOP is performed. Signed-off-by: Andrew Lunn Tested-by: Gregory CLEMENT Acked-by: Kishon Vijay Abraham I Signed-off-by: Jason Cooper --- Documentation/phy.txt | 6 ++++++ drivers/phy/phy-core.c | 17 ++++++++++++++++- 2 files changed, 22 insertions(+), 1 deletion(-) diff --git a/Documentation/phy.txt b/Documentation/phy.txt index 0103e4b15b0e..2e24b993e95f 100644 --- a/Documentation/phy.txt +++ b/Documentation/phy.txt @@ -84,6 +84,12 @@ The only difference between the two APIs is that devm_phy_get associates the device with the PHY using devres on successful PHY get. On driver detach, release function is invoked on the the devres data and devres data is freed. +It should be noted that NULL is a valid phy reference. All phy +consumer calls on the NULL phy become NOPs. That is the release calls, +the phy_init() and phy_exit() calls, and phy_power_on() and +phy_power_off() calls are all NOP when applied to a NULL phy. The NULL +phy is useful in devices for handling optional phy devices. + 5. Releasing a reference to the PHY When the controller no longer needs the PHY, it has to release the reference diff --git a/drivers/phy/phy-core.c b/drivers/phy/phy-core.c index 645c867c1257..a9cdeee20d91 100644 --- a/drivers/phy/phy-core.c +++ b/drivers/phy/phy-core.c @@ -162,6 +162,9 @@ int phy_init(struct phy *phy) { int ret; + if (!phy) + return 0; + ret = phy_pm_runtime_get_sync(phy); if (ret < 0 && ret != -ENOTSUPP) return ret; @@ -187,6 +190,9 @@ int phy_exit(struct phy *phy) { int ret; + if (!phy) + return 0; + ret = phy_pm_runtime_get_sync(phy); if (ret < 0 && ret != -ENOTSUPP) return ret; @@ -212,6 +218,9 @@ int phy_power_on(struct phy *phy) { int ret; + if (!phy) + return 0; + ret = phy_pm_runtime_get_sync(phy); if (ret < 0 && ret != -ENOTSUPP) return ret; @@ -240,6 +249,9 @@ int phy_power_off(struct phy *phy) { int ret; + if (!phy) + return 0; + mutex_lock(&phy->mutex); if (phy->power_count == 1 && phy->ops->power_off) { ret = phy->ops->power_off(phy); @@ -308,7 +320,7 @@ static struct phy *of_phy_get(struct device *dev, int index) */ void phy_put(struct phy *phy) { - if (IS_ERR(phy)) + if (!phy || IS_ERR(phy)) return; module_put(phy->ops->owner); @@ -328,6 +340,9 @@ void devm_phy_put(struct device *dev, struct phy *phy) { int r; + if (!phy) + return; + r = devres_destroy(dev, devm_phy_release, devm_phy_match, phy); dev_WARN_ONCE(dev, r, "couldn't find PHY resource\n"); } From 788a4d56ff378bff0b8e685d03a962b36903a149 Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Tue, 4 Feb 2014 18:33:12 +0100 Subject: [PATCH 208/596] drivers: phy: Add support for optional phys Add devm_phy_optional_get and phy_optional_get, which should be used when the phy is optional. They does not return an error when the phy does not exist, rather they returns NULL, which is considered as a valid phy, but results in NOPs when used with the consumer API. Signed-off-by: Andrew Lunn Tested-by: Gregory CLEMENT Acked-by: Kishon Vijay Abraham I Signed-off-by: Jason Cooper --- Documentation/phy.txt | 18 +++++++++++------ drivers/phy/phy-core.c | 45 +++++++++++++++++++++++++++++++++++++++++ include/linux/phy/phy.h | 14 +++++++++++++ 3 files changed, 71 insertions(+), 6 deletions(-) diff --git a/Documentation/phy.txt b/Documentation/phy.txt index 2e24b993e95f..ebff6ee52441 100644 --- a/Documentation/phy.txt +++ b/Documentation/phy.txt @@ -75,14 +75,20 @@ Before the controller can make use of the PHY, it has to get a reference to it. This framework provides the following APIs to get a reference to the PHY. struct phy *phy_get(struct device *dev, const char *string); +struct phy *phy_optional_get(struct device *dev, const char *string); struct phy *devm_phy_get(struct device *dev, const char *string); +struct phy *devm_phy_optional_get(struct device *dev, const char *string); -phy_get and devm_phy_get can be used to get the PHY. In the case of dt boot, -the string arguments should contain the phy name as given in the dt data and -in the case of non-dt boot, it should contain the label of the PHY. -The only difference between the two APIs is that devm_phy_get associates the -device with the PHY using devres on successful PHY get. On driver detach, -release function is invoked on the the devres data and devres data is freed. +phy_get, phy_optional_get, devm_phy_get and devm_phy_optional_get can +be used to get the PHY. In the case of dt boot, the string arguments +should contain the phy name as given in the dt data and in the case of +non-dt boot, it should contain the label of the PHY. The two +devm_phy_get associates the device with the PHY using devres on +successful PHY get. On driver detach, release function is invoked on +the the devres data and devres data is freed. phy_optional_get and +devm_phy_optional_get should be used when the phy is optional. These +two functions will never return -ENODEV, but instead returns NULL when +the phy cannot be found. It should be noted that NULL is a valid phy reference. All phy consumer calls on the NULL phy become NOPs. That is the release calls, diff --git a/drivers/phy/phy-core.c b/drivers/phy/phy-core.c index a9cdeee20d91..5f5b0f4be5be 100644 --- a/drivers/phy/phy-core.c +++ b/drivers/phy/phy-core.c @@ -425,6 +425,27 @@ struct phy *phy_get(struct device *dev, const char *string) } EXPORT_SYMBOL_GPL(phy_get); +/** + * phy_optional_get() - lookup and obtain a reference to an optional phy. + * @dev: device that requests this phy + * @string: the phy name as given in the dt data or the name of the controller + * port for non-dt case + * + * Returns the phy driver, after getting a refcount to it; or + * NULL if there is no such phy. The caller is responsible for + * calling phy_put() to release that count. + */ +struct phy *phy_optional_get(struct device *dev, const char *string) +{ + struct phy *phy = phy_get(dev, string); + + if (PTR_ERR(phy) == -ENODEV) + phy = NULL; + + return phy; +} +EXPORT_SYMBOL_GPL(phy_optional_get); + /** * devm_phy_get() - lookup and obtain a reference to a phy. * @dev: device that requests this phy @@ -455,6 +476,30 @@ struct phy *devm_phy_get(struct device *dev, const char *string) } EXPORT_SYMBOL_GPL(devm_phy_get); +/** + * devm_phy_optional_get() - lookup and obtain a reference to an optional phy. + * @dev: device that requests this phy + * @string: the phy name as given in the dt data or phy device name + * for non-dt case + * + * Gets the phy using phy_get(), and associates a device with it using + * devres. On driver detach, release function is invoked on the devres + * data, then, devres data is freed. This differs to devm_phy_get() in + * that if the phy does not exist, it is not considered an error and + * -ENODEV will not be returned. Instead the NULL phy is returned, + * which can be passed to all other phy consumer calls. + */ +struct phy *devm_phy_optional_get(struct device *dev, const char *string) +{ + struct phy *phy = devm_phy_get(dev, string); + + if (PTR_ERR(phy) == -ENODEV) + phy = NULL; + + return phy; +} +EXPORT_SYMBOL_GPL(devm_phy_optional_get); + /** * phy_create() - create a new phy * @dev: device that is creating the new phy diff --git a/include/linux/phy/phy.h b/include/linux/phy/phy.h index e273e5ac19c9..3f83459dbb20 100644 --- a/include/linux/phy/phy.h +++ b/include/linux/phy/phy.h @@ -146,7 +146,9 @@ static inline void phy_set_bus_width(struct phy *phy, int bus_width) phy->attrs.bus_width = bus_width; } struct phy *phy_get(struct device *dev, const char *string); +struct phy *phy_optional_get(struct device *dev, const char *string); struct phy *devm_phy_get(struct device *dev, const char *string); +struct phy *devm_phy_optional_get(struct device *dev, const char *string); void phy_put(struct phy *phy); void devm_phy_put(struct device *dev, struct phy *phy); struct phy *of_phy_simple_xlate(struct device *dev, @@ -232,11 +234,23 @@ static inline struct phy *phy_get(struct device *dev, const char *string) return ERR_PTR(-ENOSYS); } +static inline struct phy *phy_optional_get(struct device *dev, + const char *string) +{ + return ERR_PTR(-ENOSYS); +} + static inline struct phy *devm_phy_get(struct device *dev, const char *string) { return ERR_PTR(-ENOSYS); } +static inline struct phy *devm_phy_optional_get(struct device *dev, + const char *string) +{ + return ERR_PTR(-ENOSYS); +} + static inline void phy_put(struct phy *phy) { } From 90aa2997029fa623fe9e3ec3a469a00a34130237 Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Tue, 4 Feb 2014 18:33:13 +0100 Subject: [PATCH 209/596] ata: sata_mv: Fix probe failures with optional phys Make use of devm_phy_optional_get() in order to fix probe failures on Armada 370, XP and others, when there is no phy driver available. Signed-off-by: Andrew Lunn Tested-by: Gregory CLEMENT Acked-by: Tejun Heo Acked-by: Kishon Vijay Abraham I Signed-off-by: Jason Cooper --- drivers/ata/sata_mv.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/drivers/ata/sata_mv.c b/drivers/ata/sata_mv.c index 20a7517bd339..52b8181ddafd 100644 --- a/drivers/ata/sata_mv.c +++ b/drivers/ata/sata_mv.c @@ -4126,12 +4126,14 @@ static int mv_platform_probe(struct platform_device *pdev) clk_prepare_enable(hpriv->port_clks[port]); sprintf(port_number, "port%d", port); - hpriv->port_phys[port] = devm_phy_get(&pdev->dev, port_number); + hpriv->port_phys[port] = devm_phy_optional_get(&pdev->dev, + port_number); if (IS_ERR(hpriv->port_phys[port])) { rc = PTR_ERR(hpriv->port_phys[port]); hpriv->port_phys[port] = NULL; - if ((rc != -EPROBE_DEFER) && (rc != -ENODEV)) - dev_warn(&pdev->dev, "error getting phy"); + if (rc != -EPROBE_DEFER) + dev_warn(&pdev->dev, "error getting phy %d", + rc); goto err; } else phy_power_on(hpriv->port_phys[port]); From d3c56568f43807135f2c2a09582a69f809f0d8b7 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Mon, 3 Feb 2014 09:56:13 +0100 Subject: [PATCH 210/596] ALSA: hda/realtek - Avoid invalid COEFs for ALC271X We've seen often problems after suspend/resume on Acer Aspire One AO725 with ALC271X codec as reported in kernel bugzilla, and it turned out that some COEFs doesn't work and triggers the codec communication stall. Since these magic COEF setups are specific to ALC269VB for some PLL configurations, the machine works even without these manual adjustment. So, let's simply avoid applying them for ALC271X. Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=52181 Cc: Signed-off-by: Takashi Iwai --- sound/pci/hda/patch_realtek.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index 56a8f1876603..e16a71e586f0 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -4427,6 +4427,9 @@ static void alc269_fill_coef(struct hda_codec *codec) if (spec->codec_variant != ALC269_TYPE_ALC269VB) return; + /* ALC271X doesn't seem to support these COEFs (bko#52181) */ + if (!strcmp(codec->chip_name, "ALC271X")) + return; if ((alc_get_coef0(codec) & 0x00ff) < 0x015) { alc_write_coef_idx(codec, 0xf, 0x960b); From aea0bb4f8ee513537ad84b9f3f609f96e272d98e Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Tue, 14 Jan 2014 16:27:49 +0000 Subject: [PATCH 211/596] openvswitch: Pad OVS_PACKET_ATTR_PACKET if linear copy was performed While the zerocopy method is correctly omitted if user space does not support unaligned Netlink messages. The attribute is still not padded correctly as skb_zerocopy() will not ensure padding and the attribute size is no longer pre calculated though nla_reserve() which ensured padding previously. This patch applies appropriate padding if a linear data copy was performed in skb_zerocopy(). Signed-off-by: Thomas Graf Acked-by: Zoltan Kiss Signed-off-by: Jesse Gross --- net/openvswitch/datapath.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c index df4692826ead..d1a73a6102f8 100644 --- a/net/openvswitch/datapath.c +++ b/net/openvswitch/datapath.c @@ -466,6 +466,14 @@ static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb, skb_zerocopy(user_skb, skb, skb->len, hlen); + /* Pad OVS_PACKET_ATTR_PACKET if linear copy was performed */ + if (!(dp->user_features & OVS_DP_F_UNALIGNED)) { + size_t plen = NLA_ALIGN(user_skb->len) - user_skb->len; + + if (plen > 0) + memset(skb_put(user_skb, plen), 0, plen); + } + ((struct nlmsghdr *) user_skb->data)->nlmsg_len = user_skb->len; err = genlmsg_unicast(ovs_dp_get_net(dp), user_skb, upcall_info->portid); From e80857cce82da31e41a6599fc888dfc92e0167cc Mon Sep 17 00:00:00 2001 From: Andy Zhou Date: Tue, 21 Jan 2014 09:31:04 -0800 Subject: [PATCH 212/596] openvswitch: Fix kernel panic on ovs_flow_free Both mega flow mask's reference counter and per flow table mask list should only be accessed when holding ovs_mutex() lock. However this is not true with ovs_flow_table_flush(). The patch fixes this bug. Reported-by: Joe Stringer Signed-off-by: Andy Zhou Signed-off-by: Jesse Gross --- net/openvswitch/datapath.c | 9 ++-- net/openvswitch/flow_table.c | 86 +++++++++++++++++------------------- net/openvswitch/flow_table.h | 2 +- 3 files changed, 48 insertions(+), 49 deletions(-) diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c index d1a73a6102f8..e1b337e0bf4d 100644 --- a/net/openvswitch/datapath.c +++ b/net/openvswitch/datapath.c @@ -55,6 +55,7 @@ #include "datapath.h" #include "flow.h" +#include "flow_table.h" #include "flow_netlink.h" #include "vport-internal_dev.h" #include "vport-netdev.h" @@ -160,7 +161,6 @@ static void destroy_dp_rcu(struct rcu_head *rcu) { struct datapath *dp = container_of(rcu, struct datapath, rcu); - ovs_flow_tbl_destroy(&dp->table); free_percpu(dp->stats_percpu); release_net(ovs_dp_get_net(dp)); kfree(dp->ports); @@ -1287,7 +1287,7 @@ static int ovs_dp_cmd_new(struct sk_buff *skb, struct genl_info *info) err_destroy_percpu: free_percpu(dp->stats_percpu); err_destroy_table: - ovs_flow_tbl_destroy(&dp->table); + ovs_flow_tbl_destroy(&dp->table, false); err_free_dp: release_net(ovs_dp_get_net(dp)); kfree(dp); @@ -1314,10 +1314,13 @@ static void __dp_destroy(struct datapath *dp) list_del_rcu(&dp->list_node); /* OVSP_LOCAL is datapath internal port. We need to make sure that - * all port in datapath are destroyed first before freeing datapath. + * all ports in datapath are destroyed first before freeing datapath. */ ovs_dp_detach_port(ovs_vport_ovsl(dp, OVSP_LOCAL)); + /* RCU destroy the flow table */ + ovs_flow_tbl_destroy(&dp->table, true); + call_rcu(&dp->rcu, destroy_dp_rcu); } diff --git a/net/openvswitch/flow_table.c b/net/openvswitch/flow_table.c index c58a0fe3c889..bd14052ed342 100644 --- a/net/openvswitch/flow_table.c +++ b/net/openvswitch/flow_table.c @@ -153,29 +153,27 @@ static void rcu_free_flow_callback(struct rcu_head *rcu) flow_free(flow); } -static void flow_mask_del_ref(struct sw_flow_mask *mask, bool deferred) -{ - if (!mask) - return; - - BUG_ON(!mask->ref_count); - mask->ref_count--; - - if (!mask->ref_count) { - list_del_rcu(&mask->list); - if (deferred) - kfree_rcu(mask, rcu); - else - kfree(mask); - } -} - void ovs_flow_free(struct sw_flow *flow, bool deferred) { if (!flow) return; - flow_mask_del_ref(flow->mask, deferred); + ASSERT_OVSL(); + + if (flow->mask) { + struct sw_flow_mask *mask = flow->mask; + + BUG_ON(!mask->ref_count); + mask->ref_count--; + + if (!mask->ref_count) { + list_del_rcu(&mask->list); + if (deferred) + kfree_rcu(mask, rcu); + else + kfree(mask); + } + } if (deferred) call_rcu(&flow->rcu, rcu_free_flow_callback); @@ -188,26 +186,9 @@ static void free_buckets(struct flex_array *buckets) flex_array_free(buckets); } + static void __table_instance_destroy(struct table_instance *ti) { - int i; - - if (ti->keep_flows) - goto skip_flows; - - for (i = 0; i < ti->n_buckets; i++) { - struct sw_flow *flow; - struct hlist_head *head = flex_array_get(ti->buckets, i); - struct hlist_node *n; - int ver = ti->node_ver; - - hlist_for_each_entry_safe(flow, n, head, hash_node[ver]) { - hlist_del(&flow->hash_node[ver]); - ovs_flow_free(flow, false); - } - } - -skip_flows: free_buckets(ti->buckets); kfree(ti); } @@ -258,20 +239,38 @@ static void flow_tbl_destroy_rcu_cb(struct rcu_head *rcu) static void table_instance_destroy(struct table_instance *ti, bool deferred) { + int i; + if (!ti) return; + if (ti->keep_flows) + goto skip_flows; + + for (i = 0; i < ti->n_buckets; i++) { + struct sw_flow *flow; + struct hlist_head *head = flex_array_get(ti->buckets, i); + struct hlist_node *n; + int ver = ti->node_ver; + + hlist_for_each_entry_safe(flow, n, head, hash_node[ver]) { + hlist_del_rcu(&flow->hash_node[ver]); + ovs_flow_free(flow, deferred); + } + } + +skip_flows: if (deferred) call_rcu(&ti->rcu, flow_tbl_destroy_rcu_cb); else __table_instance_destroy(ti); } -void ovs_flow_tbl_destroy(struct flow_table *table) +void ovs_flow_tbl_destroy(struct flow_table *table, bool deferred) { struct table_instance *ti = ovsl_dereference(table->ti); - table_instance_destroy(ti, false); + table_instance_destroy(ti, deferred); } struct sw_flow *ovs_flow_tbl_dump_next(struct table_instance *ti, @@ -504,16 +503,11 @@ static struct sw_flow_mask *mask_alloc(void) mask = kmalloc(sizeof(*mask), GFP_KERNEL); if (mask) - mask->ref_count = 0; + mask->ref_count = 1; return mask; } -static void mask_add_ref(struct sw_flow_mask *mask) -{ - mask->ref_count++; -} - static bool mask_equal(const struct sw_flow_mask *a, const struct sw_flow_mask *b) { @@ -554,9 +548,11 @@ static int flow_mask_insert(struct flow_table *tbl, struct sw_flow *flow, mask->key = new->key; mask->range = new->range; list_add_rcu(&mask->list, &tbl->mask_list); + } else { + BUG_ON(!mask->ref_count); + mask->ref_count++; } - mask_add_ref(mask); flow->mask = mask; return 0; } diff --git a/net/openvswitch/flow_table.h b/net/openvswitch/flow_table.h index 1996e34c0fd8..baaeb101924d 100644 --- a/net/openvswitch/flow_table.h +++ b/net/openvswitch/flow_table.h @@ -60,7 +60,7 @@ void ovs_flow_free(struct sw_flow *, bool deferred); int ovs_flow_tbl_init(struct flow_table *); int ovs_flow_tbl_count(struct flow_table *table); -void ovs_flow_tbl_destroy(struct flow_table *table); +void ovs_flow_tbl_destroy(struct flow_table *table, bool deferred); int ovs_flow_tbl_flush(struct flow_table *flow_table); int ovs_flow_tbl_insert(struct flow_table *table, struct sw_flow *flow, From 45fb9c35b27c9982e9a55d04ed0a5230a2d0b306 Mon Sep 17 00:00:00 2001 From: Daniele Di Proietto Date: Thu, 23 Jan 2014 10:47:35 -0800 Subject: [PATCH 213/596] openvswitch: Fix ovs_dp_cmd_msg_size() commit 43d4be9cb55f3bac5253e9289996fd9d735531db (openvswitch: Allow user space to announce ability to accept unaligned Netlink messages) introduced OVS_DP_ATTR_USER_FEATURES netlink attribute in datapath responses, but the attribute size was not taken into account in ovs_dp_cmd_msg_size(). Signed-off-by: Daniele Di Proietto Signed-off-by: Jesse Gross --- net/openvswitch/datapath.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c index e1b337e0bf4d..58689dda8377 100644 --- a/net/openvswitch/datapath.c +++ b/net/openvswitch/datapath.c @@ -1087,6 +1087,7 @@ static size_t ovs_dp_cmd_msg_size(void) msgsize += nla_total_size(IFNAMSIZ); msgsize += nla_total_size(sizeof(struct ovs_dp_stats)); msgsize += nla_total_size(sizeof(struct ovs_dp_megaflow_stats)); + msgsize += nla_total_size(sizeof(u32)); /* OVS_DP_ATTR_USER_FEATURES */ return msgsize; } From e4c6d7595403e943a3afc334eb8c0efcd840043a Mon Sep 17 00:00:00 2001 From: Pravin B Shelar Date: Fri, 31 Jan 2014 09:43:23 -0800 Subject: [PATCH 214/596] openvswitch: Fix ovs_flow_free() ovs-lock assert. ovs_flow_free() is not called under ovs-lock during packet execute path (ovs_packet_cmd_execute()). Since packet execute does not touch flow->mask, there is no need to take that lock either. So move assert in case where flow->mask is checked. Found by code inspection. Signed-off-by: Pravin B Shelar Signed-off-by: Jesse Gross --- net/openvswitch/flow_table.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/net/openvswitch/flow_table.c b/net/openvswitch/flow_table.c index bd14052ed342..3c268b3d71c3 100644 --- a/net/openvswitch/flow_table.c +++ b/net/openvswitch/flow_table.c @@ -158,11 +158,13 @@ void ovs_flow_free(struct sw_flow *flow, bool deferred) if (!flow) return; - ASSERT_OVSL(); - if (flow->mask) { struct sw_flow_mask *mask = flow->mask; + /* ovs-lock is required to protect mask-refcount and + * mask list. + */ + ASSERT_OVSL(); BUG_ON(!mask->ref_count); mask->ref_count--; From c7579fed1f1b2567529aea64ef19871337403ab3 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Wed, 5 Feb 2014 07:28:10 +0100 Subject: [PATCH 215/596] ALSA: hda - Add missing mixer widget for AD1983 The mixer widget on AD1983 at NID 0x0e was missing in the commit [f2f8be43c5c9: ALSA: hda - Add aamix NID to AD codecs]. Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=70011 Cc: Signed-off-by: Takashi Iwai --- sound/pci/hda/patch_analog.c | 1 + 1 file changed, 1 insertion(+) diff --git a/sound/pci/hda/patch_analog.c b/sound/pci/hda/patch_analog.c index 50b2427f19ca..195cd62cdce5 100644 --- a/sound/pci/hda/patch_analog.c +++ b/sound/pci/hda/patch_analog.c @@ -498,6 +498,7 @@ static int patch_ad1983(struct hda_codec *codec) return err; spec = codec->spec; + spec->gen.mixer_nid = 0x0e; spec->gen.beep_nid = 0x10; set_beep_amp(spec, 0x10, 0, HDA_OUTPUT); err = ad198x_parse_auto_config(codec, false); From c14e0953ca51dbcb8d1ac92acbdcff23d0caa158 Mon Sep 17 00:00:00 2001 From: Andy Zhou Date: Sun, 2 Feb 2014 17:08:06 -0800 Subject: [PATCH 216/596] openvswitch: Suppress error messages on megaflow updates With subfacets, we'd expect megaflow updates message to carry the original micro flow. If not, EINVAL is returned and kernel logs an error message. Now that the user space subfacet layer is removed, it is expected that flow updates can arrive with a micro flow other than the original. Change the return code to EEXIST and remove the kernel error log message. Reported-by: Ben Pfaff Signed-off-by: Andy Zhou Signed-off-by: Jesse Gross --- net/openvswitch/datapath.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c index 58689dda8377..e9a48baf8551 100644 --- a/net/openvswitch/datapath.c +++ b/net/openvswitch/datapath.c @@ -860,11 +860,8 @@ static int ovs_flow_cmd_new_or_set(struct sk_buff *skb, struct genl_info *info) goto err_unlock_ovs; /* The unmasked key has to be the same for flow updates. */ - error = -EINVAL; - if (!ovs_flow_cmp_unmasked_key(flow, &match)) { - OVS_NLERR("Flow modification message rejected, unmasked key does not match.\n"); + if (!ovs_flow_cmp_unmasked_key(flow, &match)) goto err_unlock_ovs; - } /* Update actions. */ old_acts = ovsl_dereference(flow->sf_acts); From c20f31ec421ea4fabea5e95a6afd46c5f41e5599 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Mon, 3 Feb 2014 11:02:10 +0100 Subject: [PATCH 217/596] ALSA: hda - Fix missing VREF setup for Mac Pro 1,1 Mac Pro 1,1 with ALC889A codec needs the VREF setup on NID 0x18 to VREF50, in order to make the speaker working. The same fixup was already needed for MacBook Air 1,1, so we can reuse it. Reported-by: Nicolai Beuermann Cc: Signed-off-by: Takashi Iwai --- sound/pci/hda/patch_realtek.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index e16a71e586f0..d9693ca9546f 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -1821,6 +1821,7 @@ enum { ALC889_FIXUP_IMAC91_VREF, ALC889_FIXUP_MBA11_VREF, ALC889_FIXUP_MBA21_VREF, + ALC889_FIXUP_MP11_VREF, ALC882_FIXUP_INV_DMIC, ALC882_FIXUP_NO_PRIMARY_HP, ALC887_FIXUP_ASUS_BASS, @@ -2190,6 +2191,12 @@ static const struct hda_fixup alc882_fixups[] = { .chained = true, .chain_id = ALC889_FIXUP_MBP_VREF, }, + [ALC889_FIXUP_MP11_VREF] = { + .type = HDA_FIXUP_FUNC, + .v.func = alc889_fixup_mba11_vref, + .chained = true, + .chain_id = ALC885_FIXUP_MACPRO_GPIO, + }, [ALC882_FIXUP_INV_DMIC] = { .type = HDA_FIXUP_FUNC, .v.func = alc_fixup_inv_dmic_0x12, @@ -2253,7 +2260,7 @@ static const struct snd_pci_quirk alc882_fixup_tbl[] = { SND_PCI_QUIRK(0x106b, 0x00a0, "MacBookPro 3,1", ALC889_FIXUP_MBP_VREF), SND_PCI_QUIRK(0x106b, 0x00a1, "Macbook", ALC889_FIXUP_MBP_VREF), SND_PCI_QUIRK(0x106b, 0x00a4, "MacbookPro 4,1", ALC889_FIXUP_MBP_VREF), - SND_PCI_QUIRK(0x106b, 0x0c00, "Mac Pro", ALC885_FIXUP_MACPRO_GPIO), + SND_PCI_QUIRK(0x106b, 0x0c00, "Mac Pro", ALC889_FIXUP_MP11_VREF), SND_PCI_QUIRK(0x106b, 0x1000, "iMac 24", ALC885_FIXUP_MACPRO_GPIO), SND_PCI_QUIRK(0x106b, 0x2800, "AppleTV", ALC885_FIXUP_MACPRO_GPIO), SND_PCI_QUIRK(0x106b, 0x2c00, "MacbookPro rev3", ALC889_FIXUP_MBP_VREF), From 76c7d18bcddc9794f898ebdee44a3160c636da9c Mon Sep 17 00:00:00 2001 From: Thomas Hellstrom Date: Thu, 30 Jan 2014 10:46:12 +0100 Subject: [PATCH 218/596] drm/vmwgfx: Don't commit staged bindings if execbuf fails If execbuf fails and binding commands are never sent to the device, don't commit the staged context bindings to the tracker. Signed-off-by: Thomas Hellstrom Reviewed-by: Jakob Bornecrantz --- drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c index 7a5f1eb55c5a..3f0b4d1450ff 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c @@ -114,8 +114,10 @@ static void vmw_resource_list_unreserve(struct list_head *list, * persistent context binding tracker. */ if (unlikely(val->staged_bindings)) { - vmw_context_binding_state_transfer - (val->res, val->staged_bindings); + if (!backoff) { + vmw_context_binding_state_transfer + (val->res, val->staged_bindings); + } kfree(val->staged_bindings); val->staged_bindings = NULL; } From cf5e3413337309050c05e13dcebe85b7194a21e5 Mon Sep 17 00:00:00 2001 From: Thomas Hellstrom Date: Thu, 30 Jan 2014 10:58:19 +0100 Subject: [PATCH 219/596] drm/vmwgfx: Fix regression caused by "drm/ttm: make ttm reservation calls behave like reservation calls" The call to ttm_eu_backoff_reservation() as part of an error path would cause a lock imbalance if the reservation ticket was not initialized. This error is easily triggered from user-space by submitting a bogus command stream. Signed-off-by: Thomas Hellstrom Reviewed-by: Jakob Bornecrantz Cc: stable@vger.kernel.org Cc: Maarten Lankhorst Cc: Jerome Glisse Cc: Dave Airlie --- drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c index 3f0b4d1450ff..dafa139c0ca7 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c @@ -2195,11 +2195,11 @@ int vmw_execbuf_process(struct drm_file *file_priv, ret = vmw_cmd_check_all(dev_priv, sw_context, kernel_commands, command_size); if (unlikely(ret != 0)) - goto out_err; + goto out_err_nores; ret = vmw_resources_reserve(sw_context); if (unlikely(ret != 0)) - goto out_err; + goto out_err_nores; ret = ttm_eu_reserve_buffers(&ticket, &sw_context->validate_nodes); if (unlikely(ret != 0)) @@ -2291,10 +2291,11 @@ int vmw_execbuf_process(struct drm_file *file_priv, out_unlock_binding: mutex_unlock(&dev_priv->binding_mutex); out_err: + ttm_eu_backoff_reservation(&ticket, &sw_context->validate_nodes); +out_err_nores: + vmw_resource_list_unreserve(&sw_context->resource_list, true); vmw_resource_relocations_free(&sw_context->res_relocations); vmw_free_relocations(sw_context); - ttm_eu_backoff_reservation(&ticket, &sw_context->validate_nodes); - vmw_resource_list_unreserve(&sw_context->resource_list, true); vmw_clear_validations(sw_context); if (unlikely(dev_priv->pinned_bo != NULL && !dev_priv->query_cid_valid)) From 0ccbbae43c2dfe45ded1d7ed59b8fc7ac8214fb0 Mon Sep 17 00:00:00 2001 From: Thomas Hellstrom Date: Thu, 30 Jan 2014 11:13:43 +0100 Subject: [PATCH 220/596] drm/vmwgfx: Fix SET_SHADER_CONST emulation on guest-backed devices Emulate the SET_SHADER_CONST legacy command on guest-backed devices by issuing a SET_GB_SHADERCONSTS_INLINE command. Signed-off-by: Thomas Hellstrom Reviewed-by: Jakob Bornecrantz --- drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c | 37 +++++++++++++++++++++++-- 1 file changed, 35 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c index dafa139c0ca7..9441825c7860 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c @@ -1528,6 +1528,39 @@ static int vmw_cmd_set_shader(struct vmw_private *dev_priv, return 0; } +/** + * vmw_cmd_set_shader_const - Validate an SVGA_3D_CMD_SET_SHADER_CONST + * command + * + * @dev_priv: Pointer to a device private struct. + * @sw_context: The software context being used for this batch. + * @header: Pointer to the command header in the command stream. + */ +static int vmw_cmd_set_shader_const(struct vmw_private *dev_priv, + struct vmw_sw_context *sw_context, + SVGA3dCmdHeader *header) +{ + struct vmw_set_shader_const_cmd { + SVGA3dCmdHeader header; + SVGA3dCmdSetShaderConst body; + } *cmd; + int ret; + + cmd = container_of(header, struct vmw_set_shader_const_cmd, + header); + + ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_context, + user_context_converter, &cmd->body.cid, + NULL); + if (unlikely(ret != 0)) + return ret; + + if (dev_priv->has_mob) + header->id = SVGA_3D_CMD_SET_GB_SHADERCONSTS_INLINE; + + return 0; +} + /** * vmw_cmd_bind_gb_shader - Validate an SVGA_3D_CMD_BIND_GB_SHADER * command @@ -1642,8 +1675,8 @@ static const struct vmw_cmd_entry const vmw_cmd_entries[SVGA_3D_CMD_MAX] = { true, true, false), VMW_CMD_DEF(SVGA_3D_CMD_SET_SHADER, &vmw_cmd_set_shader, true, false, false), - VMW_CMD_DEF(SVGA_3D_CMD_SET_SHADER_CONST, &vmw_cmd_cid_check, - true, true, false), + VMW_CMD_DEF(SVGA_3D_CMD_SET_SHADER_CONST, &vmw_cmd_set_shader_const, + true, false, false), VMW_CMD_DEF(SVGA_3D_CMD_DRAW_PRIMITIVES, &vmw_cmd_draw, true, false, false), VMW_CMD_DEF(SVGA_3D_CMD_SETSCISSORRECT, &vmw_cmd_cid_check, From c1a21373d2cb94a7808161a8c237b249cd799ce7 Mon Sep 17 00:00:00 2001 From: Thomas Hellstrom Date: Thu, 30 Jan 2014 11:18:38 +0100 Subject: [PATCH 221/596] drm/vmwgfx: Fix legacy surface reference size copyback Surfaces created using the guest-backed surface interface only keeps the base mip size, so only copy that if the legacy surface reference ioctl requests the size information. Signed-off-by: Thomas Hellstrom Reviewed-by: Jakob Bornecrantz --- drivers/gpu/drm/vmwgfx/vmwgfx_surface.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c b/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c index 979da1c246a5..ec2b99833fce 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c @@ -908,8 +908,8 @@ int vmw_surface_reference_ioctl(struct drm_device *dev, void *data, rep->size_addr; if (user_sizes) - ret = copy_to_user(user_sizes, srf->sizes, - srf->num_sizes * sizeof(*srf->sizes)); + ret = copy_to_user(user_sizes, &srf->base_size, + sizeof(srf->base_size)); if (unlikely(ret != 0)) { DRM_ERROR("copy_to_user failed %p %u\n", user_sizes, srf->num_sizes); From d5bde956630b86462ee22055f5816a04290aed57 Mon Sep 17 00:00:00 2001 From: Thomas Hellstrom Date: Fri, 31 Jan 2014 10:12:10 +0100 Subject: [PATCH 222/596] drm/vmwgfx: Emulate legacy shaders on guest-backed devices v2 Command stream legacy shader creation and destruction is replaced by NOPs in the command stream, and instead guest-backed shaders are created and destroyed as part of the command validation process. v2: Removed some stray debug messages. Signed-off-by: Thomas Hellstrom Reviewed-by: Jakob Bornecrantz --- drivers/gpu/drm/vmwgfx/vmwgfx_drv.c | 7 + drivers/gpu/drm/vmwgfx/vmwgfx_drv.h | 29 +- drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c | 197 ++++++++-- drivers/gpu/drm/vmwgfx/vmwgfx_shader.c | 489 +++++++++++++++++++++--- 4 files changed, 632 insertions(+), 90 deletions(-) diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c index 9893328f8fdc..3bdc0adc656d 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c @@ -941,6 +941,7 @@ static void vmw_postclose(struct drm_device *dev, drm_master_put(&vmw_fp->locked_master); } + vmw_compat_shader_man_destroy(vmw_fp->shman); ttm_object_file_release(&vmw_fp->tfile); kfree(vmw_fp); } @@ -960,11 +961,17 @@ static int vmw_driver_open(struct drm_device *dev, struct drm_file *file_priv) if (unlikely(vmw_fp->tfile == NULL)) goto out_no_tfile; + vmw_fp->shman = vmw_compat_shader_man_create(dev_priv); + if (IS_ERR(vmw_fp->shman)) + goto out_no_shman; + file_priv->driver_priv = vmw_fp; dev_priv->bdev.dev_mapping = dev->dev_mapping; return 0; +out_no_shman: + ttm_object_file_release(&vmw_fp->tfile); out_no_tfile: kfree(vmw_fp); return ret; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h index 554e7fa33082..cef0ff7ac738 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h @@ -75,10 +75,14 @@ #define VMW_RES_FENCE ttm_driver_type3 #define VMW_RES_SHADER ttm_driver_type4 +struct vmw_compat_shader_manager; + struct vmw_fpriv { struct drm_master *locked_master; struct ttm_object_file *tfile; struct list_head fence_events; + bool gb_aware; + struct vmw_compat_shader_manager *shman; }; struct vmw_dma_buffer { @@ -318,7 +322,7 @@ struct vmw_sw_context{ struct drm_open_hash res_ht; bool res_ht_initialized; bool kernel; /**< is the called made from the kernel */ - struct ttm_object_file *tfile; + struct vmw_fpriv *fp; struct list_head validate_nodes; struct vmw_relocation relocs[VMWGFX_MAX_RELOCATIONS]; uint32_t cur_reloc; @@ -336,6 +340,7 @@ struct vmw_sw_context{ bool needs_post_query_barrier; struct vmw_resource *error_resource; struct vmw_ctx_binding_state staged_bindings; + struct list_head staged_shaders; }; struct vmw_legacy_display; @@ -991,6 +996,28 @@ extern int vmw_shader_define_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv); extern int vmw_shader_destroy_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv); +extern int vmw_compat_shader_lookup(struct vmw_compat_shader_manager *man, + SVGA3dShaderType shader_type, + u32 *user_key); +extern void vmw_compat_shaders_commit(struct vmw_compat_shader_manager *man, + struct list_head *list); +extern void vmw_compat_shaders_revert(struct vmw_compat_shader_manager *man, + struct list_head *list); +extern int vmw_compat_shader_remove(struct vmw_compat_shader_manager *man, + u32 user_key, + SVGA3dShaderType shader_type, + struct list_head *list); +extern int vmw_compat_shader_add(struct vmw_compat_shader_manager *man, + u32 user_key, const void *bytecode, + SVGA3dShaderType shader_type, + size_t size, + struct ttm_object_file *tfile, + struct list_head *list); +extern struct vmw_compat_shader_manager * +vmw_compat_shader_man_create(struct vmw_private *dev_priv); +extern void +vmw_compat_shader_man_destroy(struct vmw_compat_shader_manager *man); + /** * Inline helper functions diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c index 9441825c7860..352224b9d667 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c @@ -235,8 +235,12 @@ static void vmw_resource_relocations_apply(uint32_t *cb, { struct vmw_resource_relocation *rel; - list_for_each_entry(rel, list, head) - cb[rel->offset] = rel->res->id; + list_for_each_entry(rel, list, head) { + if (likely(rel->res != NULL)) + cb[rel->offset] = rel->res->id; + else + cb[rel->offset] = SVGA_3D_CMD_NOP; + } } static int vmw_cmd_invalid(struct vmw_private *dev_priv, @@ -381,22 +385,27 @@ static int vmw_resources_validate(struct vmw_sw_context *sw_context) } /** - * vmw_cmd_res_check - Check that a resource is present and if so, put it + * vmw_cmd_compat_res_check - Check that a resource is present and if so, put it * on the resource validate list unless it's already there. * * @dev_priv: Pointer to a device private structure. * @sw_context: Pointer to the software context. * @res_type: Resource type. * @converter: User-space visisble type specific information. - * @id: Pointer to the location in the command buffer currently being + * @id: user-space resource id handle. + * @id_loc: Pointer to the location in the command buffer currently being * parsed from where the user-space resource id handle is located. + * @p_val: Pointer to pointer to resource validalidation node. Populated + * on exit. */ -static int vmw_cmd_res_check(struct vmw_private *dev_priv, - struct vmw_sw_context *sw_context, - enum vmw_res_type res_type, - const struct vmw_user_resource_conv *converter, - uint32_t *id, - struct vmw_resource_val_node **p_val) +static int +vmw_cmd_compat_res_check(struct vmw_private *dev_priv, + struct vmw_sw_context *sw_context, + enum vmw_res_type res_type, + const struct vmw_user_resource_conv *converter, + uint32_t id, + uint32_t *id_loc, + struct vmw_resource_val_node **p_val) { struct vmw_res_cache_entry *rcache = &sw_context->res_cache[res_type]; @@ -404,7 +413,7 @@ static int vmw_cmd_res_check(struct vmw_private *dev_priv, struct vmw_resource_val_node *node; int ret; - if (*id == SVGA3D_INVALID_ID) { + if (id == SVGA3D_INVALID_ID) { if (p_val) *p_val = NULL; if (res_type == vmw_res_context) { @@ -419,7 +428,7 @@ static int vmw_cmd_res_check(struct vmw_private *dev_priv, * resource */ - if (likely(rcache->valid && *id == rcache->handle)) { + if (likely(rcache->valid && id == rcache->handle)) { const struct vmw_resource *res = rcache->res; rcache->node->first_usage = false; @@ -428,28 +437,28 @@ static int vmw_cmd_res_check(struct vmw_private *dev_priv, return vmw_resource_relocation_add (&sw_context->res_relocations, res, - id - sw_context->buf_start); + id_loc - sw_context->buf_start); } ret = vmw_user_resource_lookup_handle(dev_priv, - sw_context->tfile, - *id, + sw_context->fp->tfile, + id, converter, &res); if (unlikely(ret != 0)) { DRM_ERROR("Could not find or use resource 0x%08x.\n", - (unsigned) *id); + (unsigned) id); dump_stack(); return ret; } rcache->valid = true; rcache->res = res; - rcache->handle = *id; + rcache->handle = id; ret = vmw_resource_relocation_add(&sw_context->res_relocations, res, - id - sw_context->buf_start); + id_loc - sw_context->buf_start); if (unlikely(ret != 0)) goto out_no_reloc; @@ -482,6 +491,31 @@ static int vmw_cmd_res_check(struct vmw_private *dev_priv, return ret; } +/** + * vmw_cmd_res_check - Check that a resource is present and if so, put it + * on the resource validate list unless it's already there. + * + * @dev_priv: Pointer to a device private structure. + * @sw_context: Pointer to the software context. + * @res_type: Resource type. + * @converter: User-space visisble type specific information. + * @id_loc: Pointer to the location in the command buffer currently being + * parsed from where the user-space resource id handle is located. + * @p_val: Pointer to pointer to resource validalidation node. Populated + * on exit. + */ +static int +vmw_cmd_res_check(struct vmw_private *dev_priv, + struct vmw_sw_context *sw_context, + enum vmw_res_type res_type, + const struct vmw_user_resource_conv *converter, + uint32_t *id_loc, + struct vmw_resource_val_node **p_val) +{ + return vmw_cmd_compat_res_check(dev_priv, sw_context, res_type, + converter, *id_loc, id_loc, p_val); +} + /** * vmw_cmd_cid_check - Check a command header for valid context information. * @@ -769,7 +803,7 @@ static int vmw_translate_mob_ptr(struct vmw_private *dev_priv, struct vmw_relocation *reloc; int ret; - ret = vmw_user_dmabuf_lookup(sw_context->tfile, handle, &vmw_bo); + ret = vmw_user_dmabuf_lookup(sw_context->fp->tfile, handle, &vmw_bo); if (unlikely(ret != 0)) { DRM_ERROR("Could not find or use MOB buffer.\n"); return -EINVAL; @@ -830,7 +864,7 @@ static int vmw_translate_guest_ptr(struct vmw_private *dev_priv, struct vmw_relocation *reloc; int ret; - ret = vmw_user_dmabuf_lookup(sw_context->tfile, handle, &vmw_bo); + ret = vmw_user_dmabuf_lookup(sw_context->fp->tfile, handle, &vmw_bo); if (unlikely(ret != 0)) { DRM_ERROR("Could not find or use GMR region.\n"); return -EINVAL; @@ -1129,7 +1163,8 @@ static int vmw_cmd_dma(struct vmw_private *dev_priv, srf = vmw_res_to_srf(sw_context->res_cache[vmw_res_surface].res); - vmw_kms_cursor_snoop(srf, sw_context->tfile, &vmw_bo->base, header); + vmw_kms_cursor_snoop(srf, sw_context->fp->tfile, &vmw_bo->base, + header); out_no_surface: vmw_dmabuf_unreference(&vmw_bo); @@ -1480,6 +1515,98 @@ static int vmw_cmd_invalidate_gb_surface(struct vmw_private *dev_priv, &cmd->body.sid, NULL); } + +/** + * vmw_cmd_shader_define - Validate an SVGA_3D_CMD_SHADER_DEFINE + * command + * + * @dev_priv: Pointer to a device private struct. + * @sw_context: The software context being used for this batch. + * @header: Pointer to the command header in the command stream. + */ +static int vmw_cmd_shader_define(struct vmw_private *dev_priv, + struct vmw_sw_context *sw_context, + SVGA3dCmdHeader *header) +{ + struct vmw_shader_define_cmd { + SVGA3dCmdHeader header; + SVGA3dCmdDefineShader body; + } *cmd; + int ret; + size_t size; + + cmd = container_of(header, struct vmw_shader_define_cmd, + header); + + ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_context, + user_context_converter, &cmd->body.cid, + NULL); + if (unlikely(ret != 0)) + return ret; + + if (unlikely(!dev_priv->has_mob)) + return 0; + + size = cmd->header.size - sizeof(cmd->body); + ret = vmw_compat_shader_add(sw_context->fp->shman, + cmd->body.shid, cmd + 1, + cmd->body.type, size, + sw_context->fp->tfile, + &sw_context->staged_shaders); + if (unlikely(ret != 0)) + return ret; + + return vmw_resource_relocation_add(&sw_context->res_relocations, + NULL, &cmd->header.id - + sw_context->buf_start); + + return 0; +} + +/** + * vmw_cmd_shader_destroy - Validate an SVGA_3D_CMD_SHADER_DESTROY + * command + * + * @dev_priv: Pointer to a device private struct. + * @sw_context: The software context being used for this batch. + * @header: Pointer to the command header in the command stream. + */ +static int vmw_cmd_shader_destroy(struct vmw_private *dev_priv, + struct vmw_sw_context *sw_context, + SVGA3dCmdHeader *header) +{ + struct vmw_shader_destroy_cmd { + SVGA3dCmdHeader header; + SVGA3dCmdDestroyShader body; + } *cmd; + int ret; + + cmd = container_of(header, struct vmw_shader_destroy_cmd, + header); + + ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_context, + user_context_converter, &cmd->body.cid, + NULL); + if (unlikely(ret != 0)) + return ret; + + if (unlikely(!dev_priv->has_mob)) + return 0; + + ret = vmw_compat_shader_remove(sw_context->fp->shman, + cmd->body.shid, + cmd->body.type, + &sw_context->staged_shaders); + if (unlikely(ret != 0)) + return ret; + + return vmw_resource_relocation_add(&sw_context->res_relocations, + NULL, &cmd->header.id - + sw_context->buf_start); + + return 0; +} + /** * vmw_cmd_set_shader - Validate an SVGA_3D_CMD_SET_SHADER * command @@ -1511,10 +1638,17 @@ static int vmw_cmd_set_shader(struct vmw_private *dev_priv, if (dev_priv->has_mob) { struct vmw_ctx_bindinfo bi; struct vmw_resource_val_node *res_node; + u32 shid = cmd->body.shid; - ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_shader, - user_shader_converter, - &cmd->body.shid, &res_node); + (void) vmw_compat_shader_lookup(sw_context->fp->shman, + cmd->body.type, + &shid); + + ret = vmw_cmd_compat_res_check(dev_priv, sw_context, + vmw_res_shader, + user_shader_converter, + shid, + &cmd->body.shid, &res_node); if (unlikely(ret != 0)) return ret; @@ -1669,10 +1803,10 @@ static const struct vmw_cmd_entry const vmw_cmd_entries[SVGA_3D_CMD_MAX] = { true, false, false), VMW_CMD_DEF(SVGA_3D_CMD_PRESENT, &vmw_cmd_present_check, false, false, false), - VMW_CMD_DEF(SVGA_3D_CMD_SHADER_DEFINE, &vmw_cmd_cid_check, - true, true, false), - VMW_CMD_DEF(SVGA_3D_CMD_SHADER_DESTROY, &vmw_cmd_cid_check, - true, true, false), + VMW_CMD_DEF(SVGA_3D_CMD_SHADER_DEFINE, &vmw_cmd_shader_define, + true, false, false), + VMW_CMD_DEF(SVGA_3D_CMD_SHADER_DESTROY, &vmw_cmd_shader_destroy, + true, false, false), VMW_CMD_DEF(SVGA_3D_CMD_SET_SHADER, &vmw_cmd_set_shader, true, false, false), VMW_CMD_DEF(SVGA_3D_CMD_SET_SHADER_CONST, &vmw_cmd_set_shader_const, @@ -2206,7 +2340,7 @@ int vmw_execbuf_process(struct drm_file *file_priv, } else sw_context->kernel = true; - sw_context->tfile = vmw_fpriv(file_priv)->tfile; + sw_context->fp = vmw_fpriv(file_priv); sw_context->cur_reloc = 0; sw_context->cur_val_buf = 0; sw_context->fence_flags = 0; @@ -2223,6 +2357,7 @@ int vmw_execbuf_process(struct drm_file *file_priv, goto out_unlock; sw_context->res_ht_initialized = true; } + INIT_LIST_HEAD(&sw_context->staged_shaders); INIT_LIST_HEAD(&resource_list); ret = vmw_cmd_check_all(dev_priv, sw_context, kernel_commands, @@ -2311,6 +2446,8 @@ int vmw_execbuf_process(struct drm_file *file_priv, } list_splice_init(&sw_context->resource_list, &resource_list); + vmw_compat_shaders_commit(sw_context->fp->shman, + &sw_context->staged_shaders); mutex_unlock(&dev_priv->cmdbuf_mutex); /* @@ -2337,6 +2474,8 @@ int vmw_execbuf_process(struct drm_file *file_priv, list_splice_init(&sw_context->resource_list, &resource_list); error_resource = sw_context->error_resource; sw_context->error_resource = NULL; + vmw_compat_shaders_revert(sw_context->fp->shman, + &sw_context->staged_shaders); mutex_unlock(&dev_priv->cmdbuf_mutex); /* diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c b/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c index 1457ec4b7125..be85d7fdfb5b 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c @@ -29,6 +29,8 @@ #include "vmwgfx_resource_priv.h" #include "ttm/ttm_placement.h" +#define VMW_COMPAT_SHADER_HT_ORDER 12 + struct vmw_shader { struct vmw_resource res; SVGA3dShaderType type; @@ -40,6 +42,50 @@ struct vmw_user_shader { struct vmw_shader shader; }; +/** + * enum vmw_compat_shader_state - Staging state for compat shaders + */ +enum vmw_compat_shader_state { + VMW_COMPAT_COMMITED, + VMW_COMPAT_ADD, + VMW_COMPAT_DEL +}; + +/** + * struct vmw_compat_shader - Metadata for compat shaders. + * + * @handle: The TTM handle of the guest backed shader. + * @tfile: The struct ttm_object_file the guest backed shader is registered + * with. + * @hash: Hash item for lookup. + * @head: List head for staging lists or the compat shader manager list. + * @state: Staging state. + * + * The structure is protected by the cmdbuf lock. + */ +struct vmw_compat_shader { + u32 handle; + struct ttm_object_file *tfile; + struct drm_hash_item hash; + struct list_head head; + enum vmw_compat_shader_state state; +}; + +/** + * struct vmw_compat_shader_manager - Compat shader manager. + * + * @shaders: Hash table containing staged and commited compat shaders + * @list: List of commited shaders. + * @dev_priv: Pointer to a device private structure. + * + * @shaders and @list are protected by the cmdbuf mutex for now. + */ +struct vmw_compat_shader_manager { + struct drm_open_hash shaders; + struct list_head list; + struct vmw_private *dev_priv; +}; + static void vmw_user_shader_free(struct vmw_resource *res); static struct vmw_resource * vmw_user_shader_base_to_res(struct ttm_base_object *base); @@ -325,13 +371,81 @@ int vmw_shader_destroy_ioctl(struct drm_device *dev, void *data, TTM_REF_USAGE); } +int vmw_shader_alloc(struct vmw_private *dev_priv, + struct vmw_dma_buffer *buffer, + size_t shader_size, + size_t offset, + SVGA3dShaderType shader_type, + struct ttm_object_file *tfile, + u32 *handle) +{ + struct vmw_user_shader *ushader; + struct vmw_resource *res, *tmp; + int ret; + + /* + * Approximate idr memory usage with 128 bytes. It will be limited + * by maximum number_of shaders anyway. + */ + if (unlikely(vmw_user_shader_size == 0)) + vmw_user_shader_size = + ttm_round_pot(sizeof(struct vmw_user_shader)) + 128; + + ret = ttm_mem_global_alloc(vmw_mem_glob(dev_priv), + vmw_user_shader_size, + false, true); + if (unlikely(ret != 0)) { + if (ret != -ERESTARTSYS) + DRM_ERROR("Out of graphics memory for shader " + "creation.\n"); + goto out; + } + + ushader = kzalloc(sizeof(*ushader), GFP_KERNEL); + if (unlikely(ushader == NULL)) { + ttm_mem_global_free(vmw_mem_glob(dev_priv), + vmw_user_shader_size); + ret = -ENOMEM; + goto out; + } + + res = &ushader->shader.res; + ushader->base.shareable = false; + ushader->base.tfile = NULL; + + /* + * From here on, the destructor takes over resource freeing. + */ + + ret = vmw_gb_shader_init(dev_priv, res, shader_size, + offset, shader_type, buffer, + vmw_user_shader_free); + if (unlikely(ret != 0)) + goto out; + + tmp = vmw_resource_reference(res); + ret = ttm_base_object_init(tfile, &ushader->base, false, + VMW_RES_SHADER, + &vmw_user_shader_base_release, NULL); + + if (unlikely(ret != 0)) { + vmw_resource_unreference(&tmp); + goto out_err; + } + + if (handle) + *handle = ushader->base.hash.key; +out_err: + vmw_resource_unreference(&res); +out: + return ret; +} + + int vmw_shader_define_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv) { struct vmw_private *dev_priv = vmw_priv(dev); - struct vmw_user_shader *ushader; - struct vmw_resource *res; - struct vmw_resource *tmp; struct drm_vmw_shader_create_arg *arg = (struct drm_vmw_shader_create_arg *)data; struct ttm_object_file *tfile = vmw_fpriv(file_priv)->tfile; @@ -373,69 +487,324 @@ int vmw_shader_define_ioctl(struct drm_device *dev, void *data, goto out_bad_arg; } - /* - * Approximate idr memory usage with 128 bytes. It will be limited - * by maximum number_of shaders anyway. - */ - - if (unlikely(vmw_user_shader_size == 0)) - vmw_user_shader_size = ttm_round_pot(sizeof(*ushader)) - + 128; - ret = ttm_read_lock(&vmaster->lock, true); if (unlikely(ret != 0)) - return ret; + goto out_bad_arg; - ret = ttm_mem_global_alloc(vmw_mem_glob(dev_priv), - vmw_user_shader_size, - false, true); - if (unlikely(ret != 0)) { - if (ret != -ERESTARTSYS) - DRM_ERROR("Out of graphics memory for shader" - " creation.\n"); - goto out_unlock; - } + ret = vmw_shader_alloc(dev_priv, buffer, arg->size, arg->offset, + shader_type, tfile, &arg->shader_handle); - ushader = kzalloc(sizeof(*ushader), GFP_KERNEL); - if (unlikely(ushader == NULL)) { - ttm_mem_global_free(vmw_mem_glob(dev_priv), - vmw_user_shader_size); - ret = -ENOMEM; - goto out_unlock; - } - - res = &ushader->shader.res; - ushader->base.shareable = false; - ushader->base.tfile = NULL; - - /* - * From here on, the destructor takes over resource freeing. - */ - - ret = vmw_gb_shader_init(dev_priv, res, arg->size, - arg->offset, shader_type, buffer, - vmw_user_shader_free); - if (unlikely(ret != 0)) - goto out_unlock; - - tmp = vmw_resource_reference(res); - ret = ttm_base_object_init(tfile, &ushader->base, false, - VMW_RES_SHADER, - &vmw_user_shader_base_release, NULL); - - if (unlikely(ret != 0)) { - vmw_resource_unreference(&tmp); - goto out_err; - } - - arg->shader_handle = ushader->base.hash.key; -out_err: - vmw_resource_unreference(&res); -out_unlock: ttm_read_unlock(&vmaster->lock); out_bad_arg: vmw_dmabuf_unreference(&buffer); - return ret; - +} + +/** + * vmw_compat_shader_lookup - Look up a compat shader + * + * @man: Pointer to the compat shader manager. + * @shader_type: The shader type, that combined with the user_key identifies + * the shader. + * @user_key: On entry, this should be a pointer to the user_key. + * On successful exit, it will contain the guest-backed shader's TTM handle. + * + * Returns 0 on success. Non-zero on failure, in which case the value pointed + * to by @user_key is unmodified. + */ +int vmw_compat_shader_lookup(struct vmw_compat_shader_manager *man, + SVGA3dShaderType shader_type, + u32 *user_key) +{ + struct drm_hash_item *hash; + int ret; + unsigned long key = *user_key | (shader_type << 24); + + ret = drm_ht_find_item(&man->shaders, key, &hash); + if (unlikely(ret != 0)) + return ret; + + *user_key = drm_hash_entry(hash, struct vmw_compat_shader, + hash)->handle; + + return 0; +} + +/** + * vmw_compat_shader_free - Free a compat shader. + * + * @man: Pointer to the compat shader manager. + * @entry: Pointer to a struct vmw_compat_shader. + * + * Frees a struct vmw_compat_shder entry and drops its reference to the + * guest backed shader. + */ +static void vmw_compat_shader_free(struct vmw_compat_shader_manager *man, + struct vmw_compat_shader *entry) +{ + list_del(&entry->head); + WARN_ON(drm_ht_remove_item(&man->shaders, &entry->hash)); + WARN_ON(ttm_ref_object_base_unref(entry->tfile, entry->handle, + TTM_REF_USAGE)); + kfree(entry); +} + +/** + * vmw_compat_shaders_commit - Commit a list of compat shader actions. + * + * @man: Pointer to the compat shader manager. + * @list: Caller's list of compat shader actions. + * + * This function commits a list of compat shader additions or removals. + * It is typically called when the execbuf ioctl call triggering these + * actions has commited the fifo contents to the device. + */ +void vmw_compat_shaders_commit(struct vmw_compat_shader_manager *man, + struct list_head *list) +{ + struct vmw_compat_shader *entry, *next; + + list_for_each_entry_safe(entry, next, list, head) { + list_del(&entry->head); + switch (entry->state) { + case VMW_COMPAT_ADD: + entry->state = VMW_COMPAT_COMMITED; + list_add_tail(&entry->head, &man->list); + break; + case VMW_COMPAT_DEL: + ttm_ref_object_base_unref(entry->tfile, entry->handle, + TTM_REF_USAGE); + kfree(entry); + break; + default: + BUG(); + break; + } + } +} + +/** + * vmw_compat_shaders_revert - Revert a list of compat shader actions + * + * @man: Pointer to the compat shader manager. + * @list: Caller's list of compat shader actions. + * + * This function reverts a list of compat shader additions or removals. + * It is typically called when the execbuf ioctl call triggering these + * actions failed for some reason, and the command stream was never + * submitted. + */ +void vmw_compat_shaders_revert(struct vmw_compat_shader_manager *man, + struct list_head *list) +{ + struct vmw_compat_shader *entry, *next; + int ret; + + list_for_each_entry_safe(entry, next, list, head) { + switch (entry->state) { + case VMW_COMPAT_ADD: + vmw_compat_shader_free(man, entry); + break; + case VMW_COMPAT_DEL: + ret = drm_ht_insert_item(&man->shaders, &entry->hash); + list_del(&entry->head); + list_add_tail(&entry->head, &man->list); + entry->state = VMW_COMPAT_COMMITED; + break; + default: + BUG(); + break; + } + } +} + +/** + * vmw_compat_shader_remove - Stage a compat shader for removal. + * + * @man: Pointer to the compat shader manager + * @user_key: The key that is used to identify the shader. The key is + * unique to the shader type. + * @shader_type: Shader type. + * @list: Caller's list of staged shader actions. + * + * This function stages a compat shader for removal and removes the key from + * the shader manager's hash table. If the shader was previously only staged + * for addition it is completely removed (But the execbuf code may keep a + * reference if it was bound to a context between addition and removal). If + * it was previously commited to the manager, it is staged for removal. + */ +int vmw_compat_shader_remove(struct vmw_compat_shader_manager *man, + u32 user_key, SVGA3dShaderType shader_type, + struct list_head *list) +{ + struct vmw_compat_shader *entry; + struct drm_hash_item *hash; + int ret; + + ret = drm_ht_find_item(&man->shaders, user_key | (shader_type << 24), + &hash); + if (likely(ret != 0)) + return -EINVAL; + + entry = drm_hash_entry(hash, struct vmw_compat_shader, hash); + + switch (entry->state) { + case VMW_COMPAT_ADD: + vmw_compat_shader_free(man, entry); + break; + case VMW_COMPAT_COMMITED: + (void) drm_ht_remove_item(&man->shaders, &entry->hash); + list_del(&entry->head); + entry->state = VMW_COMPAT_DEL; + list_add_tail(&entry->head, list); + break; + default: + BUG(); + break; + } + + return 0; +} + +/** + * vmw_compat_shader_add - Create a compat shader and add the + * key to the manager + * + * @man: Pointer to the compat shader manager + * @user_key: The key that is used to identify the shader. The key is + * unique to the shader type. + * @bytecode: Pointer to the bytecode of the shader. + * @shader_type: Shader type. + * @tfile: Pointer to a struct ttm_object_file that the guest-backed shader is + * to be created with. + * @list: Caller's list of staged shader actions. + * + * Note that only the key is added to the shader manager's hash table. + * The shader is not yet added to the shader manager's list of shaders. + */ +int vmw_compat_shader_add(struct vmw_compat_shader_manager *man, + u32 user_key, const void *bytecode, + SVGA3dShaderType shader_type, + size_t size, + struct ttm_object_file *tfile, + struct list_head *list) +{ + struct vmw_dma_buffer *buf; + struct ttm_bo_kmap_obj map; + bool is_iomem; + struct vmw_compat_shader *compat; + u32 handle; + int ret; + + if (user_key > ((1 << 24) - 1) || (unsigned) shader_type > 16) + return -EINVAL; + + /* Allocate and pin a DMA buffer */ + buf = kzalloc(sizeof(*buf), GFP_KERNEL); + if (unlikely(buf == NULL)) + return -ENOMEM; + + ret = vmw_dmabuf_init(man->dev_priv, buf, size, &vmw_sys_ne_placement, + true, vmw_dmabuf_bo_free); + if (unlikely(ret != 0)) + goto out; + + ret = ttm_bo_reserve(&buf->base, false, true, false, NULL); + if (unlikely(ret != 0)) + goto no_reserve; + + /* Map and copy shader bytecode. */ + ret = ttm_bo_kmap(&buf->base, 0, PAGE_ALIGN(size) >> PAGE_SHIFT, + &map); + if (unlikely(ret != 0)) { + ttm_bo_unreserve(&buf->base); + goto no_reserve; + } + + memcpy(ttm_kmap_obj_virtual(&map, &is_iomem), bytecode, size); + WARN_ON(is_iomem); + + ttm_bo_kunmap(&map); + ret = ttm_bo_validate(&buf->base, &vmw_sys_placement, false, true); + WARN_ON(ret != 0); + ttm_bo_unreserve(&buf->base); + + /* Create a guest-backed shader container backed by the dma buffer */ + ret = vmw_shader_alloc(man->dev_priv, buf, size, 0, shader_type, + tfile, &handle); + vmw_dmabuf_unreference(&buf); + if (unlikely(ret != 0)) + goto no_reserve; + /* + * Create a compat shader structure and stage it for insertion + * in the manager + */ + compat = kzalloc(sizeof(*compat), GFP_KERNEL); + if (compat == NULL) + goto no_compat; + + compat->hash.key = user_key | (shader_type << 24); + ret = drm_ht_insert_item(&man->shaders, &compat->hash); + if (unlikely(ret != 0)) + goto out_invalid_key; + + compat->state = VMW_COMPAT_ADD; + compat->handle = handle; + compat->tfile = tfile; + list_add_tail(&compat->head, list); + + return 0; + +out_invalid_key: + kfree(compat); +no_compat: + ttm_ref_object_base_unref(tfile, handle, TTM_REF_USAGE); +no_reserve: +out: + return ret; +} + +/** + * vmw_compat_shader_man_create - Create a compat shader manager + * + * @dev_priv: Pointer to a device private structure. + * + * Typically done at file open time. If successful returns a pointer to a + * compat shader manager. Otherwise returns an error pointer. + */ +struct vmw_compat_shader_manager * +vmw_compat_shader_man_create(struct vmw_private *dev_priv) +{ + struct vmw_compat_shader_manager *man; + int ret; + + man = kzalloc(sizeof(*man), GFP_KERNEL); + + man->dev_priv = dev_priv; + INIT_LIST_HEAD(&man->list); + ret = drm_ht_create(&man->shaders, VMW_COMPAT_SHADER_HT_ORDER); + if (ret == 0) + return man; + + kfree(man); + return ERR_PTR(ret); +} + +/** + * vmw_compat_shader_man_destroy - Destroy a compat shader manager + * + * @man: Pointer to the shader manager to destroy. + * + * Typically done at file close time. + */ +void vmw_compat_shader_man_destroy(struct vmw_compat_shader_manager *man) +{ + struct vmw_compat_shader *entry, *next; + + mutex_lock(&man->dev_priv->cmdbuf_mutex); + list_for_each_entry_safe(entry, next, &man->list, head) + vmw_compat_shader_free(man, entry); + + mutex_unlock(&man->dev_priv->cmdbuf_mutex); + kfree(man); } From a6fc955ff9fc60dcffc80003410a85d874e4f1e3 Mon Sep 17 00:00:00 2001 From: Thomas Hellstrom Date: Fri, 31 Jan 2014 10:21:10 +0100 Subject: [PATCH 223/596] drm/vmwgfx: Detect old user-space drivers and set up legacy emulation v2 GB aware mesa userspace drivers are detected by the fact that they are calling the vmw getparam ioctl querying DRM_VMW_PARAM_HW_CAPS to detect whether the device is Guest-backed object capable. For other drivers, lie about hardware version and send the 3D capabilities in a format they expect. v2: Use DRM_VMW_PARAM_MAX_MOB_MEMORY to detect gb awareness, Make sure we don't ovwerwrite bounce buffer or write past user-space buffer indicated size. Signed-off-by: Thomas Hellstrom Reviewed-by: Jakob Bornecrantz --- drivers/gpu/drm/vmwgfx/svga3d_reg.h | 24 +++++++ drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c | 93 ++++++++++++++++++++++----- 2 files changed, 101 insertions(+), 16 deletions(-) diff --git a/drivers/gpu/drm/vmwgfx/svga3d_reg.h b/drivers/gpu/drm/vmwgfx/svga3d_reg.h index d95335cb90bd..b645647b7776 100644 --- a/drivers/gpu/drm/vmwgfx/svga3d_reg.h +++ b/drivers/gpu/drm/vmwgfx/svga3d_reg.h @@ -2583,4 +2583,28 @@ typedef union { float f; } SVGA3dDevCapResult; +typedef enum { + SVGA3DCAPS_RECORD_UNKNOWN = 0, + SVGA3DCAPS_RECORD_DEVCAPS_MIN = 0x100, + SVGA3DCAPS_RECORD_DEVCAPS = 0x100, + SVGA3DCAPS_RECORD_DEVCAPS_MAX = 0x1ff, +} SVGA3dCapsRecordType; + +typedef +struct SVGA3dCapsRecordHeader { + uint32 length; + SVGA3dCapsRecordType type; +} +SVGA3dCapsRecordHeader; + +typedef +struct SVGA3dCapsRecord { + SVGA3dCapsRecordHeader header; + uint32 data[1]; +} +SVGA3dCapsRecord; + + +typedef uint32 SVGA3dCapPair[2]; + #endif /* _SVGA3D_REG_H_ */ diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c index 116c49736763..f9881f9e62bd 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c @@ -29,12 +29,18 @@ #include #include "vmwgfx_kms.h" +struct svga_3d_compat_cap { + SVGA3dCapsRecordHeader header; + SVGA3dCapPair pairs[SVGA3D_DEVCAP_MAX]; +}; + int vmw_getparam_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv) { struct vmw_private *dev_priv = vmw_priv(dev); struct drm_vmw_getparam_arg *param = (struct drm_vmw_getparam_arg *)data; + struct vmw_fpriv *vmw_fp = vmw_fpriv(file_priv); switch (param->param) { case DRM_VMW_PARAM_NUM_STREAMS: @@ -60,6 +66,11 @@ int vmw_getparam_ioctl(struct drm_device *dev, void *data, __le32 __iomem *fifo_mem = dev_priv->mmio_virt; const struct vmw_fifo_state *fifo = &dev_priv->fifo; + if ((dev_priv->capabilities & SVGA_CAP_GBOBJECTS)) { + param->value = SVGA3D_HWVERSION_WS8_B1; + break; + } + param->value = ioread32(fifo_mem + ((fifo->capabilities & @@ -69,17 +80,26 @@ int vmw_getparam_ioctl(struct drm_device *dev, void *data, break; } case DRM_VMW_PARAM_MAX_SURF_MEMORY: - param->value = dev_priv->memory_size; + if ((dev_priv->capabilities & SVGA_CAP_GBOBJECTS) && + !vmw_fp->gb_aware) + param->value = dev_priv->max_mob_pages * PAGE_SIZE / 2; + else + param->value = dev_priv->memory_size; break; case DRM_VMW_PARAM_3D_CAPS_SIZE: - if (dev_priv->capabilities & SVGA_CAP_GBOBJECTS) - param->value = SVGA3D_DEVCAP_MAX; + if ((dev_priv->capabilities & SVGA_CAP_GBOBJECTS) && + vmw_fp->gb_aware) + param->value = SVGA3D_DEVCAP_MAX * sizeof(uint32_t); + else if (dev_priv->capabilities & SVGA_CAP_GBOBJECTS) + param->value = sizeof(struct svga_3d_compat_cap) + + sizeof(uint32_t); else param->value = (SVGA_FIFO_3D_CAPS_LAST - - SVGA_FIFO_3D_CAPS + 1); - param->value *= sizeof(uint32_t); + SVGA_FIFO_3D_CAPS + 1) * + sizeof(uint32_t); break; case DRM_VMW_PARAM_MAX_MOB_MEMORY: + vmw_fp->gb_aware = true; param->value = dev_priv->max_mob_pages * PAGE_SIZE; break; default: @@ -91,6 +111,38 @@ int vmw_getparam_ioctl(struct drm_device *dev, void *data, return 0; } +static int vmw_fill_compat_cap(struct vmw_private *dev_priv, void *bounce, + size_t size) +{ + struct svga_3d_compat_cap *compat_cap = + (struct svga_3d_compat_cap *) bounce; + unsigned int i; + size_t pair_offset = offsetof(struct svga_3d_compat_cap, pairs); + unsigned int max_size; + + if (size < pair_offset) + return -EINVAL; + + max_size = (size - pair_offset) / sizeof(SVGA3dCapPair); + + if (max_size > SVGA3D_DEVCAP_MAX) + max_size = SVGA3D_DEVCAP_MAX; + + compat_cap->header.length = + (pair_offset + max_size * sizeof(SVGA3dCapPair)) / sizeof(u32); + compat_cap->header.type = SVGA3DCAPS_RECORD_DEVCAPS; + + mutex_lock(&dev_priv->hw_mutex); + for (i = 0; i < max_size; ++i) { + vmw_write(dev_priv, SVGA_REG_DEV_CAP, i); + compat_cap->pairs[i][0] = i; + compat_cap->pairs[i][1] = vmw_read(dev_priv, SVGA_REG_DEV_CAP); + } + mutex_unlock(&dev_priv->hw_mutex); + + return 0; +} + int vmw_get_cap_3d_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv) @@ -104,41 +156,49 @@ int vmw_get_cap_3d_ioctl(struct drm_device *dev, void *data, void *bounce; int ret; bool gb_objects = !!(dev_priv->capabilities & SVGA_CAP_GBOBJECTS); + struct vmw_fpriv *vmw_fp = vmw_fpriv(file_priv); if (unlikely(arg->pad64 != 0)) { DRM_ERROR("Illegal GET_3D_CAP argument.\n"); return -EINVAL; } - if (gb_objects) - size = SVGA3D_DEVCAP_MAX; + if (gb_objects && vmw_fp->gb_aware) + size = SVGA3D_DEVCAP_MAX * sizeof(uint32_t); + else if (gb_objects) + size = sizeof(struct svga_3d_compat_cap) + sizeof(uint32_t); else - size = (SVGA_FIFO_3D_CAPS_LAST - SVGA_FIFO_3D_CAPS + 1); - - size *= sizeof(uint32_t); + size = (SVGA_FIFO_3D_CAPS_LAST - SVGA_FIFO_3D_CAPS + 1) * + sizeof(uint32_t); if (arg->max_size < size) size = arg->max_size; - bounce = vmalloc(size); + bounce = vzalloc(size); if (unlikely(bounce == NULL)) { DRM_ERROR("Failed to allocate bounce buffer for 3D caps.\n"); return -ENOMEM; } - if (gb_objects) { - int i; + if (gb_objects && vmw_fp->gb_aware) { + int i, num; uint32_t *bounce32 = (uint32_t *) bounce; + num = size / sizeof(uint32_t); + if (num > SVGA3D_DEVCAP_MAX) + num = SVGA3D_DEVCAP_MAX; + mutex_lock(&dev_priv->hw_mutex); - for (i = 0; i < SVGA3D_DEVCAP_MAX; ++i) { + for (i = 0; i < num; ++i) { vmw_write(dev_priv, SVGA_REG_DEV_CAP, i); *bounce32++ = vmw_read(dev_priv, SVGA_REG_DEV_CAP); } mutex_unlock(&dev_priv->hw_mutex); - + } else if (gb_objects) { + ret = vmw_fill_compat_cap(dev_priv, bounce, size); + if (unlikely(ret != 0)) + goto out_err; } else { - fifo_mem = dev_priv->mmio_virt; memcpy_fromio(bounce, &fifo_mem[SVGA_FIFO_3D_CAPS], size); } @@ -146,6 +206,7 @@ int vmw_get_cap_3d_ioctl(struct drm_device *dev, void *data, ret = copy_to_user(buffer, bounce, size); if (ret) ret = -EFAULT; +out_err: vfree(bounce); if (unlikely(ret != 0)) From 30f82d816d2dccfdc2063ac8cca994904c9b612c Mon Sep 17 00:00:00 2001 From: Thomas Hellstrom Date: Wed, 5 Feb 2014 08:13:56 +0100 Subject: [PATCH 224/596] drm/vmwgfx: Reemit context bindings when necessary v2 When a context is first referenced in the command stream, make sure that all scrubbed (as a result of eviction) bindings are re-emitted. Also make sure that all bound resources are put on the resource validate list. This is needed for legacy emulation, since legacy user-space drivers will typically not re-emit shader bindings. It also removes the requirement for user-space drivers to re-emit render-target- and texture bindings. Makes suspend and hibernate now also work with legacy user-space drivers on guest-backed devices. v2: Don't rebind on legacy devices. Signed-off-by: Thomas Hellstrom Reviewed-by: Jakob Bornecrantz --- drivers/gpu/drm/vmwgfx/vmwgfx_context.c | 144 +++++++++++++++++++---- drivers/gpu/drm/vmwgfx/vmwgfx_drv.h | 6 + drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c | 85 ++++++++++++- drivers/gpu/drm/vmwgfx/vmwgfx_resource.c | 11 +- drivers/gpu/drm/vmwgfx/vmwgfx_shader.c | 2 +- drivers/gpu/drm/vmwgfx/vmwgfx_surface.c | 2 +- 6 files changed, 222 insertions(+), 28 deletions(-) diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_context.c b/drivers/gpu/drm/vmwgfx/vmwgfx_context.c index 82c41daebc0e..9426c53fb483 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_context.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_context.c @@ -37,7 +37,7 @@ struct vmw_user_context { -typedef int (*vmw_scrub_func)(struct vmw_ctx_bindinfo *); +typedef int (*vmw_scrub_func)(struct vmw_ctx_bindinfo *, bool); static void vmw_user_context_free(struct vmw_resource *res); static struct vmw_resource * @@ -50,9 +50,11 @@ static int vmw_gb_context_unbind(struct vmw_resource *res, bool readback, struct ttm_validate_buffer *val_buf); static int vmw_gb_context_destroy(struct vmw_resource *res); -static int vmw_context_scrub_shader(struct vmw_ctx_bindinfo *bi); -static int vmw_context_scrub_render_target(struct vmw_ctx_bindinfo *bi); -static int vmw_context_scrub_texture(struct vmw_ctx_bindinfo *bi); +static int vmw_context_scrub_shader(struct vmw_ctx_bindinfo *bi, bool rebind); +static int vmw_context_scrub_render_target(struct vmw_ctx_bindinfo *bi, + bool rebind); +static int vmw_context_scrub_texture(struct vmw_ctx_bindinfo *bi, bool rebind); +static void vmw_context_binding_state_scrub(struct vmw_ctx_binding_state *cbs); static void vmw_context_binding_state_kill(struct vmw_ctx_binding_state *cbs); static uint64_t vmw_user_context_size; @@ -111,10 +113,14 @@ static void vmw_hw_context_destroy(struct vmw_resource *res) if (res->func->destroy == vmw_gb_context_destroy) { mutex_lock(&dev_priv->cmdbuf_mutex); + mutex_lock(&dev_priv->binding_mutex); + (void) vmw_context_binding_state_kill + (&container_of(res, struct vmw_user_context, res)->cbs); (void) vmw_gb_context_destroy(res); if (dev_priv->pinned_bo != NULL && !dev_priv->query_cid_valid) __vmw_execbuf_release_pinned_bo(dev_priv, NULL); + mutex_unlock(&dev_priv->binding_mutex); mutex_unlock(&dev_priv->cmdbuf_mutex); return; } @@ -328,7 +334,7 @@ static int vmw_gb_context_unbind(struct vmw_resource *res, BUG_ON(bo->mem.mem_type != VMW_PL_MOB); mutex_lock(&dev_priv->binding_mutex); - vmw_context_binding_state_kill(&uctx->cbs); + vmw_context_binding_state_scrub(&uctx->cbs); submit_size = sizeof(*cmd2) + (readback ? sizeof(*cmd1) : 0); @@ -378,10 +384,6 @@ static int vmw_gb_context_destroy(struct vmw_resource *res) SVGA3dCmdHeader header; SVGA3dCmdDestroyGBContext body; } *cmd; - struct vmw_user_context *uctx = - container_of(res, struct vmw_user_context, res); - - BUG_ON(!list_empty(&uctx->cbs.list)); if (likely(res->id == -1)) return 0; @@ -528,8 +530,9 @@ int vmw_context_define_ioctl(struct drm_device *dev, void *data, * vmw_context_scrub_shader - scrub a shader binding from a context. * * @bi: single binding information. + * @rebind: Whether to issue a bind instead of scrub command. */ -static int vmw_context_scrub_shader(struct vmw_ctx_bindinfo *bi) +static int vmw_context_scrub_shader(struct vmw_ctx_bindinfo *bi, bool rebind) { struct vmw_private *dev_priv = bi->ctx->dev_priv; struct { @@ -548,7 +551,8 @@ static int vmw_context_scrub_shader(struct vmw_ctx_bindinfo *bi) cmd->header.size = sizeof(cmd->body); cmd->body.cid = bi->ctx->id; cmd->body.type = bi->i1.shader_type; - cmd->body.shid = SVGA3D_INVALID_ID; + cmd->body.shid = + cpu_to_le32((rebind) ? bi->res->id : SVGA3D_INVALID_ID); vmw_fifo_commit(dev_priv, sizeof(*cmd)); return 0; @@ -559,8 +563,10 @@ static int vmw_context_scrub_shader(struct vmw_ctx_bindinfo *bi) * from a context. * * @bi: single binding information. + * @rebind: Whether to issue a bind instead of scrub command. */ -static int vmw_context_scrub_render_target(struct vmw_ctx_bindinfo *bi) +static int vmw_context_scrub_render_target(struct vmw_ctx_bindinfo *bi, + bool rebind) { struct vmw_private *dev_priv = bi->ctx->dev_priv; struct { @@ -579,7 +585,8 @@ static int vmw_context_scrub_render_target(struct vmw_ctx_bindinfo *bi) cmd->header.size = sizeof(cmd->body); cmd->body.cid = bi->ctx->id; cmd->body.type = bi->i1.rt_type; - cmd->body.target.sid = SVGA3D_INVALID_ID; + cmd->body.target.sid = + cpu_to_le32((rebind) ? bi->res->id : SVGA3D_INVALID_ID); cmd->body.target.face = 0; cmd->body.target.mipmap = 0; vmw_fifo_commit(dev_priv, sizeof(*cmd)); @@ -591,11 +598,13 @@ static int vmw_context_scrub_render_target(struct vmw_ctx_bindinfo *bi) * vmw_context_scrub_texture - scrub a texture binding from a context. * * @bi: single binding information. + * @rebind: Whether to issue a bind instead of scrub command. * * TODO: Possibly complement this function with a function that takes * a list of texture bindings and combines them to a single command. */ -static int vmw_context_scrub_texture(struct vmw_ctx_bindinfo *bi) +static int vmw_context_scrub_texture(struct vmw_ctx_bindinfo *bi, + bool rebind) { struct vmw_private *dev_priv = bi->ctx->dev_priv; struct { @@ -619,7 +628,8 @@ static int vmw_context_scrub_texture(struct vmw_ctx_bindinfo *bi) cmd->body.c.cid = bi->ctx->id; cmd->body.s1.stage = bi->i1.texture_stage; cmd->body.s1.name = SVGA3D_TS_BIND_TEXTURE; - cmd->body.s1.value = (uint32) SVGA3D_INVALID_ID; + cmd->body.s1.value = + cpu_to_le32((rebind) ? bi->res->id : SVGA3D_INVALID_ID); vmw_fifo_commit(dev_priv, sizeof(*cmd)); return 0; @@ -692,6 +702,7 @@ int vmw_context_binding_add(struct vmw_ctx_binding_state *cbs, vmw_context_binding_drop(loc); loc->bi = *bi; + loc->bi.scrubbed = false; list_add_tail(&loc->ctx_list, &cbs->list); INIT_LIST_HEAD(&loc->res_list); @@ -727,12 +738,11 @@ static void vmw_context_binding_transfer(struct vmw_ctx_binding_state *cbs, if (loc->bi.ctx != NULL) vmw_context_binding_drop(loc); - loc->bi = *bi; - list_add_tail(&loc->ctx_list, &cbs->list); - if (bi->res != NULL) + if (bi->res != NULL) { + loc->bi = *bi; + list_add_tail(&loc->ctx_list, &cbs->list); list_add_tail(&loc->res_list, &bi->res->binding_head); - else - INIT_LIST_HEAD(&loc->res_list); + } } /** @@ -746,7 +756,10 @@ static void vmw_context_binding_transfer(struct vmw_ctx_binding_state *cbs, */ static void vmw_context_binding_kill(struct vmw_ctx_binding *cb) { - (void) vmw_scrub_funcs[cb->bi.bt](&cb->bi); + if (!cb->bi.scrubbed) { + (void) vmw_scrub_funcs[cb->bi.bt](&cb->bi, false); + cb->bi.scrubbed = true; + } vmw_context_binding_drop(cb); } @@ -767,6 +780,27 @@ static void vmw_context_binding_state_kill(struct vmw_ctx_binding_state *cbs) vmw_context_binding_kill(entry); } +/** + * vmw_context_binding_state_scrub - Scrub all bindings associated with a + * struct vmw_ctx_binding state structure. + * + * @cbs: Pointer to the context binding state tracker. + * + * Emits commands to scrub all bindings associated with the + * context binding state tracker. + */ +static void vmw_context_binding_state_scrub(struct vmw_ctx_binding_state *cbs) +{ + struct vmw_ctx_binding *entry; + + list_for_each_entry(entry, &cbs->list, ctx_list) { + if (!entry->bi.scrubbed) { + (void) vmw_scrub_funcs[entry->bi.bt](&entry->bi, false); + entry->bi.scrubbed = true; + } + } +} + /** * vmw_context_binding_res_list_kill - Kill all bindings on a * resource binding list @@ -784,6 +818,27 @@ void vmw_context_binding_res_list_kill(struct list_head *head) vmw_context_binding_kill(entry); } +/** + * vmw_context_binding_res_list_scrub - Scrub all bindings on a + * resource binding list + * + * @head: list head of resource binding list + * + * Scrub all bindings associated with a specific resource. Typically + * called before the resource is evicted. + */ +void vmw_context_binding_res_list_scrub(struct list_head *head) +{ + struct vmw_ctx_binding *entry; + + list_for_each_entry(entry, head, res_list) { + if (!entry->bi.scrubbed) { + (void) vmw_scrub_funcs[entry->bi.bt](&entry->bi, false); + entry->bi.scrubbed = true; + } + } +} + /** * vmw_context_binding_state_transfer - Commit staged binding info * @@ -803,3 +858,50 @@ void vmw_context_binding_state_transfer(struct vmw_resource *ctx, list_for_each_entry_safe(entry, next, &from->list, ctx_list) vmw_context_binding_transfer(&uctx->cbs, &entry->bi); } + +/** + * vmw_context_rebind_all - Rebind all scrubbed bindings of a context + * + * @ctx: The context resource + * + * Walks through the context binding list and rebinds all scrubbed + * resources. + */ +int vmw_context_rebind_all(struct vmw_resource *ctx) +{ + struct vmw_ctx_binding *entry; + struct vmw_user_context *uctx = + container_of(ctx, struct vmw_user_context, res); + struct vmw_ctx_binding_state *cbs = &uctx->cbs; + int ret; + + list_for_each_entry(entry, &cbs->list, ctx_list) { + if (likely(!entry->bi.scrubbed)) + continue; + + if (WARN_ON(entry->bi.res == NULL || entry->bi.res->id == + SVGA3D_INVALID_ID)) + continue; + + ret = vmw_scrub_funcs[entry->bi.bt](&entry->bi, true); + if (unlikely(ret != 0)) + return ret; + + entry->bi.scrubbed = false; + } + + return 0; +} + +/** + * vmw_context_binding_list - Return a list of context bindings + * + * @ctx: The context resource + * + * Returns the current list of bindings of the given context. Note that + * this list becomes stale as soon as the dev_priv::binding_mutex is unlocked. + */ +struct list_head *vmw_context_binding_list(struct vmw_resource *ctx) +{ + return &(container_of(ctx, struct vmw_user_context, res)->cbs.list); +} diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h index cef0ff7ac738..ecaa302a6154 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h @@ -276,6 +276,7 @@ struct vmw_ctx_bindinfo { struct vmw_resource *ctx; struct vmw_resource *res; enum vmw_ctx_binding_type bt; + bool scrubbed; union { SVGA3dShaderType shader_type; SVGA3dRenderTargetType rt_type; @@ -574,6 +575,8 @@ struct vmw_user_resource_conv; extern void vmw_resource_unreference(struct vmw_resource **p_res); extern struct vmw_resource *vmw_resource_reference(struct vmw_resource *res); +extern struct vmw_resource * +vmw_resource_reference_unless_doomed(struct vmw_resource *res); extern int vmw_resource_validate(struct vmw_resource *res); extern int vmw_resource_reserve(struct vmw_resource *res, bool no_backup); extern bool vmw_resource_needs_backup(const struct vmw_resource *res); @@ -962,6 +965,9 @@ extern void vmw_context_binding_state_transfer(struct vmw_resource *res, struct vmw_ctx_binding_state *cbs); extern void vmw_context_binding_res_list_kill(struct list_head *head); +extern void vmw_context_binding_res_list_scrub(struct list_head *head); +extern int vmw_context_rebind_all(struct vmw_resource *ctx); +extern struct list_head *vmw_context_binding_list(struct vmw_resource *ctx); /* * Surface management - vmwgfx_surface.c diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c index 352224b9d667..269b85cc875a 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c @@ -179,6 +179,44 @@ static int vmw_resource_val_add(struct vmw_sw_context *sw_context, return 0; } +/** + * vmw_resource_context_res_add - Put resources previously bound to a context on + * the validation list + * + * @dev_priv: Pointer to a device private structure + * @sw_context: Pointer to a software context used for this command submission + * @ctx: Pointer to the context resource + * + * This function puts all resources that were previously bound to @ctx on + * the resource validation list. This is part of the context state reemission + */ +static int vmw_resource_context_res_add(struct vmw_private *dev_priv, + struct vmw_sw_context *sw_context, + struct vmw_resource *ctx) +{ + struct list_head *binding_list; + struct vmw_ctx_binding *entry; + int ret = 0; + struct vmw_resource *res; + + mutex_lock(&dev_priv->binding_mutex); + binding_list = vmw_context_binding_list(ctx); + + list_for_each_entry(entry, binding_list, ctx_list) { + res = vmw_resource_reference_unless_doomed(entry->bi.res); + if (unlikely(res == NULL)) + continue; + + ret = vmw_resource_val_add(sw_context, entry->bi.res, NULL); + vmw_resource_unreference(&res); + if (unlikely(ret != 0)) + break; + } + + mutex_unlock(&dev_priv->binding_mutex); + return ret; +} + /** * vmw_resource_relocation_add - Add a relocation to the relocation list * @@ -470,7 +508,11 @@ vmw_cmd_compat_res_check(struct vmw_private *dev_priv, if (p_val) *p_val = node; - if (node->first_usage && res_type == vmw_res_context) { + if (dev_priv->has_mob && node->first_usage && + res_type == vmw_res_context) { + ret = vmw_resource_context_res_add(dev_priv, sw_context, res); + if (unlikely(ret != 0)) + goto out_no_reloc; node->staged_bindings = kzalloc(sizeof(*node->staged_bindings), GFP_KERNEL); if (node->staged_bindings == NULL) { @@ -516,6 +558,34 @@ vmw_cmd_res_check(struct vmw_private *dev_priv, converter, *id_loc, id_loc, p_val); } +/** + * vmw_rebind_contexts - Rebind all resources previously bound to + * referenced contexts. + * + * @sw_context: Pointer to the software context. + * + * Rebind context binding points that have been scrubbed because of eviction. + */ +static int vmw_rebind_contexts(struct vmw_sw_context *sw_context) +{ + struct vmw_resource_val_node *val; + int ret; + + list_for_each_entry(val, &sw_context->resource_list, head) { + if (likely(!val->staged_bindings)) + continue; + + ret = vmw_context_rebind_all(val->res); + if (unlikely(ret != 0)) { + if (ret != -ERESTARTSYS) + DRM_ERROR("Failed to rebind context.\n"); + return ret; + } + } + + return 0; +} + /** * vmw_cmd_cid_check - Check a command header for valid context information. * @@ -1640,9 +1710,10 @@ static int vmw_cmd_set_shader(struct vmw_private *dev_priv, struct vmw_resource_val_node *res_node; u32 shid = cmd->body.shid; - (void) vmw_compat_shader_lookup(sw_context->fp->shman, - cmd->body.type, - &shid); + if (shid != SVGA3D_INVALID_ID) + (void) vmw_compat_shader_lookup(sw_context->fp->shman, + cmd->body.type, + &shid); ret = vmw_cmd_compat_res_check(dev_priv, sw_context, vmw_res_shader, @@ -2395,6 +2466,12 @@ int vmw_execbuf_process(struct drm_file *file_priv, goto out_err; } + if (dev_priv->has_mob) { + ret = vmw_rebind_contexts(sw_context); + if (unlikely(ret != 0)) + goto out_err; + } + cmd = vmw_fifo_reserve(dev_priv, command_size); if (unlikely(cmd == NULL)) { DRM_ERROR("Failed reserving fifo space for commands.\n"); diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c b/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c index 6fdd82d42f65..2aa4bc6a4d60 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c @@ -88,6 +88,11 @@ struct vmw_resource *vmw_resource_reference(struct vmw_resource *res) return res; } +struct vmw_resource * +vmw_resource_reference_unless_doomed(struct vmw_resource *res) +{ + return kref_get_unless_zero(&res->kref) ? res : NULL; +} /** * vmw_resource_release_id - release a resource id to the id manager. @@ -136,8 +141,12 @@ static void vmw_resource_release(struct kref *kref) vmw_dmabuf_unreference(&res->backup); } - if (likely(res->hw_destroy != NULL)) + if (likely(res->hw_destroy != NULL)) { res->hw_destroy(res); + mutex_lock(&dev_priv->binding_mutex); + vmw_context_binding_res_list_kill(&res->binding_head); + mutex_unlock(&dev_priv->binding_mutex); + } id = res->id; if (res->res_free != NULL) diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c b/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c index be85d7fdfb5b..217d941b8176 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c @@ -304,7 +304,7 @@ static int vmw_gb_shader_destroy(struct vmw_resource *res) return 0; mutex_lock(&dev_priv->binding_mutex); - vmw_context_binding_res_list_kill(&res->binding_head); + vmw_context_binding_res_list_scrub(&res->binding_head); cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd)); if (unlikely(cmd == NULL)) { diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c b/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c index ec2b99833fce..82468d902915 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c @@ -1111,7 +1111,7 @@ static int vmw_gb_surface_destroy(struct vmw_resource *res) return 0; mutex_lock(&dev_priv->binding_mutex); - vmw_context_binding_res_list_kill(&res->binding_head); + vmw_context_binding_res_list_scrub(&res->binding_head); cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd)); if (unlikely(cmd == NULL)) { From 276ab336b4c6e483d12fd46cbf24f97f71867710 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Wed, 5 Feb 2014 08:49:41 +0100 Subject: [PATCH 225/596] ALSA: hda - Improve loopback path lookups for AD1983 AD1983 has flexible loopback routes and the generic parser would take wrong path confusingly instead of taking individual paths via NID 0x0c and 0x0d. For avoiding it, limit the connections at these widgets so that the parser can think more straightforwardly. This fixes the regression of the missing line-in loopback on Dell machine. Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=70011 Cc: Signed-off-by: Takashi Iwai --- sound/pci/hda/patch_analog.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/sound/pci/hda/patch_analog.c b/sound/pci/hda/patch_analog.c index 195cd62cdce5..df3652ad15ef 100644 --- a/sound/pci/hda/patch_analog.c +++ b/sound/pci/hda/patch_analog.c @@ -491,6 +491,8 @@ static int ad1983_add_spdif_mux_ctl(struct hda_codec *codec) static int patch_ad1983(struct hda_codec *codec) { struct ad198x_spec *spec; + static hda_nid_t conn_0c[] = { 0x08 }; + static hda_nid_t conn_0d[] = { 0x09 }; int err; err = alloc_ad_spec(codec); @@ -501,6 +503,11 @@ static int patch_ad1983(struct hda_codec *codec) spec->gen.mixer_nid = 0x0e; spec->gen.beep_nid = 0x10; set_beep_amp(spec, 0x10, 0, HDA_OUTPUT); + + /* limit the loopback routes not to confuse the parser */ + snd_hda_override_conn_list(codec, 0x0c, ARRAY_SIZE(conn_0c), conn_0c); + snd_hda_override_conn_list(codec, 0x0d, ARRAY_SIZE(conn_0d), conn_0d); + err = ad198x_parse_auto_config(codec, false); if (err < 0) goto error; From cd9a21a831af0af7539a0e37e4455da03df7cf82 Mon Sep 17 00:00:00 2001 From: Dave Jones Date: Thu, 30 Jan 2014 21:27:25 -0500 Subject: [PATCH 226/596] vmwgfx: Fix unitialized stack read in vmw_setup_otable_base One of the error paths in vmw_setup_otable_base causes us to return with 'ret' having never been set to anything causing us to return whatever was on the stack. Found with Coverity Signed-off-by: Dave Jones Reviewed-by: Thomas Hellstrom --- drivers/gpu/drm/vmwgfx/vmwgfx_mob.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_mob.c b/drivers/gpu/drm/vmwgfx/vmwgfx_mob.c index 4910e7b81811..d4a5a19cb8c3 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_mob.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_mob.c @@ -134,6 +134,7 @@ static int vmw_setup_otable_base(struct vmw_private *dev_priv, cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd)); if (unlikely(cmd == NULL)) { DRM_ERROR("Failed reserving FIFO space for OTable setup.\n"); + ret = -ENOMEM; goto out_no_fifo; } From c66f854338253e603a4fb6817069ee23eacf0ae3 Mon Sep 17 00:00:00 2001 From: Thomas Hellstrom Date: Fri, 24 Jan 2014 08:49:45 +0100 Subject: [PATCH 227/596] drm/ttm: Fix TTM object open regression Commit drm/ttm: ttm object security fixes for render nodes introduced a regression where, if a TTM object was opened multiple times from the same open file, the caller would spin uninterruptibly in the kernel. Fix this. Signed-off-by: Thomas Hellstrom Reviewed-by: Jakob Bornecrantz --- drivers/gpu/drm/ttm/ttm_object.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/ttm/ttm_object.c b/drivers/gpu/drm/ttm/ttm_object.c index 37079859afc8..53b51c4e671a 100644 --- a/drivers/gpu/drm/ttm/ttm_object.c +++ b/drivers/gpu/drm/ttm/ttm_object.c @@ -292,7 +292,7 @@ int ttm_ref_object_add(struct ttm_object_file *tfile, if (ret == 0) { ref = drm_hash_entry(hash, struct ttm_ref_object, hash); - if (!kref_get_unless_zero(&ref->kref)) { + if (kref_get_unless_zero(&ref->kref)) { rcu_read_unlock(); break; } From 923fa4ea382f592dee2ba3b205befb90cbddf3af Mon Sep 17 00:00:00 2001 From: Nitin A Kamble Date: Thu, 30 Jan 2014 16:50:10 -0800 Subject: [PATCH 228/596] genirq: Generic irq chip requires IRQ_DOMAIN The generic_chip.c uses interfaces from irq_domain.c which is controlled by the IRQ_DOMAIN config option, but there is no Kconfig dependency so the build can fail: linux/kernel/irq/generic-chip.c:400:11: error: 'irq_domain_xlate_onetwocell' undeclared here (not in a function) Select IRQ_DOMAIN when GENERIC_IRQ_CHIP is selected. Signed-off-by: Nitin A Kamble Link: http://lkml.kernel.org/r/1391129410-54548-2-git-send-email-nitin.a.kamble@intel.com Signed-off-by: Thomas Gleixner Cc: stable@vger.kernel.org # 3.11+ --- kernel/irq/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/kernel/irq/Kconfig b/kernel/irq/Kconfig index 4a1fef09f658..07cbdfea9ae2 100644 --- a/kernel/irq/Kconfig +++ b/kernel/irq/Kconfig @@ -40,6 +40,7 @@ config IRQ_EDGE_EOI_HANDLER # Generic configurable interrupt chip implementation config GENERIC_IRQ_CHIP bool + select IRQ_DOMAIN # Generic irq_domain hw <--> linux irq number translation config IRQ_DOMAIN From 84a34575fe7c87278e4e00f016aa5b624dcb9171 Mon Sep 17 00:00:00 2001 From: "xinhui.pan" Date: Fri, 31 Jan 2014 13:08:01 -0800 Subject: [PATCH 229/596] gpio: intel-mid: fix the incorrect return of idle callback intel_gpio_runtime_idle should return correct error code if it do fail. make it more correct even though -EBUSY is the most possible return value. Signed-off-by: bo.he Signed-off-by: xinhui.pan Signed-off-by: David Cohen Reviewed-by: Felipe Balbi Signed-off-by: Linus Walleij --- drivers/gpio/gpio-intel-mid.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpio/gpio-intel-mid.c b/drivers/gpio/gpio-intel-mid.c index d1b50ef5fab8..e585163f1ad5 100644 --- a/drivers/gpio/gpio-intel-mid.c +++ b/drivers/gpio/gpio-intel-mid.c @@ -394,8 +394,8 @@ static const struct irq_domain_ops intel_gpio_irq_ops = { static int intel_gpio_runtime_idle(struct device *dev) { - pm_schedule_suspend(dev, 500); - return -EBUSY; + int err = pm_schedule_suspend(dev, 500); + return err ?: -EBUSY; } static const struct dev_pm_ops intel_gpio_pm_ops = { From 6f587c9f7bbb50a2dc5ad7bdc43392cf608e0857 Mon Sep 17 00:00:00 2001 From: Markus Mayer Date: Mon, 3 Feb 2014 13:43:00 -0800 Subject: [PATCH 230/596] gpio: bcm281xx: Update MODULE_AUTHOR Add Broadcom's upstreaming mailing list address to MODULE_AUTHOR. Signed-off-by: Markus Mayer Signed-off-by: Linus Walleij --- drivers/gpio/gpio-bcm-kona.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpio/gpio-bcm-kona.c b/drivers/gpio/gpio-bcm-kona.c index 233d088ac59f..f32357e2d78d 100644 --- a/drivers/gpio/gpio-bcm-kona.c +++ b/drivers/gpio/gpio-bcm-kona.c @@ -1,5 +1,5 @@ /* - * Copyright (C) 2012-2013 Broadcom Corporation + * Copyright (C) 2012-2014 Broadcom Corporation * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License as @@ -657,6 +657,6 @@ static struct platform_driver bcm_kona_gpio_driver = { module_platform_driver(bcm_kona_gpio_driver); -MODULE_AUTHOR("Broadcom"); +MODULE_AUTHOR("Broadcom Corporation "); MODULE_DESCRIPTION("Broadcom Kona GPIO Driver"); MODULE_LICENSE("GPL v2"); From 8d7f6690cedb83456edd41c9bd583783f0703bf0 Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Mon, 3 Feb 2014 17:37:15 +0100 Subject: [PATCH 231/596] s390: fix kernel crash due to linkage stack instructions The kernel currently crashes with a low-address-protection exception if a user space process executes an instruction that tries to use the linkage stack. Set the base-ASTE origin and the subspace-ASTE origin of the dispatchable-unit-control-table to point to a dummy ASTE. Set up control register 15 to point to an empty linkage stack with no room left. A user space process with a linkage stack instruction will still crash but with a different exception which is correctly translated to a segmentation fault instead of a kernel oops. Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/head64.S | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/arch/s390/kernel/head64.S b/arch/s390/kernel/head64.S index b9e25ae2579c..d7c00507568a 100644 --- a/arch/s390/kernel/head64.S +++ b/arch/s390/kernel/head64.S @@ -59,7 +59,7 @@ ENTRY(startup_continue) .quad 0 # cr12: tracing off .quad 0 # cr13: home space segment table .quad 0xc0000000 # cr14: machine check handling off - .quad 0 # cr15: linkage stack operations + .quad .Llinkage_stack # cr15: linkage stack operations .Lpcmsk:.quad 0x0000000180000000 .L4malign:.quad 0xffffffffffc00000 .Lscan2g:.quad 0x80000000 + 0x20000 - 8 # 2GB + 128K - 8 @@ -67,12 +67,15 @@ ENTRY(startup_continue) .Lparmaddr: .quad PARMAREA .align 64 -.Lduct: .long 0,0,0,0,.Lduald,0,0,0 +.Lduct: .long 0,.Laste,.Laste,0,.Lduald,0,0,0 .long 0,0,0,0,0,0,0,0 +.Laste: .quad 0,0xffffffffffffffff,0,0,0,0,0,0 .align 128 .Lduald:.rept 8 .long 0x80000000,0,0,0 # invalid access-list entries .endr +.Llinkage_stack: + .long 0,0,0x89000000,0,0,0,0x8a000000,0 ENTRY(_ehead) From 21708c991eaea76080a9700117181c0d467af9fc Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Tue, 4 Feb 2014 17:39:15 +0800 Subject: [PATCH 232/596] gpio: clps711x: Add module alias to support module auto loading commit 55fe14ab872a "GPIO: clps711x: Rewrite driver for using generic GPIO code" allows this driver to be built as a module. Thus add module alias to support module auto loading. Signed-off-by: Axel Lin Signed-off-by: Linus Walleij --- drivers/gpio/gpio-clps711x.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpio/gpio-clps711x.c b/drivers/gpio/gpio-clps711x.c index d3550274b8f7..3c2ba2ad0ada 100644 --- a/drivers/gpio/gpio-clps711x.c +++ b/drivers/gpio/gpio-clps711x.c @@ -97,3 +97,4 @@ module_platform_driver(clps711x_gpio_driver); MODULE_LICENSE("GPL"); MODULE_AUTHOR("Alexander Shiyan "); MODULE_DESCRIPTION("CLPS711X GPIO driver"); +MODULE_ALIAS("platform:clps711x-gpio"); From 90d16a43d50c6557ea8d6b7d3b597025c932ad92 Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Tue, 4 Feb 2014 18:59:41 +0800 Subject: [PATCH 233/596] gpio: tb10x: GPIO_TB10X needs to select GENERIC_IRQ_CHIP gpio-tb10x driver uses generic irq chip APIs (irq_alloc_domain_generic_chips, irq_remove_generic_chip), so it needs to select GENERIC_IRQ_CHIP to avoid build error. Signed-off-by: Axel Lin Signed-off-by: Linus Walleij --- drivers/gpio/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpio/Kconfig b/drivers/gpio/Kconfig index 697338772b64..903f24d28ba0 100644 --- a/drivers/gpio/Kconfig +++ b/drivers/gpio/Kconfig @@ -403,6 +403,7 @@ config GPIO_GRGPIO config GPIO_TB10X bool + select GENERIC_IRQ_CHIP select OF_GPIO comment "I2C GPIO expanders:" From 5044bad43ee573d0b6d90e3ccb7a40c2c7d25eb4 Mon Sep 17 00:00:00 2001 From: Vinayak Kale Date: Wed, 5 Feb 2014 09:34:36 +0000 Subject: [PATCH 234/596] arm64: add DSB after icache flush in __flush_icache_all() Add DSB after icache flush to complete the cache maintenance operation. The function __flush_icache_all() is used only for user space mappings and an ISB is not required because of an exception return before executing user instructions. An exception return would behave like an ISB. Signed-off-by: Vinayak Kale Acked-by: Will Deacon Cc: Signed-off-by: Catalin Marinas --- arch/arm64/include/asm/cacheflush.h | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/arm64/include/asm/cacheflush.h b/arch/arm64/include/asm/cacheflush.h index fea9ee327206..889324981aa4 100644 --- a/arch/arm64/include/asm/cacheflush.h +++ b/arch/arm64/include/asm/cacheflush.h @@ -116,6 +116,7 @@ extern void flush_dcache_page(struct page *); static inline void __flush_icache_all(void) { asm("ic ialluis"); + dsb(); } #define flush_dcache_mmap_lock(mapping) \ From ccc9e244eb1b9654915634827322932cbafd8244 Mon Sep 17 00:00:00 2001 From: Laura Abbott Date: Tue, 4 Feb 2014 23:08:57 +0000 Subject: [PATCH 235/596] arm64: Align CMA sizes to PAGE_SIZE dma_alloc_from_contiguous takes number of pages for a size. Align up the dma size passed in to page size to avoid truncation and allocation failures on sizes less than PAGE_SIZE. Cc: Will Deacon Signed-off-by: Laura Abbott Signed-off-by: Catalin Marinas --- arch/arm64/mm/dma-mapping.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/arm64/mm/dma-mapping.c b/arch/arm64/mm/dma-mapping.c index 45b5ab54c9ee..fbd76785c5db 100644 --- a/arch/arm64/mm/dma-mapping.c +++ b/arch/arm64/mm/dma-mapping.c @@ -45,6 +45,7 @@ static void *arm64_swiotlb_alloc_coherent(struct device *dev, size_t size, if (IS_ENABLED(CONFIG_DMA_CMA)) { struct page *page; + size = PAGE_ALIGN(size); page = dma_alloc_from_contiguous(dev, size >> PAGE_SHIFT, get_order(size)); if (!page) From a55f9929a9b257f84b6cc7b2397379cabd744a22 Mon Sep 17 00:00:00 2001 From: Catalin Marinas Date: Tue, 4 Feb 2014 16:01:31 +0000 Subject: [PATCH 236/596] arm64: Invalidate the TLB when replacing pmd entries during boot With the 64K page size configuration, __create_page_tables in head.S maps enough memory to get started but using 64K pages rather than 512M sections with a single pgd/pud/pmd entry pointing to a pte table. create_mapping() may override the pgd/pud/pmd table entry with a block (section) one if the RAM size is more than 512MB and aligned correctly. For the end of this block to be accessible, the old TLB entry must be invalidated. Cc: Reported-by: Mark Salter Tested-by: Mark Salter Signed-off-by: Catalin Marinas --- arch/arm64/mm/mmu.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) diff --git a/arch/arm64/mm/mmu.c b/arch/arm64/mm/mmu.c index f557ebbe7013..f8dc7e8fce6f 100644 --- a/arch/arm64/mm/mmu.c +++ b/arch/arm64/mm/mmu.c @@ -203,10 +203,18 @@ static void __init alloc_init_pmd(pud_t *pud, unsigned long addr, do { next = pmd_addr_end(addr, end); /* try section mapping first */ - if (((addr | next | phys) & ~SECTION_MASK) == 0) + if (((addr | next | phys) & ~SECTION_MASK) == 0) { + pmd_t old_pmd =*pmd; set_pmd(pmd, __pmd(phys | prot_sect_kernel)); - else + /* + * Check for previous table entries created during + * boot (__create_page_tables) and flush them. + */ + if (!pmd_none(old_pmd)) + flush_tlb_all(); + } else { alloc_init_pte(pmd, addr, next, __phys_to_pfn(phys)); + } phys += next - addr; } while (pmd++, addr = next, addr != end); } From bfb67a5606376bb32cb6f93dc05cda2e8c2038a5 Mon Sep 17 00:00:00 2001 From: Mark Rutland Date: Wed, 5 Feb 2014 10:24:12 +0000 Subject: [PATCH 237/596] arm64: fix typo: s/SERRROR/SERROR/ Somehow SERROR has acquired an additional 'R' in a couple of headers. This patch removes them before they spread further. As neither instance is in use yet, no other sites need to be fixed up. Signed-off-by: Mark Rutland Acked-by: Marc Zyngier Signed-off-by: Catalin Marinas --- arch/arm64/include/asm/esr.h | 2 +- arch/arm64/include/asm/kvm_arm.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/arm64/include/asm/esr.h b/arch/arm64/include/asm/esr.h index 78834123a32e..c4a7f940b387 100644 --- a/arch/arm64/include/asm/esr.h +++ b/arch/arm64/include/asm/esr.h @@ -42,7 +42,7 @@ #define ESR_EL1_EC_SP_ALIGN (0x26) #define ESR_EL1_EC_FP_EXC32 (0x28) #define ESR_EL1_EC_FP_EXC64 (0x2C) -#define ESR_EL1_EC_SERRROR (0x2F) +#define ESR_EL1_EC_SERROR (0x2F) #define ESR_EL1_EC_BREAKPT_EL0 (0x30) #define ESR_EL1_EC_BREAKPT_EL1 (0x31) #define ESR_EL1_EC_SOFTSTP_EL0 (0x32) diff --git a/arch/arm64/include/asm/kvm_arm.h b/arch/arm64/include/asm/kvm_arm.h index c98ef4771c73..0eb398655378 100644 --- a/arch/arm64/include/asm/kvm_arm.h +++ b/arch/arm64/include/asm/kvm_arm.h @@ -231,7 +231,7 @@ #define ESR_EL2_EC_SP_ALIGN (0x26) #define ESR_EL2_EC_FP_EXC32 (0x28) #define ESR_EL2_EC_FP_EXC64 (0x2C) -#define ESR_EL2_EC_SERRROR (0x2F) +#define ESR_EL2_EC_SERROR (0x2F) #define ESR_EL2_EC_BREAKPT (0x30) #define ESR_EL2_EC_BREAKPT_HYP (0x31) #define ESR_EL2_EC_SOFTSTP (0x32) From 883d50a0ed403446437444a495356ce31e1197a3 Mon Sep 17 00:00:00 2001 From: Mark Rutland Date: Wed, 5 Feb 2014 10:24:13 +0000 Subject: [PATCH 238/596] arm64: simplify pgd_alloc Currently pgd_alloc has a redundant NULL check in its return path that can be removed with no ill effects. With that removed it's also possible to return early and eliminate the new_pgd temporary variable. This patch applies said modifications, making the logic of pgd_alloc correspond 1-1 with that of pgd_free. Signed-off-by: Mark Rutland Cc: Will Deacon Signed-off-by: Catalin Marinas --- arch/arm64/mm/pgd.c | 11 ++--------- 1 file changed, 2 insertions(+), 9 deletions(-) diff --git a/arch/arm64/mm/pgd.c b/arch/arm64/mm/pgd.c index 7083cdada657..62c6101df260 100644 --- a/arch/arm64/mm/pgd.c +++ b/arch/arm64/mm/pgd.c @@ -32,17 +32,10 @@ pgd_t *pgd_alloc(struct mm_struct *mm) { - pgd_t *new_pgd; - if (PGD_SIZE == PAGE_SIZE) - new_pgd = (pgd_t *)get_zeroed_page(GFP_KERNEL); + return (pgd_t *)get_zeroed_page(GFP_KERNEL); else - new_pgd = kzalloc(PGD_SIZE, GFP_KERNEL); - - if (!new_pgd) - return NULL; - - return new_pgd; + return kzalloc(PGD_SIZE, GFP_KERNEL); } void pgd_free(struct mm_struct *mm, pgd_t *pgd) From 8fcfb99c8e29c73dd8945b6105ef54ca4eeb171e Mon Sep 17 00:00:00 2001 From: Toshi Kani Date: Tue, 4 Feb 2014 17:48:28 -0700 Subject: [PATCH 239/596] ACPI / hotplug: Fix panic on eject to ejected device When an eject request is sent to an ejected ACPI device, the following panic occurs: ACPI: \_SB_.SCK3.CPU3: ACPI_NOTIFY_EJECT_REQUEST event BUG: unable to handle kernel NULL pointer dereference at 0000000000000070 IP: [] acpi_device_hotplug+0x10b/0x33b : Call Trace: [] acpi_hotplug_work_fn+0x1c/0x27 [] process_one_work+0x175/0x430 [] worker_thread+0x11b/0x3a0 This is becase device->handler is NULL in acpi_device_hotplug(). This case was used to fail in acpi_hotplug_notify_cb() as the target had no acpi_deivce. However, acpi_device now exists after ejection. Added a check to verify if acpi_device->handler is valid for an eject request in acpi_hotplug_notify_cb(). Note that handler passed from an argument is still valid while acpi_device->handler is NULL. Fixes: 202317a573b2 (ACPI / scan: Add acpi_device objects for all device nodes in the namespace) Signed-off-by: Toshi Kani Signed-off-by: Rafael J. Wysocki --- drivers/acpi/scan.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/acpi/scan.c b/drivers/acpi/scan.c index 7384158c7f87..57b053f424d1 100644 --- a/drivers/acpi/scan.c +++ b/drivers/acpi/scan.c @@ -484,7 +484,6 @@ static void acpi_device_hotplug(void *data, u32 src) static void acpi_hotplug_notify_cb(acpi_handle handle, u32 type, void *data) { u32 ost_code = ACPI_OST_SC_NON_SPECIFIC_FAILURE; - struct acpi_scan_handler *handler = data; struct acpi_device *adev; acpi_status status; @@ -500,7 +499,10 @@ static void acpi_hotplug_notify_cb(acpi_handle handle, u32 type, void *data) break; case ACPI_NOTIFY_EJECT_REQUEST: acpi_handle_debug(handle, "ACPI_NOTIFY_EJECT_REQUEST event\n"); - if (!handler->hotplug.enabled) { + if (!adev->handler) + goto err_out; + + if (!adev->handler->hotplug.enabled) { acpi_handle_err(handle, "Eject disabled\n"); ost_code = ACPI_OST_SC_EJECT_NOT_SUPPORTED; goto err_out; From 069b918623e1510e58dacf178905a72c3baa3ae4 Mon Sep 17 00:00:00 2001 From: Nathan Lynch Date: Wed, 5 Feb 2014 05:53:04 +0000 Subject: [PATCH 240/596] arm64: vdso: fix coarse clock handling When __kernel_clock_gettime is called with a CLOCK_MONOTONIC_COARSE or CLOCK_REALTIME_COARSE clock id, it returns incorrectly to whatever the caller has placed in x2 ("ret x2" to return from the fast path). Fix this by saving x30/LR to x2 only in code that will call __do_get_tspec, restoring x30 afterward, and using a plain "ret" to return from the routine. Also: while the resulting tv_nsec value for CLOCK_REALTIME and CLOCK_MONOTONIC must be computed using intermediate values that are left-shifted by cs_shift (x12, set by __do_get_tspec), the results for coarse clocks should be calculated using unshifted values (xtime_coarse_nsec is in units of actual nanoseconds). The current code shifts intermediate values by x12 unconditionally, but x12 is uninitialized when servicing a coarse clock. Fix this by setting x12 to 0 once we know we are dealing with a coarse clock id. Signed-off-by: Nathan Lynch Acked-by: Will Deacon Cc: Signed-off-by: Catalin Marinas --- arch/arm64/kernel/vdso/gettimeofday.S | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/arch/arm64/kernel/vdso/gettimeofday.S b/arch/arm64/kernel/vdso/gettimeofday.S index f0a6d10b5211..fe652ffd34c2 100644 --- a/arch/arm64/kernel/vdso/gettimeofday.S +++ b/arch/arm64/kernel/vdso/gettimeofday.S @@ -103,6 +103,8 @@ ENTRY(__kernel_clock_gettime) bl __do_get_tspec seqcnt_check w9, 1b + mov x30, x2 + cmp w0, #CLOCK_MONOTONIC b.ne 6f @@ -118,6 +120,9 @@ ENTRY(__kernel_clock_gettime) ccmp w0, #CLOCK_MONOTONIC_COARSE, #0x4, ne b.ne 8f + /* xtime_coarse_nsec is already right-shifted */ + mov x12, #0 + /* Get coarse timespec. */ adr vdso_data, _vdso_data 3: seqcnt_acquire @@ -156,7 +161,7 @@ ENTRY(__kernel_clock_gettime) lsr x11, x11, x12 stp x10, x11, [x1, #TSPEC_TV_SEC] mov x0, xzr - ret x2 + ret 7: mov x30, x2 8: /* Syscall fallback. */ From d4022a335271a48cce49df35d825897914fbffe3 Mon Sep 17 00:00:00 2001 From: Nathan Lynch Date: Mon, 3 Feb 2014 19:48:52 +0000 Subject: [PATCH 241/596] arm64: vdso: update wtm fields for CLOCK_MONOTONIC_COARSE Update wall-to-monotonic fields in the VDSO data page unconditionally. These are used to service CLOCK_MONOTONIC_COARSE, which is not guarded by use_syscall. Signed-off-by: Nathan Lynch Acked-by: Will Deacon Cc: Signed-off-by: Catalin Marinas --- arch/arm64/kernel/vdso.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/arm64/kernel/vdso.c b/arch/arm64/kernel/vdso.c index 65d40cf6945a..a7149cae1615 100644 --- a/arch/arm64/kernel/vdso.c +++ b/arch/arm64/kernel/vdso.c @@ -238,6 +238,8 @@ void update_vsyscall(struct timekeeper *tk) vdso_data->use_syscall = use_syscall; vdso_data->xtime_coarse_sec = xtime_coarse.tv_sec; vdso_data->xtime_coarse_nsec = xtime_coarse.tv_nsec; + vdso_data->wtm_clock_sec = tk->wall_to_monotonic.tv_sec; + vdso_data->wtm_clock_nsec = tk->wall_to_monotonic.tv_nsec; if (!use_syscall) { vdso_data->cs_cycle_last = tk->clock->cycle_last; @@ -245,8 +247,6 @@ void update_vsyscall(struct timekeeper *tk) vdso_data->xtime_clock_nsec = tk->xtime_nsec; vdso_data->cs_mult = tk->mult; vdso_data->cs_shift = tk->shift; - vdso_data->wtm_clock_sec = tk->wall_to_monotonic.tv_sec; - vdso_data->wtm_clock_nsec = tk->wall_to_monotonic.tv_nsec; } smp_wmb(); From 6290b53de025dd08a79257ee84173ef7b6d926f7 Mon Sep 17 00:00:00 2001 From: Catalin Marinas Date: Wed, 5 Feb 2014 12:03:52 +0000 Subject: [PATCH 242/596] arm64: compat: Wire up new AArch32 syscalls This patch enables sys_compat, sys_finit_module, sys_sched_setattr and sys_sched_getattr for compat (AArch32) applications. Signed-off-by: Catalin Marinas --- arch/arm64/include/asm/unistd32.h | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/arch/arm64/include/asm/unistd32.h b/arch/arm64/include/asm/unistd32.h index 58125bf008d3..bb8eb8a78e67 100644 --- a/arch/arm64/include/asm/unistd32.h +++ b/arch/arm64/include/asm/unistd32.h @@ -399,7 +399,10 @@ __SYSCALL(374, compat_sys_sendmmsg) __SYSCALL(375, sys_setns) __SYSCALL(376, compat_sys_process_vm_readv) __SYSCALL(377, compat_sys_process_vm_writev) -__SYSCALL(378, sys_ni_syscall) /* 378 for kcmp */ +__SYSCALL(378, sys_kcmp) +__SYSCALL(379, sys_finit_module) +__SYSCALL(380, sys_sched_setattr) +__SYSCALL(381, sys_sched_getattr) #define __NR_compat_syscalls 379 From 2a53bfb3e0fb6aa6b1ac93e5979a040a4b57ea8b Mon Sep 17 00:00:00 2001 From: Arturo Borrero Date: Fri, 17 Jan 2014 02:28:45 +0100 Subject: [PATCH 243/596] netfilter: nft_ct: fix unconditional dump of 'dir' attr We want to make sure that the information that we get from the kernel can be reinjected without troubles. The kernel shouldn't return an attribute that is not required, or even prohibited. Dumping unconditionally NFTA_CT_DIRECTION could lead an application in userspace to interpret that the attribute was originally set, while it was not. Signed-off-by: Arturo Borrero Gonzalez Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nft_ct.c | 15 +++++++++++++-- 1 file changed, 13 insertions(+), 2 deletions(-) diff --git a/net/netfilter/nft_ct.c b/net/netfilter/nft_ct.c index 917052e20602..feaf0f354a93 100644 --- a/net/netfilter/nft_ct.c +++ b/net/netfilter/nft_ct.c @@ -311,8 +311,19 @@ static int nft_ct_get_dump(struct sk_buff *skb, const struct nft_expr *expr) goto nla_put_failure; if (nla_put_be32(skb, NFTA_CT_KEY, htonl(priv->key))) goto nla_put_failure; - if (nla_put_u8(skb, NFTA_CT_DIRECTION, priv->dir)) - goto nla_put_failure; + + switch (priv->key) { + case NFT_CT_PROTOCOL: + case NFT_CT_SRC: + case NFT_CT_DST: + case NFT_CT_PROTO_SRC: + case NFT_CT_PROTO_DST: + if (nla_put_u8(skb, NFTA_CT_DIRECTION, priv->dir)) + goto nla_put_failure; + default: + break; + } + return 0; nla_put_failure: From 3dd7279fb6db05ec5a088cd0cae6ba22580a82bd Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Sat, 25 Jan 2014 08:04:07 +0000 Subject: [PATCH 244/596] netfilter: nf_tables: fix oops when deleting a chain with references The following commands trigger an oops: # nft -i nft> add table filter nft> add chain filter input { type filter hook input priority 0; } nft> add chain filter test nft> add rule filter input jump test nft> delete chain filter test We need to check the chain use counter before allowing destruction since we might have references from sets or jump rules. Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=69341 Reported-by: Matthew Ife Tested-by: Matthew Ife Signed-off-by: Patrick McHardy Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_tables_api.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 117bbaaddde6..9ce30534f853 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -1045,7 +1045,7 @@ static int nf_tables_delchain(struct sock *nlsk, struct sk_buff *skb, if (IS_ERR(chain)) return PTR_ERR(chain); - if (!list_empty(&chain->rules)) + if (!list_empty(&chain->rules) || chain->use > 0) return -EBUSY; list_del(&chain->list); From c6825c0976fa7893692e0e43b09740b419b23c09 Mon Sep 17 00:00:00 2001 From: Andrey Vagin Date: Wed, 29 Jan 2014 19:34:14 +0100 Subject: [PATCH 245/596] netfilter: nf_conntrack: fix RCU race in nf_conntrack_find_get Lets look at destroy_conntrack: hlist_nulls_del_rcu(&ct->tuplehash[IP_CT_DIR_ORIGINAL].hnnode); ... nf_conntrack_free(ct) kmem_cache_free(net->ct.nf_conntrack_cachep, ct); net->ct.nf_conntrack_cachep is created with SLAB_DESTROY_BY_RCU. The hash is protected by rcu, so readers look up conntracks without locks. A conntrack is removed from the hash, but in this moment a few readers still can use the conntrack. Then this conntrack is released and another thread creates conntrack with the same address and the equal tuple. After this a reader starts to validate the conntrack: * It's not dying, because a new conntrack was created * nf_ct_tuple_equal() returns true. But this conntrack is not initialized yet, so it can not be used by two threads concurrently. In this case BUG_ON may be triggered from nf_nat_setup_info(). Florian Westphal suggested to check the confirm bit too. I think it's right. task 1 task 2 task 3 nf_conntrack_find_get ____nf_conntrack_find destroy_conntrack hlist_nulls_del_rcu nf_conntrack_free kmem_cache_free __nf_conntrack_alloc kmem_cache_alloc memset(&ct->tuplehash[IP_CT_DIR_MAX], if (nf_ct_is_dying(ct)) if (!nf_ct_tuple_equal() I'm not sure, that I have ever seen this race condition in a real life. Currently we are investigating a bug, which is reproduced on a few nodes. In our case one conntrack is initialized from a few tasks concurrently, we don't have any other explanation for this. <2>[46267.083061] kernel BUG at net/ipv4/netfilter/nf_nat_core.c:322! ... <4>[46267.083951] RIP: 0010:[] [] nf_nat_setup_info+0x564/0x590 [nf_nat] ... <4>[46267.085549] Call Trace: <4>[46267.085622] [] alloc_null_binding+0x5b/0xa0 [iptable_nat] <4>[46267.085697] [] nf_nat_rule_find+0x5c/0x80 [iptable_nat] <4>[46267.085770] [] nf_nat_fn+0x111/0x260 [iptable_nat] <4>[46267.085843] [] nf_nat_out+0x48/0xd0 [iptable_nat] <4>[46267.085919] [] nf_iterate+0x69/0xb0 <4>[46267.085991] [] ? ip_finish_output+0x0/0x2f0 <4>[46267.086063] [] nf_hook_slow+0x74/0x110 <4>[46267.086133] [] ? ip_finish_output+0x0/0x2f0 <4>[46267.086207] [] ? dst_output+0x0/0x20 <4>[46267.086277] [] ip_output+0xa4/0xc0 <4>[46267.086346] [] raw_sendmsg+0x8b4/0x910 <4>[46267.086419] [] inet_sendmsg+0x4a/0xb0 <4>[46267.086491] [] ? sock_update_classid+0x3a/0x50 <4>[46267.086562] [] sock_sendmsg+0x117/0x140 <4>[46267.086638] [] ? _spin_unlock_bh+0x1b/0x20 <4>[46267.086712] [] ? autoremove_wake_function+0x0/0x40 <4>[46267.086785] [] ? do_ip_setsockopt+0x90/0xd80 <4>[46267.086858] [] ? call_function_interrupt+0xe/0x20 <4>[46267.086936] [] ? ub_slab_ptr+0x20/0x90 <4>[46267.087006] [] ? ub_slab_ptr+0x20/0x90 <4>[46267.087081] [] ? kmem_cache_alloc+0xd8/0x1e0 <4>[46267.087151] [] sys_sendto+0x139/0x190 <4>[46267.087229] [] ? sock_setsockopt+0x16d/0x6f0 <4>[46267.087303] [] ? audit_syscall_entry+0x1d7/0x200 <4>[46267.087378] [] ? __audit_syscall_exit+0x265/0x290 <4>[46267.087454] [] ? compat_sys_setsockopt+0x75/0x210 <4>[46267.087531] [] compat_sys_socketcall+0x13f/0x210 <4>[46267.087607] [] ia32_sysret+0x0/0x5 <4>[46267.087676] Code: 91 20 e2 01 75 29 48 89 de 4c 89 f7 e8 56 fa ff ff 85 c0 0f 84 68 fc ff ff 0f b6 4d c6 41 8b 45 00 e9 4d fb ff ff e8 7c 19 e9 e0 <0f> 0b eb fe f6 05 17 91 20 e2 80 74 ce 80 3d 5f 2e 00 00 00 74 <1>[46267.088023] RIP [] nf_nat_setup_info+0x564/0x590 Cc: Eric Dumazet Cc: Florian Westphal Cc: Pablo Neira Ayuso Cc: Patrick McHardy Cc: Jozsef Kadlecsik Cc: "David S. Miller" Cc: Cyrill Gorcunov Signed-off-by: Andrey Vagin Acked-by: Eric Dumazet Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_conntrack_core.c | 21 +++++++++++++++++---- 1 file changed, 17 insertions(+), 4 deletions(-) diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index 8824ed0ccc9c..4d1fb5d094c3 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -312,6 +312,21 @@ static void death_by_timeout(unsigned long ul_conntrack) nf_ct_delete((struct nf_conn *)ul_conntrack, 0, 0); } +static inline bool +nf_ct_key_equal(struct nf_conntrack_tuple_hash *h, + const struct nf_conntrack_tuple *tuple, + u16 zone) +{ + struct nf_conn *ct = nf_ct_tuplehash_to_ctrack(h); + + /* A conntrack can be recreated with the equal tuple, + * so we need to check that the conntrack is confirmed + */ + return nf_ct_tuple_equal(tuple, &h->tuple) && + nf_ct_zone(ct) == zone && + nf_ct_is_confirmed(ct); +} + /* * Warning : * - Caller must take a reference on returned object @@ -333,8 +348,7 @@ ____nf_conntrack_find(struct net *net, u16 zone, local_bh_disable(); begin: hlist_nulls_for_each_entry_rcu(h, n, &net->ct.hash[bucket], hnnode) { - if (nf_ct_tuple_equal(tuple, &h->tuple) && - nf_ct_zone(nf_ct_tuplehash_to_ctrack(h)) == zone) { + if (nf_ct_key_equal(h, tuple, zone)) { NF_CT_STAT_INC(net, found); local_bh_enable(); return h; @@ -372,8 +386,7 @@ __nf_conntrack_find_get(struct net *net, u16 zone, !atomic_inc_not_zero(&ct->ct_general.use))) h = NULL; else { - if (unlikely(!nf_ct_tuple_equal(tuple, &h->tuple) || - nf_ct_zone(ct) != zone)) { + if (unlikely(!nf_ct_key_equal(h, tuple, zone))) { nf_ct_put(ct); goto begin; } From 530b099dfe8499d639e7fbcad28c4199e2a720c7 Mon Sep 17 00:00:00 2001 From: Colin Cross Date: Tue, 4 Feb 2014 02:15:32 +0000 Subject: [PATCH 246/596] security: select correct default LSM_MMAP_MIN_ADDR on arm on arm64 Binaries compiled for arm may run on arm64 if CONFIG_COMPAT is selected. Set LSM_MMAP_MIN_ADDR to 32768 if ARM64 && COMPAT to prevent selinux failures launching 32-bit static executables that are mapped at 0x8000. Signed-off-by: Colin Cross Acked-by: Will Deacon Acked-by: Eric Paris Acked-by: James Morris Signed-off-by: Catalin Marinas --- security/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/security/Kconfig b/security/Kconfig index e9c6ac724fef..beb86b500adf 100644 --- a/security/Kconfig +++ b/security/Kconfig @@ -103,7 +103,7 @@ config INTEL_TXT config LSM_MMAP_MIN_ADDR int "Low address space for LSM to protect from user allocation" depends on SECURITY && SECURITY_SELINUX - default 32768 if ARM + default 32768 if ARM || (ARM64 && COMPAT) default 65536 help This is the portion of low virtual memory which should be protected From 1b76af5ce8deb1c775ad1674bd249d782b5b13d9 Mon Sep 17 00:00:00 2001 From: Thomas Hellstrom Date: Wed, 5 Feb 2014 09:18:26 +0100 Subject: [PATCH 247/596] drm/ttm: Don't clear page metadata of imported sg pages These page pointers shouldn't be visible to TTM in the first place, but until we fix that up, don't clear the page metadata because that will upset the exporter. Reported-and-tested-by: Cristoph Haag Signed-off-by: Thomas Hellstrom Reviewed-by: Jakob Bornecrantz --- drivers/gpu/drm/ttm/ttm_tt.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/gpu/drm/ttm/ttm_tt.c b/drivers/gpu/drm/ttm/ttm_tt.c index 9af99084b344..75f319090043 100644 --- a/drivers/gpu/drm/ttm/ttm_tt.c +++ b/drivers/gpu/drm/ttm/ttm_tt.c @@ -380,6 +380,9 @@ static void ttm_tt_clear_mapping(struct ttm_tt *ttm) pgoff_t i; struct page **page = ttm->pages; + if (ttm->page_flags & TTM_PAGE_FLAG_SG) + return; + for (i = 0; i < ttm->num_pages; ++i) { (*page)->mapping = NULL; (*page++)->index = 0; From 662372e42e46d9bbfcb83e1cce81f6b33cebaddd Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Mon, 3 Feb 2014 08:53:44 -0600 Subject: [PATCH 248/596] of: restructure for_each macros to fix compile warnings Commit 00b2c76a6a "include/linux/of.h: make for_each_child_of_node() reference its args when CONFIG_OF=n" fixed warnings for unused variables, but introduced variable "used uninitialized" warnings. Simply initializing the variables would result in "set but not used" warnings with W=1. Fix both types of warnings by making all the for_each macros unconditional and rely on the dummy static inline functions to initialize and reference any variables. Reported-by: Geert Uytterhoeven Cc: David Howells Signed-off-by: Rob Herring Acked-by: Grant Likely --- include/linux/of.h | 163 +++++++++++++++++++++++++-------------------- 1 file changed, 89 insertions(+), 74 deletions(-) diff --git a/include/linux/of.h b/include/linux/of.h index 70c64ba17fa5..435cb995904d 100644 --- a/include/linux/of.h +++ b/include/linux/of.h @@ -169,35 +169,15 @@ static inline const char *of_node_full_name(const struct device_node *np) extern struct device_node *of_find_node_by_name(struct device_node *from, const char *name); -#define for_each_node_by_name(dn, name) \ - for (dn = of_find_node_by_name(NULL, name); dn; \ - dn = of_find_node_by_name(dn, name)) extern struct device_node *of_find_node_by_type(struct device_node *from, const char *type); -#define for_each_node_by_type(dn, type) \ - for (dn = of_find_node_by_type(NULL, type); dn; \ - dn = of_find_node_by_type(dn, type)) extern struct device_node *of_find_compatible_node(struct device_node *from, const char *type, const char *compat); -#define for_each_compatible_node(dn, type, compatible) \ - for (dn = of_find_compatible_node(NULL, type, compatible); dn; \ - dn = of_find_compatible_node(dn, type, compatible)) extern struct device_node *of_find_matching_node_and_match( struct device_node *from, const struct of_device_id *matches, const struct of_device_id **match); -static inline struct device_node *of_find_matching_node( - struct device_node *from, - const struct of_device_id *matches) -{ - return of_find_matching_node_and_match(from, matches, NULL); -} -#define for_each_matching_node(dn, matches) \ - for (dn = of_find_matching_node(NULL, matches); dn; \ - dn = of_find_matching_node(dn, matches)) -#define for_each_matching_node_and_match(dn, matches, match) \ - for (dn = of_find_matching_node_and_match(NULL, matches, match); \ - dn; dn = of_find_matching_node_and_match(dn, matches, match)) + extern struct device_node *of_find_node_by_path(const char *path); extern struct device_node *of_find_node_by_phandle(phandle handle); extern struct device_node *of_get_parent(const struct device_node *node); @@ -209,43 +189,11 @@ extern struct device_node *of_get_next_available_child( extern struct device_node *of_get_child_by_name(const struct device_node *node, const char *name); -#define for_each_child_of_node(parent, child) \ - for (child = of_get_next_child(parent, NULL); child != NULL; \ - child = of_get_next_child(parent, child)) - -#define for_each_available_child_of_node(parent, child) \ - for (child = of_get_next_available_child(parent, NULL); child != NULL; \ - child = of_get_next_available_child(parent, child)) - -static inline int of_get_child_count(const struct device_node *np) -{ - struct device_node *child; - int num = 0; - - for_each_child_of_node(np, child) - num++; - - return num; -} - -static inline int of_get_available_child_count(const struct device_node *np) -{ - struct device_node *child; - int num = 0; - - for_each_available_child_of_node(np, child) - num++; - - return num; -} /* cache lookup */ extern struct device_node *of_find_next_cache_node(const struct device_node *); extern struct device_node *of_find_node_with_property( struct device_node *from, const char *prop_name); -#define for_each_node_with_property(dn, prop_name) \ - for (dn = of_find_node_with_property(NULL, prop_name); dn; \ - dn = of_find_node_with_property(dn, prop_name)) extern struct property *of_find_property(const struct device_node *np, const char *name, @@ -367,27 +315,48 @@ static inline struct device_node *of_find_node_by_name(struct device_node *from, return NULL; } +static inline struct device_node *of_find_node_by_type(struct device_node *from, + const char *type) +{ + return NULL; +} + +static inline struct device_node *of_find_matching_node_and_match( + struct device_node *from, + const struct of_device_id *matches, + const struct of_device_id **match) +{ + return NULL; +} + static inline struct device_node *of_get_parent(const struct device_node *node) { return NULL; } +static inline struct device_node *of_get_next_child( + const struct device_node *node, struct device_node *prev) +{ + return NULL; +} + +static inline struct device_node *of_get_next_available_child( + const struct device_node *node, struct device_node *prev) +{ + return NULL; +} + +static inline struct device_node *of_find_node_with_property( + struct device_node *from, const char *prop_name) +{ + return NULL; +} + static inline bool of_have_populated_dt(void) { return false; } -/* Kill an unused variable warning on a device_node pointer */ -static inline void __of_use_dn(const struct device_node *np) -{ -} - -#define for_each_child_of_node(parent, child) \ - while (__of_use_dn(parent), __of_use_dn(child), 0) - -#define for_each_available_child_of_node(parent, child) \ - while (0) - static inline struct device_node *of_get_child_by_name( const struct device_node *node, const char *name) @@ -395,16 +364,6 @@ static inline struct device_node *of_get_child_by_name( return NULL; } -static inline int of_get_child_count(const struct device_node *np) -{ - return 0; -} - -static inline int of_get_available_child_count(const struct device_node *np) -{ - return 0; -} - static inline int of_device_is_compatible(const struct device_node *device, const char *name) { @@ -569,6 +528,13 @@ extern int of_node_to_nid(struct device_node *np); static inline int of_node_to_nid(struct device_node *device) { return 0; } #endif +static inline struct device_node *of_find_matching_node( + struct device_node *from, + const struct of_device_id *matches) +{ + return of_find_matching_node_and_match(from, matches, NULL); +} + /** * of_property_read_bool - Findfrom a property * @np: device node from which the property value is to be read. @@ -618,6 +584,55 @@ static inline int of_property_read_u32(const struct device_node *np, s; \ s = of_prop_next_string(prop, s)) +#define for_each_node_by_name(dn, name) \ + for (dn = of_find_node_by_name(NULL, name); dn; \ + dn = of_find_node_by_name(dn, name)) +#define for_each_node_by_type(dn, type) \ + for (dn = of_find_node_by_type(NULL, type); dn; \ + dn = of_find_node_by_type(dn, type)) +#define for_each_compatible_node(dn, type, compatible) \ + for (dn = of_find_compatible_node(NULL, type, compatible); dn; \ + dn = of_find_compatible_node(dn, type, compatible)) +#define for_each_matching_node(dn, matches) \ + for (dn = of_find_matching_node(NULL, matches); dn; \ + dn = of_find_matching_node(dn, matches)) +#define for_each_matching_node_and_match(dn, matches, match) \ + for (dn = of_find_matching_node_and_match(NULL, matches, match); \ + dn; dn = of_find_matching_node_and_match(dn, matches, match)) + +#define for_each_child_of_node(parent, child) \ + for (child = of_get_next_child(parent, NULL); child != NULL; \ + child = of_get_next_child(parent, child)) +#define for_each_available_child_of_node(parent, child) \ + for (child = of_get_next_available_child(parent, NULL); child != NULL; \ + child = of_get_next_available_child(parent, child)) + +#define for_each_node_with_property(dn, prop_name) \ + for (dn = of_find_node_with_property(NULL, prop_name); dn; \ + dn = of_find_node_with_property(dn, prop_name)) + +static inline int of_get_child_count(const struct device_node *np) +{ + struct device_node *child; + int num = 0; + + for_each_child_of_node(np, child) + num++; + + return num; +} + +static inline int of_get_available_child_count(const struct device_node *np) +{ + struct device_node *child; + int num = 0; + + for_each_available_child_of_node(np, child) + num++; + + return num; +} + #if defined(CONFIG_PROC_FS) && defined(CONFIG_PROC_DEVICETREE) extern void proc_device_tree_add_node(struct device_node *, struct proc_dir_entry *); extern void proc_device_tree_add_prop(struct proc_dir_entry *pde, struct property *prop); From 14e2abb732e485ee57d9d5b2cb8884652238e5c1 Mon Sep 17 00:00:00 2001 From: Kleber Sacilotto de Souza Date: Mon, 3 Feb 2014 13:31:03 -0200 Subject: [PATCH 249/596] of: fix PCI bus match for PCIe slots On IBM pseries systems the device_type device-tree property of a PCIe bridge contains the string "pciex". The of_bus_pci_match() function was looking only for "pci" on this property, so in such cases the bus matching code was falling back to the default bus, causing problems on functions that should be using "assigned-addresses" for region address translation. This patch fixes the problem by also looking for "pciex" on the PCI bus match function. v2: added comment Signed-off-by: Kleber Sacilotto de Souza Acked-by: Grant Likely Cc: stable@vger.kernel.org Signed-off-by: Rob Herring --- drivers/of/address.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/of/address.c b/drivers/of/address.c index d3dd41c840f1..1a54f1ffaadb 100644 --- a/drivers/of/address.c +++ b/drivers/of/address.c @@ -99,11 +99,12 @@ static unsigned int of_bus_default_get_flags(const __be32 *addr) static int of_bus_pci_match(struct device_node *np) { /* + * "pciex" is PCI Express * "vci" is for the /chaos bridge on 1st-gen PCI powermacs * "ht" is hypertransport */ - return !strcmp(np->type, "pci") || !strcmp(np->type, "vci") || - !strcmp(np->type, "ht"); + return !strcmp(np->type, "pci") || !strcmp(np->type, "pciex") || + !strcmp(np->type, "vci") || !strcmp(np->type, "ht"); } static void of_bus_pci_count_cells(struct device_node *np, From e066755163aaa4ae880a82d586e958051cf97c11 Mon Sep 17 00:00:00 2001 From: Heiko Stuebner Date: Thu, 30 Jan 2014 15:18:46 +0100 Subject: [PATCH 250/596] dt-bindings: add vendor-prefix for neonode Neonode Inc. is the Manufacturer of the zforce infraread touchscreens used in a lot of ebook readers and supported by the zforce_ts driver. Signed-off-by: Heiko Stuebner Signed-off-by: Rob Herring --- Documentation/devicetree/bindings/vendor-prefixes.txt | 1 + 1 file changed, 1 insertion(+) diff --git a/Documentation/devicetree/bindings/vendor-prefixes.txt b/Documentation/devicetree/bindings/vendor-prefixes.txt index ba43b521218f..0f6ff6109c92 100644 --- a/Documentation/devicetree/bindings/vendor-prefixes.txt +++ b/Documentation/devicetree/bindings/vendor-prefixes.txt @@ -57,6 +57,7 @@ maxim Maxim Integrated Products microchip Microchip Technology Inc. mosaixtech Mosaix Technologies, Inc. national National Semiconductor +neonode Neonode Inc. nintendo Nintendo nvidia NVIDIA nxp NXP Semiconductors From 1db73ae39a97797b7d929eca2af469f82fbb337a Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Mon, 3 Feb 2014 09:27:40 +0100 Subject: [PATCH 251/596] of/device: Nullify match table in of_match_device() for CONFIG_OF=n If the of_device_id table inside a device driver is protected by #ifdef CONFIG_OF, the driver still has to provide a dummy declaration of the table, or wrap it inside of_match_ptr(), when calling of_match_device() in the CONFIG_OF=n case, else the driver fails to compile with e.g. drivers/spi/spi-rspi.c: In function 'rspi_probe': drivers/spi/spi-rspi.c:1203:26: error: 'rspi_of_match' undeclared (first use in this function) drivers/spi/spi-rspi.c:1203:26: note: each undeclared identifier is reported only once for each function it appears in Make of_match_device() nullify the table pointer if CONFIG_OF=n to fix this. Reported-by: Yoshihiro Shimoda Signed-off-by: Geert Uytterhoeven Signed-off-by: Rob Herring --- include/linux/of_device.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/include/linux/of_device.h b/include/linux/of_device.h index 8d7dd6768cb7..ef370210ffb2 100644 --- a/include/linux/of_device.h +++ b/include/linux/of_device.h @@ -78,11 +78,13 @@ static inline int of_device_uevent_modalias(struct device *dev, static inline void of_device_node_put(struct device *dev) { } -static inline const struct of_device_id *of_match_device( +static inline const struct of_device_id *__of_match_device( const struct of_device_id *matches, const struct device *dev) { return NULL; } +#define of_match_device(matches, dev) \ + __of_match_device(of_match_ptr(matches), (dev)) static inline struct device_node *of_cpu_device_node_get(int cpu) { From 7896052d90d7de899fd3a34c0b34e91941e2f2ea Mon Sep 17 00:00:00 2001 From: Rob Clark Date: Fri, 13 Dec 2013 09:11:01 -0500 Subject: [PATCH 252/596] drm/msm: fix inconsequential typo Small typo I noticed in the mdp4_plane code.. no consequence because PIPE_SRC_XY and PIPE_DST_XY have same register layout. Signed-off-by: Rob Clark --- drivers/gpu/drm/msm/mdp/mdp4/mdp4_plane.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_plane.c b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_plane.c index 2406027200ec..1e893dd13859 100644 --- a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_plane.c +++ b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_plane.c @@ -170,8 +170,8 @@ int mdp4_plane_mode_set(struct drm_plane *plane, MDP4_PIPE_DST_SIZE_HEIGHT(crtc_h)); mdp4_write(mdp4_kms, REG_MDP4_PIPE_DST_XY(pipe), - MDP4_PIPE_SRC_XY_X(crtc_x) | - MDP4_PIPE_SRC_XY_Y(crtc_y)); + MDP4_PIPE_DST_XY_X(crtc_x) | + MDP4_PIPE_DST_XY_Y(crtc_y)); mdp4_plane_set_scanout(plane, fb); From 37033a7689b01d0c46f9cc450bdf9f02d86b7e57 Mon Sep 17 00:00:00 2001 From: Rob Clark Date: Fri, 13 Dec 2013 10:38:54 -0500 Subject: [PATCH 253/596] drm/msm/mdp5: fix ref leaks in error paths Signed-off-by: Rob Clark --- drivers/gpu/drm/msm/mdp/mdp5/mdp5_crtc.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_crtc.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_crtc.c index 71a3b2345eb3..f2794021f086 100644 --- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_crtc.c +++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_crtc.c @@ -296,6 +296,7 @@ static int mdp5_crtc_mode_set(struct drm_crtc *crtc, x << 16, y << 16, mode->hdisplay << 16, mode->vdisplay << 16); if (ret) { + drm_framebuffer_unreference(crtc->fb); dev_err(crtc->dev->dev, "%s: failed to set mode on plane: %d\n", mdp5_crtc->name, ret); return ret; @@ -343,11 +344,15 @@ static int mdp5_crtc_mode_set_base(struct drm_crtc *crtc, int x, int y, 0, 0, mode->hdisplay, mode->vdisplay, x << 16, y << 16, mode->hdisplay << 16, mode->vdisplay << 16); + if (ret) { + drm_framebuffer_unreference(crtc->fb); + return ret; + } update_fb(crtc, crtc->fb); update_scanout(crtc, crtc->fb); - return ret; + return 0; } static void mdp5_crtc_load_lut(struct drm_crtc *crtc) From b69720c0f5d417310fbfd59c2d681bd90430a4f5 Mon Sep 17 00:00:00 2001 From: Rob Clark Date: Fri, 13 Dec 2013 10:41:07 -0500 Subject: [PATCH 254/596] drm/msm/mdp4: pageflip fixes Backport a few fixes found in the course of getting mdp5 working. There is a window of time after pageflip is requested, before we start scanning out the new fb (ie. while we are waiting for gpu). During that time we need to continue holding a reference to the still-current scanout fb, to avoid the backing gem bo's from being destroyed. Possibly a common mdp_crtc parent class could be useful to share some of this logic between mdp4_crtc and mdp5_crtc. OTOH, this all can be removed from the driver once atomic is in place, as plane/crtc updates get deferred until all fb's are ready before calling in to .page_flip(), etc. Signed-off-by: Rob Clark --- drivers/gpu/drm/msm/mdp/mdp4/mdp4_crtc.c | 157 +++++++++++++++-------- 1 file changed, 100 insertions(+), 57 deletions(-) diff --git a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_crtc.c b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_crtc.c index 1964f4f0d452..ed739e887c25 100644 --- a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_crtc.c +++ b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_crtc.c @@ -57,9 +57,16 @@ struct mdp4_crtc { #define PENDING_FLIP 0x2 atomic_t pending; - /* the fb that we currently hold a scanout ref to: */ + /* the fb that we logically (from PoV of KMS API) hold a ref + * to. Which we may not yet be scanning out (we may still + * be scanning out previous in case of page_flip while waiting + * for gpu rendering to complete: + */ struct drm_framebuffer *fb; + /* the fb that we currently hold a scanout ref to: */ + struct drm_framebuffer *scanout_fb; + /* for unref'ing framebuffers after scanout completes: */ struct drm_flip_work unref_fb_work; @@ -77,24 +84,73 @@ static struct mdp4_kms *get_kms(struct drm_crtc *crtc) return to_mdp4_kms(to_mdp_kms(priv->kms)); } -static void update_fb(struct drm_crtc *crtc, bool async, - struct drm_framebuffer *new_fb) +static void request_pending(struct drm_crtc *crtc, uint32_t pending) +{ + struct mdp4_crtc *mdp4_crtc = to_mdp4_crtc(crtc); + + atomic_or(pending, &mdp4_crtc->pending); + mdp_irq_register(&get_kms(crtc)->base, &mdp4_crtc->vblank); +} + +static void crtc_flush(struct drm_crtc *crtc) +{ + struct mdp4_crtc *mdp4_crtc = to_mdp4_crtc(crtc); + struct mdp4_kms *mdp4_kms = get_kms(crtc); + uint32_t i, flush = 0; + + for (i = 0; i < ARRAY_SIZE(mdp4_crtc->planes); i++) { + struct drm_plane *plane = mdp4_crtc->planes[i]; + if (plane) { + enum mdp4_pipe pipe_id = mdp4_plane_pipe(plane); + flush |= pipe2flush(pipe_id); + } + } + flush |= ovlp2flush(mdp4_crtc->ovlp); + + DBG("%s: flush=%08x", mdp4_crtc->name, flush); + + mdp4_write(mdp4_kms, REG_MDP4_OVERLAY_FLUSH, flush); +} + +static void update_fb(struct drm_crtc *crtc, struct drm_framebuffer *new_fb) { struct mdp4_crtc *mdp4_crtc = to_mdp4_crtc(crtc); struct drm_framebuffer *old_fb = mdp4_crtc->fb; - if (old_fb) - drm_flip_work_queue(&mdp4_crtc->unref_fb_work, old_fb); - /* grab reference to incoming scanout fb: */ drm_framebuffer_reference(new_fb); mdp4_crtc->base.fb = new_fb; mdp4_crtc->fb = new_fb; - if (!async) { - /* enable vblank to pick up the old_fb */ - mdp_irq_register(&get_kms(crtc)->base, &mdp4_crtc->vblank); - } + if (old_fb) + drm_flip_work_queue(&mdp4_crtc->unref_fb_work, old_fb); +} + +/* unlike update_fb(), take a ref to the new scanout fb *before* updating + * plane, then call this. Needed to ensure we don't unref the buffer that + * is actually still being scanned out. + * + * Note that this whole thing goes away with atomic.. since we can defer + * calling into driver until rendering is done. + */ +static void update_scanout(struct drm_crtc *crtc, struct drm_framebuffer *fb) +{ + struct mdp4_crtc *mdp4_crtc = to_mdp4_crtc(crtc); + + /* flush updates, to make sure hw is updated to new scanout fb, + * so that we can safely queue unref to current fb (ie. next + * vblank we know hw is done w/ previous scanout_fb). + */ + crtc_flush(crtc); + + if (mdp4_crtc->scanout_fb) + drm_flip_work_queue(&mdp4_crtc->unref_fb_work, + mdp4_crtc->scanout_fb); + + mdp4_crtc->scanout_fb = fb; + + /* enable vblank to complete flip: */ + request_pending(crtc, PENDING_FLIP); } /* if file!=NULL, this is preclose potential cancel-flip path */ @@ -120,34 +176,6 @@ static void complete_flip(struct drm_crtc *crtc, struct drm_file *file) spin_unlock_irqrestore(&dev->event_lock, flags); } -static void crtc_flush(struct drm_crtc *crtc) -{ - struct mdp4_crtc *mdp4_crtc = to_mdp4_crtc(crtc); - struct mdp4_kms *mdp4_kms = get_kms(crtc); - uint32_t i, flush = 0; - - for (i = 0; i < ARRAY_SIZE(mdp4_crtc->planes); i++) { - struct drm_plane *plane = mdp4_crtc->planes[i]; - if (plane) { - enum mdp4_pipe pipe_id = mdp4_plane_pipe(plane); - flush |= pipe2flush(pipe_id); - } - } - flush |= ovlp2flush(mdp4_crtc->ovlp); - - DBG("%s: flush=%08x", mdp4_crtc->name, flush); - - mdp4_write(mdp4_kms, REG_MDP4_OVERLAY_FLUSH, flush); -} - -static void request_pending(struct drm_crtc *crtc, uint32_t pending) -{ - struct mdp4_crtc *mdp4_crtc = to_mdp4_crtc(crtc); - - atomic_or(pending, &mdp4_crtc->pending); - mdp_irq_register(&get_kms(crtc)->base, &mdp4_crtc->vblank); -} - static void pageflip_cb(struct msm_fence_cb *cb) { struct mdp4_crtc *mdp4_crtc = @@ -158,11 +186,9 @@ static void pageflip_cb(struct msm_fence_cb *cb) if (!fb) return; + drm_framebuffer_reference(fb); mdp4_plane_set_scanout(mdp4_crtc->plane, fb); - crtc_flush(crtc); - - /* enable vblank to complete flip: */ - request_pending(crtc, PENDING_FLIP); + update_scanout(crtc, fb); } static void unref_fb_worker(struct drm_flip_work *work, void *val) @@ -320,6 +346,20 @@ static int mdp4_crtc_mode_set(struct drm_crtc *crtc, mode->vsync_end, mode->vtotal, mode->type, mode->flags); + /* grab extra ref for update_scanout() */ + drm_framebuffer_reference(crtc->fb); + + ret = mdp4_plane_mode_set(mdp4_crtc->plane, crtc, crtc->fb, + 0, 0, mode->hdisplay, mode->vdisplay, + x << 16, y << 16, + mode->hdisplay << 16, mode->vdisplay << 16); + if (ret) { + drm_framebuffer_unreference(crtc->fb); + dev_err(crtc->dev->dev, "%s: failed to set mode on plane: %d\n", + mdp4_crtc->name, ret); + return ret; + } + mdp4_write(mdp4_kms, REG_MDP4_DMA_SRC_SIZE(dma), MDP4_DMA_SRC_SIZE_WIDTH(mode->hdisplay) | MDP4_DMA_SRC_SIZE_HEIGHT(mode->vdisplay)); @@ -341,24 +381,15 @@ static int mdp4_crtc_mode_set(struct drm_crtc *crtc, mdp4_write(mdp4_kms, REG_MDP4_OVLP_CFG(ovlp), 1); - update_fb(crtc, false, crtc->fb); - - ret = mdp4_plane_mode_set(mdp4_crtc->plane, crtc, crtc->fb, - 0, 0, mode->hdisplay, mode->vdisplay, - x << 16, y << 16, - mode->hdisplay << 16, mode->vdisplay << 16); - if (ret) { - dev_err(crtc->dev->dev, "%s: failed to set mode on plane: %d\n", - mdp4_crtc->name, ret); - return ret; - } - if (dma == DMA_E) { mdp4_write(mdp4_kms, REG_MDP4_DMA_E_QUANT(0), 0x00ff0000); mdp4_write(mdp4_kms, REG_MDP4_DMA_E_QUANT(1), 0x00ff0000); mdp4_write(mdp4_kms, REG_MDP4_DMA_E_QUANT(2), 0x00ff0000); } + update_fb(crtc, crtc->fb); + update_scanout(crtc, crtc->fb); + return 0; } @@ -385,13 +416,24 @@ static int mdp4_crtc_mode_set_base(struct drm_crtc *crtc, int x, int y, struct mdp4_crtc *mdp4_crtc = to_mdp4_crtc(crtc); struct drm_plane *plane = mdp4_crtc->plane; struct drm_display_mode *mode = &crtc->mode; + int ret; - update_fb(crtc, false, crtc->fb); + /* grab extra ref for update_scanout() */ + drm_framebuffer_reference(crtc->fb); - return mdp4_plane_mode_set(plane, crtc, crtc->fb, + ret = mdp4_plane_mode_set(plane, crtc, crtc->fb, 0, 0, mode->hdisplay, mode->vdisplay, x << 16, y << 16, mode->hdisplay << 16, mode->vdisplay << 16); + if (ret) { + drm_framebuffer_unreference(crtc->fb); + return ret; + } + + update_fb(crtc, crtc->fb); + update_scanout(crtc, crtc->fb); + + return 0; } static void mdp4_crtc_load_lut(struct drm_crtc *crtc) @@ -419,7 +461,7 @@ static int mdp4_crtc_page_flip(struct drm_crtc *crtc, mdp4_crtc->event = event; spin_unlock_irqrestore(&dev->event_lock, flags); - update_fb(crtc, true, new_fb); + update_fb(crtc, new_fb); return msm_gem_queue_inactive_cb(obj, &mdp4_crtc->pageflip_cb); } @@ -713,6 +755,7 @@ struct drm_crtc *mdp4_crtc_init(struct drm_device *dev, crtc = &mdp4_crtc->base; mdp4_crtc->plane = plane; + mdp4_crtc->id = id; mdp4_crtc->ovlp = ovlp_id; mdp4_crtc->dma = dma_id; From aa1b0e59d3096f4602afeb114a3b3be505889a26 Mon Sep 17 00:00:00 2001 From: Rob Clark Date: Fri, 13 Dec 2013 10:49:25 -0500 Subject: [PATCH 255/596] drm/msm/mdp4: cursor fixes It seems we need to update all cursor registers from vblank. This appears to be the cause of intermittent underflows when enabling/ disabling cursor. Signed-off-by: Rob Clark --- drivers/gpu/drm/msm/mdp/mdp4/mdp4_crtc.c | 22 ++++++++++++++++------ 1 file changed, 16 insertions(+), 6 deletions(-) diff --git a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_crtc.c b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_crtc.c index ed739e887c25..84c5b13b33c9 100644 --- a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_crtc.c +++ b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_crtc.c @@ -39,6 +39,7 @@ struct mdp4_crtc { spinlock_t lock; bool stale; uint32_t width, height; + uint32_t x, y; /* next cursor to scan-out: */ uint32_t next_iova; @@ -484,12 +485,12 @@ static int mdp4_crtc_set_property(struct drm_crtc *crtc, static void update_cursor(struct drm_crtc *crtc) { struct mdp4_crtc *mdp4_crtc = to_mdp4_crtc(crtc); + struct mdp4_kms *mdp4_kms = get_kms(crtc); enum mdp4_dma dma = mdp4_crtc->dma; unsigned long flags; spin_lock_irqsave(&mdp4_crtc->cursor.lock, flags); if (mdp4_crtc->cursor.stale) { - struct mdp4_kms *mdp4_kms = get_kms(crtc); struct drm_gem_object *next_bo = mdp4_crtc->cursor.next_bo; struct drm_gem_object *prev_bo = mdp4_crtc->cursor.scanout_bo; uint32_t iova = mdp4_crtc->cursor.next_iova; @@ -521,6 +522,11 @@ static void update_cursor(struct drm_crtc *crtc) mdp4_crtc->cursor.scanout_bo = next_bo; mdp4_crtc->cursor.stale = false; } + + mdp4_write(mdp4_kms, REG_MDP4_DMA_CURSOR_POS(dma), + MDP4_DMA_CURSOR_POS_X(mdp4_crtc->cursor.x) | + MDP4_DMA_CURSOR_POS_Y(mdp4_crtc->cursor.y)); + spin_unlock_irqrestore(&mdp4_crtc->cursor.lock, flags); } @@ -572,6 +578,7 @@ static int mdp4_crtc_cursor_set(struct drm_crtc *crtc, drm_gem_object_unreference_unlocked(old_bo); } + crtc_flush(crtc); request_pending(crtc, PENDING_CURSOR); return 0; @@ -584,12 +591,15 @@ static int mdp4_crtc_cursor_set(struct drm_crtc *crtc, static int mdp4_crtc_cursor_move(struct drm_crtc *crtc, int x, int y) { struct mdp4_crtc *mdp4_crtc = to_mdp4_crtc(crtc); - struct mdp4_kms *mdp4_kms = get_kms(crtc); - enum mdp4_dma dma = mdp4_crtc->dma; + unsigned long flags; - mdp4_write(mdp4_kms, REG_MDP4_DMA_CURSOR_POS(dma), - MDP4_DMA_CURSOR_POS_X(x) | - MDP4_DMA_CURSOR_POS_Y(y)); + spin_lock_irqsave(&mdp4_crtc->cursor.lock, flags); + mdp4_crtc->cursor.x = x; + mdp4_crtc->cursor.y = y; + spin_unlock_irqrestore(&mdp4_crtc->cursor.lock, flags); + + crtc_flush(crtc); + request_pending(crtc, PENDING_CURSOR); return 0; } From 9999f105e76977d97304d7cb2030a52d96800d69 Mon Sep 17 00:00:00 2001 From: Rob Clark Date: Tue, 4 Feb 2014 14:17:32 -0500 Subject: [PATCH 256/596] drm/msm: fix deadlock in bo create fail path We already hold struct_mutex here. Signed-off-by: Rob Clark --- drivers/gpu/drm/msm/msm_gem.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/msm/msm_gem.c b/drivers/gpu/drm/msm/msm_gem.c index d8d60c969ac7..3da8264d3039 100644 --- a/drivers/gpu/drm/msm/msm_gem.c +++ b/drivers/gpu/drm/msm/msm_gem.c @@ -644,7 +644,7 @@ struct drm_gem_object *msm_gem_new(struct drm_device *dev, fail: if (obj) - drm_gem_object_unreference_unlocked(obj); + drm_gem_object_unreference(obj); return ERR_PTR(ret); } From 860a445c25aa2f99aa5881603a1f4ed2cec64025 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Tue, 4 Feb 2014 22:59:09 +0100 Subject: [PATCH 257/596] DT: Add vendor prefix for Spansion Inc. This is already in active use in several DTS files. Signed-off-by: Geert Uytterhoeven Signed-off-by: Rob Herring --- Documentation/devicetree/bindings/vendor-prefixes.txt | 1 + 1 file changed, 1 insertion(+) diff --git a/Documentation/devicetree/bindings/vendor-prefixes.txt b/Documentation/devicetree/bindings/vendor-prefixes.txt index 0f6ff6109c92..40ce2df0e0e9 100644 --- a/Documentation/devicetree/bindings/vendor-prefixes.txt +++ b/Documentation/devicetree/bindings/vendor-prefixes.txt @@ -81,6 +81,7 @@ silabs Silicon Laboratories simtek sirf SiRF Technology, Inc. snps Synopsys, Inc. +spansion Spansion Inc. st STMicroelectronics ste ST-Ericsson stericsson ST-Ericsson From 829d9315c46a2be57a8fb40c89aeb7db61513d96 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Mon, 3 Feb 2014 13:07:24 +0100 Subject: [PATCH 258/596] netfilter: nf_nat_h323: fix crash in nf_ct_unlink_expect_report() Similar bug fixed in SIP module in 3f509c6 ("netfilter: nf_nat_sip: fix incorrect handling of EBUSY for RTCP expectation"). BUG: unable to handle kernel paging request at 00100104 IP: [] nf_ct_unlink_expect_report+0x57/0xf0 [nf_conntrack] ... Call Trace: [] ? del_timer+0x48/0x70 [] nf_ct_remove_expectations+0x47/0x60 [nf_conntrack] [] nf_ct_delete_from_lists+0x59/0x90 [nf_conntrack] [] death_by_timeout+0x14e/0x1c0 [nf_conntrack] [] ? nf_conntrack_set_hashsize+0x190/0x190 [nf_conntrack] [] call_timer_fn+0x1d/0x80 [] run_timer_softirq+0x18e/0x1a0 [] ? nf_conntrack_set_hashsize+0x190/0x190 [nf_conntrack] [] __do_softirq+0xa3/0x170 [] ? __local_bh_enable+0x70/0x70 [] ? irq_exit+0x67/0xa0 [] ? do_IRQ+0x46/0xb0 [] ? clockevents_notify+0x35/0x110 [] ? common_interrupt+0x2c/0x40 [] ? cpuidle_enter_state+0x41/0xf0 [] ? cpuidle_idle_call+0x8b/0x100 [] ? arch_cpu_idle+0x8/0x30 [] ? cpu_idle_loop+0x4b/0x140 [] ? cpu_startup_entry+0x18/0x20 [] ? rest_init+0x5d/0x70 [] ? start_kernel+0x2ec/0x2f2 [] ? repair_env_string+0x5b/0x5b [] ? i386_start_kernel+0x33/0x35 Signed-off-by: Alexey Dobriyan Signed-off-by: Pablo Neira Ayuso --- net/ipv4/netfilter/nf_nat_h323.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/net/ipv4/netfilter/nf_nat_h323.c b/net/ipv4/netfilter/nf_nat_h323.c index 9eea059dd621..574f7ebba0b6 100644 --- a/net/ipv4/netfilter/nf_nat_h323.c +++ b/net/ipv4/netfilter/nf_nat_h323.c @@ -229,7 +229,10 @@ static int nat_rtp_rtcp(struct sk_buff *skb, struct nf_conn *ct, ret = nf_ct_expect_related(rtcp_exp); if (ret == 0) break; - else if (ret != -EBUSY) { + else if (ret == -EBUSY) { + nf_ct_unexpect_related(rtp_exp); + continue; + } else if (ret < 0) { nf_ct_unexpect_related(rtp_exp); nated_port = 0; break; From e53376bef2cd97d3e3f61fdc677fb8da7d03d0da Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Mon, 3 Feb 2014 20:01:53 +0100 Subject: [PATCH 259/596] netfilter: nf_conntrack: don't release a conntrack with non-zero refcnt With this patch, the conntrack refcount is initially set to zero and it is bumped once it is added to any of the list, so we fulfill Eric's golden rule which is that all released objects always have a refcount that equals zero. Andrey Vagin reports that nf_conntrack_free can't be called for a conntrack with non-zero ref-counter, because it can race with nf_conntrack_find_get(). A conntrack slab is created with SLAB_DESTROY_BY_RCU. Non-zero ref-counter says that this conntrack is used. So when we release a conntrack with non-zero counter, we break this assumption. CPU1 CPU2 ____nf_conntrack_find() nf_ct_put() destroy_conntrack() ... init_conntrack __nf_conntrack_alloc (set use = 1) atomic_inc_not_zero(&ct->use) (use = 2) if (!l4proto->new(ct, skb, dataoff, timeouts)) nf_conntrack_free(ct); (use = 2 !!!) ... __nf_conntrack_alloc (set use = 1) if (!nf_ct_key_equal(h, tuple, zone)) nf_ct_put(ct); (use = 0) destroy_conntrack() /* continue to work with CT */ After applying the path "[PATCH] netfilter: nf_conntrack: fix RCU race in nf_conntrack_find_get" another bug was triggered in destroy_conntrack(): <4>[67096.759334] ------------[ cut here ]------------ <2>[67096.759353] kernel BUG at net/netfilter/nf_conntrack_core.c:211! ... <4>[67096.759837] Pid: 498649, comm: atdd veid: 666 Tainted: G C --------------- 2.6.32-042stab084.18 #1 042stab084_18 /DQ45CB <4>[67096.759932] RIP: 0010:[] [] destroy_conntrack+0x15c/0x190 [nf_conntrack] <4>[67096.760255] Call Trace: <4>[67096.760255] [] nf_conntrack_destroy+0x17/0x30 <4>[67096.760255] [] nf_conntrack_find_get+0x85/0x130 [nf_conntrack] <4>[67096.760255] [] nf_conntrack_in+0x352/0xb60 [nf_conntrack] <4>[67096.760255] [] ipv4_conntrack_local+0x51/0x60 [nf_conntrack_ipv4] <4>[67096.760255] [] nf_iterate+0x69/0xb0 <4>[67096.760255] [] ? dst_output+0x0/0x20 <4>[67096.760255] [] nf_hook_slow+0x74/0x110 <4>[67096.760255] [] ? dst_output+0x0/0x20 <4>[67096.760255] [] raw_sendmsg+0x775/0x910 <4>[67096.760255] [] ? flush_tlb_others_ipi+0x128/0x130 <4>[67096.760255] [] ? apic_timer_interrupt+0xe/0x20 <4>[67096.760255] [] ? apic_timer_interrupt+0xe/0x20 <4>[67096.760255] [] inet_sendmsg+0x4a/0xb0 <4>[67096.760255] [] ? sock_sendmsg+0x13/0x140 <4>[67096.760255] [] sock_sendmsg+0x117/0x140 <4>[67096.760255] [] ? native_smp_send_reschedule+0x49/0x60 <4>[67096.760255] [] ? _spin_unlock_bh+0x1b/0x20 <4>[67096.760255] [] ? autoremove_wake_function+0x0/0x40 <4>[67096.760255] [] ? do_ip_setsockopt+0x90/0xd80 <4>[67096.760255] [] ? apic_timer_interrupt+0xe/0x20 <4>[67096.760255] [] ? apic_timer_interrupt+0xe/0x20 <4>[67096.760255] [] sys_sendto+0x139/0x190 <4>[67096.760255] [] ? audit_syscall_entry+0x1d7/0x200 <4>[67096.760255] [] ? __audit_syscall_exit+0x265/0x290 <4>[67096.760255] [] compat_sys_socketcall+0x13f/0x210 <4>[67096.760255] [] ia32_sysret+0x0/0x5 I have reused the original title for the RFC patch that Andrey posted and most of the original patch description. Cc: Eric Dumazet Cc: Andrew Vagin Cc: Florian Westphal Reported-by: Andrew Vagin Signed-off-by: Pablo Neira Ayuso Reviewed-by: Eric Dumazet Acked-by: Andrew Vagin --- include/net/netfilter/nf_conntrack.h | 2 ++ net/netfilter/nf_conntrack_core.c | 34 ++++++++++++++++++++++++---- net/netfilter/nf_synproxy_core.c | 5 ++-- net/netfilter/xt_CT.c | 7 +----- 4 files changed, 34 insertions(+), 14 deletions(-) diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h index 01ea6eed1bb1..b2ac6246b7e0 100644 --- a/include/net/netfilter/nf_conntrack.h +++ b/include/net/netfilter/nf_conntrack.h @@ -284,6 +284,8 @@ extern unsigned int nf_conntrack_max; extern unsigned int nf_conntrack_hash_rnd; void init_nf_conntrack_hash_rnd(void); +void nf_conntrack_tmpl_insert(struct net *net, struct nf_conn *tmpl); + #define NF_CT_STAT_INC(net, count) __this_cpu_inc((net)->ct.stat->count) #define NF_CT_STAT_INC_ATOMIC(net, count) this_cpu_inc((net)->ct.stat->count) diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index 4d1fb5d094c3..356bef519fe5 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -448,7 +448,9 @@ nf_conntrack_hash_check_insert(struct nf_conn *ct) goto out; add_timer(&ct->timeout); - nf_conntrack_get(&ct->ct_general); + smp_wmb(); + /* The caller holds a reference to this object */ + atomic_set(&ct->ct_general.use, 2); __nf_conntrack_hash_insert(ct, hash, repl_hash); NF_CT_STAT_INC(net, insert); spin_unlock_bh(&nf_conntrack_lock); @@ -462,6 +464,21 @@ nf_conntrack_hash_check_insert(struct nf_conn *ct) } EXPORT_SYMBOL_GPL(nf_conntrack_hash_check_insert); +/* deletion from this larval template list happens via nf_ct_put() */ +void nf_conntrack_tmpl_insert(struct net *net, struct nf_conn *tmpl) +{ + __set_bit(IPS_TEMPLATE_BIT, &tmpl->status); + __set_bit(IPS_CONFIRMED_BIT, &tmpl->status); + nf_conntrack_get(&tmpl->ct_general); + + spin_lock_bh(&nf_conntrack_lock); + /* Overload tuple linked list to put us in template list. */ + hlist_nulls_add_head_rcu(&tmpl->tuplehash[IP_CT_DIR_ORIGINAL].hnnode, + &net->ct.tmpl); + spin_unlock_bh(&nf_conntrack_lock); +} +EXPORT_SYMBOL_GPL(nf_conntrack_tmpl_insert); + /* Confirm a connection given skb; places it in hash table */ int __nf_conntrack_confirm(struct sk_buff *skb) @@ -733,11 +750,10 @@ __nf_conntrack_alloc(struct net *net, u16 zone, nf_ct_zone->id = zone; } #endif - /* - * changes to lookup keys must be done before setting refcnt to 1 + /* Because we use RCU lookups, we set ct_general.use to zero before + * this is inserted in any list. */ - smp_wmb(); - atomic_set(&ct->ct_general.use, 1); + atomic_set(&ct->ct_general.use, 0); return ct; #ifdef CONFIG_NF_CONNTRACK_ZONES @@ -761,6 +777,11 @@ void nf_conntrack_free(struct nf_conn *ct) { struct net *net = nf_ct_net(ct); + /* A freed object has refcnt == 0, that's + * the golden rule for SLAB_DESTROY_BY_RCU + */ + NF_CT_ASSERT(atomic_read(&ct->ct_general.use) == 0); + nf_ct_ext_destroy(ct); nf_ct_ext_free(ct); kmem_cache_free(net->ct.nf_conntrack_cachep, ct); @@ -856,6 +877,9 @@ init_conntrack(struct net *net, struct nf_conn *tmpl, NF_CT_STAT_INC(net, new); } + /* Now it is inserted into the unconfirmed list, bump refcount */ + nf_conntrack_get(&ct->ct_general); + /* Overload tuple linked list to put us in unconfirmed list. */ hlist_nulls_add_head_rcu(&ct->tuplehash[IP_CT_DIR_ORIGINAL].hnnode, &net->ct.unconfirmed); diff --git a/net/netfilter/nf_synproxy_core.c b/net/netfilter/nf_synproxy_core.c index 9858e3e51a3a..52e20c9a46a5 100644 --- a/net/netfilter/nf_synproxy_core.c +++ b/net/netfilter/nf_synproxy_core.c @@ -363,9 +363,8 @@ static int __net_init synproxy_net_init(struct net *net) goto err2; if (!nfct_synproxy_ext_add(ct)) goto err2; - __set_bit(IPS_TEMPLATE_BIT, &ct->status); - __set_bit(IPS_CONFIRMED_BIT, &ct->status); + nf_conntrack_tmpl_insert(net, ct); snet->tmpl = ct; snet->stats = alloc_percpu(struct synproxy_stats); @@ -390,7 +389,7 @@ static void __net_exit synproxy_net_exit(struct net *net) { struct synproxy_net *snet = synproxy_pernet(net); - nf_conntrack_free(snet->tmpl); + nf_ct_put(snet->tmpl); synproxy_proc_exit(net); free_percpu(snet->stats); } diff --git a/net/netfilter/xt_CT.c b/net/netfilter/xt_CT.c index 5929be622c5c..75747aecdebe 100644 --- a/net/netfilter/xt_CT.c +++ b/net/netfilter/xt_CT.c @@ -228,12 +228,7 @@ static int xt_ct_tg_check(const struct xt_tgchk_param *par, goto err3; } - __set_bit(IPS_TEMPLATE_BIT, &ct->status); - __set_bit(IPS_CONFIRMED_BIT, &ct->status); - - /* Overload tuple linked list to put us in template list. */ - hlist_nulls_add_head_rcu(&ct->tuplehash[IP_CT_DIR_ORIGINAL].hnnode, - &par->net->ct.tmpl); + nf_conntrack_tmpl_insert(par->net, ct); out: info->ct = ct; return 0; From 53b70287ddf487a38b7cbf0a10db28f40714b799 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 5 Feb 2014 12:26:22 +0100 Subject: [PATCH 260/596] netfilter: nf_tables: fix overrun in nf_tables_set_alloc_name() The map that is used to allocate anonymous sets is indeed BITS_PER_BYTE * PAGE_SIZE long. Signed-off-by: Patrick McHardy Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_tables_api.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 9ce30534f853..2a22a186eb3d 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -1989,13 +1989,13 @@ static int nf_tables_set_alloc_name(struct nft_ctx *ctx, struct nft_set *set, if (!sscanf(i->name, name, &tmp)) continue; - if (tmp < 0 || tmp > BITS_PER_LONG * PAGE_SIZE) + if (tmp < 0 || tmp >= BITS_PER_BYTE * PAGE_SIZE) continue; set_bit(tmp, inuse); } - n = find_first_zero_bit(inuse, BITS_PER_LONG * PAGE_SIZE); + n = find_first_zero_bit(inuse, BITS_PER_BYTE * PAGE_SIZE); free_page((unsigned long)inuse); } From 667a6b7a620ec6e0c1e6689b8c6531c51ce6cdfa Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Wed, 5 Feb 2014 17:56:27 +0100 Subject: [PATCH 261/596] regulator: max14577: Add missing of_node_put Decrease the reference count for 'regulators' device_node, obtained by of_get_child_by_name(). Signed-off-by: Krzysztof Kozlowski Signed-off-by: Mark Brown --- drivers/regulator/max14577.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/regulator/max14577.c b/drivers/regulator/max14577.c index b1078ba3f393..186df8785a91 100644 --- a/drivers/regulator/max14577.c +++ b/drivers/regulator/max14577.c @@ -168,10 +168,11 @@ static int max14577_regulator_dt_parse_pdata(struct platform_device *pdev) MAX14577_REG_MAX); if (ret < 0) { dev_err(&pdev->dev, "Error parsing regulator init data: %d\n", ret); - return ret; } - return 0; + of_node_put(np); + + return ret; } static inline struct regulator_init_data *match_init_data(int index) From 6a96e15096da6e7491107321cfa660c7c2aa119d Mon Sep 17 00:00:00 2001 From: Paul Moore Date: Tue, 28 Jan 2014 14:45:41 -0500 Subject: [PATCH 262/596] selinux: add SOCK_DIAG_BY_FAMILY to the list of netlink message types The SELinux AF_NETLINK/NETLINK_SOCK_DIAG socket class was missing the SOCK_DIAG_BY_FAMILY definition which caused SELINUX_ERR messages when the ss tool was run. # ss Netid State Recv-Q Send-Q Local Address:Port Peer Address:Port u_str ESTAB 0 0 * 14189 * 14190 u_str ESTAB 0 0 * 14145 * 14144 u_str ESTAB 0 0 * 14151 * 14150 {...} # ausearch -m SELINUX_ERR ---- time->Thu Jan 23 11:11:16 2014 type=SYSCALL msg=audit(1390493476.445:374): arch=c000003e syscall=44 success=yes exit=40 a0=3 a1=7fff03aa11f0 a2=28 a3=0 items=0 ppid=1852 pid=1895 auid=0 uid=0 gid=0 euid=0 suid=0 fsuid=0 egid=0 sgid=0 fsgid=0 tty=pts0 ses=1 comm="ss" exe="/usr/sbin/ss" subj=unconfined_u:unconfined_r:unconfined_t:s0-s0:c0.c1023 key=(null) type=SELINUX_ERR msg=audit(1390493476.445:374): SELinux: unrecognized netlink message type=20 for sclass=32 Signed-off-by: Paul Moore --- security/selinux/nlmsgtab.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/security/selinux/nlmsgtab.c b/security/selinux/nlmsgtab.c index 332ac8a80cf5..2df7b900e259 100644 --- a/security/selinux/nlmsgtab.c +++ b/security/selinux/nlmsgtab.c @@ -17,6 +17,7 @@ #include #include #include +#include #include "flask.h" #include "av_permissions.h" @@ -78,6 +79,7 @@ static struct nlmsg_perm nlmsg_tcpdiag_perms[] = { { TCPDIAG_GETSOCK, NETLINK_TCPDIAG_SOCKET__NLMSG_READ }, { DCCPDIAG_GETSOCK, NETLINK_TCPDIAG_SOCKET__NLMSG_READ }, + { SOCK_DIAG_BY_FAMILY, NETLINK_TCPDIAG_SOCKET__NLMSG_READ }, }; static struct nlmsg_perm nlmsg_xfrm_perms[] = From 2172fa709ab32ca60e86179dc67d0857be8e2c98 Mon Sep 17 00:00:00 2001 From: Stephen Smalley Date: Thu, 30 Jan 2014 11:26:59 -0500 Subject: [PATCH 263/596] SELinux: Fix kernel BUG on empty security contexts. Setting an empty security context (length=0) on a file will lead to incorrectly dereferencing the type and other fields of the security context structure, yielding a kernel BUG. As a zero-length security context is never valid, just reject all such security contexts whether coming from userspace via setxattr or coming from the filesystem upon a getxattr request by SELinux. Setting a security context value (empty or otherwise) unknown to SELinux in the first place is only possible for a root process (CAP_MAC_ADMIN), and, if running SELinux in enforcing mode, only if the corresponding SELinux mac_admin permission is also granted to the domain by policy. In Fedora policies, this is only allowed for specific domains such as livecd for setting down security contexts that are not defined in the build host policy. Reproducer: su setenforce 0 touch foo setfattr -n security.selinux foo Caveat: Relabeling or removing foo after doing the above may not be possible without booting with SELinux disabled. Any subsequent access to foo after doing the above will also trigger the BUG. BUG output from Matthew Thode: [ 473.893141] ------------[ cut here ]------------ [ 473.962110] kernel BUG at security/selinux/ss/services.c:654! [ 473.995314] invalid opcode: 0000 [#6] SMP [ 474.027196] Modules linked in: [ 474.058118] CPU: 0 PID: 8138 Comm: ls Tainted: G D I 3.13.0-grsec #1 [ 474.116637] Hardware name: Supermicro X8ST3/X8ST3, BIOS 2.0 07/29/10 [ 474.149768] task: ffff8805f50cd010 ti: ffff8805f50cd488 task.ti: ffff8805f50cd488 [ 474.183707] RIP: 0010:[] [] context_struct_compute_av+0xce/0x308 [ 474.219954] RSP: 0018:ffff8805c0ac3c38 EFLAGS: 00010246 [ 474.252253] RAX: 0000000000000000 RBX: ffff8805c0ac3d94 RCX: 0000000000000100 [ 474.287018] RDX: ffff8805e8aac000 RSI: 00000000ffffffff RDI: ffff8805e8aaa000 [ 474.321199] RBP: ffff8805c0ac3cb8 R08: 0000000000000010 R09: 0000000000000006 [ 474.357446] R10: 0000000000000000 R11: ffff8805c567a000 R12: 0000000000000006 [ 474.419191] R13: ffff8805c2b74e88 R14: 00000000000001da R15: 0000000000000000 [ 474.453816] FS: 00007f2e75220800(0000) GS:ffff88061fc00000(0000) knlGS:0000000000000000 [ 474.489254] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 474.522215] CR2: 00007f2e74716090 CR3: 00000005c085e000 CR4: 00000000000207f0 [ 474.556058] Stack: [ 474.584325] ffff8805c0ac3c98 ffffffff811b549b ffff8805c0ac3c98 ffff8805f1190a40 [ 474.618913] ffff8805a6202f08 ffff8805c2b74e88 00068800d0464990 ffff8805e8aac860 [ 474.653955] ffff8805c0ac3cb8 000700068113833a ffff880606c75060 ffff8805c0ac3d94 [ 474.690461] Call Trace: [ 474.723779] [] ? lookup_fast+0x1cd/0x22a [ 474.778049] [] security_compute_av+0xf4/0x20b [ 474.811398] [] avc_compute_av+0x2a/0x179 [ 474.843813] [] avc_has_perm+0x45/0xf4 [ 474.875694] [] inode_has_perm+0x2a/0x31 [ 474.907370] [] selinux_inode_getattr+0x3c/0x3e [ 474.938726] [] security_inode_getattr+0x1b/0x22 [ 474.970036] [] vfs_getattr+0x19/0x2d [ 475.000618] [] vfs_fstatat+0x54/0x91 [ 475.030402] [] vfs_lstat+0x19/0x1b [ 475.061097] [] SyS_newlstat+0x15/0x30 [ 475.094595] [] ? __audit_syscall_entry+0xa1/0xc3 [ 475.148405] [] system_call_fastpath+0x16/0x1b [ 475.179201] Code: 00 48 85 c0 48 89 45 b8 75 02 0f 0b 48 8b 45 a0 48 8b 3d 45 d0 b6 00 8b 40 08 89 c6 ff ce e8 d1 b0 06 00 48 85 c0 49 89 c7 75 02 <0f> 0b 48 8b 45 b8 4c 8b 28 eb 1e 49 8d 7d 08 be 80 01 00 00 e8 [ 475.255884] RIP [] context_struct_compute_av+0xce/0x308 [ 475.296120] RSP [ 475.328734] ---[ end trace f076482e9d754adc ]--- Reported-by: Matthew Thode Signed-off-by: Stephen Smalley Cc: stable@vger.kernel.org Signed-off-by: Paul Moore --- security/selinux/ss/services.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/security/selinux/ss/services.c b/security/selinux/ss/services.c index fc5a63a05a1c..f1e46d776544 100644 --- a/security/selinux/ss/services.c +++ b/security/selinux/ss/services.c @@ -1232,6 +1232,10 @@ static int security_context_to_sid_core(const char *scontext, u32 scontext_len, struct context context; int rc = 0; + /* An empty security context is never valid. */ + if (!scontext_len) + return -EINVAL; + if (!ss_initialized) { int i; From f948dcf9e9973c05d957bc65b3185682f45feda3 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Bj=C3=B8rn=20Mork?= Date: Tue, 4 Feb 2014 13:02:31 +0100 Subject: [PATCH 264/596] usb: qcserial: add Netgear Aircard 340U MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This device was mentioned in an OpenWRT forum. Seems to have a "standard" Sierra Wireless ifnumber to function layout: 0: qcdm 2: nmea 3: modem 8: qmi 9: storage Cc: Signed-off-by: Bjørn Mork Signed-off-by: Greg Kroah-Hartman --- drivers/usb/serial/qcserial.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/usb/serial/qcserial.c b/drivers/usb/serial/qcserial.c index c65437cfd4a2..968a40201e5f 100644 --- a/drivers/usb/serial/qcserial.c +++ b/drivers/usb/serial/qcserial.c @@ -139,6 +139,9 @@ static const struct usb_device_id id_table[] = { {USB_DEVICE_INTERFACE_NUMBER(0x1199, 0x901c, 0)}, /* Sierra Wireless EM7700 Device Management */ {USB_DEVICE_INTERFACE_NUMBER(0x1199, 0x901c, 2)}, /* Sierra Wireless EM7700 NMEA */ {USB_DEVICE_INTERFACE_NUMBER(0x1199, 0x901c, 3)}, /* Sierra Wireless EM7700 Modem */ + {USB_DEVICE_INTERFACE_NUMBER(0x1199, 0x9051, 0)}, /* Netgear AirCard 340U Device Management */ + {USB_DEVICE_INTERFACE_NUMBER(0x1199, 0x9051, 2)}, /* Netgear AirCard 340U NMEA */ + {USB_DEVICE_INTERFACE_NUMBER(0x1199, 0x9051, 3)}, /* Netgear AirCard 340U Modem */ { } /* Terminating entry */ }; From 76f24e3f39a1a94bab0d54e98899d64abcd9f69c Mon Sep 17 00:00:00 2001 From: Ulrich Hahn Date: Sun, 2 Feb 2014 14:42:52 +0100 Subject: [PATCH 265/596] USB: ftdi_sio: add Tagsys RFID Reader IDs Adding two more IDs to the ftdi_sio usb serial driver. It now connects Tagsys RFID readers. There might be more IDs out there for other Tagsys models. Signed-off-by: Ulrich Hahn Cc: Johan Hovold Cc: stable Signed-off-by: Greg Kroah-Hartman --- drivers/usb/serial/ftdi_sio.c | 2 ++ drivers/usb/serial/ftdi_sio_ids.h | 6 ++++++ 2 files changed, 8 insertions(+) diff --git a/drivers/usb/serial/ftdi_sio.c b/drivers/usb/serial/ftdi_sio.c index 0f3248a017b3..ee1f00f03c43 100644 --- a/drivers/usb/serial/ftdi_sio.c +++ b/drivers/usb/serial/ftdi_sio.c @@ -192,6 +192,8 @@ static const struct usb_device_id id_table_combined[] = { { USB_DEVICE(INTERBIOMETRICS_VID, INTERBIOMETRICS_IOBOARD_PID) }, { USB_DEVICE(INTERBIOMETRICS_VID, INTERBIOMETRICS_MINI_IOBOARD_PID) }, { USB_DEVICE(FTDI_VID, FTDI_SPROG_II) }, + { USB_DEVICE(FTDI_VID, FTDI_TAGSYS_LP101_PID) }, + { USB_DEVICE(FTDI_VID, FTDI_TAGSYS_P200X_PID) }, { USB_DEVICE(FTDI_VID, FTDI_LENZ_LIUSB_PID) }, { USB_DEVICE(FTDI_VID, FTDI_XF_632_PID) }, { USB_DEVICE(FTDI_VID, FTDI_XF_634_PID) }, diff --git a/drivers/usb/serial/ftdi_sio_ids.h b/drivers/usb/serial/ftdi_sio_ids.h index eb0302135348..1e2d369df86e 100644 --- a/drivers/usb/serial/ftdi_sio_ids.h +++ b/drivers/usb/serial/ftdi_sio_ids.h @@ -364,6 +364,12 @@ /* Sprog II (Andrew Crosland's SprogII DCC interface) */ #define FTDI_SPROG_II 0xF0C8 +/* + * Two of the Tagsys RFID Readers + */ +#define FTDI_TAGSYS_LP101_PID 0xF0E9 /* Tagsys L-P101 RFID*/ +#define FTDI_TAGSYS_P200X_PID 0xF0EE /* Tagsys Medio P200x RFID*/ + /* an infrared receiver for user access control with IR tags */ #define FTDI_PIEGROUP_PID 0xF208 /* Product Id */ From 7f196caffbf2dc96cc145bf5d2ef5ef8a7b4f687 Mon Sep 17 00:00:00 2001 From: Christian Engelmayer Date: Tue, 28 Jan 2014 22:22:27 +0100 Subject: [PATCH 266/596] usb: core: Fix potential memory leak adding dyn USBdevice IDs Fix a memory leak in the usb_store_new_id() error paths. When bailing out due to sanity checks, the function left the already allocated usb_dynid struct in place. This regression was introduced by the following commits: c63fe8f6 (usb: core: add sanity checks when using bInterfaceClass with new_id) 1b9fb31f (usb: core: check for valid id_table when using the RefId feature) 52a6966c (usb: core: bail out if user gives an unknown RefId when using new_id) Detected by Coverity: CID 1162604. Signed-off-by: Christian Engelmayer Acked-by: Wolfram Sang Signed-off-by: Greg Kroah-Hartman --- drivers/usb/core/driver.c | 24 +++++++++++++++++------- 1 file changed, 17 insertions(+), 7 deletions(-) diff --git a/drivers/usb/core/driver.c b/drivers/usb/core/driver.c index 5d01558cef66..ab90a0156828 100644 --- a/drivers/usb/core/driver.c +++ b/drivers/usb/core/driver.c @@ -63,8 +63,10 @@ ssize_t usb_store_new_id(struct usb_dynids *dynids, dynid->id.idProduct = idProduct; dynid->id.match_flags = USB_DEVICE_ID_MATCH_DEVICE; if (fields > 2 && bInterfaceClass) { - if (bInterfaceClass > 255) - return -EINVAL; + if (bInterfaceClass > 255) { + retval = -EINVAL; + goto fail; + } dynid->id.bInterfaceClass = (u8)bInterfaceClass; dynid->id.match_flags |= USB_DEVICE_ID_MATCH_INT_CLASS; @@ -73,17 +75,21 @@ ssize_t usb_store_new_id(struct usb_dynids *dynids, if (fields > 4) { const struct usb_device_id *id = id_table; - if (!id) - return -ENODEV; + if (!id) { + retval = -ENODEV; + goto fail; + } for (; id->match_flags; id++) if (id->idVendor == refVendor && id->idProduct == refProduct) break; - if (id->match_flags) + if (id->match_flags) { dynid->id.driver_info = id->driver_info; - else - return -ENODEV; + } else { + retval = -ENODEV; + goto fail; + } } spin_lock(&dynids->lock); @@ -95,6 +101,10 @@ ssize_t usb_store_new_id(struct usb_dynids *dynids, if (retval) return retval; return count; + +fail: + kfree(dynid); + return retval; } EXPORT_SYMBOL_GPL(usb_store_new_id); From da9846ae15186d491d6e21ebbb5051e1d3c7f652 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Wed, 29 Jan 2014 12:04:03 -0500 Subject: [PATCH 267/596] kernfs: make kernfs_deactivate() honor KERNFS_LOCKDEP flag kernfs_deactivate() forgot to check whether KERNFS_LOCKDEP is set before performing lockdep annotations and ends up feeding uninitialized lockdep_map to lockdep triggering warning like the following on USB stick hotunplug. usb 1-2: USB disconnect, device number 2 INFO: trying to register non-static key. the code is fine but needs lockdep annotation. turning off the locking correctness validator. CPU: 1 PID: 62 Comm: khubd Not tainted 3.13.0-work+ #82 Hardware name: empty empty/S3992, BIOS 080011 10/26/2007 ffff880065ca7f60 ffff88013a4ffa08 ffffffff81cfb6bd 0000000000000002 ffff88013a4ffac8 ffffffff810f8530 ffff88013a4fc710 0000000000000002 ffff880100000000 ffffffff82a3db50 0000000000000001 ffff88013a4fc710 Call Trace: [] dump_stack+0x4e/0x7a [] __lock_acquire+0x1910/0x1e70 [] lock_acquire+0x9a/0x1d0 [] kernfs_deactivate+0xee/0x130 [] kernfs_addrm_finish+0x38/0x60 [] kernfs_remove_by_name_ns+0x51/0xa0 [] remove_files.isra.1+0x41/0x80 [] sysfs_remove_group+0x47/0xa0 [] sysfs_remove_groups+0x33/0x50 [] device_remove_attrs+0x4d/0x80 [] device_del+0x12e/0x1d0 [] usb_disconnect+0x122/0x1a0 [] hub_thread+0x3c5/0x1290 [] kthread+0xed/0x110 [] ret_from_fork+0x7c/0xb0 Fix it by making kernfs_deactivate() perform lockdep annotations only if KERNFS_LOCKDEP is set. Signed-off-by: Tejun Heo Reported-by: Fabio Estevam Reported-by: Alan Stern Reported-by: Jiri Kosina Reported-by: Dave Jones Tested-by: Fabio Estevam Tested-by: Jiri Kosina Signed-off-by: Greg Kroah-Hartman --- fs/kernfs/dir.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/fs/kernfs/dir.c b/fs/kernfs/dir.c index 5104cf5d25c5..bd6e18be6e1a 100644 --- a/fs/kernfs/dir.c +++ b/fs/kernfs/dir.c @@ -187,19 +187,23 @@ static void kernfs_deactivate(struct kernfs_node *kn) kn->u.completion = (void *)&wait; - rwsem_acquire(&kn->dep_map, 0, 0, _RET_IP_); + if (kn->flags & KERNFS_LOCKDEP) + rwsem_acquire(&kn->dep_map, 0, 0, _RET_IP_); /* atomic_add_return() is a mb(), put_active() will always see * the updated kn->u.completion. */ v = atomic_add_return(KN_DEACTIVATED_BIAS, &kn->active); if (v != KN_DEACTIVATED_BIAS) { - lock_contended(&kn->dep_map, _RET_IP_); + if (kn->flags & KERNFS_LOCKDEP) + lock_contended(&kn->dep_map, _RET_IP_); wait_for_completion(&wait); } - lock_acquired(&kn->dep_map, _RET_IP_); - rwsem_release(&kn->dep_map, 1, _RET_IP_); + if (kn->flags & KERNFS_LOCKDEP) { + lock_acquired(&kn->dep_map, _RET_IP_); + rwsem_release(&kn->dep_map, 1, _RET_IP_); + } } /** From c4ad8f98bef77c7356aa6a9ad9188a6acc6b849d Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Wed, 5 Feb 2014 12:54:53 -0800 Subject: [PATCH 268/596] execve: use 'struct filename *' for executable name passing This changes 'do_execve()' to get the executable name as a 'struct filename', and to free it when it is done. This is what the normal users want, and it simplifies and streamlines their error handling. The controlled lifetime of the executable name also fixes a use-after-free problem with the trace_sched_process_exec tracepoint: the lifetime of the passed-in string for kernel users was not at all obvious, and the user-mode helper code used UMH_WAIT_EXEC to serialize the pathname allocation lifetime with the execve() having finished, which in turn meant that the trace point that happened after mm_release() of the old process VM ended up using already free'd memory. To solve the kernel string lifetime issue, this simply introduces "getname_kernel()" that works like the normal user-space getname() function, except with the source coming from kernel memory. As Oleg points out, this also means that we could drop the tcomm[] array from 'struct linux_binprm', since the pathname lifetime now covers setup_new_exec(). That would be a separate cleanup. Reported-by: Igor Zhbanov Tested-by: Steven Rostedt Cc: Oleg Nesterov Cc: Al Viro Signed-off-by: Linus Torvalds --- arch/parisc/hpux/fs.c | 15 +------------- fs/exec.c | 45 +++++++++++++++++++---------------------- fs/namei.c | 30 +++++++++++++++++++++++++++ include/linux/binfmts.h | 1 - include/linux/fs.h | 1 + include/linux/sched.h | 3 ++- init/main.c | 2 +- kernel/auditsc.c | 2 +- kernel/kmod.c | 2 +- 9 files changed, 58 insertions(+), 43 deletions(-) diff --git a/arch/parisc/hpux/fs.c b/arch/parisc/hpux/fs.c index 88d0962de65a..2bedafea3d94 100644 --- a/arch/parisc/hpux/fs.c +++ b/arch/parisc/hpux/fs.c @@ -33,22 +33,9 @@ int hpux_execve(struct pt_regs *regs) { - int error; - struct filename *filename; - - filename = getname((const char __user *) regs->gr[26]); - error = PTR_ERR(filename); - if (IS_ERR(filename)) - goto out; - - error = do_execve(filename->name, + return do_execve(getname((const char __user *) regs->gr[26]), (const char __user *const __user *) regs->gr[25], (const char __user *const __user *) regs->gr[24]); - - putname(filename); - -out: - return error; } struct hpux_dirent { diff --git a/fs/exec.c b/fs/exec.c index e1529b4c79b1..3d78fccdd723 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -748,11 +748,10 @@ EXPORT_SYMBOL(setup_arg_pages); #endif /* CONFIG_MMU */ -struct file *open_exec(const char *name) +static struct file *do_open_exec(struct filename *name) { struct file *file; int err; - struct filename tmp = { .name = name }; static const struct open_flags open_exec_flags = { .open_flag = O_LARGEFILE | O_RDONLY | __FMODE_EXEC, .acc_mode = MAY_EXEC | MAY_OPEN, @@ -760,7 +759,7 @@ struct file *open_exec(const char *name) .lookup_flags = LOOKUP_FOLLOW, }; - file = do_filp_open(AT_FDCWD, &tmp, &open_exec_flags); + file = do_filp_open(AT_FDCWD, name, &open_exec_flags); if (IS_ERR(file)) goto out; @@ -784,6 +783,12 @@ struct file *open_exec(const char *name) fput(file); return ERR_PTR(err); } + +struct file *open_exec(const char *name) +{ + struct filename tmp = { .name = name }; + return do_open_exec(&tmp); +} EXPORT_SYMBOL(open_exec); int kernel_read(struct file *file, loff_t offset, @@ -1162,7 +1167,7 @@ int prepare_bprm_creds(struct linux_binprm *bprm) return -ENOMEM; } -void free_bprm(struct linux_binprm *bprm) +static void free_bprm(struct linux_binprm *bprm) { free_arg_pages(bprm); if (bprm->cred) { @@ -1432,7 +1437,7 @@ static int exec_binprm(struct linux_binprm *bprm) /* * sys_execve() executes a new program. */ -static int do_execve_common(const char *filename, +static int do_execve_common(struct filename *filename, struct user_arg_ptr argv, struct user_arg_ptr envp) { @@ -1441,6 +1446,9 @@ static int do_execve_common(const char *filename, struct files_struct *displaced; int retval; + if (IS_ERR(filename)) + return PTR_ERR(filename); + /* * We move the actual failure in case of RLIMIT_NPROC excess from * set*uid() to execve() because too many poorly written programs @@ -1473,7 +1481,7 @@ static int do_execve_common(const char *filename, check_unsafe_exec(bprm); current->in_execve = 1; - file = open_exec(filename); + file = do_open_exec(filename); retval = PTR_ERR(file); if (IS_ERR(file)) goto out_unmark; @@ -1481,8 +1489,7 @@ static int do_execve_common(const char *filename, sched_exec(); bprm->file = file; - bprm->filename = filename; - bprm->interp = filename; + bprm->filename = bprm->interp = filename->name; retval = bprm_mm_init(bprm); if (retval) @@ -1523,6 +1530,7 @@ static int do_execve_common(const char *filename, acct_update_integrals(current); task_numa_free(current); free_bprm(bprm); + putname(filename); if (displaced) put_files_struct(displaced); return retval; @@ -1544,10 +1552,11 @@ static int do_execve_common(const char *filename, if (displaced) reset_files_struct(displaced); out_ret: + putname(filename); return retval; } -int do_execve(const char *filename, +int do_execve(struct filename *filename, const char __user *const __user *__argv, const char __user *const __user *__envp) { @@ -1557,7 +1566,7 @@ int do_execve(const char *filename, } #ifdef CONFIG_COMPAT -static int compat_do_execve(const char *filename, +static int compat_do_execve(struct filename *filename, const compat_uptr_t __user *__argv, const compat_uptr_t __user *__envp) { @@ -1607,25 +1616,13 @@ SYSCALL_DEFINE3(execve, const char __user *const __user *, argv, const char __user *const __user *, envp) { - struct filename *path = getname(filename); - int error = PTR_ERR(path); - if (!IS_ERR(path)) { - error = do_execve(path->name, argv, envp); - putname(path); - } - return error; + return do_execve(getname(filename), argv, envp); } #ifdef CONFIG_COMPAT asmlinkage long compat_sys_execve(const char __user * filename, const compat_uptr_t __user * argv, const compat_uptr_t __user * envp) { - struct filename *path = getname(filename); - int error = PTR_ERR(path); - if (!IS_ERR(path)) { - error = compat_do_execve(path->name, argv, envp); - putname(path); - } - return error; + return compat_do_execve(getname(filename), argv, envp); } #endif diff --git a/fs/namei.c b/fs/namei.c index d580df2e6804..385f7817bfcc 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -196,6 +196,7 @@ getname_flags(const char __user *filename, int flags, int *empty) goto error; result->uptr = filename; + result->aname = NULL; audit_getname(result); return result; @@ -210,6 +211,35 @@ getname(const char __user * filename) return getname_flags(filename, 0, NULL); } +/* + * The "getname_kernel()" interface doesn't do pathnames longer + * than EMBEDDED_NAME_MAX. Deal with it - you're a kernel user. + */ +struct filename * +getname_kernel(const char * filename) +{ + struct filename *result; + char *kname; + int len; + + len = strlen(filename); + if (len >= EMBEDDED_NAME_MAX) + return ERR_PTR(-ENAMETOOLONG); + + result = __getname(); + if (unlikely(!result)) + return ERR_PTR(-ENOMEM); + + kname = (char *)result + sizeof(*result); + result->name = kname; + result->uptr = NULL; + result->aname = NULL; + result->separate = false; + + strlcpy(kname, filename, EMBEDDED_NAME_MAX); + return result; +} + #ifdef CONFIG_AUDITSYSCALL void putname(struct filename *name) { diff --git a/include/linux/binfmts.h b/include/linux/binfmts.h index fd8bf3219ef7..b4a745d7d9a9 100644 --- a/include/linux/binfmts.h +++ b/include/linux/binfmts.h @@ -115,7 +115,6 @@ extern int copy_strings_kernel(int argc, const char *const *argv, extern int prepare_bprm_creds(struct linux_binprm *bprm); extern void install_exec_creds(struct linux_binprm *bprm); extern void set_binfmt(struct linux_binfmt *new); -extern void free_bprm(struct linux_binprm *); extern ssize_t read_code(struct file *, unsigned long, loff_t, size_t); #endif /* _LINUX_BINFMTS_H */ diff --git a/include/linux/fs.h b/include/linux/fs.h index 09f553c59813..d79678c188ad 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -2079,6 +2079,7 @@ extern struct file * dentry_open(const struct path *, int, const struct cred *); extern int filp_close(struct file *, fl_owner_t id); extern struct filename *getname(const char __user *); +extern struct filename *getname_kernel(const char *); enum { FILE_CREATED = 1, diff --git a/include/linux/sched.h b/include/linux/sched.h index 68a0e84463a0..a781dec1cd0b 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -128,6 +128,7 @@ struct bio_list; struct fs_struct; struct perf_event_context; struct blk_plug; +struct filename; /* * List of flags we want to share for kernel threads, @@ -2311,7 +2312,7 @@ extern void do_group_exit(int); extern int allow_signal(int); extern int disallow_signal(int); -extern int do_execve(const char *, +extern int do_execve(struct filename *, const char __user * const __user *, const char __user * const __user *); extern long do_fork(unsigned long, unsigned long, unsigned long, int __user *, int __user *); diff --git a/init/main.c b/init/main.c index 2fd9cef70ee8..eb03090cdced 100644 --- a/init/main.c +++ b/init/main.c @@ -812,7 +812,7 @@ void __init load_default_modules(void) static int run_init_process(const char *init_filename) { argv_init[0] = init_filename; - return do_execve(init_filename, + return do_execve(getname_kernel(init_filename), (const char __user *const __user *)argv_init, (const char __user *const __user *)envp_init); } diff --git a/kernel/auditsc.c b/kernel/auditsc.c index 10176cd5956a..7aef2f4b6c64 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -1719,7 +1719,7 @@ void audit_putname(struct filename *name) struct audit_context *context = current->audit_context; BUG_ON(!context); - if (!context->in_syscall) { + if (!name->aname || !context->in_syscall) { #if AUDIT_DEBUG == 2 printk(KERN_ERR "%s:%d(:%d): final_putname(%p)\n", __FILE__, __LINE__, context->serial, name); diff --git a/kernel/kmod.c b/kernel/kmod.c index b086006c59e7..6b375af4958d 100644 --- a/kernel/kmod.c +++ b/kernel/kmod.c @@ -239,7 +239,7 @@ static int ____call_usermodehelper(void *data) commit_creds(new); - retval = do_execve(sub_info->path, + retval = do_execve(getname_kernel(sub_info->path), (const char __user *const __user *)sub_info->argv, (const char __user *const __user *)sub_info->envp); if (!retval) From f8f202348208fa8a2d817b42f250e145fa885620 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Wed, 5 Feb 2014 06:51:37 +0100 Subject: [PATCH 269/596] x86: Disable CONFIG_X86_DECODER_SELFTEST in allmod/allyesconfigs It can take some time to validate the image, make sure {allyes|allmod}config doesn't enable it. I'd say randconfig will cover it often enough, and the failure is also borderline build coverage related: you cannot really make the decoder test fail via source level changes, only with changes in the build environment, so I agree with Andi that we can disable this one too. Signed-off-by: Ingo Molnar Acked-by: Paul Gortmaker paul.gortmaker@windriver.com> Suggested-and-acked-by: Andi Kleen andi@firstfloor.org> Signed-off-by: Linus Torvalds --- arch/x86/Kconfig.debug | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/x86/Kconfig.debug b/arch/x86/Kconfig.debug index 0f3621ed1db6..321a52ccf63a 100644 --- a/arch/x86/Kconfig.debug +++ b/arch/x86/Kconfig.debug @@ -184,6 +184,7 @@ config HAVE_MMIOTRACE_SUPPORT config X86_DECODER_SELFTEST bool "x86 instruction decoder selftest" depends on DEBUG_KERNEL && KPROBES + depends on !COMPILE_TEST ---help--- Perform x86 instruction decoder selftests at build time. This option is useful for checking the sanity of x86 instruction From ec2c9935688fbd5eaa7c975e3e21562c3da77363 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 5 Feb 2014 15:03:35 +0000 Subject: [PATCH 270/596] netfilter: nf_tables: fix potential oops when dumping sets Commit c9c8e48597 (netfilter: nf_tables: dump sets in all existing families) changed nft_ctx_init_from_setattr() to only look up the address family if it is not NFPROTO_UNSPEC. However if it is NFPROTO_UNSPEC and a table attribute is given, nftables_afinfo_lookup() will dereference the NULL afi pointer. Fix by checking for non-NULL afi and also move a check added by that commit to the proper position. Signed-off-by: Patrick McHardy Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_tables_api.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 2a22a186eb3d..3c5a219f4242 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -1943,6 +1943,9 @@ static int nft_ctx_init_from_setattr(struct nft_ctx *ctx, } if (nla[NFTA_SET_TABLE] != NULL) { + if (afi == NULL) + return -EAFNOSUPPORT; + table = nf_tables_table_lookup(afi, nla[NFTA_SET_TABLE]); if (IS_ERR(table)) return PTR_ERR(table); @@ -2428,6 +2431,8 @@ static int nf_tables_delset(struct sock *nlsk, struct sk_buff *skb, struct nft_ctx ctx; int err; + if (nfmsg->nfgen_family == NFPROTO_UNSPEC) + return -EAFNOSUPPORT; if (nla[NFTA_SET_TABLE] == NULL) return -EINVAL; @@ -2435,9 +2440,6 @@ static int nf_tables_delset(struct sock *nlsk, struct sk_buff *skb, if (err < 0) return err; - if (nfmsg->nfgen_family == NFPROTO_UNSPEC) - return -EAFNOSUPPORT; - set = nf_tables_set_lookup(ctx.table, nla[NFTA_SET_NAME]); if (IS_ERR(set)) return PTR_ERR(set); From 51292c0735eb2d9e29115cbf6264845e19a6c77d Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 5 Feb 2014 15:03:36 +0000 Subject: [PATCH 271/596] netfilter: nft_ct: fix missing NFT_CT_L3PROTOCOL key in validity checks The key was missing in the list of valid keys, add it. Signed-off-by: Patrick McHardy Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nft_ct.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/netfilter/nft_ct.c b/net/netfilter/nft_ct.c index feaf0f354a93..46e275403838 100644 --- a/net/netfilter/nft_ct.c +++ b/net/netfilter/nft_ct.c @@ -226,6 +226,7 @@ static int nft_ct_init_validate_get(const struct nft_expr *expr, if (tb[NFTA_CT_DIRECTION] != NULL) return -EINVAL; break; + case NFT_CT_L3PROTOCOL: case NFT_CT_PROTOCOL: case NFT_CT_SRC: case NFT_CT_DST: From 64d46806b6218c97f68742c5663a8ae3a5fbe838 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 5 Feb 2014 15:03:37 +0000 Subject: [PATCH 272/596] netfilter: nf_tables: add AF specific expression support For the reject module, we need to add AF-specific implementations to get rid of incorrect module dependencies. Try to load an AF-specific module first and fall back to generic modules. Signed-off-by: Patrick McHardy Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_tables.h | 5 +++++ net/netfilter/nf_tables_api.c | 22 ++++++++++++++++------ 2 files changed, 21 insertions(+), 6 deletions(-) diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h index 57c8ff7955df..0f68e47d3e5e 100644 --- a/include/net/netfilter/nf_tables.h +++ b/include/net/netfilter/nf_tables.h @@ -252,6 +252,7 @@ void nf_tables_unbind_set(const struct nft_ctx *ctx, struct nft_set *set, * @owner: module reference * @policy: netlink attribute policy * @maxattr: highest netlink attribute number + * @family: address family for AF-specific types */ struct nft_expr_type { const struct nft_expr_ops *(*select_ops)(const struct nft_ctx *, @@ -262,6 +263,7 @@ struct nft_expr_type { struct module *owner; const struct nla_policy *policy; unsigned int maxattr; + u8 family; }; /** @@ -529,6 +531,9 @@ void nft_unregister_expr(struct nft_expr_type *); #define MODULE_ALIAS_NFT_CHAIN(family, name) \ MODULE_ALIAS("nft-chain-" __stringify(family) "-" name) +#define MODULE_ALIAS_NFT_AF_EXPR(family, name) \ + MODULE_ALIAS("nft-expr-" __stringify(family) "-" name) + #define MODULE_ALIAS_NFT_EXPR(name) \ MODULE_ALIAS("nft-expr-" name) diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 3c5a219f4242..113c469c7579 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -1114,35 +1114,45 @@ void nft_unregister_expr(struct nft_expr_type *type) } EXPORT_SYMBOL_GPL(nft_unregister_expr); -static const struct nft_expr_type *__nft_expr_type_get(struct nlattr *nla) +static const struct nft_expr_type *__nft_expr_type_get(u8 family, + struct nlattr *nla) { const struct nft_expr_type *type; list_for_each_entry(type, &nf_tables_expressions, list) { - if (!nla_strcmp(nla, type->name)) + if (!nla_strcmp(nla, type->name) && + (!type->family || type->family == family)) return type; } return NULL; } -static const struct nft_expr_type *nft_expr_type_get(struct nlattr *nla) +static const struct nft_expr_type *nft_expr_type_get(u8 family, + struct nlattr *nla) { const struct nft_expr_type *type; if (nla == NULL) return ERR_PTR(-EINVAL); - type = __nft_expr_type_get(nla); + type = __nft_expr_type_get(family, nla); if (type != NULL && try_module_get(type->owner)) return type; #ifdef CONFIG_MODULES if (type == NULL) { + nfnl_unlock(NFNL_SUBSYS_NFTABLES); + request_module("nft-expr-%u-%.*s", family, + nla_len(nla), (char *)nla_data(nla)); + nfnl_lock(NFNL_SUBSYS_NFTABLES); + if (__nft_expr_type_get(family, nla)) + return ERR_PTR(-EAGAIN); + nfnl_unlock(NFNL_SUBSYS_NFTABLES); request_module("nft-expr-%.*s", nla_len(nla), (char *)nla_data(nla)); nfnl_lock(NFNL_SUBSYS_NFTABLES); - if (__nft_expr_type_get(nla)) + if (__nft_expr_type_get(family, nla)) return ERR_PTR(-EAGAIN); } #endif @@ -1193,7 +1203,7 @@ static int nf_tables_expr_parse(const struct nft_ctx *ctx, if (err < 0) return err; - type = nft_expr_type_get(tb[NFTA_EXPR_NAME]); + type = nft_expr_type_get(ctx->afi->family, tb[NFTA_EXPR_NAME]); if (IS_ERR(type)) return PTR_ERR(type); From 081cd62a010f97b5bc1d2b0cd123c5abc692b68a Mon Sep 17 00:00:00 2001 From: Matt Fleming Date: Tue, 14 Jan 2014 12:40:09 +0000 Subject: [PATCH 273/596] x86/efi: Allow mapping BGRT on x86-32 CONFIG_X86_32 doesn't map the boot services regions into the EFI memory map (see commit 700870119f49 ("x86, efi: Don't map Boot Services on i386")), and so efi_lookup_mapped_addr() will fail to return a valid address. Executing the ioremap() path in efi_bgrt_init() causes the following warning on x86-32 because we're trying to ioremap() RAM, WARNING: CPU: 0 PID: 0 at arch/x86/mm/ioremap.c:102 __ioremap_caller+0x2ad/0x2c0() Modules linked in: CPU: 0 PID: 0 Comm: swapper/0 Not tainted 3.13.0-0.rc5.git0.1.2.fc21.i686 #1 Hardware name: DellInc. Venue 8 Pro 5830/09RP78, BIOS A02 10/17/2013 00000000 00000000 c0c0df08 c09a5196 00000000 c0c0df38 c0448c1e c0b41310 00000000 00000000 c0b37bc1 00000066 c043bbfd c043bbfd 00e7dfe0 00073eff 00073eff c0c0df48 c0448ce2 00000009 00000000 c0c0df9c c043bbfd 00078d88 Call Trace: [] dump_stack+0x41/0x52 [] warn_slowpath_common+0x7e/0xa0 [] ? __ioremap_caller+0x2ad/0x2c0 [] ? __ioremap_caller+0x2ad/0x2c0 [] warn_slowpath_null+0x22/0x30 [] __ioremap_caller+0x2ad/0x2c0 [] ? acpi_tb_verify_table+0x1c/0x43 [] ? acpi_get_table_with_size+0x63/0xb5 [] ? efi_lookup_mapped_addr+0xe/0xf0 [] ioremap_nocache+0x1b/0x20 [] ? efi_bgrt_init+0x83/0x10c [] efi_bgrt_init+0x83/0x10c [] efi_late_init+0x8/0xa [] start_kernel+0x3ae/0x3c3 [] ? repair_env_string+0x51/0x51 [] i386_start_kernel+0x12e/0x131 Switch to using early_memremap(), which won't trigger this warning, and has the added benefit of more accurately conveying what we're trying to do - map a chunk of memory. This patch addresses the following bug report, https://bugzilla.kernel.org/show_bug.cgi?id=67911 Reported-by: Adam Williamson Cc: Josh Triplett Cc: Matthew Garrett Cc: Rafael J. Wysocki Signed-off-by: Matt Fleming --- arch/x86/platform/efi/efi-bgrt.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/arch/x86/platform/efi/efi-bgrt.c b/arch/x86/platform/efi/efi-bgrt.c index 7145ec63c520..4df9591eadad 100644 --- a/arch/x86/platform/efi/efi-bgrt.c +++ b/arch/x86/platform/efi/efi-bgrt.c @@ -49,7 +49,8 @@ void __init efi_bgrt_init(void) image = efi_lookup_mapped_addr(bgrt_tab->image_address); if (!image) { - image = ioremap(bgrt_tab->image_address, sizeof(bmp_header)); + image = early_memremap(bgrt_tab->image_address, + sizeof(bmp_header)); ioremapped = true; if (!image) return; @@ -57,7 +58,7 @@ void __init efi_bgrt_init(void) memcpy_fromio(&bmp_header, image, sizeof(bmp_header)); if (ioremapped) - iounmap(image); + early_iounmap(image, sizeof(bmp_header)); bgrt_image_size = bmp_header.size; bgrt_image = kmalloc(bgrt_image_size, GFP_KERNEL); @@ -65,7 +66,8 @@ void __init efi_bgrt_init(void) return; if (ioremapped) { - image = ioremap(bgrt_tab->image_address, bmp_header.size); + image = early_memremap(bgrt_tab->image_address, + bmp_header.size); if (!image) { kfree(bgrt_image); bgrt_image = NULL; @@ -75,5 +77,5 @@ void __init efi_bgrt_init(void) memcpy_fromio(bgrt_image, image, bgrt_image_size); if (ioremapped) - iounmap(image); + early_iounmap(image, bmp_header.size); } From 9ab9831b4c59d379a17f92a6157c7e921811dea2 Mon Sep 17 00:00:00 2001 From: Zoltan Kiss Date: Tue, 4 Feb 2014 19:54:37 +0000 Subject: [PATCH 274/596] xen-netback: Fix Rx stall due to race condition The recent patch to fix receive side flow control (11b57f90257c1d6a91cee720151b69e0c2020cf6: xen-netback: stop vif thread spinning if frontend is unresponsive) solved the spinning thread problem, however caused an another one. The receive side can stall, if: - [THREAD] xenvif_rx_action sets rx_queue_stopped to true - [INTERRUPT] interrupt happens, and sets rx_event to true - [THREAD] then xenvif_kthread sets rx_event to false - [THREAD] rx_work_todo doesn't return true anymore Also, if interrupt sent but there is still no room in the ring, it take quite a long time until xenvif_rx_action realize it. This patch ditch that two variable, and rework rx_work_todo. If the thread finds it can't fit more skb's into the ring, it saves the last slot estimation into rx_last_skb_slots, otherwise it's kept as 0. Then rx_work_todo will check if: - there is something to send to the ring (like before) - there is space for the topmost packet in the queue I think that's more natural and optimal thing to test than two bool which are set somewhere else. Signed-off-by: Zoltan Kiss Reviewed-by: Paul Durrant Acked-by: Wei Liu Signed-off-by: David S. Miller --- drivers/net/xen-netback/common.h | 6 +----- drivers/net/xen-netback/interface.c | 1 - drivers/net/xen-netback/netback.c | 16 ++++++---------- 3 files changed, 7 insertions(+), 16 deletions(-) diff --git a/drivers/net/xen-netback/common.h b/drivers/net/xen-netback/common.h index 4c76bcb9a879..ae413a2cbee7 100644 --- a/drivers/net/xen-netback/common.h +++ b/drivers/net/xen-netback/common.h @@ -143,11 +143,7 @@ struct xenvif { char rx_irq_name[IFNAMSIZ+4]; /* DEVNAME-rx */ struct xen_netif_rx_back_ring rx; struct sk_buff_head rx_queue; - bool rx_queue_stopped; - /* Set when the RX interrupt is triggered by the frontend. - * The worker thread may need to wake the queue. - */ - bool rx_event; + RING_IDX rx_last_skb_slots; /* This array is allocated seperately as it is large */ struct gnttab_copy *grant_copy_op; diff --git a/drivers/net/xen-netback/interface.c b/drivers/net/xen-netback/interface.c index b9de31ea7fc4..7669d49a67e2 100644 --- a/drivers/net/xen-netback/interface.c +++ b/drivers/net/xen-netback/interface.c @@ -100,7 +100,6 @@ static irqreturn_t xenvif_rx_interrupt(int irq, void *dev_id) { struct xenvif *vif = dev_id; - vif->rx_event = true; xenvif_kick_thread(vif); return IRQ_HANDLED; diff --git a/drivers/net/xen-netback/netback.c b/drivers/net/xen-netback/netback.c index 6b62c3eb8e18..e5284bca2d90 100644 --- a/drivers/net/xen-netback/netback.c +++ b/drivers/net/xen-netback/netback.c @@ -476,7 +476,6 @@ static void xenvif_rx_action(struct xenvif *vif) unsigned long offset; struct skb_cb_overlay *sco; bool need_to_notify = false; - bool ring_full = false; struct netrx_pending_operations npo = { .copy = vif->grant_copy_op, @@ -486,7 +485,7 @@ static void xenvif_rx_action(struct xenvif *vif) skb_queue_head_init(&rxq); while ((skb = skb_dequeue(&vif->rx_queue)) != NULL) { - int max_slots_needed; + RING_IDX max_slots_needed; int i; /* We need a cheap worse case estimate for the number of @@ -509,9 +508,10 @@ static void xenvif_rx_action(struct xenvif *vif) if (!xenvif_rx_ring_slots_available(vif, max_slots_needed)) { skb_queue_head(&vif->rx_queue, skb); need_to_notify = true; - ring_full = true; + vif->rx_last_skb_slots = max_slots_needed; break; - } + } else + vif->rx_last_skb_slots = 0; sco = (struct skb_cb_overlay *)skb->cb; sco->meta_slots_used = xenvif_gop_skb(skb, &npo); @@ -522,8 +522,6 @@ static void xenvif_rx_action(struct xenvif *vif) BUG_ON(npo.meta_prod > ARRAY_SIZE(vif->meta)); - vif->rx_queue_stopped = !npo.copy_prod && ring_full; - if (!npo.copy_prod) goto done; @@ -1473,8 +1471,8 @@ static struct xen_netif_rx_response *make_rx_response(struct xenvif *vif, static inline int rx_work_todo(struct xenvif *vif) { - return (!skb_queue_empty(&vif->rx_queue) && !vif->rx_queue_stopped) || - vif->rx_event; + return !skb_queue_empty(&vif->rx_queue) && + xenvif_rx_ring_slots_available(vif, vif->rx_last_skb_slots); } static inline int tx_work_todo(struct xenvif *vif) @@ -1560,8 +1558,6 @@ int xenvif_kthread(void *data) if (!skb_queue_empty(&vif->rx_queue)) xenvif_rx_action(vif); - vif->rx_event = false; - if (skb_queue_empty(&vif->rx_queue) && netif_queue_stopped(vif->dev)) xenvif_start_queue(vif); From 8b7ad1bb3d440da888f2a939dc870eba429b9192 Mon Sep 17 00:00:00 2001 From: Dave Airlie Date: Wed, 5 Feb 2014 14:47:45 +1000 Subject: [PATCH 275/596] drm/mgag200,ast,cirrus: fix regression with drm_can_sleep conversion I totally sign inverted my way out of this one. Cc: stable@vger.kernel.org Reported-by: "Sabrina Dubroca" Signed-off-by: Dave Airlie --- drivers/gpu/drm/ast/ast_fb.c | 2 +- drivers/gpu/drm/cirrus/cirrus_fbdev.c | 2 +- drivers/gpu/drm/mgag200/mgag200_fb.c | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/ast/ast_fb.c b/drivers/gpu/drm/ast/ast_fb.c index 3f65dd6676b2..a28640f47c27 100644 --- a/drivers/gpu/drm/ast/ast_fb.c +++ b/drivers/gpu/drm/ast/ast_fb.c @@ -65,7 +65,7 @@ static void ast_dirty_update(struct ast_fbdev *afbdev, * then the BO is being moved and we should * store up the damage until later. */ - if (!drm_can_sleep()) + if (drm_can_sleep()) ret = ast_bo_reserve(bo, true); if (ret) { if (ret != -EBUSY) diff --git a/drivers/gpu/drm/cirrus/cirrus_fbdev.c b/drivers/gpu/drm/cirrus/cirrus_fbdev.c index 2fd4a92162cb..32bbba0a787b 100644 --- a/drivers/gpu/drm/cirrus/cirrus_fbdev.c +++ b/drivers/gpu/drm/cirrus/cirrus_fbdev.c @@ -39,7 +39,7 @@ static void cirrus_dirty_update(struct cirrus_fbdev *afbdev, * then the BO is being moved and we should * store up the damage until later. */ - if (!drm_can_sleep()) + if (drm_can_sleep()) ret = cirrus_bo_reserve(bo, true); if (ret) { if (ret != -EBUSY) diff --git a/drivers/gpu/drm/mgag200/mgag200_fb.c b/drivers/gpu/drm/mgag200/mgag200_fb.c index f9adc27ef32a..13b7dd83faa9 100644 --- a/drivers/gpu/drm/mgag200/mgag200_fb.c +++ b/drivers/gpu/drm/mgag200/mgag200_fb.c @@ -41,7 +41,7 @@ static void mga_dirty_update(struct mga_fbdev *mfbdev, * then the BO is being moved and we should * store up the damage until later. */ - if (!drm_can_sleep()) + if (drm_can_sleep()) ret = mgag200_bo_reserve(bo, true); if (ret) { if (ret != -EBUSY) From 7c4c62a04a2a80e3feb5d6c97aca1e413b11c790 Mon Sep 17 00:00:00 2001 From: Dave Airlie Date: Thu, 30 Jan 2014 14:11:12 +1000 Subject: [PATCH 276/596] drm/radeon: allow geom rings to be setup on r600/r700 (v2) the evergreen CS parser has allowed this for a while, just port the code to the r600 one. This is required before geom shaders can be made work. v2: agd5f: minor cleanup and add additional 7xx reg. Signed-off-by: Dave Airlie Signed-off-by: Alex Deucher Signed-off-by: Dave Airlie --- drivers/gpu/drm/radeon/r600_cs.c | 18 ++++++++++++++++-- drivers/gpu/drm/radeon/radeon_drv.c | 3 ++- drivers/gpu/drm/radeon/reg_srcs/r600 | 1 + 3 files changed, 19 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/radeon/r600_cs.c b/drivers/gpu/drm/radeon/r600_cs.c index 7b399dc5fd54..2812c7d1ae6f 100644 --- a/drivers/gpu/drm/radeon/r600_cs.c +++ b/drivers/gpu/drm/radeon/r600_cs.c @@ -1007,8 +1007,22 @@ static int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx) case R_008C64_SQ_VSTMP_RING_SIZE: case R_0288C8_SQ_GS_VERT_ITEMSIZE: /* get value to populate the IB don't remove */ - tmp =radeon_get_ib_value(p, idx); - ib[idx] = 0; + /*tmp =radeon_get_ib_value(p, idx); + ib[idx] = 0;*/ + break; + case SQ_ESGS_RING_BASE: + case SQ_GSVS_RING_BASE: + case SQ_ESTMP_RING_BASE: + case SQ_GSTMP_RING_BASE: + case SQ_PSTMP_RING_BASE: + case SQ_VSTMP_RING_BASE: + r = radeon_cs_packet_next_reloc(p, &reloc, 0); + if (r) { + dev_warn(p->dev, "bad SET_CONTEXT_REG " + "0x%04X\n", reg); + return -EINVAL; + } + ib[idx] += (u32)((reloc->lobj.gpu_offset >> 8) & 0xffffffff); break; case SQ_CONFIG: track->sq_config = radeon_get_ib_value(p, idx); diff --git a/drivers/gpu/drm/radeon/radeon_drv.c b/drivers/gpu/drm/radeon/radeon_drv.c index ec8c388eec17..84a1bbb75f91 100644 --- a/drivers/gpu/drm/radeon/radeon_drv.c +++ b/drivers/gpu/drm/radeon/radeon_drv.c @@ -78,9 +78,10 @@ * 2.34.0 - Add CIK tiling mode array query * 2.35.0 - Add CIK macrotile mode array query * 2.36.0 - Fix CIK DCE tiling setup + * 2.37.0 - allow GS ring setup on r6xx/r7xx */ #define KMS_DRIVER_MAJOR 2 -#define KMS_DRIVER_MINOR 36 +#define KMS_DRIVER_MINOR 37 #define KMS_DRIVER_PATCHLEVEL 0 int radeon_driver_load_kms(struct drm_device *dev, unsigned long flags); int radeon_driver_unload_kms(struct drm_device *dev); diff --git a/drivers/gpu/drm/radeon/reg_srcs/r600 b/drivers/gpu/drm/radeon/reg_srcs/r600 index 20bfbda7b3f1..ec0c6829c1dc 100644 --- a/drivers/gpu/drm/radeon/reg_srcs/r600 +++ b/drivers/gpu/drm/radeon/reg_srcs/r600 @@ -18,6 +18,7 @@ r600 0x9400 0x00028A3C VGT_GROUP_VECT_1_FMT_CNTL 0x00028A40 VGT_GS_MODE 0x00028A6C VGT_GS_OUT_PRIM_TYPE +0x00028B38 VGT_GS_MAX_VERT_OUT 0x000088C8 VGT_GS_PER_ES 0x000088E8 VGT_GS_PER_VS 0x000088D4 VGT_GS_VERTEX_REUSE From cc4723ca316742891954efa346298e7c747c0d17 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 5 Feb 2014 15:03:38 +0000 Subject: [PATCH 277/596] netfilter: nft_reject: split up reject module into IPv4 and IPv6 specifc parts Currently the nft_reject module depends on symbols from ipv6. This is wrong since no generic module should force IPv6 support to be loaded. Split up the module into AF-specific and a generic part. Signed-off-by: Patrick McHardy Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nft_reject.h | 17 ++++++ net/ipv4/netfilter/Kconfig | 5 ++ net/ipv4/netfilter/Makefile | 1 + net/ipv4/netfilter/nft_reject_ipv4.c | 74 +++++++++++++++++++++++ net/ipv6/netfilter/Kconfig | 5 ++ net/ipv6/netfilter/Makefile | 1 + net/ipv6/netfilter/nft_reject_ipv6.c | 75 +++++++++++++++++++++++ net/netfilter/Kconfig | 1 - net/netfilter/nft_reject.c | 89 +++------------------------- 9 files changed, 187 insertions(+), 81 deletions(-) create mode 100644 include/net/netfilter/nft_reject.h create mode 100644 net/ipv4/netfilter/nft_reject_ipv4.c create mode 100644 net/ipv6/netfilter/nft_reject_ipv6.c diff --git a/include/net/netfilter/nft_reject.h b/include/net/netfilter/nft_reject.h new file mode 100644 index 000000000000..ecda75945e77 --- /dev/null +++ b/include/net/netfilter/nft_reject.h @@ -0,0 +1,17 @@ +#ifndef _NFT_REJECT_H_ +#define _NFT_REJECT_H_ + +struct nft_reject { + enum nft_reject_types type:8; + u8 icmp_code; +}; + +extern const struct nla_policy nft_reject_policy[]; + +int nft_reject_init(const struct nft_ctx *ctx, + const struct nft_expr *expr, + const struct nlattr * const tb[]); + +int nft_reject_dump(struct sk_buff *skb, const struct nft_expr *expr); + +#endif diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig index 81c6910cfa92..a26ce035e3fa 100644 --- a/net/ipv4/netfilter/Kconfig +++ b/net/ipv4/netfilter/Kconfig @@ -61,6 +61,11 @@ config NFT_CHAIN_NAT_IPV4 packet transformations such as the source, destination address and source and destination ports. +config NFT_REJECT_IPV4 + depends on NF_TABLES_IPV4 + default NFT_REJECT + tristate + config NF_TABLES_ARP depends on NF_TABLES tristate "ARP nf_tables support" diff --git a/net/ipv4/netfilter/Makefile b/net/ipv4/netfilter/Makefile index c16be9d58420..90b82405331e 100644 --- a/net/ipv4/netfilter/Makefile +++ b/net/ipv4/netfilter/Makefile @@ -30,6 +30,7 @@ obj-$(CONFIG_NF_NAT_PROTO_GRE) += nf_nat_proto_gre.o obj-$(CONFIG_NF_TABLES_IPV4) += nf_tables_ipv4.o obj-$(CONFIG_NFT_CHAIN_ROUTE_IPV4) += nft_chain_route_ipv4.o obj-$(CONFIG_NFT_CHAIN_NAT_IPV4) += nft_chain_nat_ipv4.o +obj-$(CONFIG_NFT_REJECT_IPV4) += nft_reject_ipv4.o obj-$(CONFIG_NF_TABLES_ARP) += nf_tables_arp.o # generic IP tables diff --git a/net/ipv4/netfilter/nft_reject_ipv4.c b/net/ipv4/netfilter/nft_reject_ipv4.c new file mode 100644 index 000000000000..e935d8de1182 --- /dev/null +++ b/net/ipv4/netfilter/nft_reject_ipv4.c @@ -0,0 +1,74 @@ +/* + * Copyright (c) 2008-2009 Patrick McHardy + * Copyright (c) 2013 Eric Leblond + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * Development of this code funded by Astaro AG (http://www.astaro.com/) + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +static void nft_reject_ipv4_eval(const struct nft_expr *expr, + struct nft_data data[NFT_REG_MAX + 1], + const struct nft_pktinfo *pkt) +{ + struct nft_reject *priv = nft_expr_priv(expr); + + switch (priv->type) { + case NFT_REJECT_ICMP_UNREACH: + nf_send_unreach(pkt->skb, priv->icmp_code); + break; + case NFT_REJECT_TCP_RST: + nf_send_reset(pkt->skb, pkt->ops->hooknum); + break; + } + + data[NFT_REG_VERDICT].verdict = NF_DROP; +} + +static struct nft_expr_type nft_reject_ipv4_type; +static const struct nft_expr_ops nft_reject_ipv4_ops = { + .type = &nft_reject_ipv4_type, + .size = NFT_EXPR_SIZE(sizeof(struct nft_reject)), + .eval = nft_reject_ipv4_eval, + .init = nft_reject_init, + .dump = nft_reject_dump, +}; + +static struct nft_expr_type nft_reject_ipv4_type __read_mostly = { + .family = NFPROTO_IPV4, + .name = "reject", + .ops = &nft_reject_ipv4_ops, + .policy = nft_reject_policy, + .maxattr = NFTA_REJECT_MAX, + .owner = THIS_MODULE, +}; + +static int __init nft_reject_ipv4_module_init(void) +{ + return nft_register_expr(&nft_reject_ipv4_type); +} + +static void __exit nft_reject_ipv4_module_exit(void) +{ + nft_unregister_expr(&nft_reject_ipv4_type); +} + +module_init(nft_reject_ipv4_module_init); +module_exit(nft_reject_ipv4_module_exit); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Patrick McHardy "); +MODULE_ALIAS_NFT_AF_EXPR(AF_INET, "reject"); diff --git a/net/ipv6/netfilter/Kconfig b/net/ipv6/netfilter/Kconfig index 35750df744dc..4bff1f297e39 100644 --- a/net/ipv6/netfilter/Kconfig +++ b/net/ipv6/netfilter/Kconfig @@ -50,6 +50,11 @@ config NFT_CHAIN_NAT_IPV6 packet transformations such as the source, destination address and source and destination ports. +config NFT_REJECT_IPV6 + depends on NF_TABLES_IPV6 + default NFT_REJECT + tristate + config IP6_NF_IPTABLES tristate "IP6 tables support (required for filtering)" depends on INET && IPV6 diff --git a/net/ipv6/netfilter/Makefile b/net/ipv6/netfilter/Makefile index d1b4928f34f7..70d3dd66f2cd 100644 --- a/net/ipv6/netfilter/Makefile +++ b/net/ipv6/netfilter/Makefile @@ -27,6 +27,7 @@ obj-$(CONFIG_NF_DEFRAG_IPV6) += nf_defrag_ipv6.o obj-$(CONFIG_NF_TABLES_IPV6) += nf_tables_ipv6.o obj-$(CONFIG_NFT_CHAIN_ROUTE_IPV6) += nft_chain_route_ipv6.o obj-$(CONFIG_NFT_CHAIN_NAT_IPV6) += nft_chain_nat_ipv6.o +obj-$(CONFIG_NFT_REJECT_IPV6) += nft_reject_ipv6.o # matches obj-$(CONFIG_IP6_NF_MATCH_AH) += ip6t_ah.o diff --git a/net/ipv6/netfilter/nft_reject_ipv6.c b/net/ipv6/netfilter/nft_reject_ipv6.c new file mode 100644 index 000000000000..f73285924144 --- /dev/null +++ b/net/ipv6/netfilter/nft_reject_ipv6.c @@ -0,0 +1,75 @@ +/* + * Copyright (c) 2008-2009 Patrick McHardy + * Copyright (c) 2013 Eric Leblond + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * Development of this code funded by Astaro AG (http://www.astaro.com/) + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include + +static void nft_reject_ipv6_eval(const struct nft_expr *expr, + struct nft_data data[NFT_REG_MAX + 1], + const struct nft_pktinfo *pkt) +{ + struct nft_reject *priv = nft_expr_priv(expr); + struct net *net = dev_net((pkt->in != NULL) ? pkt->in : pkt->out); + + switch (priv->type) { + case NFT_REJECT_ICMP_UNREACH: + nf_send_unreach6(net, pkt->skb, priv->icmp_code, + pkt->ops->hooknum); + break; + case NFT_REJECT_TCP_RST: + nf_send_reset6(net, pkt->skb, pkt->ops->hooknum); + break; + } + + data[NFT_REG_VERDICT].verdict = NF_DROP; +} + +static struct nft_expr_type nft_reject_ipv6_type; +static const struct nft_expr_ops nft_reject_ipv6_ops = { + .type = &nft_reject_ipv6_type, + .size = NFT_EXPR_SIZE(sizeof(struct nft_reject)), + .eval = nft_reject_ipv6_eval, + .init = nft_reject_init, + .dump = nft_reject_dump, +}; + +static struct nft_expr_type nft_reject_ipv6_type __read_mostly = { + .family = NFPROTO_IPV6, + .name = "reject", + .ops = &nft_reject_ipv6_ops, + .policy = nft_reject_policy, + .maxattr = NFTA_REJECT_MAX, + .owner = THIS_MODULE, +}; + +static int __init nft_reject_ipv6_module_init(void) +{ + return nft_register_expr(&nft_reject_ipv6_type); +} + +static void __exit nft_reject_ipv6_module_exit(void) +{ + nft_unregister_expr(&nft_reject_ipv6_type); +} + +module_init(nft_reject_ipv6_module_init); +module_exit(nft_reject_ipv6_module_exit); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Patrick McHardy "); +MODULE_ALIAS_NFT_AF_EXPR(AF_INET6, "reject"); diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index c37467562fd0..ed8b50e62276 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -513,7 +513,6 @@ config NFT_QUEUE config NFT_REJECT depends on NF_TABLES - depends on NF_TABLES_IPV6 || !NF_TABLES_IPV6 default m if NETFILTER_ADVANCED=n tristate "Netfilter nf_tables reject support" help diff --git a/net/netfilter/nft_reject.c b/net/netfilter/nft_reject.c index 5e204711d704..f3448c296446 100644 --- a/net/netfilter/nft_reject.c +++ b/net/netfilter/nft_reject.c @@ -16,65 +16,23 @@ #include #include #include -#include -#include +#include -#if IS_ENABLED(CONFIG_NF_TABLES_IPV6) -#include -#endif - -struct nft_reject { - enum nft_reject_types type:8; - u8 icmp_code; - u8 family; -}; - -static void nft_reject_eval(const struct nft_expr *expr, - struct nft_data data[NFT_REG_MAX + 1], - const struct nft_pktinfo *pkt) -{ - struct nft_reject *priv = nft_expr_priv(expr); -#if IS_ENABLED(CONFIG_NF_TABLES_IPV6) - struct net *net = dev_net((pkt->in != NULL) ? pkt->in : pkt->out); -#endif - switch (priv->type) { - case NFT_REJECT_ICMP_UNREACH: - if (priv->family == NFPROTO_IPV4) - nf_send_unreach(pkt->skb, priv->icmp_code); -#if IS_ENABLED(CONFIG_NF_TABLES_IPV6) - else if (priv->family == NFPROTO_IPV6) - nf_send_unreach6(net, pkt->skb, priv->icmp_code, - pkt->ops->hooknum); -#endif - break; - case NFT_REJECT_TCP_RST: - if (priv->family == NFPROTO_IPV4) - nf_send_reset(pkt->skb, pkt->ops->hooknum); -#if IS_ENABLED(CONFIG_NF_TABLES_IPV6) - else if (priv->family == NFPROTO_IPV6) - nf_send_reset6(net, pkt->skb, pkt->ops->hooknum); -#endif - break; - } - - data[NFT_REG_VERDICT].verdict = NF_DROP; -} - -static const struct nla_policy nft_reject_policy[NFTA_REJECT_MAX + 1] = { +const struct nla_policy nft_reject_policy[NFTA_REJECT_MAX + 1] = { [NFTA_REJECT_TYPE] = { .type = NLA_U32 }, [NFTA_REJECT_ICMP_CODE] = { .type = NLA_U8 }, }; +EXPORT_SYMBOL_GPL(nft_reject_policy); -static int nft_reject_init(const struct nft_ctx *ctx, - const struct nft_expr *expr, - const struct nlattr * const tb[]) +int nft_reject_init(const struct nft_ctx *ctx, + const struct nft_expr *expr, + const struct nlattr * const tb[]) { struct nft_reject *priv = nft_expr_priv(expr); if (tb[NFTA_REJECT_TYPE] == NULL) return -EINVAL; - priv->family = ctx->afi->family; priv->type = ntohl(nla_get_be32(tb[NFTA_REJECT_TYPE])); switch (priv->type) { case NFT_REJECT_ICMP_UNREACH: @@ -89,8 +47,9 @@ static int nft_reject_init(const struct nft_ctx *ctx, return 0; } +EXPORT_SYMBOL_GPL(nft_reject_init); -static int nft_reject_dump(struct sk_buff *skb, const struct nft_expr *expr) +int nft_reject_dump(struct sk_buff *skb, const struct nft_expr *expr) { const struct nft_reject *priv = nft_expr_priv(expr); @@ -109,37 +68,7 @@ static int nft_reject_dump(struct sk_buff *skb, const struct nft_expr *expr) nla_put_failure: return -1; } - -static struct nft_expr_type nft_reject_type; -static const struct nft_expr_ops nft_reject_ops = { - .type = &nft_reject_type, - .size = NFT_EXPR_SIZE(sizeof(struct nft_reject)), - .eval = nft_reject_eval, - .init = nft_reject_init, - .dump = nft_reject_dump, -}; - -static struct nft_expr_type nft_reject_type __read_mostly = { - .name = "reject", - .ops = &nft_reject_ops, - .policy = nft_reject_policy, - .maxattr = NFTA_REJECT_MAX, - .owner = THIS_MODULE, -}; - -static int __init nft_reject_module_init(void) -{ - return nft_register_expr(&nft_reject_type); -} - -static void __exit nft_reject_module_exit(void) -{ - nft_unregister_expr(&nft_reject_type); -} - -module_init(nft_reject_module_init); -module_exit(nft_reject_module_exit); +EXPORT_SYMBOL_GPL(nft_reject_dump); MODULE_LICENSE("GPL"); MODULE_AUTHOR("Patrick McHardy "); -MODULE_ALIAS_NFT_EXPR("reject"); From 05513e9e33dbded8124567466a444d32173eecc6 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 5 Feb 2014 15:03:39 +0000 Subject: [PATCH 278/596] netfilter: nf_tables: add reject module for NFPROTO_INET Add a reject module for NFPROTO_INET. It does nothing but dispatch to the AF-specific modules based on the hook family. Signed-off-by: Patrick McHardy Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nft_reject.h | 8 ++++ net/ipv4/netfilter/nft_reject_ipv4.c | 7 ++-- net/ipv6/netfilter/nft_reject_ipv6.c | 7 ++-- net/netfilter/Kconfig | 5 +++ net/netfilter/Makefile | 1 + net/netfilter/nft_reject_inet.c | 63 ++++++++++++++++++++++++++++ 6 files changed, 85 insertions(+), 6 deletions(-) create mode 100644 net/netfilter/nft_reject_inet.c diff --git a/include/net/netfilter/nft_reject.h b/include/net/netfilter/nft_reject.h index ecda75945e77..36b0da2d55bb 100644 --- a/include/net/netfilter/nft_reject.h +++ b/include/net/netfilter/nft_reject.h @@ -14,4 +14,12 @@ int nft_reject_init(const struct nft_ctx *ctx, int nft_reject_dump(struct sk_buff *skb, const struct nft_expr *expr); +void nft_reject_ipv4_eval(const struct nft_expr *expr, + struct nft_data data[NFT_REG_MAX + 1], + const struct nft_pktinfo *pkt); + +void nft_reject_ipv6_eval(const struct nft_expr *expr, + struct nft_data data[NFT_REG_MAX + 1], + const struct nft_pktinfo *pkt); + #endif diff --git a/net/ipv4/netfilter/nft_reject_ipv4.c b/net/ipv4/netfilter/nft_reject_ipv4.c index e935d8de1182..e79718a382f2 100644 --- a/net/ipv4/netfilter/nft_reject_ipv4.c +++ b/net/ipv4/netfilter/nft_reject_ipv4.c @@ -20,9 +20,9 @@ #include #include -static void nft_reject_ipv4_eval(const struct nft_expr *expr, - struct nft_data data[NFT_REG_MAX + 1], - const struct nft_pktinfo *pkt) +void nft_reject_ipv4_eval(const struct nft_expr *expr, + struct nft_data data[NFT_REG_MAX + 1], + const struct nft_pktinfo *pkt) { struct nft_reject *priv = nft_expr_priv(expr); @@ -37,6 +37,7 @@ static void nft_reject_ipv4_eval(const struct nft_expr *expr, data[NFT_REG_VERDICT].verdict = NF_DROP; } +EXPORT_SYMBOL_GPL(nft_reject_ipv4_eval); static struct nft_expr_type nft_reject_ipv4_type; static const struct nft_expr_ops nft_reject_ipv4_ops = { diff --git a/net/ipv6/netfilter/nft_reject_ipv6.c b/net/ipv6/netfilter/nft_reject_ipv6.c index f73285924144..0bc19fa87821 100644 --- a/net/ipv6/netfilter/nft_reject_ipv6.c +++ b/net/ipv6/netfilter/nft_reject_ipv6.c @@ -19,9 +19,9 @@ #include #include -static void nft_reject_ipv6_eval(const struct nft_expr *expr, - struct nft_data data[NFT_REG_MAX + 1], - const struct nft_pktinfo *pkt) +void nft_reject_ipv6_eval(const struct nft_expr *expr, + struct nft_data data[NFT_REG_MAX + 1], + const struct nft_pktinfo *pkt) { struct nft_reject *priv = nft_expr_priv(expr); struct net *net = dev_net((pkt->in != NULL) ? pkt->in : pkt->out); @@ -38,6 +38,7 @@ static void nft_reject_ipv6_eval(const struct nft_expr *expr, data[NFT_REG_VERDICT].verdict = NF_DROP; } +EXPORT_SYMBOL_GPL(nft_reject_ipv6_eval); static struct nft_expr_type nft_reject_ipv6_type; static const struct nft_expr_ops nft_reject_ipv6_ops = { diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index ed8b50e62276..e9410d17619d 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -520,6 +520,11 @@ config NFT_REJECT explicitly deny and notify via TCP reset/ICMP informational errors unallowed traffic. +config NFT_REJECT_INET + depends on NF_TABLES_INET + default NFT_REJECT + tristate + config NFT_COMPAT depends on NF_TABLES depends on NETFILTER_XTABLES diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index ee9c4de5f8ed..bffdad774da7 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -79,6 +79,7 @@ obj-$(CONFIG_NFT_LIMIT) += nft_limit.o obj-$(CONFIG_NFT_NAT) += nft_nat.o obj-$(CONFIG_NFT_QUEUE) += nft_queue.o obj-$(CONFIG_NFT_REJECT) += nft_reject.o +obj-$(CONFIG_NFT_REJECT_INET) += nft_reject_inet.o obj-$(CONFIG_NFT_RBTREE) += nft_rbtree.o obj-$(CONFIG_NFT_HASH) += nft_hash.o obj-$(CONFIG_NFT_COUNTER) += nft_counter.o diff --git a/net/netfilter/nft_reject_inet.c b/net/netfilter/nft_reject_inet.c new file mode 100644 index 000000000000..8a310f239c93 --- /dev/null +++ b/net/netfilter/nft_reject_inet.c @@ -0,0 +1,63 @@ +/* + * Copyright (c) 2014 Patrick McHardy + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include +#include +#include +#include +#include +#include +#include +#include + +static void nft_reject_inet_eval(const struct nft_expr *expr, + struct nft_data data[NFT_REG_MAX + 1], + const struct nft_pktinfo *pkt) +{ + switch (pkt->ops->pf) { + case NFPROTO_IPV4: + nft_reject_ipv4_eval(expr, data, pkt); + case NFPROTO_IPV6: + nft_reject_ipv6_eval(expr, data, pkt); + } +} + +static struct nft_expr_type nft_reject_inet_type; +static const struct nft_expr_ops nft_reject_inet_ops = { + .type = &nft_reject_inet_type, + .size = NFT_EXPR_SIZE(sizeof(struct nft_reject)), + .eval = nft_reject_inet_eval, + .init = nft_reject_init, + .dump = nft_reject_dump, +}; + +static struct nft_expr_type nft_reject_inet_type __read_mostly = { + .family = NFPROTO_INET, + .name = "reject", + .ops = &nft_reject_inet_ops, + .policy = nft_reject_policy, + .maxattr = NFTA_REJECT_MAX, + .owner = THIS_MODULE, +}; + +static int __init nft_reject_inet_module_init(void) +{ + return nft_register_expr(&nft_reject_inet_type); +} + +static void __exit nft_reject_inet_module_exit(void) +{ + nft_unregister_expr(&nft_reject_inet_type); +} + +module_init(nft_reject_inet_module_init); +module_exit(nft_reject_inet_module_exit); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Patrick McHardy "); +MODULE_ALIAS_NFT_AF_EXPR(1, "reject"); From 2f617435c3a6fe3f39efb9ae2baa77de2d6c97b8 Mon Sep 17 00:00:00 2001 From: Eliad Peller Date: Sun, 12 Jan 2014 11:06:37 +0200 Subject: [PATCH 279/596] mac80211: move roc cookie assignment earlier ieee80211_start_roc_work() might add a new roc to existing roc, and tell cfg80211 it has already started. However, this might happen before the roc cookie was set, resulting in REMAIN_ON_CHANNEL (started) event with null cookie. Consequently, it can make wpa_supplicant go out of sync. Fix it by setting the roc cookie earlier. Cc: stable@vger.kernel.org Signed-off-by: Eliad Peller Signed-off-by: Johannes Berg --- net/mac80211/cfg.c | 36 ++++++++++++++++++------------------ 1 file changed, 18 insertions(+), 18 deletions(-) diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c index f9ae9b85d4c1..94b4acb5aabb 100644 --- a/net/mac80211/cfg.c +++ b/net/mac80211/cfg.c @@ -2638,6 +2638,24 @@ static int ieee80211_start_roc_work(struct ieee80211_local *local, INIT_DELAYED_WORK(&roc->work, ieee80211_sw_roc_work); INIT_LIST_HEAD(&roc->dependents); + /* + * cookie is either the roc cookie (for normal roc) + * or the SKB (for mgmt TX) + */ + if (!txskb) { + /* local->mtx protects this */ + local->roc_cookie_counter++; + roc->cookie = local->roc_cookie_counter; + /* wow, you wrapped 64 bits ... more likely a bug */ + if (WARN_ON(roc->cookie == 0)) { + roc->cookie = 1; + local->roc_cookie_counter++; + } + *cookie = roc->cookie; + } else { + *cookie = (unsigned long)txskb; + } + /* if there's one pending or we're scanning, queue this one */ if (!list_empty(&local->roc_list) || local->scanning || local->radar_detect_enabled) @@ -2772,24 +2790,6 @@ static int ieee80211_start_roc_work(struct ieee80211_local *local, if (!queued) list_add_tail(&roc->list, &local->roc_list); - /* - * cookie is either the roc cookie (for normal roc) - * or the SKB (for mgmt TX) - */ - if (!txskb) { - /* local->mtx protects this */ - local->roc_cookie_counter++; - roc->cookie = local->roc_cookie_counter; - /* wow, you wrapped 64 bits ... more likely a bug */ - if (WARN_ON(roc->cookie == 0)) { - roc->cookie = 1; - local->roc_cookie_counter++; - } - *cookie = roc->cookie; - } else { - *cookie = (unsigned long)txskb; - } - return 0; } From f12cb2893069495726c21a4b0178705dacfecfe0 Mon Sep 17 00:00:00 2001 From: Pontus Fuchs Date: Thu, 16 Jan 2014 15:00:40 +0100 Subject: [PATCH 280/596] nl80211: Reset split_start when netlink skb is exhausted When the netlink skb is exhausted split_start is left set. In the subsequent retry, with a larger buffer, the dump is continued from the failing point instead of from the beginning. This was causing my rt28xx based USB dongle to now show up when running "iw list" with an old iw version without split dump support. Cc: stable@vger.kernel.org Fixes: 3713b4e364ef ("nl80211: allow splitting wiphy information in dumps") Signed-off-by: Pontus Fuchs [avoid the entire workaround when state->split is set] Signed-off-by: Johannes Berg --- net/wireless/nl80211.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index 7a742594916e..6ea960b1a8eb 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -1719,9 +1719,10 @@ static int nl80211_dump_wiphy(struct sk_buff *skb, struct netlink_callback *cb) * We can then retry with the larger buffer. */ if ((ret == -ENOBUFS || ret == -EMSGSIZE) && - !skb->len && + !skb->len && !state->split && cb->min_dump_alloc < 4096) { cb->min_dump_alloc = 4096; + state->split_start = 0; rtnl_unlock(); return 1; } From 5a6aa705ffdd97552ff756bbfa7d5a3b62a6c690 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Thu, 23 Jan 2014 16:32:29 +0100 Subject: [PATCH 281/596] cfg80211: re-enable 5/10 MHz support Unfortunately I forgot this during the merge window, but the patch seems small enough to go in as a fix. The userspace API bug that was the reason for disabling it has long been fixed. Signed-off-by: Johannes Berg --- net/wireless/core.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/net/wireless/core.c b/net/wireless/core.c index d89dee2259b5..77fe4c791269 100644 --- a/net/wireless/core.c +++ b/net/wireless/core.c @@ -440,9 +440,6 @@ int wiphy_register(struct wiphy *wiphy) int i; u16 ifmodes = wiphy->interface_modes; - /* support for 5/10 MHz is broken due to nl80211 API mess - disable */ - wiphy->flags &= ~WIPHY_FLAG_SUPPORTS_5_10_MHZ; - /* * There are major locking problems in nl80211/mac80211 for CSA, * disable for all drivers until this has been reworked. From 8ffcc704c963b4157391bd87a4544cdfd18b574d Mon Sep 17 00:00:00 2001 From: Emmanuel Grumbach Date: Thu, 23 Jan 2014 14:28:16 +0200 Subject: [PATCH 282/596] mac80211: avoid deadlock revealed by lockdep MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit sdata->u.ap.request_smps_work can’t be flushed synchronously under wdev_lock(wdev) since ieee80211_request_smps_ap_work itself locks the same lock. While at it, reset the driver_smps_mode when the ap is stopped to its default: OFF. This solves: ====================================================== [ INFO: possible circular locking dependency detected ] 3.12.0-ipeer+ #2 Tainted: G O ------------------------------------------------------- rmmod/2867 is trying to acquire lock: ((&sdata->u.ap.request_smps_work)){+.+...}, at: [] flush_work+0x0/0x90 but task is already holding lock: (&wdev->mtx){+.+.+.}, at: [] cfg80211_stop_ap+0x26/0x230 [cfg80211] which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #1 (&wdev->mtx){+.+.+.}: [] lock_acquire+0x79/0xe0 [] mutex_lock_nested+0x4a/0x360 [] ieee80211_request_smps_ap_work+0x2b/0x50 [mac80211] [] process_one_work+0x198/0x450 [] worker_thread+0xf9/0x320 [] kthread+0x9f/0xb0 [] ret_from_kernel_thread+0x1b/0x28 -> #0 ((&sdata->u.ap.request_smps_work)){+.+...}: [] __lock_acquire+0x183f/0x1910 [] lock_acquire+0x79/0xe0 [] flush_work+0x47/0x90 [] __cancel_work_timer+0x67/0xe0 [] cancel_work_sync+0xf/0x20 [] ieee80211_stop_ap+0x8c/0x340 [mac80211] [] cfg80211_stop_ap+0x8c/0x230 [cfg80211] [] cfg80211_leave+0x79/0x100 [cfg80211] [] cfg80211_netdev_notifier_call+0xf2/0x4f0 [cfg80211] [] notifier_call_chain+0x59/0x130 [] __raw_notifier_call_chain+0x1e/0x30 [] raw_notifier_call_chain+0x1f/0x30 [] call_netdevice_notifiers_info+0x33/0x70 [] call_netdevice_notifiers+0x13/0x20 [] __dev_close_many+0x34/0xb0 [] dev_close_many+0x6e/0xc0 [] rollback_registered_many+0xa7/0x1f0 [] unregister_netdevice_many+0x14/0x60 [] ieee80211_remove_interfaces+0xe9/0x170 [mac80211] [] ieee80211_unregister_hw+0x56/0x110 [mac80211] [] iwl_op_mode_mvm_stop+0x26/0xe0 [iwlmvm] [] _iwl_op_mode_stop+0x3a/0x70 [iwlwifi] [] iwl_opmode_deregister+0x6f/0x90 [iwlwifi] [] __exit_compat+0xd/0x19 [iwlmvm] [] SyS_delete_module+0x179/0x2b0 [] sysenter_do_call+0x12/0x32 Fixes: 687da132234f ("mac80211: implement SMPS for AP") Cc: [3.13] Reported-by: Ilan Peer Signed-off-by: Emmanuel Grumbach Signed-off-by: Johannes Berg --- net/mac80211/cfg.c | 3 +-- net/mac80211/ht.c | 4 +++- net/mac80211/iface.c | 15 +++++++++++---- 3 files changed, 15 insertions(+), 7 deletions(-) diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c index 94b4acb5aabb..33acdca4a1df 100644 --- a/net/mac80211/cfg.c +++ b/net/mac80211/cfg.c @@ -1090,8 +1090,6 @@ static int ieee80211_stop_ap(struct wiphy *wiphy, struct net_device *dev) kfree(sdata->u.ap.next_beacon); sdata->u.ap.next_beacon = NULL; - cancel_work_sync(&sdata->u.ap.request_smps_work); - /* turn off carrier for this interface and dependent VLANs */ list_for_each_entry(vlan, &sdata->u.ap.vlans, u.vlan.list) netif_carrier_off(vlan->dev); @@ -1103,6 +1101,7 @@ static int ieee80211_stop_ap(struct wiphy *wiphy, struct net_device *dev) kfree_rcu(old_beacon, rcu_head); if (old_probe_resp) kfree_rcu(old_probe_resp, rcu_head); + sdata->u.ap.driver_smps_mode = IEEE80211_SMPS_OFF; __sta_info_flush(sdata, true); ieee80211_free_keys(sdata, true); diff --git a/net/mac80211/ht.c b/net/mac80211/ht.c index fab7b91923e0..70dd013de836 100644 --- a/net/mac80211/ht.c +++ b/net/mac80211/ht.c @@ -466,7 +466,9 @@ void ieee80211_request_smps_ap_work(struct work_struct *work) u.ap.request_smps_work); sdata_lock(sdata); - __ieee80211_request_smps_ap(sdata, sdata->u.ap.driver_smps_mode); + if (sdata_dereference(sdata->u.ap.beacon, sdata)) + __ieee80211_request_smps_ap(sdata, + sdata->u.ap.driver_smps_mode); sdata_unlock(sdata); } diff --git a/net/mac80211/iface.c b/net/mac80211/iface.c index 3dfd20a453ab..ae2eb148a028 100644 --- a/net/mac80211/iface.c +++ b/net/mac80211/iface.c @@ -770,12 +770,19 @@ static void ieee80211_do_stop(struct ieee80211_sub_if_data *sdata, ieee80211_roc_purge(local, sdata); - if (sdata->vif.type == NL80211_IFTYPE_STATION) + switch (sdata->vif.type) { + case NL80211_IFTYPE_STATION: ieee80211_mgd_stop(sdata); - - if (sdata->vif.type == NL80211_IFTYPE_ADHOC) + break; + case NL80211_IFTYPE_ADHOC: ieee80211_ibss_stop(sdata); - + break; + case NL80211_IFTYPE_AP: + cancel_work_sync(&sdata->u.ap.request_smps_work); + break; + default: + break; + } /* * Remove all stations associated with this interface. From a617302c531eaf497ccd02a61d380efc119ba999 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 22 Jan 2014 11:14:18 +0200 Subject: [PATCH 283/596] cfg80211: fix scan done race When an interface/wdev is removed, any ongoing scan should be cancelled by the driver. This will make it call cfg80211, which only queues a work struct. If interface/wdev removal is quick enough, this can leave the scan request pending and processed only after the interface is gone, causing a use-after-free. Fix this by making sure the scan request is not pending after the interface is destroyed. We can't flush or cancel the work item due to locking concerns, but when it'll run it shouldn't find anything to do. This leaves a potential issue, if a new scan gets requested before the work runs, it prematurely stops the running scan, potentially causing another crash. I'll fix that in the next patch. This was particularly observed with P2P_DEVICE wdevs, likely because freeing them is quicker than freeing netdevs. Reported-by: Andrei Otcheretianski Fixes: 4a58e7c38443 ("cfg80211: don't "leak" uncompleted scans") Signed-off-by: Johannes Berg --- net/wireless/core.c | 14 ++++++++++---- 1 file changed, 10 insertions(+), 4 deletions(-) diff --git a/net/wireless/core.c b/net/wireless/core.c index 77fe4c791269..02ed00dbf2df 100644 --- a/net/wireless/core.c +++ b/net/wireless/core.c @@ -203,8 +203,11 @@ void cfg80211_stop_p2p_device(struct cfg80211_registered_device *rdev, rdev->opencount--; - WARN_ON(rdev->scan_req && rdev->scan_req->wdev == wdev && - !rdev->scan_req->notified); + if (rdev->scan_req && rdev->scan_req->wdev == wdev) { + if (WARN_ON(!rdev->scan_req->notified)) + rdev->scan_req->aborted = true; + ___cfg80211_scan_done(rdev); + } } static int cfg80211_rfkill_set_block(void *data, bool blocked) @@ -856,8 +859,11 @@ static int cfg80211_netdev_notifier_call(struct notifier_block *nb, break; case NETDEV_DOWN: cfg80211_update_iface_num(rdev, wdev->iftype, -1); - WARN_ON(rdev->scan_req && rdev->scan_req->wdev == wdev && - !rdev->scan_req->notified); + if (rdev->scan_req && rdev->scan_req->wdev == wdev) { + if (WARN_ON(!rdev->scan_req->notified)) + rdev->scan_req->aborted = true; + ___cfg80211_scan_done(rdev); + } if (WARN_ON(rdev->sched_scan_req && rdev->sched_scan_req->dev == wdev->netdev)) { From f9d15d162b3acf28f85b3ac05c4883e5ed588d28 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 22 Jan 2014 11:14:19 +0200 Subject: [PATCH 284/596] cfg80211: send scan results from work queue Due to the previous commit, when a scan finishes, it is in theory possible to hit the following sequence: 1. interface starts being removed 2. scan is cancelled by driver and cfg80211 is notified 3. scan done work is scheduled 4. interface is removed completely, rdev->scan_req is freed, event sent to userspace but scan done work remains pending 5. new scan is requested on another virtual interface 6. scan done work runs, freeing the still-running scan To fix this situation, hang on to the scan done message and block new scans while that is the case, and only send the message from the work function, regardless of whether the scan_req is already freed from interface removal. This makes step 5 above impossible and changes step 6 to be 5. scan done work runs, sending the scan done message As this can't work for wext, so we send the message immediately, but this shouldn't be an issue since we still return -EBUSY. Signed-off-by: Johannes Berg --- net/wireless/core.c | 4 ++-- net/wireless/core.h | 4 +++- net/wireless/nl80211.c | 29 ++++++++++------------------- net/wireless/nl80211.h | 8 ++++---- net/wireless/scan.c | 40 +++++++++++++++++++++++++--------------- net/wireless/sme.c | 2 +- 6 files changed, 45 insertions(+), 42 deletions(-) diff --git a/net/wireless/core.c b/net/wireless/core.c index 02ed00dbf2df..010892b81a06 100644 --- a/net/wireless/core.c +++ b/net/wireless/core.c @@ -206,7 +206,7 @@ void cfg80211_stop_p2p_device(struct cfg80211_registered_device *rdev, if (rdev->scan_req && rdev->scan_req->wdev == wdev) { if (WARN_ON(!rdev->scan_req->notified)) rdev->scan_req->aborted = true; - ___cfg80211_scan_done(rdev); + ___cfg80211_scan_done(rdev, false); } } @@ -862,7 +862,7 @@ static int cfg80211_netdev_notifier_call(struct notifier_block *nb, if (rdev->scan_req && rdev->scan_req->wdev == wdev) { if (WARN_ON(!rdev->scan_req->notified)) rdev->scan_req->aborted = true; - ___cfg80211_scan_done(rdev); + ___cfg80211_scan_done(rdev, false); } if (WARN_ON(rdev->sched_scan_req && diff --git a/net/wireless/core.h b/net/wireless/core.h index 37ec16d7bb1a..f1d193b557b6 100644 --- a/net/wireless/core.h +++ b/net/wireless/core.h @@ -62,6 +62,7 @@ struct cfg80211_registered_device { struct rb_root bss_tree; u32 bss_generation; struct cfg80211_scan_request *scan_req; /* protected by RTNL */ + struct sk_buff *scan_msg; struct cfg80211_sched_scan_request *sched_scan_req; unsigned long suspend_at; struct work_struct scan_done_wk; @@ -361,7 +362,8 @@ int cfg80211_validate_key_settings(struct cfg80211_registered_device *rdev, struct key_params *params, int key_idx, bool pairwise, const u8 *mac_addr); void __cfg80211_scan_done(struct work_struct *wk); -void ___cfg80211_scan_done(struct cfg80211_registered_device *rdev); +void ___cfg80211_scan_done(struct cfg80211_registered_device *rdev, + bool send_message); void __cfg80211_sched_scan_results(struct work_struct *wk); int __cfg80211_stop_sched_scan(struct cfg80211_registered_device *rdev, bool driver_initiated); diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index 6ea960b1a8eb..4fe2e6e2bc76 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -5245,7 +5245,7 @@ static int nl80211_trigger_scan(struct sk_buff *skb, struct genl_info *info) if (!rdev->ops->scan) return -EOPNOTSUPP; - if (rdev->scan_req) { + if (rdev->scan_req || rdev->scan_msg) { err = -EBUSY; goto unlock; } @@ -10012,40 +10012,31 @@ void nl80211_send_scan_start(struct cfg80211_registered_device *rdev, NL80211_MCGRP_SCAN, GFP_KERNEL); } -void nl80211_send_scan_done(struct cfg80211_registered_device *rdev, - struct wireless_dev *wdev) +struct sk_buff *nl80211_build_scan_msg(struct cfg80211_registered_device *rdev, + struct wireless_dev *wdev, bool aborted) { struct sk_buff *msg; msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); if (!msg) - return; + return NULL; if (nl80211_send_scan_msg(msg, rdev, wdev, 0, 0, 0, - NL80211_CMD_NEW_SCAN_RESULTS) < 0) { + aborted ? NL80211_CMD_SCAN_ABORTED : + NL80211_CMD_NEW_SCAN_RESULTS) < 0) { nlmsg_free(msg); - return; + return NULL; } - genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0, - NL80211_MCGRP_SCAN, GFP_KERNEL); + return msg; } -void nl80211_send_scan_aborted(struct cfg80211_registered_device *rdev, - struct wireless_dev *wdev) +void nl80211_send_scan_result(struct cfg80211_registered_device *rdev, + struct sk_buff *msg) { - struct sk_buff *msg; - - msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); if (!msg) return; - if (nl80211_send_scan_msg(msg, rdev, wdev, 0, 0, 0, - NL80211_CMD_SCAN_ABORTED) < 0) { - nlmsg_free(msg); - return; - } - genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0, NL80211_MCGRP_SCAN, GFP_KERNEL); } diff --git a/net/wireless/nl80211.h b/net/wireless/nl80211.h index b1b231324e10..75799746d845 100644 --- a/net/wireless/nl80211.h +++ b/net/wireless/nl80211.h @@ -8,10 +8,10 @@ void nl80211_exit(void); void nl80211_notify_dev_rename(struct cfg80211_registered_device *rdev); void nl80211_send_scan_start(struct cfg80211_registered_device *rdev, struct wireless_dev *wdev); -void nl80211_send_scan_done(struct cfg80211_registered_device *rdev, - struct wireless_dev *wdev); -void nl80211_send_scan_aborted(struct cfg80211_registered_device *rdev, - struct wireless_dev *wdev); +struct sk_buff *nl80211_build_scan_msg(struct cfg80211_registered_device *rdev, + struct wireless_dev *wdev, bool aborted); +void nl80211_send_scan_result(struct cfg80211_registered_device *rdev, + struct sk_buff *msg); void nl80211_send_sched_scan(struct cfg80211_registered_device *rdev, struct net_device *netdev, u32 cmd); void nl80211_send_sched_scan_results(struct cfg80211_registered_device *rdev, diff --git a/net/wireless/scan.c b/net/wireless/scan.c index b528e31da2cf..d1ed4aebbbb7 100644 --- a/net/wireless/scan.c +++ b/net/wireless/scan.c @@ -161,18 +161,25 @@ static void __cfg80211_bss_expire(struct cfg80211_registered_device *dev, dev->bss_generation++; } -void ___cfg80211_scan_done(struct cfg80211_registered_device *rdev) +void ___cfg80211_scan_done(struct cfg80211_registered_device *rdev, + bool send_message) { struct cfg80211_scan_request *request; struct wireless_dev *wdev; + struct sk_buff *msg; #ifdef CONFIG_CFG80211_WEXT union iwreq_data wrqu; #endif ASSERT_RTNL(); - request = rdev->scan_req; + if (rdev->scan_msg) { + nl80211_send_scan_result(rdev, rdev->scan_msg); + rdev->scan_msg = NULL; + return; + } + request = rdev->scan_req; if (!request) return; @@ -186,18 +193,16 @@ void ___cfg80211_scan_done(struct cfg80211_registered_device *rdev) if (wdev->netdev) cfg80211_sme_scan_done(wdev->netdev); - if (request->aborted) { - nl80211_send_scan_aborted(rdev, wdev); - } else { - if (request->flags & NL80211_SCAN_FLAG_FLUSH) { - /* flush entries from previous scans */ - spin_lock_bh(&rdev->bss_lock); - __cfg80211_bss_expire(rdev, request->scan_start); - spin_unlock_bh(&rdev->bss_lock); - } - nl80211_send_scan_done(rdev, wdev); + if (!request->aborted && + request->flags & NL80211_SCAN_FLAG_FLUSH) { + /* flush entries from previous scans */ + spin_lock_bh(&rdev->bss_lock); + __cfg80211_bss_expire(rdev, request->scan_start); + spin_unlock_bh(&rdev->bss_lock); } + msg = nl80211_build_scan_msg(rdev, wdev, request->aborted); + #ifdef CONFIG_CFG80211_WEXT if (wdev->netdev && !request->aborted) { memset(&wrqu, 0, sizeof(wrqu)); @@ -211,6 +216,11 @@ void ___cfg80211_scan_done(struct cfg80211_registered_device *rdev) rdev->scan_req = NULL; kfree(request); + + if (!send_message) + rdev->scan_msg = msg; + else + nl80211_send_scan_result(rdev, msg); } void __cfg80211_scan_done(struct work_struct *wk) @@ -221,7 +231,7 @@ void __cfg80211_scan_done(struct work_struct *wk) scan_done_wk); rtnl_lock(); - ___cfg80211_scan_done(rdev); + ___cfg80211_scan_done(rdev, true); rtnl_unlock(); } @@ -1079,7 +1089,7 @@ int cfg80211_wext_siwscan(struct net_device *dev, if (IS_ERR(rdev)) return PTR_ERR(rdev); - if (rdev->scan_req) { + if (rdev->scan_req || rdev->scan_msg) { err = -EBUSY; goto out; } @@ -1481,7 +1491,7 @@ int cfg80211_wext_giwscan(struct net_device *dev, if (IS_ERR(rdev)) return PTR_ERR(rdev); - if (rdev->scan_req) + if (rdev->scan_req || rdev->scan_msg) return -EAGAIN; res = ieee80211_scan_results(rdev, info, extra, data->length); diff --git a/net/wireless/sme.c b/net/wireless/sme.c index a63509118508..f04d4c32e96e 100644 --- a/net/wireless/sme.c +++ b/net/wireless/sme.c @@ -67,7 +67,7 @@ static int cfg80211_conn_scan(struct wireless_dev *wdev) ASSERT_RDEV_LOCK(rdev); ASSERT_WDEV_LOCK(wdev); - if (rdev->scan_req) + if (rdev->scan_req || rdev->scan_msg) return -EBUSY; if (wdev->conn->params.channel) From 0297ea17bf7879fb5846fafd1be4c0471e72848d Mon Sep 17 00:00:00 2001 From: Emmanuel Grumbach Date: Mon, 27 Jan 2014 11:07:42 +0200 Subject: [PATCH 285/596] mac80211: release the channel in error path in start_ap When the driver cannot start the AP or when the assignement of the beacon goes wrong, we need to unassign the vif. Cc: stable@vger.kernel.org Signed-off-by: Emmanuel Grumbach Signed-off-by: Johannes Berg --- net/mac80211/cfg.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c index 33acdca4a1df..453e974287d1 100644 --- a/net/mac80211/cfg.c +++ b/net/mac80211/cfg.c @@ -1021,8 +1021,10 @@ static int ieee80211_start_ap(struct wiphy *wiphy, struct net_device *dev, IEEE80211_P2P_OPPPS_ENABLE_BIT; err = ieee80211_assign_beacon(sdata, ¶ms->beacon); - if (err < 0) + if (err < 0) { + ieee80211_vif_release_channel(sdata); return err; + } changed |= err; err = drv_start_ap(sdata->local, sdata); @@ -1032,6 +1034,7 @@ static int ieee80211_start_ap(struct wiphy *wiphy, struct net_device *dev, if (old) kfree_rcu(old, rcu_head); RCU_INIT_POINTER(sdata->u.ap.beacon, NULL); + ieee80211_vif_release_channel(sdata); return err; } From d4c80d9df6d1e4473b1409e4d220ca3d1612125c Mon Sep 17 00:00:00 2001 From: Sujith Manoharan Date: Thu, 30 Jan 2014 14:17:28 +0530 Subject: [PATCH 286/596] mac80211: Fix IBSS disconnect Currently, when a station leaves an IBSS network, the corresponding BSS is not dropped from cfg80211 if there are other active stations in the network. But, the small window that is present when trying to determine a station's status based on IEEE80211_IBSS_MERGE_INTERVAL introduces a race. Instead of trying to keep the BSS, always remove it when leaving an IBSS network. There is not much benefit to retain the BSS entry since it will be added with a subsequent join operation. This fixes an issue where a dangling BSS entry causes ath9k to wait for a beacon indefinitely. Cc: Reported-by: Simon Wunderlich Signed-off-by: Sujith Manoharan Signed-off-by: Johannes Berg --- net/mac80211/ibss.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/net/mac80211/ibss.c b/net/mac80211/ibss.c index 771080ec7212..2796a198728f 100644 --- a/net/mac80211/ibss.c +++ b/net/mac80211/ibss.c @@ -695,12 +695,9 @@ static void ieee80211_ibss_disconnect(struct ieee80211_sub_if_data *sdata) struct cfg80211_bss *cbss; struct beacon_data *presp; struct sta_info *sta; - int active_ibss; u16 capability; - active_ibss = ieee80211_sta_active_ibss(sdata); - - if (!active_ibss && !is_zero_ether_addr(ifibss->bssid)) { + if (!is_zero_ether_addr(ifibss->bssid)) { capability = WLAN_CAPABILITY_IBSS; if (ifibss->privacy) From 338f977f4eb441e69bb9a46eaa0ac715c931a67f Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Sat, 1 Feb 2014 00:16:23 +0100 Subject: [PATCH 287/596] mac80211: fix fragmentation code, particularly for encryption The "new" fragmentation code (since my rewrite almost 5 years ago) erroneously sets skb->len rather than using skb_trim() to adjust the length of the first fragment after copying out all the others. This leaves the skb tail pointer pointing to after where the data originally ended, and thus causes the encryption MIC to be written at that point, rather than where it belongs: immediately after the data. The impact of this is that if software encryption is done, then a) encryption doesn't work for the first fragment, the connection becomes unusable as the first fragment will never be properly verified at the receiver, the MIC is practically guaranteed to be wrong b) we leak up to 8 bytes of plaintext (!) of the packet out into the air This is only mitigated by the fact that many devices are capable of doing encryption in hardware, in which case this can't happen as the tail pointer is irrelevant in that case. Additionally, fragmentation is not used very frequently and would normally have to be configured manually. Fix this by using skb_trim() properly. Cc: stable@vger.kernel.org Fixes: 2de8e0d999b8 ("mac80211: rewrite fragmentation") Reported-by: Jouni Malinen Signed-off-by: Johannes Berg --- net/mac80211/tx.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index 27c990bf2320..97a02d3f7d87 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -878,7 +878,7 @@ static int ieee80211_fragment(struct ieee80211_tx_data *tx, } /* adjust first fragment's length */ - skb->len = hdrlen + per_fragm; + skb_trim(skb, hdrlen + per_fragm); return 0; } From fab57a6cc227468ca9e6a4c7ff8d3b10727785ee Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 29 Jan 2014 13:28:02 +0100 Subject: [PATCH 288/596] mac80211: fix virtual monitor interface iteration During channel context assignment, the interface should be found by interface iteration, so we need to assign the pointer before the channel context. Reported-by: Emmanuel Grumbach Tested-by: Emmanuel Grumbach Signed-off-by: Johannes Berg --- net/mac80211/iface.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/net/mac80211/iface.c b/net/mac80211/iface.c index ae2eb148a028..d6d1f1df9119 100644 --- a/net/mac80211/iface.c +++ b/net/mac80211/iface.c @@ -418,20 +418,24 @@ int ieee80211_add_virtual_monitor(struct ieee80211_local *local) return ret; } + mutex_lock(&local->iflist_mtx); + rcu_assign_pointer(local->monitor_sdata, sdata); + mutex_unlock(&local->iflist_mtx); + mutex_lock(&local->mtx); ret = ieee80211_vif_use_channel(sdata, &local->monitor_chandef, IEEE80211_CHANCTX_EXCLUSIVE); mutex_unlock(&local->mtx); if (ret) { + mutex_lock(&local->iflist_mtx); + rcu_assign_pointer(local->monitor_sdata, NULL); + mutex_unlock(&local->iflist_mtx); + synchronize_net(); drv_remove_interface(local, sdata); kfree(sdata); return ret; } - mutex_lock(&local->iflist_mtx); - rcu_assign_pointer(local->monitor_sdata, sdata); - mutex_unlock(&local->iflist_mtx); - return 0; } From a3485d088516be4c2ae2890e8ad64321481f2857 Mon Sep 17 00:00:00 2001 From: Lars-Peter Clausen Date: Tue, 4 Feb 2014 13:42:10 +0100 Subject: [PATCH 289/596] gpio: consumer.h: Move forward declarations outside #ifdef Make sure that the forward declared structs in gpio/consumer.h are also visible on the else branch of the CONFIG_GPIOLIB #ifdef. Fixes the following warnings and their associated errors when CONFIG_GPIOLIB is not selected: include/linux/gpio/consumer.h:67:14: warning: 'struct device' declared inside parameter list include/linux/gpio/consumer.h:67:14: warning: its scope is only this definition or declaration, which is probably not what you want [...] Signed-off-by: Lars-Peter Clausen Reviewed-by: Alexandre Courbot Signed-off-by: Linus Walleij --- include/linux/gpio/consumer.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/include/linux/gpio/consumer.h b/include/linux/gpio/consumer.h index 4d34dbbbad4d..7a8144fef406 100644 --- a/include/linux/gpio/consumer.h +++ b/include/linux/gpio/consumer.h @@ -4,8 +4,6 @@ #include #include -#ifdef CONFIG_GPIOLIB - struct device; struct gpio_chip; @@ -18,6 +16,8 @@ struct gpio_chip; */ struct gpio_desc; +#ifdef CONFIG_GPIOLIB + /* Acquire and dispose GPIOs */ struct gpio_desc *__must_check gpiod_get(struct device *dev, const char *con_id); From 1bc8927cc5b57c4d9198e4c4ac6b426b7c7d72c7 Mon Sep 17 00:00:00 2001 From: Sebastian Ott Date: Wed, 5 Feb 2014 13:36:05 +0100 Subject: [PATCH 290/596] s390/cio: improve cio_commit_config The modify subchannel wrapper cio_commit_config can fail when (unexpected) status is pending on the subchannel. Callers of cio_commit_config (that operated on enabled subchannels) needed to do error handling for that case (clear the unexpected status with test subchannel and retry). This error handling is missing in some code paths and caused online setting of devices to fail. Fix this for all callers by moving the error handling inside cio_commit_config. Reviewed-by: Peter Oberparleiter Signed-off-by: Sebastian Ott Signed-off-by: Martin Schwidefsky --- drivers/s390/cio/cio.c | 40 ++++++++++++++-------------------------- 1 file changed, 14 insertions(+), 26 deletions(-) diff --git a/drivers/s390/cio/cio.c b/drivers/s390/cio/cio.c index 88e35d85d205..8ee88c4ebd83 100644 --- a/drivers/s390/cio/cio.c +++ b/drivers/s390/cio/cio.c @@ -342,8 +342,9 @@ static int cio_check_config(struct subchannel *sch, struct schib *schib) */ int cio_commit_config(struct subchannel *sch) { - struct schib schib; int ccode, retry, ret = 0; + struct schib schib; + struct irb irb; if (stsch_err(sch->schid, &schib) || !css_sch_is_valid(&schib)) return -ENODEV; @@ -367,7 +368,10 @@ int cio_commit_config(struct subchannel *sch) ret = -EAGAIN; break; case 1: /* status pending */ - return -EBUSY; + ret = -EBUSY; + if (tsch(sch->schid, &irb)) + return ret; + break; case 2: /* busy */ udelay(100); /* allow for recovery */ ret = -EBUSY; @@ -403,7 +407,6 @@ EXPORT_SYMBOL_GPL(cio_update_schib); */ int cio_enable_subchannel(struct subchannel *sch, u32 intparm) { - int retry; int ret; CIO_TRACE_EVENT(2, "ensch"); @@ -418,20 +421,14 @@ int cio_enable_subchannel(struct subchannel *sch, u32 intparm) sch->config.isc = sch->isc; sch->config.intparm = intparm; - for (retry = 0; retry < 3; retry++) { + ret = cio_commit_config(sch); + if (ret == -EIO) { + /* + * Got a program check in msch. Try without + * the concurrent sense bit the next time. + */ + sch->config.csense = 0; ret = cio_commit_config(sch); - if (ret == -EIO) { - /* - * Got a program check in msch. Try without - * the concurrent sense bit the next time. - */ - sch->config.csense = 0; - } else if (ret == -EBUSY) { - struct irb irb; - if (tsch(sch->schid, &irb) != 0) - break; - } else - break; } CIO_HEX_EVENT(2, &ret, sizeof(ret)); return ret; @@ -444,7 +441,6 @@ EXPORT_SYMBOL_GPL(cio_enable_subchannel); */ int cio_disable_subchannel(struct subchannel *sch) { - int retry; int ret; CIO_TRACE_EVENT(2, "dissch"); @@ -456,16 +452,8 @@ int cio_disable_subchannel(struct subchannel *sch) return -ENODEV; sch->config.ena = 0; + ret = cio_commit_config(sch); - for (retry = 0; retry < 3; retry++) { - ret = cio_commit_config(sch); - if (ret == -EBUSY) { - struct irb irb; - if (tsch(sch->schid, &irb) != 0) - break; - } else - break; - } CIO_HEX_EVENT(2, &ret, sizeof(ret)); return ret; } From b8ecbee67c732ef9fc47fcf50aed6b7bb6231d98 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Thu, 6 Feb 2014 09:17:41 +0000 Subject: [PATCH 291/596] netfilter: nf_tables: fix log/queue expressions for NFPROTO_INET The log and queue expressions both store the family during ->init() and use it to deliver packets. This is wrong when used in NFPROTO_INET since they should both deliver to the actual AF of the packet, not the dummy NFPROTO_INET. Use the family from the hook ops to fix this. Signed-off-by: Patrick McHardy Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nft_log.c | 5 +---- net/netfilter/nft_queue.c | 4 +--- 2 files changed, 2 insertions(+), 7 deletions(-) diff --git a/net/netfilter/nft_log.c b/net/netfilter/nft_log.c index 5af790123ad8..26c5154e05f3 100644 --- a/net/netfilter/nft_log.c +++ b/net/netfilter/nft_log.c @@ -23,7 +23,6 @@ static const char *nft_log_null_prefix = ""; struct nft_log { struct nf_loginfo loginfo; char *prefix; - int family; }; static void nft_log_eval(const struct nft_expr *expr, @@ -33,7 +32,7 @@ static void nft_log_eval(const struct nft_expr *expr, const struct nft_log *priv = nft_expr_priv(expr); struct net *net = dev_net(pkt->in ? pkt->in : pkt->out); - nf_log_packet(net, priv->family, pkt->ops->hooknum, pkt->skb, pkt->in, + nf_log_packet(net, pkt->ops->pf, pkt->ops->hooknum, pkt->skb, pkt->in, pkt->out, &priv->loginfo, "%s", priv->prefix); } @@ -52,8 +51,6 @@ static int nft_log_init(const struct nft_ctx *ctx, struct nf_loginfo *li = &priv->loginfo; const struct nlattr *nla; - priv->family = ctx->afi->family; - nla = tb[NFTA_LOG_PREFIX]; if (nla != NULL) { priv->prefix = kmalloc(nla_len(nla) + 1, GFP_KERNEL); diff --git a/net/netfilter/nft_queue.c b/net/netfilter/nft_queue.c index cbea473d69e9..e8ae2f6bf232 100644 --- a/net/netfilter/nft_queue.c +++ b/net/netfilter/nft_queue.c @@ -25,7 +25,6 @@ struct nft_queue { u16 queuenum; u16 queues_total; u16 flags; - u8 family; }; static void nft_queue_eval(const struct nft_expr *expr, @@ -43,7 +42,7 @@ static void nft_queue_eval(const struct nft_expr *expr, queue = priv->queuenum + cpu % priv->queues_total; } else { queue = nfqueue_hash(pkt->skb, queue, - priv->queues_total, priv->family, + priv->queues_total, pkt->ops->pf, jhash_initval); } } @@ -71,7 +70,6 @@ static int nft_queue_init(const struct nft_ctx *ctx, return -EINVAL; init_hashrandom(&jhash_initval); - priv->family = ctx->afi->family; priv->queuenum = ntohs(nla_get_be16(tb[NFTA_QUEUE_NUM])); if (tb[NFTA_QUEUE_TOTAL] != NULL) From 0165d9325d6a3cf856e2cbbe64a0f4635ac75893 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Sat, 25 Jan 2014 14:03:51 +0100 Subject: [PATCH 292/596] netfilter: nf_tables: fix racy rule deletion We may lost race if we flush the rule-set (which happens asynchronously via call_rcu) and we try to remove the table (that userspace assumes to be empty). Fix this by recovering synchronous rule and chain deletion. This was introduced time ago before we had no batch support, and synchronous rule deletion performance was not good. Now that we have the batch support, we can just postpone the purge of old rule in a second step in the commit phase. All object deletions are synchronous after this patch. As a side effect, we save memory as we don't need rcu_head per rule anymore. Cc: Patrick McHardy Reported-by: Arturo Borrero Gonzalez Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_tables.h | 4 ---- net/netfilter/nf_tables_api.c | 40 ++++++++++++++++++------------- 2 files changed, 23 insertions(+), 21 deletions(-) diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h index 0f68e47d3e5e..e7e14ffe0f6a 100644 --- a/include/net/netfilter/nf_tables.h +++ b/include/net/netfilter/nf_tables.h @@ -322,7 +322,6 @@ static inline void *nft_expr_priv(const struct nft_expr *expr) * struct nft_rule - nf_tables rule * * @list: used internally - * @rcu_head: used internally for rcu * @handle: rule handle * @genmask: generation mask * @dlen: length of expression data @@ -330,7 +329,6 @@ static inline void *nft_expr_priv(const struct nft_expr *expr) */ struct nft_rule { struct list_head list; - struct rcu_head rcu_head; u64 handle:46, genmask:2, dlen:16; @@ -391,7 +389,6 @@ enum nft_chain_flags { * * @rules: list of rules in the chain * @list: used internally - * @rcu_head: used internally * @net: net namespace that this chain belongs to * @table: table that this chain belongs to * @handle: chain handle @@ -403,7 +400,6 @@ enum nft_chain_flags { struct nft_chain { struct list_head rules; struct list_head list; - struct rcu_head rcu_head; struct net *net; struct nft_table *table; u64 handle; diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 113c469c7579..3a2e4800b415 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -1008,10 +1008,8 @@ static int nf_tables_newchain(struct sock *nlsk, struct sk_buff *skb, return 0; } -static void nf_tables_rcu_chain_destroy(struct rcu_head *head) +static void nf_tables_chain_destroy(struct nft_chain *chain) { - struct nft_chain *chain = container_of(head, struct nft_chain, rcu_head); - BUG_ON(chain->use > 0); if (chain->flags & NFT_BASE_CHAIN) { @@ -1059,7 +1057,9 @@ static int nf_tables_delchain(struct sock *nlsk, struct sk_buff *skb, family); /* Make sure all rule references are gone before this is released */ - call_rcu(&chain->rcu_head, nf_tables_rcu_chain_destroy); + synchronize_rcu(); + + nf_tables_chain_destroy(chain); return 0; } @@ -1531,9 +1531,8 @@ static int nf_tables_getrule(struct sock *nlsk, struct sk_buff *skb, return err; } -static void nf_tables_rcu_rule_destroy(struct rcu_head *head) +static void nf_tables_rule_destroy(struct nft_rule *rule) { - struct nft_rule *rule = container_of(head, struct nft_rule, rcu_head); struct nft_expr *expr; /* @@ -1548,11 +1547,6 @@ static void nf_tables_rcu_rule_destroy(struct rcu_head *head) kfree(rule); } -static void nf_tables_rule_destroy(struct nft_rule *rule) -{ - call_rcu(&rule->rcu_head, nf_tables_rcu_rule_destroy); -} - #define NFT_RULE_MAXEXPRS 128 static struct nft_expr_info *info; @@ -1819,9 +1813,6 @@ static int nf_tables_commit(struct sk_buff *skb) synchronize_rcu(); list_for_each_entry_safe(rupd, tmp, &net->nft.commit_list, list) { - /* Delete this rule from the dirty list */ - list_del(&rupd->list); - /* This rule was inactive in the past and just became active. * Clear the next bit of the genmask since its meaning has * changed, now it is the future. @@ -1832,6 +1823,7 @@ static int nf_tables_commit(struct sk_buff *skb) rupd->chain, rupd->rule, NFT_MSG_NEWRULE, 0, rupd->family); + list_del(&rupd->list); kfree(rupd); continue; } @@ -1841,7 +1833,15 @@ static int nf_tables_commit(struct sk_buff *skb) nf_tables_rule_notify(skb, rupd->nlh, rupd->table, rupd->chain, rupd->rule, NFT_MSG_DELRULE, 0, rupd->family); + } + + /* Make sure we don't see any packet traversing old rules */ + synchronize_rcu(); + + /* Now we can safely release unused old rules */ + list_for_each_entry_safe(rupd, tmp, &net->nft.commit_list, list) { nf_tables_rule_destroy(rupd->rule); + list_del(&rupd->list); kfree(rupd); } @@ -1854,20 +1854,26 @@ static int nf_tables_abort(struct sk_buff *skb) struct nft_rule_trans *rupd, *tmp; list_for_each_entry_safe(rupd, tmp, &net->nft.commit_list, list) { - /* Delete all rules from the dirty list */ - list_del(&rupd->list); - if (!nft_rule_is_active_next(net, rupd->rule)) { nft_rule_clear(net, rupd->rule); + list_del(&rupd->list); kfree(rupd); continue; } /* This rule is inactive, get rid of it */ list_del_rcu(&rupd->rule->list); + } + + /* Make sure we don't see any packet accessing aborted rules */ + synchronize_rcu(); + + list_for_each_entry_safe(rupd, tmp, &net->nft.commit_list, list) { nf_tables_rule_destroy(rupd->rule); + list_del(&rupd->list); kfree(rupd); } + return 0; } From 4a7ac12eedd190cdf071e61145defa73df1675c0 Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Thu, 6 Feb 2014 11:30:48 +0000 Subject: [PATCH 293/596] arm64: barriers: allow dsb macro to take option parameter The dsb instruction takes an option specifying both the target access types and shareability domain. This patch allows such an option to be passed to the dsb macro, resulting in potentially more efficient code. Currently the option is ignored until all callers are updated (unlike ARM, the option is mandated by the assembler). Signed-off-by: Will Deacon Signed-off-by: Catalin Marinas --- arch/arm64/include/asm/barrier.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm64/include/asm/barrier.h b/arch/arm64/include/asm/barrier.h index 78e20ba8806b..409ca370cfe2 100644 --- a/arch/arm64/include/asm/barrier.h +++ b/arch/arm64/include/asm/barrier.h @@ -25,7 +25,7 @@ #define wfi() asm volatile("wfi" : : : "memory") #define isb() asm volatile("isb" : : : "memory") -#define dsb() asm volatile("dsb sy" : : : "memory") +#define dsb(opt) asm volatile("dsb sy" : : : "memory") #define mb() dsb() #define rmb() asm volatile("dsb ld" : : : "memory") From 97b8b16bfcb7f5ccc1b3c10c08580f9f1acb61ac Mon Sep 17 00:00:00 2001 From: Aaro Koskinen Date: Wed, 5 Feb 2014 22:05:44 +0200 Subject: [PATCH 294/596] MIPS: fpu.h: Fix build when CONFIG_BUG is not set __enable_fpu produces a build failure when CONFIG_BUG is not set: In file included from arch/mips/kernel/cpu-probe.c:24:0: arch/mips/include/asm/fpu.h: In function '__enable_fpu': arch/mips/include/asm/fpu.h:77:1: error: control reaches end of non-void function [-Werror=return-type] This is regression introduced in 3.14-rc1. Fix that. Signed-off-by: Aaro Koskinen Acked-by: Paul Burton Cc: John Crispin Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/6504/ Signed-off-by: Ralf Baechle --- arch/mips/include/asm/fpu.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/mips/include/asm/fpu.h b/arch/mips/include/asm/fpu.h index cfe092fc720d..6b9749540edf 100644 --- a/arch/mips/include/asm/fpu.h +++ b/arch/mips/include/asm/fpu.h @@ -74,6 +74,8 @@ static inline int __enable_fpu(enum fpu_mode mode) default: BUG(); } + + return SIGFPE; } #define __disable_fpu() \ From e7f2a444891cb39f11d5429467d0fd7e011fe7fe Mon Sep 17 00:00:00 2001 From: Florian Vaussard Date: Wed, 5 Feb 2014 07:51:22 +0100 Subject: [PATCH 295/596] pinctrl: do not init debugfs entries for unimplemented functionalities Commit c420619 "pinctrl: pinconf: remove checks on ops->pin_config_get" removed the check on (ops != NULL) when performing pinconf_pins_show() or pinconf_groups_show(). As these entries are always enabled, even if pinconf is not supported, reading will result in an oops due to NULL ops. Instead of checking for ops, remove the corresponding debugfs entries if pinconf and/or pinmux are not implemented. Tested on OMAP3 (pinctrl-single). Cc: stable@vger.kernel.org Signed-off-by: Florian Vaussard Signed-off-by: Linus Walleij --- drivers/pinctrl/core.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/pinctrl/core.c b/drivers/pinctrl/core.c index cab020a58bf5..c0fe6091566a 100644 --- a/drivers/pinctrl/core.c +++ b/drivers/pinctrl/core.c @@ -1644,8 +1644,10 @@ static void pinctrl_init_device_debugfs(struct pinctrl_dev *pctldev) device_root, pctldev, &pinctrl_groups_ops); debugfs_create_file("gpio-ranges", S_IFREG | S_IRUGO, device_root, pctldev, &pinctrl_gpioranges_ops); - pinmux_init_device_debugfs(device_root, pctldev); - pinconf_init_device_debugfs(device_root, pctldev); + if (pctldev->desc->pmxops) + pinmux_init_device_debugfs(device_root, pctldev); + if (pctldev->desc->confops) + pinconf_init_device_debugfs(device_root, pctldev); } static void pinctrl_remove_device_debugfs(struct pinctrl_dev *pctldev) From 5b232c5addc02980cfce451575f1d1f975bdb04e Mon Sep 17 00:00:00 2001 From: Laxman Dewangan Date: Wed, 5 Feb 2014 19:11:34 +0530 Subject: [PATCH 296/596] pinctrl: tegra: return correct error type When memory allocation failed, drive should return error as ENOMEM. Signed-off-by: Laxman Dewangan Acked-by: Stephen Warren Signed-off-by: Linus Walleij --- drivers/pinctrl/pinctrl-tegra.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/pinctrl/pinctrl-tegra.c b/drivers/pinctrl/pinctrl-tegra.c index a2e93a2b5ff4..e767355ab0ad 100644 --- a/drivers/pinctrl/pinctrl-tegra.c +++ b/drivers/pinctrl/pinctrl-tegra.c @@ -645,7 +645,7 @@ int tegra_pinctrl_probe(struct platform_device *pdev, GFP_KERNEL); if (!pmx->regs) { dev_err(&pdev->dev, "Can't alloc regs pointer\n"); - return -ENODEV; + return -ENOMEM; } for (i = 0; i < pmx->nbanks; i++) { From 80d767d770fd9c697e434fd080c2db7b5c60c6dd Mon Sep 17 00:00:00 2001 From: Mikulas Patocka Date: Fri, 24 Jan 2014 16:41:36 -0500 Subject: [PATCH 297/596] time: Fix overflow when HZ is smaller than 60 When compiling for the IA-64 ski emulator, HZ is set to 32 because the emulation is slow and we don't want to waste too many cycles processing timers. Alpha also has an option to set HZ to 32. This causes integer underflow in kernel/time/jiffies.c: kernel/time/jiffies.c:66:2: warning: large integer implicitly truncated to unsigned type [-Woverflow] .mult = NSEC_PER_JIFFY << JIFFIES_SHIFT, /* details above */ ^ This patch reduces the JIFFIES_SHIFT value to avoid the overflow. Signed-off-by: Mikulas Patocka Link: http://lkml.kernel.org/r/alpine.LRH.2.02.1401241639100.23871@file01.intranet.prod.int.rdu2.redhat.com Cc: stable@vger.kernel.org Signed-off-by: Thomas Gleixner --- kernel/time/jiffies.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/kernel/time/jiffies.c b/kernel/time/jiffies.c index 7a925ba456fb..a6a5bf53e86d 100644 --- a/kernel/time/jiffies.c +++ b/kernel/time/jiffies.c @@ -51,7 +51,13 @@ * HZ shrinks, so values greater than 8 overflow 32bits when * HZ=100. */ +#if HZ < 34 +#define JIFFIES_SHIFT 6 +#elif HZ < 67 +#define JIFFIES_SHIFT 7 +#else #define JIFFIES_SHIFT 8 +#endif static cycle_t jiffies_read(struct clocksource *cs) { From ad037c1f4ae6f9258f27fddb679ceb703ff07a88 Mon Sep 17 00:00:00 2001 From: Tim Kryger Date: Tue, 4 Feb 2014 16:15:04 -0800 Subject: [PATCH 298/596] clocksource: Kona: Print warning rather than panic Since there may be other clocksources available, this driver should not trigger a panic simply because it can not determine the frequency of an external clock. This change refactors the driver to allow a warning to be printed in this case instead. Signed-off-by: Tim Kryger Reviewed-by: Markus Mayer Reviewed-by: Matt Porter Cc: Christian Daudt Cc: Daniel Lezcano Link: http://lkml.kernel.org/r/1391559304-26558-1-git-send-email-tim.kryger@linaro.org Signed-off-by: Thomas Gleixner --- drivers/clocksource/bcm_kona_timer.c | 54 ++++++++++++++-------------- 1 file changed, 28 insertions(+), 26 deletions(-) diff --git a/drivers/clocksource/bcm_kona_timer.c b/drivers/clocksource/bcm_kona_timer.c index 974b2db2fe10..0595dc6c453e 100644 --- a/drivers/clocksource/bcm_kona_timer.c +++ b/drivers/clocksource/bcm_kona_timer.c @@ -99,31 +99,6 @@ kona_timer_get_counter(void *timer_base, uint32_t *msw, uint32_t *lsw) return; } -static void __init kona_timers_init(struct device_node *node) -{ - u32 freq; - struct clk *external_clk; - - external_clk = of_clk_get_by_name(node, NULL); - - if (!IS_ERR(external_clk)) { - arch_timer_rate = clk_get_rate(external_clk); - clk_prepare_enable(external_clk); - } else if (!of_property_read_u32(node, "clock-frequency", &freq)) { - arch_timer_rate = freq; - } else { - panic("unable to determine clock-frequency"); - } - - /* Setup IRQ numbers */ - timers.tmr_irq = irq_of_parse_and_map(node, 0); - - /* Setup IO addresses */ - timers.tmr_regs = of_iomap(node, 0); - - kona_timer_disable_and_clear(timers.tmr_regs); -} - static int kona_timer_set_next_event(unsigned long clc, struct clock_event_device *unused) { @@ -198,7 +173,34 @@ static struct irqaction kona_timer_irq = { static void __init kona_timer_init(struct device_node *node) { - kona_timers_init(node); + u32 freq; + struct clk *external_clk; + + if (!of_device_is_available(node)) { + pr_info("Kona Timer v1 marked as disabled in device tree\n"); + return; + } + + external_clk = of_clk_get_by_name(node, NULL); + + if (!IS_ERR(external_clk)) { + arch_timer_rate = clk_get_rate(external_clk); + clk_prepare_enable(external_clk); + } else if (!of_property_read_u32(node, "clock-frequency", &freq)) { + arch_timer_rate = freq; + } else { + pr_err("Kona Timer v1 unable to determine clock-frequency"); + return; + } + + /* Setup IRQ numbers */ + timers.tmr_irq = irq_of_parse_and_map(node, 0); + + /* Setup IO addresses */ + timers.tmr_regs = of_iomap(node, 0); + + kona_timer_disable_and_clear(timers.tmr_regs); + kona_timer_clockevents_init(); setup_irq(timers.tmr_irq, &kona_timer_irq); kona_timer_set_next_event((arch_timer_rate / HZ), NULL); From 6583327c4dd55acbbf2a6f25e775b28b3abf9a42 Mon Sep 17 00:00:00 2001 From: Peter Oberparleiter Date: Thu, 6 Feb 2014 15:58:20 +0100 Subject: [PATCH 299/596] x86, hweight: Fix BUG when booting with CONFIG_GCOV_PROFILE_ALL=y Commit d61931d89b, "x86: Add optimized popcnt variants" introduced compile flag -fcall-saved-rdi for lib/hweight.c. When combined with options -fprofile-arcs and -O2, this flag causes gcc to generate broken constructor code. As a result, a 64 bit x86 kernel compiled with CONFIG_GCOV_PROFILE_ALL=y prints message "gcov: could not create file" and runs into sproadic BUGs during boot. The gcc people indicate that these kinds of problems are endemic when using ad hoc calling conventions. It is therefore best to treat any file compiled with ad hoc calling conventions as an isolated environment and avoid things like profiling or coverage analysis, since those subsystems assume a "normal" calling conventions. This patch avoids the bug by excluding lib/hweight.o from coverage profiling. Reported-by: Meelis Roos Cc: Andrew Morton Signed-off-by: Peter Oberparleiter Link: http://lkml.kernel.org/r/52F3A30C.7050205@linux.vnet.ibm.com Signed-off-by: H. Peter Anvin Cc: --- lib/Makefile | 1 + 1 file changed, 1 insertion(+) diff --git a/lib/Makefile b/lib/Makefile index a459c31e8c6b..04944e9993ed 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -43,6 +43,7 @@ obj-$(CONFIG_HAS_IOMEM) += iomap_copy.o devres.o obj-$(CONFIG_CHECK_SIGNATURE) += check_signature.o obj-$(CONFIG_DEBUG_LOCKING_API_SELFTESTS) += locking-selftest.o +GCOV_PROFILE_hweight.o := n CFLAGS_hweight.o = $(subst $(quote),,$(CONFIG_ARCH_HWEIGHT_CFLAGS)) obj-$(CONFIG_GENERIC_HWEIGHT) += hweight.o From b927e1c20462c1ad9caf4c4fa3a30e838a2d4037 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Christian=20K=C3=B6nig?= Date: Thu, 30 Jan 2014 19:01:16 +0100 Subject: [PATCH 300/596] drm/radeon: fix UVD IRQ support on SI MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Otherwise decoding isn't really useable. bug: https://bugs.freedesktop.org/show_bug.cgi?id=71448 Signed-off-by: Christian König Cc: stable@vger.kernel.org Signed-off-by: Alex Deucher --- drivers/gpu/drm/radeon/si.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/gpu/drm/radeon/si.c b/drivers/gpu/drm/radeon/si.c index 09ec4f6c53bb..83578324e5d1 100644 --- a/drivers/gpu/drm/radeon/si.c +++ b/drivers/gpu/drm/radeon/si.c @@ -6338,6 +6338,10 @@ int si_irq_process(struct radeon_device *rdev) break; } break; + case 124: /* UVD */ + DRM_DEBUG("IH: UVD int: 0x%08x\n", src_data); + radeon_fence_process(rdev, R600_RING_TYPE_UVD_INDEX); + break; case 146: case 147: addr = RREG32(VM_CONTEXT1_PROTECTION_FAULT_ADDR); From 858a41c853cef2cb01de34dae334c19c1c15b237 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Thu, 30 Jan 2014 14:35:04 -0500 Subject: [PATCH 301/596] drm/radeon: fix UVD IRQ support on 7xx Otherwise decoding isn't really useable. Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org --- drivers/gpu/drm/radeon/r600.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c index 56140b4e5bb2..cdbc4171fe73 100644 --- a/drivers/gpu/drm/radeon/r600.c +++ b/drivers/gpu/drm/radeon/r600.c @@ -3991,6 +3991,10 @@ int r600_irq_process(struct radeon_device *rdev) break; } break; + case 124: /* UVD */ + DRM_DEBUG("IH: UVD int: 0x%08x\n", src_data); + radeon_fence_process(rdev, R600_RING_TYPE_UVD_INDEX); + break; case 176: /* CP_INT in ring buffer */ case 177: /* CP_INT in IB1 */ case 178: /* CP_INT in IB2 */ From 9f3f63f24c901cad831e78bbb738df61a1c4ff02 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Thu, 30 Jan 2014 11:19:22 -0500 Subject: [PATCH 302/596] drm/radeon/dpm: use the driver state for dpm debugfs For btc and newer, we may modify the power state depending on the circumstances. Use the modified state rather than the base state. Signed-off-by: Alex Deucher --- drivers/gpu/drm/radeon/btc_dpm.c | 31 ++++++++++++++++++++++++++++ drivers/gpu/drm/radeon/btcd.h | 4 ++++ drivers/gpu/drm/radeon/ni_dpm.c | 3 ++- drivers/gpu/drm/radeon/radeon_asic.c | 2 +- drivers/gpu/drm/radeon/radeon_asic.h | 2 ++ drivers/gpu/drm/radeon/si_dpm.c | 3 ++- drivers/gpu/drm/radeon/sumo_dpm.c | 2 +- drivers/gpu/drm/radeon/trinity_dpm.c | 3 ++- 8 files changed, 45 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/radeon/btc_dpm.c b/drivers/gpu/drm/radeon/btc_dpm.c index 0fbd36f3d4e9..05ad46982a44 100644 --- a/drivers/gpu/drm/radeon/btc_dpm.c +++ b/drivers/gpu/drm/radeon/btc_dpm.c @@ -2756,6 +2756,37 @@ void btc_dpm_fini(struct radeon_device *rdev) r600_free_extended_power_table(rdev); } +void btc_dpm_debugfs_print_current_performance_level(struct radeon_device *rdev, + struct seq_file *m) +{ + struct evergreen_power_info *eg_pi = evergreen_get_pi(rdev); + struct radeon_ps *rps = &eg_pi->current_rps; + struct rv7xx_ps *ps = rv770_get_ps(rps); + struct rv7xx_pl *pl; + u32 current_index = + (RREG32(TARGET_AND_CURRENT_PROFILE_INDEX) & CURRENT_PROFILE_INDEX_MASK) >> + CURRENT_PROFILE_INDEX_SHIFT; + + if (current_index > 2) { + seq_printf(m, "invalid dpm profile %d\n", current_index); + } else { + if (current_index == 0) + pl = &ps->low; + else if (current_index == 1) + pl = &ps->medium; + else /* current_index == 2 */ + pl = &ps->high; + seq_printf(m, "uvd vclk: %d dclk: %d\n", rps->vclk, rps->dclk); + if (rdev->family >= CHIP_CEDAR) { + seq_printf(m, "power level %d sclk: %u mclk: %u vddc: %u vddci: %u\n", + current_index, pl->sclk, pl->mclk, pl->vddc, pl->vddci); + } else { + seq_printf(m, "power level %d sclk: %u mclk: %u vddc: %u\n", + current_index, pl->sclk, pl->mclk, pl->vddc); + } + } +} + u32 btc_dpm_get_sclk(struct radeon_device *rdev, bool low) { struct evergreen_power_info *eg_pi = evergreen_get_pi(rdev); diff --git a/drivers/gpu/drm/radeon/btcd.h b/drivers/gpu/drm/radeon/btcd.h index 29e32de7e025..9c65be2d55a9 100644 --- a/drivers/gpu/drm/radeon/btcd.h +++ b/drivers/gpu/drm/radeon/btcd.h @@ -44,6 +44,10 @@ # define DYN_SPREAD_SPECTRUM_EN (1 << 23) # define AC_DC_SW (1 << 24) +#define TARGET_AND_CURRENT_PROFILE_INDEX 0x66c +# define CURRENT_PROFILE_INDEX_MASK (0xf << 4) +# define CURRENT_PROFILE_INDEX_SHIFT 4 + #define CG_BIF_REQ_AND_RSP 0x7f4 #define CG_CLIENT_REQ(x) ((x) << 0) #define CG_CLIENT_REQ_MASK (0xff << 0) diff --git a/drivers/gpu/drm/radeon/ni_dpm.c b/drivers/gpu/drm/radeon/ni_dpm.c index c351226ecb31..22c339195f84 100644 --- a/drivers/gpu/drm/radeon/ni_dpm.c +++ b/drivers/gpu/drm/radeon/ni_dpm.c @@ -4322,7 +4322,8 @@ void ni_dpm_print_power_state(struct radeon_device *rdev, void ni_dpm_debugfs_print_current_performance_level(struct radeon_device *rdev, struct seq_file *m) { - struct radeon_ps *rps = rdev->pm.dpm.current_ps; + struct evergreen_power_info *eg_pi = evergreen_get_pi(rdev); + struct radeon_ps *rps = &eg_pi->current_rps; struct ni_ps *ps = ni_get_ps(rps); struct rv7xx_pl *pl; u32 current_index = diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c index f74db43346fd..dda02bfc10a4 100644 --- a/drivers/gpu/drm/radeon/radeon_asic.c +++ b/drivers/gpu/drm/radeon/radeon_asic.c @@ -1555,7 +1555,7 @@ static struct radeon_asic btc_asic = { .get_sclk = &btc_dpm_get_sclk, .get_mclk = &btc_dpm_get_mclk, .print_power_state = &rv770_dpm_print_power_state, - .debugfs_print_current_performance_level = &rv770_dpm_debugfs_print_current_performance_level, + .debugfs_print_current_performance_level = &btc_dpm_debugfs_print_current_performance_level, .force_performance_level = &rv770_dpm_force_performance_level, .vblank_too_short = &btc_dpm_vblank_too_short, }, diff --git a/drivers/gpu/drm/radeon/radeon_asic.h b/drivers/gpu/drm/radeon/radeon_asic.h index b3bc433eed4c..ae637cfda783 100644 --- a/drivers/gpu/drm/radeon/radeon_asic.h +++ b/drivers/gpu/drm/radeon/radeon_asic.h @@ -551,6 +551,8 @@ void btc_dpm_fini(struct radeon_device *rdev); u32 btc_dpm_get_sclk(struct radeon_device *rdev, bool low); u32 btc_dpm_get_mclk(struct radeon_device *rdev, bool low); bool btc_dpm_vblank_too_short(struct radeon_device *rdev); +void btc_dpm_debugfs_print_current_performance_level(struct radeon_device *rdev, + struct seq_file *m); int sumo_dpm_init(struct radeon_device *rdev); int sumo_dpm_enable(struct radeon_device *rdev); int sumo_dpm_late_enable(struct radeon_device *rdev); diff --git a/drivers/gpu/drm/radeon/si_dpm.c b/drivers/gpu/drm/radeon/si_dpm.c index 0471501338fb..eafb0e6bc67e 100644 --- a/drivers/gpu/drm/radeon/si_dpm.c +++ b/drivers/gpu/drm/radeon/si_dpm.c @@ -6472,7 +6472,8 @@ void si_dpm_fini(struct radeon_device *rdev) void si_dpm_debugfs_print_current_performance_level(struct radeon_device *rdev, struct seq_file *m) { - struct radeon_ps *rps = rdev->pm.dpm.current_ps; + struct evergreen_power_info *eg_pi = evergreen_get_pi(rdev); + struct radeon_ps *rps = &eg_pi->current_rps; struct ni_ps *ps = ni_get_ps(rps); struct rv7xx_pl *pl; u32 current_index = diff --git a/drivers/gpu/drm/radeon/sumo_dpm.c b/drivers/gpu/drm/radeon/sumo_dpm.c index f121efe12dc5..8b47b3cd0357 100644 --- a/drivers/gpu/drm/radeon/sumo_dpm.c +++ b/drivers/gpu/drm/radeon/sumo_dpm.c @@ -1807,7 +1807,7 @@ void sumo_dpm_debugfs_print_current_performance_level(struct radeon_device *rdev struct seq_file *m) { struct sumo_power_info *pi = sumo_get_pi(rdev); - struct radeon_ps *rps = rdev->pm.dpm.current_ps; + struct radeon_ps *rps = &pi->current_rps; struct sumo_ps *ps = sumo_get_ps(rps); struct sumo_pl *pl; u32 current_index = diff --git a/drivers/gpu/drm/radeon/trinity_dpm.c b/drivers/gpu/drm/radeon/trinity_dpm.c index 2d447192d6f7..2da0e17eb960 100644 --- a/drivers/gpu/drm/radeon/trinity_dpm.c +++ b/drivers/gpu/drm/radeon/trinity_dpm.c @@ -1926,7 +1926,8 @@ void trinity_dpm_print_power_state(struct radeon_device *rdev, void trinity_dpm_debugfs_print_current_performance_level(struct radeon_device *rdev, struct seq_file *m) { - struct radeon_ps *rps = rdev->pm.dpm.current_ps; + struct trinity_power_info *pi = trinity_get_pi(rdev); + struct radeon_ps *rps = &pi->current_rps; struct trinity_ps *ps = trinity_get_ps(rps); struct trinity_pl *pl; u32 current_index = From 299302294e109aec5e7f54ea61a9816c555c2106 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Thu, 30 Jan 2014 13:11:47 -0500 Subject: [PATCH 303/596] drm/radeon/dpm: use stored max_vddc rather than looking it up When we parse the power tables use the stored mac_vddc value rather than lookig it up manually each time. Signed-off-by: Alex Deucher --- drivers/gpu/drm/radeon/ni_dpm.c | 5 ++--- drivers/gpu/drm/radeon/rv770_dpm.c | 5 ++--- 2 files changed, 4 insertions(+), 6 deletions(-) diff --git a/drivers/gpu/drm/radeon/ni_dpm.c b/drivers/gpu/drm/radeon/ni_dpm.c index 22c339195f84..1217fbcbdcca 100644 --- a/drivers/gpu/drm/radeon/ni_dpm.c +++ b/drivers/gpu/drm/radeon/ni_dpm.c @@ -3945,7 +3945,6 @@ static void ni_parse_pplib_clock_info(struct radeon_device *rdev, struct rv7xx_power_info *pi = rv770_get_pi(rdev); struct evergreen_power_info *eg_pi = evergreen_get_pi(rdev); struct ni_ps *ps = ni_get_ps(rps); - u16 vddc; struct rv7xx_pl *pl = &ps->performance_levels[index]; ps->performance_level_count = index + 1; @@ -3961,8 +3960,8 @@ static void ni_parse_pplib_clock_info(struct radeon_device *rdev, /* patch up vddc if necessary */ if (pl->vddc == 0xff01) { - if (radeon_atom_get_max_vddc(rdev, 0, 0, &vddc) == 0) - pl->vddc = vddc; + if (pi->max_vddc) + pl->vddc = pi->max_vddc; } if (rps->class & ATOM_PPLIB_CLASSIFICATION_ACPI) { diff --git a/drivers/gpu/drm/radeon/rv770_dpm.c b/drivers/gpu/drm/radeon/rv770_dpm.c index 80c595aba359..5b2ea8ac0731 100644 --- a/drivers/gpu/drm/radeon/rv770_dpm.c +++ b/drivers/gpu/drm/radeon/rv770_dpm.c @@ -2174,7 +2174,6 @@ static void rv7xx_parse_pplib_clock_info(struct radeon_device *rdev, struct evergreen_power_info *eg_pi = evergreen_get_pi(rdev); struct rv7xx_ps *ps = rv770_get_ps(rps); u32 sclk, mclk; - u16 vddc; struct rv7xx_pl *pl; switch (index) { @@ -2214,8 +2213,8 @@ static void rv7xx_parse_pplib_clock_info(struct radeon_device *rdev, /* patch up vddc if necessary */ if (pl->vddc == 0xff01) { - if (radeon_atom_get_max_vddc(rdev, 0, 0, &vddc) == 0) - pl->vddc = vddc; + if (pi->max_vddc) + pl->vddc = pi->max_vddc; } if (rps->class & ATOM_PPLIB_CLASSIFICATION_ACPI) { From 4dbffb8f21999c2156ddac00186810f1dfd9eb1a Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Thu, 30 Jan 2014 14:06:05 -0500 Subject: [PATCH 304/596] drm/radeon: remove useless return Signed-off-by: Alex Deucher --- drivers/gpu/drm/radeon/uvd_v2_2.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/gpu/drm/radeon/uvd_v2_2.c b/drivers/gpu/drm/radeon/uvd_v2_2.c index 824550db3fed..d1771004cb52 100644 --- a/drivers/gpu/drm/radeon/uvd_v2_2.c +++ b/drivers/gpu/drm/radeon/uvd_v2_2.c @@ -57,7 +57,6 @@ void uvd_v2_2_fence_emit(struct radeon_device *rdev, radeon_ring_write(ring, 0); radeon_ring_write(ring, PACKET0(UVD_GPCOM_VCPU_CMD, 0)); radeon_ring_write(ring, 2); - return; } /** From 7c7e867cf0f74c4c79649605b964510041f1a8bf Mon Sep 17 00:00:00 2001 From: Dave Jones Date: Thu, 30 Jan 2014 21:17:30 -0500 Subject: [PATCH 305/596] drm/radeon/dpm: fix uninitialized read from stack in kv_dpm_late_enable If we take the false branch of the if quoted in the diff below, we end up doing a return ret, without ever having initialized it. Picked up by coverity. Signed-off-by: Dave Jones Signed-off-by: Alex Deucher --- drivers/gpu/drm/radeon/kv_dpm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/radeon/kv_dpm.c b/drivers/gpu/drm/radeon/kv_dpm.c index b6e01d5d2cce..351db361239d 100644 --- a/drivers/gpu/drm/radeon/kv_dpm.c +++ b/drivers/gpu/drm/radeon/kv_dpm.c @@ -1223,7 +1223,7 @@ int kv_dpm_enable(struct radeon_device *rdev) int kv_dpm_late_enable(struct radeon_device *rdev) { - int ret; + int ret = 0; if (rdev->irq.installed && r600_is_internal_thermal_sensor(rdev->pm.int_thermal_type)) { From 75a1ba5b2c529db60ca49626bcaf0bddf4548438 Mon Sep 17 00:00:00 2001 From: Borislav Petkov Date: Mon, 3 Feb 2014 21:41:44 +0100 Subject: [PATCH 306/596] x86, microcode, AMD: Unify valid container checks For additional coverage, BorisO and friends unknowlingly did swap AMD microcode with Intel microcode blobs in order to see what happens. What did happen on 32-bit was [ 5.722656] BUG: unable to handle kernel paging request at be3a6008 [ 5.722693] IP: [] load_microcode_amd+0x24/0x3f0 [ 5.722716] *pdpt = 0000000000000000 *pde = 0000000000000000 because there was a valid initrd there but without valid microcode in it and the container check happened *after* the relocated ramdisk handling on 32-bit, which was clearly wrong. While at it, take care of the ramdisk relocation on both 32- and 64-bit as it is done on both. Also, comment what we're doing because this code is a bit tricky. Reported-and-tested-by: Boris Ostrovsky Signed-off-by: Borislav Petkov Link: http://lkml.kernel.org/r/1391460104-7261-1-git-send-email-bp@alien8.de Signed-off-by: H. Peter Anvin --- arch/x86/kernel/cpu/microcode/amd_early.c | 43 +++++++++++++++-------- 1 file changed, 29 insertions(+), 14 deletions(-) diff --git a/arch/x86/kernel/cpu/microcode/amd_early.c b/arch/x86/kernel/cpu/microcode/amd_early.c index 8384c0fa206f..617a9e284245 100644 --- a/arch/x86/kernel/cpu/microcode/amd_early.c +++ b/arch/x86/kernel/cpu/microcode/amd_early.c @@ -285,6 +285,15 @@ static void __init collect_cpu_sig_on_bsp(void *arg) uci->cpu_sig.sig = cpuid_eax(0x00000001); } + +static void __init get_bsp_sig(void) +{ + unsigned int bsp = boot_cpu_data.cpu_index; + struct ucode_cpu_info *uci = ucode_cpu_info + bsp; + + if (!uci->cpu_sig.sig) + smp_call_function_single(bsp, collect_cpu_sig_on_bsp, NULL, 1); +} #else void load_ucode_amd_ap(void) { @@ -337,31 +346,37 @@ void load_ucode_amd_ap(void) int __init save_microcode_in_initrd_amd(void) { + unsigned long cont; enum ucode_state ret; u32 eax; -#ifdef CONFIG_X86_32 - unsigned int bsp = boot_cpu_data.cpu_index; - struct ucode_cpu_info *uci = ucode_cpu_info + bsp; + if (!container) + return -EINVAL; - if (!uci->cpu_sig.sig) - smp_call_function_single(bsp, collect_cpu_sig_on_bsp, NULL, 1); +#ifdef CONFIG_X86_32 + get_bsp_sig(); + cont = (unsigned long)container; +#else + /* + * We need the physical address of the container for both bitness since + * boot_params.hdr.ramdisk_image is a physical address. + */ + cont = __pa(container); +#endif /* - * Take into account the fact that the ramdisk might get relocated - * and therefore we need to recompute the container's position in - * virtual memory space. + * Take into account the fact that the ramdisk might get relocated and + * therefore we need to recompute the container's position in virtual + * memory space. */ - container = (u8 *)(__va((u32)relocated_ramdisk) + - ((u32)container - boot_params.hdr.ramdisk_image)); -#endif + if (relocated_ramdisk) + container = (u8 *)(__va(relocated_ramdisk) + + (cont - boot_params.hdr.ramdisk_image)); + if (ucode_new_rev) pr_info("microcode: updated early to new patch_level=0x%08x\n", ucode_new_rev); - if (!container) - return -EINVAL; - eax = cpuid_eax(0x00000001); eax = ((eax >> 8) & 0xf) + ((eax >> 20) & 0xff); From 2b4db05e7e2f1efc71e36a9dc1adf5ba164a2330 Mon Sep 17 00:00:00 2001 From: "andrea.merello" Date: Wed, 5 Feb 2014 22:38:05 +0100 Subject: [PATCH 307/596] rtl8180: Add error check for pci_map_single return value in RX path In original code the old RX DMA buffer is unmapped and processed and at the end of the isr a new buffer is mapped with pci_map_single and attached to the RX descriptor. If pci_map_single fails then the RX descriptor remains with no valid DMA buffer attached. In this condition the DMA will target where it shouldn't with obvious evil consequences. Simply avoiding re-arming the descriptor will prevent buggy DMA but it will result soon in RX stuck. This patch move the DMA mapping of the new buffer at the beginning of the ISR (and it adds error check for pci_map_single success/fail). If the DMA mapping fails then we do not unmap the old buffer and we re-arm the descriptor without processing it, with the old DMA buffer still attached. In this way we lose the currently RX-ed packet, but whenever next calls to pci_map_single will succeed again,then the RX process will go on without stuck. Signed-off-by: andrea merello Signed-off-by: John W. Linville --- drivers/net/wireless/rtl818x/rtl8180/dev.c | 16 +++++++++++++--- 1 file changed, 13 insertions(+), 3 deletions(-) diff --git a/drivers/net/wireless/rtl818x/rtl8180/dev.c b/drivers/net/wireless/rtl818x/rtl8180/dev.c index 8ec17aad0e52..fad616a0fa37 100644 --- a/drivers/net/wireless/rtl818x/rtl8180/dev.c +++ b/drivers/net/wireless/rtl818x/rtl8180/dev.c @@ -107,6 +107,7 @@ static void rtl8180_handle_rx(struct ieee80211_hw *dev) struct rtl8180_priv *priv = dev->priv; unsigned int count = 32; u8 signal, agc, sq; + dma_addr_t mapping; while (count--) { struct rtl8180_rx_desc *entry = &priv->rx_ring[priv->rx_idx]; @@ -128,6 +129,17 @@ static void rtl8180_handle_rx(struct ieee80211_hw *dev) if (unlikely(!new_skb)) goto done; + mapping = pci_map_single(priv->pdev, + skb_tail_pointer(new_skb), + MAX_RX_SIZE, PCI_DMA_FROMDEVICE); + + if (pci_dma_mapping_error(priv->pdev, mapping)) { + kfree_skb(new_skb); + dev_err(&priv->pdev->dev, "RX DMA map error\n"); + + goto done; + } + pci_unmap_single(priv->pdev, *((dma_addr_t *)skb->cb), MAX_RX_SIZE, PCI_DMA_FROMDEVICE); @@ -158,9 +170,7 @@ static void rtl8180_handle_rx(struct ieee80211_hw *dev) skb = new_skb; priv->rx_buf[priv->rx_idx] = skb; - *((dma_addr_t *) skb->cb) = - pci_map_single(priv->pdev, skb_tail_pointer(skb), - MAX_RX_SIZE, PCI_DMA_FROMDEVICE); + *((dma_addr_t *) skb->cb) = mapping; } done: From 348f7d4adee97f222e3ad9ff97956ca3793d11c6 Mon Sep 17 00:00:00 2001 From: "andrea.merello" Date: Wed, 5 Feb 2014 22:38:06 +0100 Subject: [PATCH 308/596] rtl8180: Add error check for pci_map_single return value in TX path Orignal code will not detect a DMA mapping failure, causing the HW to attempt a DMA from an invalid address. This patch add the error check and eventually simply drops the TX packet if we can't map it for DMA. Signed-off-by: andrea merello Signed-off-by: John W. Linville --- drivers/net/wireless/rtl818x/rtl8180/dev.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/net/wireless/rtl818x/rtl8180/dev.c b/drivers/net/wireless/rtl818x/rtl8180/dev.c index fad616a0fa37..3867d1470b36 100644 --- a/drivers/net/wireless/rtl818x/rtl8180/dev.c +++ b/drivers/net/wireless/rtl818x/rtl8180/dev.c @@ -276,6 +276,13 @@ static void rtl8180_tx(struct ieee80211_hw *dev, mapping = pci_map_single(priv->pdev, skb->data, skb->len, PCI_DMA_TODEVICE); + if (pci_dma_mapping_error(priv->pdev, mapping)) { + kfree_skb(skb); + dev_err(&priv->pdev->dev, "TX DMA mapping error\n"); + return; + + } + tx_flags = RTL818X_TX_DESC_FLAG_OWN | RTL818X_TX_DESC_FLAG_FS | RTL818X_TX_DESC_FLAG_LS | (ieee80211_get_tx_rate(dev, info)->hw_value << 24) | From d02f8575f1a38b3180a76a8ae0857dfde67f5ead Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Thu, 6 Feb 2014 14:22:21 -0500 Subject: [PATCH 309/596] drm/radeon: add missing include in btc_dpm.c Fixes a compile error with debugfs disabled. Signed-off-by: Alex Deucher --- drivers/gpu/drm/radeon/btc_dpm.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/radeon/btc_dpm.c b/drivers/gpu/drm/radeon/btc_dpm.c index 05ad46982a44..ea103ccdf4bd 100644 --- a/drivers/gpu/drm/radeon/btc_dpm.c +++ b/drivers/gpu/drm/radeon/btc_dpm.c @@ -29,6 +29,7 @@ #include "cypress_dpm.h" #include "btc_dpm.h" #include "atom.h" +#include #define MC_CG_ARB_FREQ_F0 0x0a #define MC_CG_ARB_FREQ_F1 0x0b From 277cba1d28b99169f2a056d0d6f98a4039531cb8 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Thu, 6 Feb 2014 12:04:19 -0800 Subject: [PATCH 310/596] Documentation/kernel-parameters.txt: fix memmap= language Clean up descriptions of memmap= boot options. Add periods (full stops), drop commas, change "used" to "reserved" or "marked". Signed-off-by: Randy Dunlap Cc: Andiry Xu Acked-by: David Rientjes Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/kernel-parameters.txt | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 8f441dab0396..7116fda7077f 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -1726,16 +1726,16 @@ bytes respectively. Such letter suffixes can also be entirely omitted. option description. memmap=nn[KMG]@ss[KMG] - [KNL] Force usage of a specific region of memory - Region of memory to be used, from ss to ss+nn. + [KNL] Force usage of a specific region of memory. + Region of memory to be used is from ss to ss+nn. memmap=nn[KMG]#ss[KMG] [KNL,ACPI] Mark specific memory as ACPI data. - Region of memory to be used, from ss to ss+nn. + Region of memory to be marked is from ss to ss+nn. memmap=nn[KMG]$ss[KMG] [KNL,ACPI] Mark specific memory as reserved. - Region of memory to be used, from ss to ss+nn. + Region of memory to be reserved is from ss to ss+nn. Example: Exclude memory from 0x18690000-0x1869ffff memmap=64K$0x18690000 or From fb951eb5e167de9f07973ce0dfff674a2019bfab Mon Sep 17 00:00:00 2001 From: Zongxun Wang Date: Thu, 6 Feb 2014 12:04:20 -0800 Subject: [PATCH 311/596] ocfs2: free allocated clusters if error occurs after ocfs2_claim_clusters Even if using the same jbd2 handle, we cannot rollback a transaction. So once some error occurs after successfully allocating clusters, the allocated clusters will never be used and it means they are lost. For example, call ocfs2_claim_clusters successfully when expanding a file, but failed in ocfs2_insert_extent. So we need free the allocated clusters if they are not used indeed. Signed-off-by: Zongxun Wang Signed-off-by: Joseph Qi Acked-by: Joel Becker Cc: Mark Fasheh Cc: Li Zefan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ocfs2/alloc.c | 38 +++++++++++++++++++++++++++++++++++--- fs/ocfs2/localalloc.c | 42 ++++++++++++++++++++++++++++++++++++++++++ fs/ocfs2/localalloc.h | 6 ++++++ 3 files changed, 83 insertions(+), 3 deletions(-) diff --git a/fs/ocfs2/alloc.c b/fs/ocfs2/alloc.c index 8750ae1b8636..aada5801567a 100644 --- a/fs/ocfs2/alloc.c +++ b/fs/ocfs2/alloc.c @@ -4742,6 +4742,7 @@ int ocfs2_add_clusters_in_btree(handle_t *handle, enum ocfs2_alloc_restarted *reason_ret) { int status = 0, err = 0; + int need_free = 0; int free_extents; enum ocfs2_alloc_restarted reason = RESTART_NONE; u32 bit_off, num_bits; @@ -4796,7 +4797,8 @@ int ocfs2_add_clusters_in_btree(handle_t *handle, OCFS2_JOURNAL_ACCESS_WRITE); if (status < 0) { mlog_errno(status); - goto leave; + need_free = 1; + goto bail; } block = ocfs2_clusters_to_blocks(osb->sb, bit_off); @@ -4807,7 +4809,8 @@ int ocfs2_add_clusters_in_btree(handle_t *handle, num_bits, flags, meta_ac); if (status < 0) { mlog_errno(status); - goto leave; + need_free = 1; + goto bail; } ocfs2_journal_dirty(handle, et->et_root_bh); @@ -4821,6 +4824,19 @@ int ocfs2_add_clusters_in_btree(handle_t *handle, reason = RESTART_TRANS; } +bail: + if (need_free) { + if (data_ac->ac_which == OCFS2_AC_USE_LOCAL) + ocfs2_free_local_alloc_bits(osb, handle, data_ac, + bit_off, num_bits); + else + ocfs2_free_clusters(handle, + data_ac->ac_inode, + data_ac->ac_bh, + ocfs2_clusters_to_blocks(osb->sb, bit_off), + num_bits); + } + leave: if (reason_ret) *reason_ret = reason; @@ -6805,6 +6821,8 @@ int ocfs2_convert_inline_data_to_extents(struct inode *inode, struct buffer_head *di_bh) { int ret, i, has_data, num_pages = 0; + int need_free = 0; + u32 bit_off, num; handle_t *handle; u64 uninitialized_var(block); struct ocfs2_inode_info *oi = OCFS2_I(inode); @@ -6850,7 +6868,6 @@ int ocfs2_convert_inline_data_to_extents(struct inode *inode, } if (has_data) { - u32 bit_off, num; unsigned int page_end; u64 phys; @@ -6886,6 +6903,7 @@ int ocfs2_convert_inline_data_to_extents(struct inode *inode, ret = ocfs2_grab_eof_pages(inode, 0, end, pages, &num_pages); if (ret) { mlog_errno(ret); + need_free = 1; goto out_commit; } @@ -6896,6 +6914,7 @@ int ocfs2_convert_inline_data_to_extents(struct inode *inode, ret = ocfs2_read_inline_data(inode, pages[0], di_bh); if (ret) { mlog_errno(ret); + need_free = 1; goto out_commit; } @@ -6927,6 +6946,7 @@ int ocfs2_convert_inline_data_to_extents(struct inode *inode, ret = ocfs2_insert_extent(handle, &et, 0, block, 1, 0, NULL); if (ret) { mlog_errno(ret); + need_free = 1; goto out_commit; } @@ -6938,6 +6958,18 @@ int ocfs2_convert_inline_data_to_extents(struct inode *inode, dquot_free_space_nodirty(inode, ocfs2_clusters_to_bytes(osb->sb, 1)); + if (need_free) { + if (data_ac->ac_which == OCFS2_AC_USE_LOCAL) + ocfs2_free_local_alloc_bits(osb, handle, data_ac, + bit_off, num); + else + ocfs2_free_clusters(handle, + data_ac->ac_inode, + data_ac->ac_bh, + ocfs2_clusters_to_blocks(osb->sb, bit_off), + num); + } + ocfs2_commit_trans(osb, handle); out_unlock: diff --git a/fs/ocfs2/localalloc.c b/fs/ocfs2/localalloc.c index cd5496b7a0a3..044013455621 100644 --- a/fs/ocfs2/localalloc.c +++ b/fs/ocfs2/localalloc.c @@ -781,6 +781,48 @@ int ocfs2_claim_local_alloc_bits(struct ocfs2_super *osb, return status; } +int ocfs2_free_local_alloc_bits(struct ocfs2_super *osb, + handle_t *handle, + struct ocfs2_alloc_context *ac, + u32 bit_off, + u32 num_bits) +{ + int status, start; + u32 clear_bits; + struct inode *local_alloc_inode; + void *bitmap; + struct ocfs2_dinode *alloc; + struct ocfs2_local_alloc *la; + + BUG_ON(ac->ac_which != OCFS2_AC_USE_LOCAL); + + local_alloc_inode = ac->ac_inode; + alloc = (struct ocfs2_dinode *) osb->local_alloc_bh->b_data; + la = OCFS2_LOCAL_ALLOC(alloc); + + bitmap = la->la_bitmap; + start = bit_off - le32_to_cpu(la->la_bm_off); + clear_bits = num_bits; + + status = ocfs2_journal_access_di(handle, + INODE_CACHE(local_alloc_inode), + osb->local_alloc_bh, + OCFS2_JOURNAL_ACCESS_WRITE); + if (status < 0) { + mlog_errno(status); + goto bail; + } + + while (clear_bits--) + ocfs2_clear_bit(start++, bitmap); + + le32_add_cpu(&alloc->id1.bitmap1.i_used, -num_bits); + ocfs2_journal_dirty(handle, osb->local_alloc_bh); + +bail: + return status; +} + static u32 ocfs2_local_alloc_count_bits(struct ocfs2_dinode *alloc) { u32 count; diff --git a/fs/ocfs2/localalloc.h b/fs/ocfs2/localalloc.h index 1be9b5864460..44a7d1fb2dec 100644 --- a/fs/ocfs2/localalloc.h +++ b/fs/ocfs2/localalloc.h @@ -55,6 +55,12 @@ int ocfs2_claim_local_alloc_bits(struct ocfs2_super *osb, u32 *bit_off, u32 *num_bits); +int ocfs2_free_local_alloc_bits(struct ocfs2_super *osb, + handle_t *handle, + struct ocfs2_alloc_context *ac, + u32 bit_off, + u32 num_bits); + void ocfs2_local_alloc_seen_free_bits(struct ocfs2_super *osb, unsigned int num_clusters); void ocfs2_la_enable_worker(struct work_struct *work); From 579f82901f6f41256642936d7e632f3979ad76d4 Mon Sep 17 00:00:00 2001 From: Shaohua Li Date: Thu, 6 Feb 2014 12:04:21 -0800 Subject: [PATCH 312/596] swap: add a simple detector for inappropriate swapin readahead This is a patch to improve swap readahead algorithm. It's from Hugh and I slightly changed it. Hugh's original changelog: swapin readahead does a blind readahead, whether or not the swapin is sequential. This may be ok on harddisk, because large reads have relatively small costs, and if the readahead pages are unneeded they can be reclaimed easily - though, what if their allocation forced reclaim of useful pages? But on SSD devices large reads are more expensive than small ones: if the readahead pages are unneeded, reading them in caused significant overhead. This patch adds very simplistic random read detection. Stealing the PageReadahead technique from Konstantin Khlebnikov's patch, avoiding the vma/anon_vma sophistications of Shaohua Li's patch, swapin_nr_pages() simply looks at readahead's current success rate, and narrows or widens its readahead window accordingly. There is little science to its heuristic: it's about as stupid as can be whilst remaining effective. The table below shows elapsed times (in centiseconds) when running a single repetitive swapping load across a 1000MB mapping in 900MB ram with 1GB swap (the harddisk tests had taken painfully too long when I used mem=500M, but SSD shows similar results for that). Vanilla is the 3.6-rc7 kernel on which I started; Shaohua denotes his Sep 3 patch in mmotm and linux-next; HughOld denotes my Oct 1 patch which Shaohua showed to be defective; HughNew this Nov 14 patch, with page_cluster as usual at default of 3 (8-page reads); HughPC4 this same patch with page_cluster 4 (16-page reads); HughPC0 with page_cluster 0 (1-page reads: no readahead). HDD for swapping to harddisk, SSD for swapping to VertexII SSD. Seq for sequential access to the mapping, cycling five times around; Rand for the same number of random touches. Anon for a MAP_PRIVATE anon mapping; Shmem for a MAP_SHARED anon mapping, equivalent to tmpfs. One weakness of Shaohua's vma/anon_vma approach was that it did not optimize Shmem: seen below. Konstantin's approach was perhaps mistuned, 50% slower on Seq: did not compete and is not shown below. HDD Vanilla Shaohua HughOld HughNew HughPC4 HughPC0 Seq Anon 73921 76210 75611 76904 78191 121542 Seq Shmem 73601 73176 73855 72947 74543 118322 Rand Anon 895392 831243 871569 845197 846496 841680 Rand Shmem 1058375 1053486 827935 764955 764376 756489 SSD Vanilla Shaohua HughOld HughNew HughPC4 HughPC0 Seq Anon 24634 24198 24673 25107 21614 70018 Seq Shmem 24959 24932 25052 25703 22030 69678 Rand Anon 43014 26146 28075 25989 26935 25901 Rand Shmem 45349 45215 28249 24268 24138 24332 These tests are, of course, two extremes of a very simple case: under heavier mixed loads I've not yet observed any consistent improvement or degradation, and wider testing would be welcome. Shaohua Li: Test shows Vanilla is slightly better in sequential workload than Hugh's patch. I observed with Hugh's patch sometimes the readahead size is shrinked too fast (from 8 to 1 immediately) in sequential workload if there is no hit. And in such case, continuing doing readahead is good actually. I don't prepare a sophisticated algorithm for the sequential workload because so far we can't guarantee sequential accessed pages are swap out sequentially. So I slightly change Hugh's heuristic - don't shrink readahead size too fast. Here is my test result (unit second, 3 runs average): Vanilla Hugh New Seq 356 370 360 Random 4525 2447 2444 Attached graph is the swapin/swapout throughput I collected with 'vmstat 2'. The first part is running a random workload (till around 1200 of the x-axis) and the second part is running a sequential workload. swapin and swapout throughput are almost identical in steady state in both workloads. These are expected behavior. while in Vanilla, swapin is much bigger than swapout especially in random workload (because wrong readahead). Original patches by: Shaohua Li and Konstantin Khlebnikov. [fengguang.wu@intel.com: swapin_nr_pages() can be static] Signed-off-by: Hugh Dickins Signed-off-by: Shaohua Li Signed-off-by: Fengguang Wu Cc: Rik van Riel Cc: Wu Fengguang Cc: Minchan Kim Cc: Konstantin Khlebnikov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/page-flags.h | 4 +-- mm/swap_state.c | 63 ++++++++++++++++++++++++++++++++++++-- 2 files changed, 62 insertions(+), 5 deletions(-) diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h index e464b4e987e8..d1fe1a761047 100644 --- a/include/linux/page-flags.h +++ b/include/linux/page-flags.h @@ -228,9 +228,9 @@ PAGEFLAG(OwnerPriv1, owner_priv_1) TESTCLEARFLAG(OwnerPriv1, owner_priv_1) TESTPAGEFLAG(Writeback, writeback) TESTSCFLAG(Writeback, writeback) PAGEFLAG(MappedToDisk, mappedtodisk) -/* PG_readahead is only used for file reads; PG_reclaim is only for writes */ +/* PG_readahead is only used for reads; PG_reclaim is only for writes */ PAGEFLAG(Reclaim, reclaim) TESTCLEARFLAG(Reclaim, reclaim) -PAGEFLAG(Readahead, reclaim) /* Reminder to do async read-ahead */ +PAGEFLAG(Readahead, reclaim) TESTCLEARFLAG(Readahead, reclaim) #ifdef CONFIG_HIGHMEM /* diff --git a/mm/swap_state.c b/mm/swap_state.c index 98e85e9c2b2d..e76ace30d436 100644 --- a/mm/swap_state.c +++ b/mm/swap_state.c @@ -63,6 +63,8 @@ unsigned long total_swapcache_pages(void) return ret; } +static atomic_t swapin_readahead_hits = ATOMIC_INIT(4); + void show_swap_cache_info(void) { printk("%lu pages in swap cache\n", total_swapcache_pages()); @@ -286,8 +288,11 @@ struct page * lookup_swap_cache(swp_entry_t entry) page = find_get_page(swap_address_space(entry), entry.val); - if (page) + if (page) { INC_CACHE_INFO(find_success); + if (TestClearPageReadahead(page)) + atomic_inc(&swapin_readahead_hits); + } INC_CACHE_INFO(find_total); return page; @@ -389,6 +394,50 @@ struct page *read_swap_cache_async(swp_entry_t entry, gfp_t gfp_mask, return found_page; } +static unsigned long swapin_nr_pages(unsigned long offset) +{ + static unsigned long prev_offset; + unsigned int pages, max_pages, last_ra; + static atomic_t last_readahead_pages; + + max_pages = 1 << ACCESS_ONCE(page_cluster); + if (max_pages <= 1) + return 1; + + /* + * This heuristic has been found to work well on both sequential and + * random loads, swapping to hard disk or to SSD: please don't ask + * what the "+ 2" means, it just happens to work well, that's all. + */ + pages = atomic_xchg(&swapin_readahead_hits, 0) + 2; + if (pages == 2) { + /* + * We can have no readahead hits to judge by: but must not get + * stuck here forever, so check for an adjacent offset instead + * (and don't even bother to check whether swap type is same). + */ + if (offset != prev_offset + 1 && offset != prev_offset - 1) + pages = 1; + prev_offset = offset; + } else { + unsigned int roundup = 4; + while (roundup < pages) + roundup <<= 1; + pages = roundup; + } + + if (pages > max_pages) + pages = max_pages; + + /* Don't shrink readahead too fast */ + last_ra = atomic_read(&last_readahead_pages) / 2; + if (pages < last_ra) + pages = last_ra; + atomic_set(&last_readahead_pages, pages); + + return pages; +} + /** * swapin_readahead - swap in pages in hope we need them soon * @entry: swap entry of this memory @@ -412,11 +461,16 @@ struct page *swapin_readahead(swp_entry_t entry, gfp_t gfp_mask, struct vm_area_struct *vma, unsigned long addr) { struct page *page; - unsigned long offset = swp_offset(entry); + unsigned long entry_offset = swp_offset(entry); + unsigned long offset = entry_offset; unsigned long start_offset, end_offset; - unsigned long mask = (1UL << page_cluster) - 1; + unsigned long mask; struct blk_plug plug; + mask = swapin_nr_pages(offset) - 1; + if (!mask) + goto skip; + /* Read a page_cluster sized and aligned cluster around offset. */ start_offset = offset & ~mask; end_offset = offset | mask; @@ -430,10 +484,13 @@ struct page *swapin_readahead(swp_entry_t entry, gfp_t gfp_mask, gfp_mask, vma, addr); if (!page) continue; + if (offset != entry_offset) + SetPageReadahead(page); page_cache_release(page); } blk_finish_plug(&plug); lru_add_drain(); /* Push any new pages onto the LRU now */ +skip: return read_swap_cache_async(entry, gfp_mask, vma, addr); } From f893ab41e4dae2fe8991faf5d86d029068d1ef3a Mon Sep 17 00:00:00 2001 From: Weijie Yang Date: Thu, 6 Feb 2014 12:04:23 -0800 Subject: [PATCH 313/596] mm/swap: fix race on swap_info reuse between swapoff and swapon swapoff clear swap_info's SWP_USED flag prematurely and free its resources after that. A concurrent swapon will reuse this swap_info while its previous resources are not cleared completely. These late freed resources are: - p->percpu_cluster - swap_cgroup_ctrl[type] - block_device setting - inode->i_flags &= ~S_SWAPFILE This patch clears the SWP_USED flag after all its resources are freed, so that swapon can reuse this swap_info by alloc_swap_info() safely. [akpm@linux-foundation.org: tidy up code comment] Signed-off-by: Weijie Yang Acked-by: Hugh Dickins Cc: Krzysztof Kozlowski Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/swapfile.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/mm/swapfile.c b/mm/swapfile.c index c6c13b050a58..4a7f7e6992b6 100644 --- a/mm/swapfile.c +++ b/mm/swapfile.c @@ -1923,7 +1923,6 @@ SYSCALL_DEFINE1(swapoff, const char __user *, specialfile) p->swap_map = NULL; cluster_info = p->cluster_info; p->cluster_info = NULL; - p->flags = 0; frontswap_map = frontswap_map_get(p); spin_unlock(&p->lock); spin_unlock(&swap_lock); @@ -1949,6 +1948,16 @@ SYSCALL_DEFINE1(swapoff, const char __user *, specialfile) mutex_unlock(&inode->i_mutex); } filp_close(swap_file, NULL); + + /* + * Clear the SWP_USED flag after all resources are freed so that swapon + * can reuse this swap_info in alloc_swap_info() safely. It is ok to + * not hold p->lock after we cleared its SWP_WRITEOK. + */ + spin_lock(&swap_lock); + p->flags = 0; + spin_unlock(&swap_lock); + err = 0; atomic_inc(&proc_poll_event); wake_up_interruptible(&proc_poll_wait); From a85d9df1ea1d23682a0ed1e100e6965006595d06 Mon Sep 17 00:00:00 2001 From: KOSAKI Motohiro Date: Thu, 6 Feb 2014 12:04:24 -0800 Subject: [PATCH 314/596] mm: __set_page_dirty_nobuffers() uses spin_lock_irqsave() instead of spin_lock_irq() During aio stress test, we observed the following lockdep warning. This mean AIO+numa_balancing is currently deadlockable. The problem is, aio_migratepage disable interrupt, but __set_page_dirty_nobuffers unintentionally enable it again. Generally, all helper function should use spin_lock_irqsave() instead of spin_lock_irq() because they don't know caller at all. other info that might help us debug this: Possible unsafe locking scenario: CPU0 ---- lock(&(&ctx->completion_lock)->rlock); lock(&(&ctx->completion_lock)->rlock); *** DEADLOCK *** dump_stack+0x19/0x1b print_usage_bug+0x1f7/0x208 mark_lock+0x21d/0x2a0 mark_held_locks+0xb9/0x140 trace_hardirqs_on_caller+0x105/0x1d0 trace_hardirqs_on+0xd/0x10 _raw_spin_unlock_irq+0x2c/0x50 __set_page_dirty_nobuffers+0x8c/0xf0 migrate_page_copy+0x434/0x540 aio_migratepage+0xb1/0x140 move_to_new_page+0x7d/0x230 migrate_pages+0x5e5/0x700 migrate_misplaced_page+0xbc/0xf0 do_numa_page+0x102/0x190 handle_pte_fault+0x241/0x970 handle_mm_fault+0x265/0x370 __do_page_fault+0x172/0x5a0 do_page_fault+0x1a/0x70 page_fault+0x28/0x30 Signed-off-by: KOSAKI Motohiro Cc: Larry Woodman Cc: Rik van Riel Cc: Johannes Weiner Acked-by: David Rientjes Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/page-writeback.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/mm/page-writeback.c b/mm/page-writeback.c index 2d30e2cfe804..7106cb1aca8e 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -2173,11 +2173,12 @@ int __set_page_dirty_nobuffers(struct page *page) if (!TestSetPageDirty(page)) { struct address_space *mapping = page_mapping(page); struct address_space *mapping2; + unsigned long flags; if (!mapping) return 1; - spin_lock_irq(&mapping->tree_lock); + spin_lock_irqsave(&mapping->tree_lock, flags); mapping2 = page_mapping(page); if (mapping2) { /* Race with truncate? */ BUG_ON(mapping2 != mapping); @@ -2186,7 +2187,7 @@ int __set_page_dirty_nobuffers(struct page *page) radix_tree_tag_set(&mapping->page_tree, page_index(page), PAGECACHE_TAG_DIRTY); } - spin_unlock_irq(&mapping->tree_lock); + spin_unlock_irqrestore(&mapping->tree_lock, flags); if (mapping->host) { /* !PageAnon && !swapper_space */ __mark_inode_dirty(mapping->host, I_DIRTY_PAGES); From 017c217a26e9bf6948482f751b30d0507e30a7d0 Mon Sep 17 00:00:00 2001 From: Tang Chen Date: Thu, 6 Feb 2014 12:04:25 -0800 Subject: [PATCH 315/596] arch/x86/mm/numa.c: initialize numa_kernel_nodes in numa_clear_kernel_node_hotplug() On-stack variable numa_kernel_nodes in numa_clear_kernel_node_hotplug() was not initialized. So we need to initialize it. [akpm@linux-foundation.org: use NODE_MASK_NONE, per David] Signed-off-by: Tang Chen Tested-by: Gu Zheng Reported-by: Dave Jones Reported-by: David Rientjes Tested-by: Dave Jones Cc: Ingo Molnar Cc: Thomas Gleixner Cc: "H. Peter Anvin" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86/mm/numa.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/mm/numa.c b/arch/x86/mm/numa.c index 81b2750f3666..45ec9d72b6dd 100644 --- a/arch/x86/mm/numa.c +++ b/arch/x86/mm/numa.c @@ -565,7 +565,7 @@ static void __init numa_init_array(void) static void __init numa_clear_kernel_node_hotplug(void) { int i, nid; - nodemask_t numa_kernel_nodes; + nodemask_t numa_kernel_nodes = NODE_MASK_NONE; unsigned long start, end; struct memblock_type *type = &memblock.reserved; From 7bc35fdde6724549a0239b71e08b9f33d8bf2bfb Mon Sep 17 00:00:00 2001 From: Tang Chen Date: Thu, 6 Feb 2014 12:04:27 -0800 Subject: [PATCH 316/596] arch/x86/mm/numa.c: fix array index overflow when synchronizing nid to memblock.reserved. The following path will cause array out of bound. memblock_add_region() will always set nid in memblock.reserved to MAX_NUMNODES. In numa_register_memblks(), after we set all nid to correct valus in memblock.reserved, we called setup_node_data(), and used memblock_alloc_nid() to allocate memory, with nid set to MAX_NUMNODES. The nodemask_t type can be seen as a bit array. And the index is 0 ~ MAX_NUMNODES-1. After that, when we call node_set() in numa_clear_kernel_node_hotplug(), the nodemask_t got an index of value MAX_NUMNODES, which is out of [0 ~ MAX_NUMNODES-1]. See below: numa_init() |---> numa_register_memblks() | |---> memblock_set_node(memory) set correct nid in memblock.memory | |---> memblock_set_node(reserved) set correct nid in memblock.reserved | |...... | |---> setup_node_data() | |---> memblock_alloc_nid() here, nid is set to MAX_NUMNODES (1024) |...... |---> numa_clear_kernel_node_hotplug() |---> node_set() here, we have an index 1024, and overflowed This patch moves nid setting to numa_clear_kernel_node_hotplug() to fix this problem. Reported-by: Dave Jones Signed-off-by: Tang Chen Tested-by: Gu Zheng Reported-by: Dave Jones Cc: David Rientjes Tested-by: Dave Jones Cc: Ingo Molnar Cc: Thomas Gleixner Cc: "H. Peter Anvin" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86/mm/numa.c | 19 +++++++++++-------- 1 file changed, 11 insertions(+), 8 deletions(-) diff --git a/arch/x86/mm/numa.c b/arch/x86/mm/numa.c index 45ec9d72b6dd..27aa0455fab3 100644 --- a/arch/x86/mm/numa.c +++ b/arch/x86/mm/numa.c @@ -493,14 +493,6 @@ static int __init numa_register_memblks(struct numa_meminfo *mi) struct numa_memblk *mb = &mi->blk[i]; memblock_set_node(mb->start, mb->end - mb->start, &memblock.memory, mb->nid); - - /* - * At this time, all memory regions reserved by memblock are - * used by the kernel. Set the nid in memblock.reserved will - * mark out all the nodes the kernel resides in. - */ - memblock_set_node(mb->start, mb->end - mb->start, - &memblock.reserved, mb->nid); } /* @@ -569,6 +561,17 @@ static void __init numa_clear_kernel_node_hotplug(void) unsigned long start, end; struct memblock_type *type = &memblock.reserved; + /* + * At this time, all memory regions reserved by memblock are + * used by the kernel. Set the nid in memblock.reserved will + * mark out all the nodes the kernel resides in. + */ + for (i = 0; i < numa_meminfo.nr_blks; i++) { + struct numa_memblk *mb = &numa_meminfo.blk[i]; + memblock_set_node(mb->start, mb->end - mb->start, + &memblock.reserved, mb->nid); + } + /* Mark all kernel nodes. */ for (i = 0; i < type->cnt; i++) node_set(type->regions[i].nid, numa_kernel_nodes); From 227d53b397a32a7614667b3ecaf1d89902fb6c12 Mon Sep 17 00:00:00 2001 From: KOSAKI Motohiro Date: Thu, 6 Feb 2014 12:04:28 -0800 Subject: [PATCH 317/596] mm: __set_page_dirty uses spin_lock_irqsave instead of spin_lock_irq To use spin_{un}lock_irq is dangerous if caller disabled interrupt. During aio buffer migration, we have a possibility to see the following call stack. aio_migratepage [disable interrupt] migrate_page_copy clear_page_dirty_for_io set_page_dirty __set_page_dirty_buffers __set_page_dirty spin_lock_irq This mean, current aio migration is a deadlockable. spin_lock_irqsave is a safer alternative and we should use it. Signed-off-by: KOSAKI Motohiro Reported-by: David Rientjes rientjes@google.com> Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/buffer.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/fs/buffer.c b/fs/buffer.c index 651dba10b9c2..27265a8b43c1 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -654,14 +654,16 @@ EXPORT_SYMBOL(mark_buffer_dirty_inode); static void __set_page_dirty(struct page *page, struct address_space *mapping, int warn) { - spin_lock_irq(&mapping->tree_lock); + unsigned long flags; + + spin_lock_irqsave(&mapping->tree_lock, flags); if (page->mapping) { /* Race with truncate? */ WARN_ON_ONCE(warn && !PageUptodate(page)); account_page_dirtied(page, mapping); radix_tree_tag_set(&mapping->page_tree, page_index(page), PAGECACHE_TAG_DIRTY); } - spin_unlock_irq(&mapping->tree_lock); + spin_unlock_irqrestore(&mapping->tree_lock, flags); __mark_inode_dirty(mapping->host, I_DIRTY_PAGES); } From 4f545a4ba158600b7a780b9daaf9ffabb934cdb6 Mon Sep 17 00:00:00 2001 From: Adam Thomson Date: Thu, 6 Feb 2014 18:03:17 +0000 Subject: [PATCH 318/596] hwmon: (da9055) Remove use of regmap_irq_get_virq() Remove use of regmap_irq_get_virq() in driver probe which was conflicting with use of platform_get_irq_byname(). platform_get_irq_byname() already returns the VIRQ number due to MFD core translation so using regmap_irq_get_virq() on that returned value results in an incorrect IRQ being requested. The driver probes then fail because of this. Signed-off-by: Adam Thomson Signed-off-by: Guenter Roeck --- drivers/hwmon/da9055-hwmon.c | 4 ---- 1 file changed, 4 deletions(-) diff --git a/drivers/hwmon/da9055-hwmon.c b/drivers/hwmon/da9055-hwmon.c index 029ecabc4380..73b3865f1207 100644 --- a/drivers/hwmon/da9055-hwmon.c +++ b/drivers/hwmon/da9055-hwmon.c @@ -278,10 +278,6 @@ static int da9055_hwmon_probe(struct platform_device *pdev) if (hwmon_irq < 0) return hwmon_irq; - hwmon_irq = regmap_irq_get_virq(hwmon->da9055->irq_data, hwmon_irq); - if (hwmon_irq < 0) - return hwmon_irq; - ret = devm_request_threaded_irq(&pdev->dev, hwmon_irq, NULL, da9055_auxadc_irq, IRQF_TRIGGER_HIGH | IRQF_ONESHOT, From 6ca605f7c70895a35737435f17ae9cc5e36f1466 Mon Sep 17 00:00:00 2001 From: Sachin Kamat Date: Thu, 16 Jan 2014 11:31:26 +0530 Subject: [PATCH 319/596] drm/exynos: Fix freeing issues in exynos_drm_drv.c Fixes the following errors: drivers/gpu/drm/exynos/exynos_drm_drv.c:182 exynos_drm_open() error: double free of 'file_priv' drivers/gpu/drm/exynos/exynos_drm_drv.c:188 exynos_drm_open() error: dereferencing freed memory 'file_priv' Signed-off-by: Sachin Kamat Signed-off-by: Inki Dae --- drivers/gpu/drm/exynos/exynos_drm_drv.c | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-) diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.c b/drivers/gpu/drm/exynos/exynos_drm_drv.c index 9d096a0c5f8d..215131ab1dd2 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_drv.c +++ b/drivers/gpu/drm/exynos/exynos_drm_drv.c @@ -171,21 +171,23 @@ static int exynos_drm_open(struct drm_device *dev, struct drm_file *file) file->driver_priv = file_priv; ret = exynos_drm_subdrv_open(dev, file); - if (ret) { - kfree(file_priv); - file->driver_priv = NULL; - } + if (ret) + goto out; anon_filp = anon_inode_getfile("exynos_gem", &exynos_drm_gem_fops, NULL, 0); if (IS_ERR(anon_filp)) { - kfree(file_priv); - return PTR_ERR(anon_filp); + ret = PTR_ERR(anon_filp); + goto out; } anon_filp->f_mode = FMODE_READ | FMODE_WRITE; file_priv->anon_filp = anon_filp; + return ret; +out: + kfree(file_priv); + file->driver_priv = NULL; return ret; } From 86ac5b84efa87c3ed741dd73597f8c29bba53da2 Mon Sep 17 00:00:00 2001 From: Tushar Behera Date: Thu, 16 Jan 2014 11:57:57 +0530 Subject: [PATCH 320/596] drm/exynos: Fix multiplatform breakage for ipp/gsc There is no need to include "plat/map-base.h" in ipp driver. Remove this and enable this driver for multi-platform. However gsc driver is not multiplatform compliant yet, so make the compilation conditional upon !ARCH_MULTIPLATFORM. Signed-off-by: Tushar Behera Signed-off-by: Inki Dae --- drivers/gpu/drm/exynos/Kconfig | 4 ++-- drivers/gpu/drm/exynos/exynos_drm_ipp.c | 1 - 2 files changed, 2 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/exynos/Kconfig b/drivers/gpu/drm/exynos/Kconfig index f227f544aa36..6e1a1a20cf6b 100644 --- a/drivers/gpu/drm/exynos/Kconfig +++ b/drivers/gpu/drm/exynos/Kconfig @@ -51,7 +51,7 @@ config DRM_EXYNOS_G2D config DRM_EXYNOS_IPP bool "Exynos DRM IPP" - depends on DRM_EXYNOS && !ARCH_MULTIPLATFORM + depends on DRM_EXYNOS help Choose this option if you want to use IPP feature for DRM. @@ -69,6 +69,6 @@ config DRM_EXYNOS_ROTATOR config DRM_EXYNOS_GSC bool "Exynos DRM GSC" - depends on DRM_EXYNOS_IPP && ARCH_EXYNOS5 + depends on DRM_EXYNOS_IPP && ARCH_EXYNOS5 && !ARCH_MULTIPLATFORM help Choose this option if you want to use Exynos GSC for DRM. diff --git a/drivers/gpu/drm/exynos/exynos_drm_ipp.c b/drivers/gpu/drm/exynos/exynos_drm_ipp.c index d519a4e5fe40..eefb429d62a1 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_ipp.c +++ b/drivers/gpu/drm/exynos/exynos_drm_ipp.c @@ -16,7 +16,6 @@ #include #include #include -#include #include #include From 5cdbc8d993074ffda3d3be68935b8afe3ff96b9e Mon Sep 17 00:00:00 2001 From: Sachin Kamat Date: Thu, 16 Jan 2014 10:00:22 +0530 Subject: [PATCH 321/596] drm/exynos: Remove unnecessary semicolon Semicolon after a switch statement is not needed. Signed-off-by: Sachin Kamat Signed-off-by: Inki Dae --- drivers/gpu/drm/exynos/exynos_drm_g2d.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/exynos/exynos_drm_g2d.c b/drivers/gpu/drm/exynos/exynos_drm_g2d.c index 380aec28840b..6c1885eedfdf 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_g2d.c +++ b/drivers/gpu/drm/exynos/exynos_drm_g2d.c @@ -607,7 +607,7 @@ static enum g2d_reg_type g2d_get_reg_type(int reg_offset) reg_type = REG_TYPE_NONE; DRM_ERROR("Unknown register offset![%d]\n", reg_offset); break; - }; + } return reg_type; } From 4fe25b822bca31aa9bf97c1a56747443f0d7d239 Mon Sep 17 00:00:00 2001 From: Sachin Kamat Date: Thu, 16 Jan 2014 10:00:23 +0530 Subject: [PATCH 322/596] drm/exynos: Fix trivial typo Changed quf -> qbuf. Signed-off-by: Sachin Kamat Signed-off-by: Inki Dae --- drivers/gpu/drm/exynos/exynos_drm_ipp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/exynos/exynos_drm_ipp.c b/drivers/gpu/drm/exynos/exynos_drm_ipp.c index eefb429d62a1..09312b877470 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_ipp.c +++ b/drivers/gpu/drm/exynos/exynos_drm_ipp.c @@ -825,7 +825,7 @@ static void ipp_put_event(struct drm_exynos_ipp_cmd_node *c_node, DRM_DEBUG_KMS("count[%d]e[0x%x]\n", count++, (int)e); /* - * quf == NULL condition means all event deletion. + * qbuf == NULL condition means all event deletion. * stop operations want to delete all event list. * another case delete only same buf id. */ From d34d59bda242cf97b90161ea5fe339f0daf0cc33 Mon Sep 17 00:00:00 2001 From: Sachin Kamat Date: Tue, 4 Feb 2014 08:40:18 +0530 Subject: [PATCH 323/596] drm/exynos: Convert to use the standard hdmi.h header Remove local definitions and use the ones provided by hdmi.h. Signed-off-by: Sachin Kamat Signed-off-by: Inki Dae --- drivers/gpu/drm/exynos/exynos_hdmi.c | 66 +++++++++++----------------- 1 file changed, 26 insertions(+), 40 deletions(-) diff --git a/drivers/gpu/drm/exynos/exynos_hdmi.c b/drivers/gpu/drm/exynos/exynos_hdmi.c index a0e10aeb0e67..c021ddc1ffb4 100644 --- a/drivers/gpu/drm/exynos/exynos_hdmi.c +++ b/drivers/gpu/drm/exynos/exynos_hdmi.c @@ -34,6 +34,7 @@ #include #include #include +#include #include @@ -59,19 +60,6 @@ #define HDMI_AUI_VERSION 0x01 #define HDMI_AUI_LENGTH 0x0A -/* HDMI infoframe to configure HDMI out packet header, AUI and AVI */ -enum HDMI_PACKET_TYPE { - /* refer to Table 5-8 Packet Type in HDMI specification v1.4a */ - /* InfoFrame packet type */ - HDMI_PACKET_TYPE_INFOFRAME = 0x80, - /* Vendor-Specific InfoFrame */ - HDMI_PACKET_TYPE_VSI = HDMI_PACKET_TYPE_INFOFRAME + 1, - /* Auxiliary Video information InfoFrame */ - HDMI_PACKET_TYPE_AVI = HDMI_PACKET_TYPE_INFOFRAME + 2, - /* Audio information InfoFrame */ - HDMI_PACKET_TYPE_AUI = HDMI_PACKET_TYPE_INFOFRAME + 4 -}; - enum hdmi_type { HDMI_TYPE13, HDMI_TYPE14, @@ -379,12 +367,6 @@ static const struct hdmiphy_config hdmiphy_v14_configs[] = { }, }; -struct hdmi_infoframe { - enum HDMI_PACKET_TYPE type; - u8 ver; - u8 len; -}; - static inline u32 hdmi_reg_read(struct hdmi_context *hdata, u32 reg_id) { return readl(hdata->regs + reg_id); @@ -682,7 +664,7 @@ static u8 hdmi_chksum(struct hdmi_context *hdata, } static void hdmi_reg_infoframe(struct hdmi_context *hdata, - struct hdmi_infoframe *infoframe) + union hdmi_infoframe *infoframe) { u32 hdr_sum; u8 chksum; @@ -700,13 +682,15 @@ static void hdmi_reg_infoframe(struct hdmi_context *hdata, return; } - switch (infoframe->type) { - case HDMI_PACKET_TYPE_AVI: + switch (infoframe->any.type) { + case HDMI_INFOFRAME_TYPE_AVI: hdmi_reg_writeb(hdata, HDMI_AVI_CON, HDMI_AVI_CON_EVERY_VSYNC); - hdmi_reg_writeb(hdata, HDMI_AVI_HEADER0, infoframe->type); - hdmi_reg_writeb(hdata, HDMI_AVI_HEADER1, infoframe->ver); - hdmi_reg_writeb(hdata, HDMI_AVI_HEADER2, infoframe->len); - hdr_sum = infoframe->type + infoframe->ver + infoframe->len; + hdmi_reg_writeb(hdata, HDMI_AVI_HEADER0, infoframe->any.type); + hdmi_reg_writeb(hdata, HDMI_AVI_HEADER1, + infoframe->any.version); + hdmi_reg_writeb(hdata, HDMI_AVI_HEADER2, infoframe->any.length); + hdr_sum = infoframe->any.type + infoframe->any.version + + infoframe->any.length; /* Output format zero hardcoded ,RGB YBCR selection */ hdmi_reg_writeb(hdata, HDMI_AVI_BYTE(1), 0 << 5 | @@ -722,18 +706,20 @@ static void hdmi_reg_infoframe(struct hdmi_context *hdata, hdmi_reg_writeb(hdata, HDMI_AVI_BYTE(4), vic); chksum = hdmi_chksum(hdata, HDMI_AVI_BYTE(1), - infoframe->len, hdr_sum); + infoframe->any.length, hdr_sum); DRM_DEBUG_KMS("AVI checksum = 0x%x\n", chksum); hdmi_reg_writeb(hdata, HDMI_AVI_CHECK_SUM, chksum); break; - case HDMI_PACKET_TYPE_AUI: + case HDMI_INFOFRAME_TYPE_AUDIO: hdmi_reg_writeb(hdata, HDMI_AUI_CON, 0x02); - hdmi_reg_writeb(hdata, HDMI_AUI_HEADER0, infoframe->type); - hdmi_reg_writeb(hdata, HDMI_AUI_HEADER1, infoframe->ver); - hdmi_reg_writeb(hdata, HDMI_AUI_HEADER2, infoframe->len); - hdr_sum = infoframe->type + infoframe->ver + infoframe->len; + hdmi_reg_writeb(hdata, HDMI_AUI_HEADER0, infoframe->any.type); + hdmi_reg_writeb(hdata, HDMI_AUI_HEADER1, + infoframe->any.version); + hdmi_reg_writeb(hdata, HDMI_AUI_HEADER2, infoframe->any.length); + hdr_sum = infoframe->any.type + infoframe->any.version + + infoframe->any.length; chksum = hdmi_chksum(hdata, HDMI_AUI_BYTE(1), - infoframe->len, hdr_sum); + infoframe->any.length, hdr_sum); DRM_DEBUG_KMS("AUI checksum = 0x%x\n", chksum); hdmi_reg_writeb(hdata, HDMI_AUI_CHECK_SUM, chksum); break; @@ -985,7 +971,7 @@ static void hdmi_conf_reset(struct hdmi_context *hdata) static void hdmi_conf_init(struct hdmi_context *hdata) { - struct hdmi_infoframe infoframe; + union hdmi_infoframe infoframe; /* disable HPD interrupts from HDMI IP block, use GPIO instead */ hdmi_reg_writemask(hdata, HDMI_INTC_CON, 0, HDMI_INTC_EN_GLOBAL | @@ -1021,14 +1007,14 @@ static void hdmi_conf_init(struct hdmi_context *hdata) hdmi_reg_writeb(hdata, HDMI_V13_AUI_CON, 0x02); hdmi_reg_writeb(hdata, HDMI_V13_ACR_CON, 0x04); } else { - infoframe.type = HDMI_PACKET_TYPE_AVI; - infoframe.ver = HDMI_AVI_VERSION; - infoframe.len = HDMI_AVI_LENGTH; + infoframe.any.type = HDMI_INFOFRAME_TYPE_AVI; + infoframe.any.version = HDMI_AVI_VERSION; + infoframe.any.length = HDMI_AVI_LENGTH; hdmi_reg_infoframe(hdata, &infoframe); - infoframe.type = HDMI_PACKET_TYPE_AUI; - infoframe.ver = HDMI_AUI_VERSION; - infoframe.len = HDMI_AUI_LENGTH; + infoframe.any.type = HDMI_INFOFRAME_TYPE_AUDIO; + infoframe.any.version = HDMI_AUI_VERSION; + infoframe.any.length = HDMI_AUI_LENGTH; hdmi_reg_infoframe(hdata, &infoframe); /* enable AVI packet every vsync, fixes purple line problem */ From 4dae1686782d52084893421b89ff1aa61e223aaf Mon Sep 17 00:00:00 2001 From: Maxime Ripard Date: Sun, 2 Feb 2014 14:49:11 +0100 Subject: [PATCH 324/596] net: ethernet: sunxi: Add new compatibles The Allwinner A10 compatibles were following a slightly different compatible patterns than the rest of the SoCs for historical reasons. Add compatibles matching the other pattern to the ethernet driver for consistency, and keep the older one for backward compatibility. Signed-off-by: Maxime Ripard Signed-off-by: David S. Miller --- .../devicetree/bindings/net/allwinner,sun4i-emac.txt | 5 +++-- drivers/net/ethernet/allwinner/sun4i-emac.c | 3 +++ 2 files changed, 6 insertions(+), 2 deletions(-) diff --git a/Documentation/devicetree/bindings/net/allwinner,sun4i-emac.txt b/Documentation/devicetree/bindings/net/allwinner,sun4i-emac.txt index b90bfcd138ff..863d5b8155c7 100644 --- a/Documentation/devicetree/bindings/net/allwinner,sun4i-emac.txt +++ b/Documentation/devicetree/bindings/net/allwinner,sun4i-emac.txt @@ -1,7 +1,8 @@ * Allwinner EMAC ethernet controller Required properties: -- compatible: should be "allwinner,sun4i-emac". +- compatible: should be "allwinner,sun4i-a10-emac" (Deprecated: + "allwinner,sun4i-emac") - reg: address and length of the register set for the device. - interrupts: interrupt for the device - phy: A phandle to a phy node defining the PHY address (as the reg @@ -14,7 +15,7 @@ Optional properties: Example: emac: ethernet@01c0b000 { - compatible = "allwinner,sun4i-emac"; + compatible = "allwinner,sun4i-a10-emac"; reg = <0x01c0b000 0x1000>; interrupts = <55>; clocks = <&ahb_gates 17>; diff --git a/drivers/net/ethernet/allwinner/sun4i-emac.c b/drivers/net/ethernet/allwinner/sun4i-emac.c index 0cc21437478c..511f6eecd58b 100644 --- a/drivers/net/ethernet/allwinner/sun4i-emac.c +++ b/drivers/net/ethernet/allwinner/sun4i-emac.c @@ -929,6 +929,9 @@ static int emac_resume(struct platform_device *dev) } static const struct of_device_id emac_of_match[] = { + {.compatible = "allwinner,sun4i-a10-emac",}, + + /* Deprecated */ {.compatible = "allwinner,sun4i-emac",}, {}, }; From efe205704fa8435b7bca495138dd2dbd973e0c28 Mon Sep 17 00:00:00 2001 From: Maxime Ripard Date: Sun, 2 Feb 2014 14:49:12 +0100 Subject: [PATCH 325/596] net: phy: sunxi: Add new compatibles The Allwinner A10 compatibles were following a slightly different compatible patterns than the rest of the SoCs for historical reasons. Add compatibles matching the other pattern to the mdio driver for consistency, and keep the older one for backward compatibility. Signed-off-by: Maxime Ripard Signed-off-by: David S. Miller --- .../devicetree/bindings/net/allwinner,sun4i-mdio.txt | 5 +++-- drivers/net/phy/mdio-sun4i.c | 3 +++ 2 files changed, 6 insertions(+), 2 deletions(-) diff --git a/Documentation/devicetree/bindings/net/allwinner,sun4i-mdio.txt b/Documentation/devicetree/bindings/net/allwinner,sun4i-mdio.txt index 00b9f9a3ec1d..4ec56413779d 100644 --- a/Documentation/devicetree/bindings/net/allwinner,sun4i-mdio.txt +++ b/Documentation/devicetree/bindings/net/allwinner,sun4i-mdio.txt @@ -1,7 +1,8 @@ * Allwinner A10 MDIO Ethernet Controller interface Required properties: -- compatible: should be "allwinner,sun4i-mdio". +- compatible: should be "allwinner,sun4i-a10-mdio" + (Deprecated: "allwinner,sun4i-mdio"). - reg: address and length of the register set for the device. Optional properties: @@ -9,7 +10,7 @@ Optional properties: Example at the SoC level: mdio@01c0b080 { - compatible = "allwinner,sun4i-mdio"; + compatible = "allwinner,sun4i-a10-mdio"; reg = <0x01c0b080 0x14>; #address-cells = <1>; #size-cells = <0>; diff --git a/drivers/net/phy/mdio-sun4i.c b/drivers/net/phy/mdio-sun4i.c index bb88bc7d81fb..9367acc84fbb 100644 --- a/drivers/net/phy/mdio-sun4i.c +++ b/drivers/net/phy/mdio-sun4i.c @@ -170,6 +170,9 @@ static int sun4i_mdio_remove(struct platform_device *pdev) } static const struct of_device_id sun4i_mdio_dt_ids[] = { + { .compatible = "allwinner,sun4i-a10-mdio" }, + + /* Deprecated */ { .compatible = "allwinner,sun4i-mdio" }, { } }; From 1c70e099ac80ee779ffac66a6169ad936425fe41 Mon Sep 17 00:00:00 2001 From: Maxime Ripard Date: Sun, 2 Feb 2014 14:49:13 +0100 Subject: [PATCH 326/596] ARM: sunxi: dt: Convert to the new net compatibles Switch the device tree to the new compatibles introduced in the ethernet and mdio drivers to have a common pattern accross all Allwinner SoCs. Signed-off-by: Maxime Ripard Signed-off-by: David S. Miller --- arch/arm/boot/dts/sun4i-a10.dtsi | 4 ++-- arch/arm/boot/dts/sun5i-a10s.dtsi | 4 ++-- arch/arm/boot/dts/sun7i-a20.dtsi | 4 ++-- 3 files changed, 6 insertions(+), 6 deletions(-) diff --git a/arch/arm/boot/dts/sun4i-a10.dtsi b/arch/arm/boot/dts/sun4i-a10.dtsi index 040bb0eba152..10666ca8aee1 100644 --- a/arch/arm/boot/dts/sun4i-a10.dtsi +++ b/arch/arm/boot/dts/sun4i-a10.dtsi @@ -315,7 +315,7 @@ soc@01c00000 { ranges; emac: ethernet@01c0b000 { - compatible = "allwinner,sun4i-emac"; + compatible = "allwinner,sun4i-a10-emac"; reg = <0x01c0b000 0x1000>; interrupts = <55>; clocks = <&ahb_gates 17>; @@ -323,7 +323,7 @@ emac: ethernet@01c0b000 { }; mdio@01c0b080 { - compatible = "allwinner,sun4i-mdio"; + compatible = "allwinner,sun4i-a10-mdio"; reg = <0x01c0b080 0x14>; status = "disabled"; #address-cells = <1>; diff --git a/arch/arm/boot/dts/sun5i-a10s.dtsi b/arch/arm/boot/dts/sun5i-a10s.dtsi index ea16054857a4..64961595e8d6 100644 --- a/arch/arm/boot/dts/sun5i-a10s.dtsi +++ b/arch/arm/boot/dts/sun5i-a10s.dtsi @@ -278,7 +278,7 @@ soc@01c00000 { ranges; emac: ethernet@01c0b000 { - compatible = "allwinner,sun4i-emac"; + compatible = "allwinner,sun4i-a10-emac"; reg = <0x01c0b000 0x1000>; interrupts = <55>; clocks = <&ahb_gates 17>; @@ -286,7 +286,7 @@ emac: ethernet@01c0b000 { }; mdio@01c0b080 { - compatible = "allwinner,sun4i-mdio"; + compatible = "allwinner,sun4i-a10-mdio"; reg = <0x01c0b080 0x14>; status = "disabled"; #address-cells = <1>; diff --git a/arch/arm/boot/dts/sun7i-a20.dtsi b/arch/arm/boot/dts/sun7i-a20.dtsi index 119f066f0d98..9ff09484847b 100644 --- a/arch/arm/boot/dts/sun7i-a20.dtsi +++ b/arch/arm/boot/dts/sun7i-a20.dtsi @@ -340,7 +340,7 @@ soc@01c00000 { ranges; emac: ethernet@01c0b000 { - compatible = "allwinner,sun4i-emac"; + compatible = "allwinner,sun4i-a10-emac"; reg = <0x01c0b000 0x1000>; interrupts = <0 55 4>; clocks = <&ahb_gates 17>; @@ -348,7 +348,7 @@ emac: ethernet@01c0b000 { }; mdio@01c0b080 { - compatible = "allwinner,sun4i-mdio"; + compatible = "allwinner,sun4i-a10-mdio"; reg = <0x01c0b080 0x14>; status = "disabled"; #address-cells = <1>; From 597b506816337842ae76e739982b1b6fc19f35df Mon Sep 17 00:00:00 2001 From: Alexander Shiyan Date: Tue, 4 Feb 2014 19:43:32 +0400 Subject: [PATCH 327/596] net: irda: ep7211-sir: Remove driver This patch removes old and unsupported CLPS711X IrDA driver. Support for IrDA for CLPS711X serial port now provided by commit 4a33f1f59abd (serial: clps711x: Add support for N_IRDA line discipline), so IrDA-mode can be turned ON with "irattach" tool through "irtty" driver. Signed-off-by: Alexander Shiyan Signed-off-by: David S. Miller --- drivers/net/irda/Kconfig | 7 ---- drivers/net/irda/Makefile | 1 - drivers/net/irda/ep7211-sir.c | 70 ----------------------------------- 3 files changed, 78 deletions(-) delete mode 100644 drivers/net/irda/ep7211-sir.c diff --git a/drivers/net/irda/Kconfig b/drivers/net/irda/Kconfig index 2dc82f1d2e70..3da44d5d9149 100644 --- a/drivers/net/irda/Kconfig +++ b/drivers/net/irda/Kconfig @@ -210,13 +210,6 @@ config KINGSUN_DONGLE To compile it as a module, choose M here: the module will be called kingsun-sir. -config EP7211_DONGLE - tristate "Cirrus Logic clps711x I/R support" - depends on IRTTY_SIR && ARCH_CLPS711X && IRDA - help - Say Y here if you want to build support for the Cirrus logic - EP7211 chipset's infrared module. - config KSDAZZLE_DONGLE tristate "KingSun Dazzle IrDA-USB dongle" depends on IRDA && USB diff --git a/drivers/net/irda/Makefile b/drivers/net/irda/Makefile index dfc64537f62f..be8ab5b9a4a2 100644 --- a/drivers/net/irda/Makefile +++ b/drivers/net/irda/Makefile @@ -35,7 +35,6 @@ obj-$(CONFIG_MCP2120_DONGLE) += mcp2120-sir.o obj-$(CONFIG_ACT200L_DONGLE) += act200l-sir.o obj-$(CONFIG_MA600_DONGLE) += ma600-sir.o obj-$(CONFIG_TOIM3232_DONGLE) += toim3232-sir.o -obj-$(CONFIG_EP7211_DONGLE) += ep7211-sir.o obj-$(CONFIG_KINGSUN_DONGLE) += kingsun-sir.o obj-$(CONFIG_KSDAZZLE_DONGLE) += ksdazzle-sir.o obj-$(CONFIG_KS959_DONGLE) += ks959-sir.o diff --git a/drivers/net/irda/ep7211-sir.c b/drivers/net/irda/ep7211-sir.c deleted file mode 100644 index 5fe1f4dd3369..000000000000 --- a/drivers/net/irda/ep7211-sir.c +++ /dev/null @@ -1,70 +0,0 @@ -/* - * IR port driver for the Cirrus Logic CLPS711X processors - * - * Copyright 2001, Blue Mug Inc. All rights reserved. - * Copyright 2007, Samuel Ortiz - */ - -#include -#include - -#include - -#include "sir-dev.h" - -static int clps711x_dongle_open(struct sir_dev *dev) -{ - unsigned int syscon; - - /* Turn on the SIR encoder. */ - syscon = clps_readl(SYSCON1); - syscon |= SYSCON1_SIREN; - clps_writel(syscon, SYSCON1); - - return 0; -} - -static int clps711x_dongle_close(struct sir_dev *dev) -{ - unsigned int syscon; - - /* Turn off the SIR encoder. */ - syscon = clps_readl(SYSCON1); - syscon &= ~SYSCON1_SIREN; - clps_writel(syscon, SYSCON1); - - return 0; -} - -static struct dongle_driver clps711x_dongle = { - .owner = THIS_MODULE, - .driver_name = "EP7211 IR driver", - .type = IRDA_EP7211_DONGLE, - .open = clps711x_dongle_open, - .close = clps711x_dongle_close, -}; - -static int clps711x_sir_probe(struct platform_device *pdev) -{ - return irda_register_dongle(&clps711x_dongle); -} - -static int clps711x_sir_remove(struct platform_device *pdev) -{ - return irda_unregister_dongle(&clps711x_dongle); -} - -static struct platform_driver clps711x_sir_driver = { - .driver = { - .name = "sir-clps711x", - .owner = THIS_MODULE, - }, - .probe = clps711x_sir_probe, - .remove = clps711x_sir_remove, -}; -module_platform_driver(clps711x_sir_driver); - -MODULE_AUTHOR("Samuel Ortiz "); -MODULE_DESCRIPTION("EP7211 IR dongle driver"); -MODULE_LICENSE("GPL"); -MODULE_ALIAS("irda-dongle-13"); /* IRDA_EP7211_DONGLE */ From 1c8bb760b632b4328f06066a6813f3758e54d8cb Mon Sep 17 00:00:00 2001 From: James M Leddy Date: Tue, 4 Feb 2014 15:10:59 -0500 Subject: [PATCH 328/596] bnx2[x]: Make module parameters readable Occasionally users want to know what parameters their Broadcom drivers are running with. For example, a user may want to know if MSI is disabled. This patch has been compile tested. Signed-off-by: James M Leddy Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnx2.c | 2 +- drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c | 12 ++++++------ 2 files changed, 7 insertions(+), 7 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnx2.c b/drivers/net/ethernet/broadcom/bnx2.c index 9d2dedadf2df..cda25ac45b47 100644 --- a/drivers/net/ethernet/broadcom/bnx2.c +++ b/drivers/net/ethernet/broadcom/bnx2.c @@ -85,7 +85,7 @@ MODULE_FIRMWARE(FW_RV2P_FILE_09_Ax); static int disable_msi = 0; -module_param(disable_msi, int, 0); +module_param(disable_msi, int, S_IRUGO); MODULE_PARM_DESC(disable_msi, "Disable Message Signaled Interrupt (MSI)"); typedef enum { diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c index c9c445e7b4a5..7d4382286457 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c @@ -95,29 +95,29 @@ MODULE_FIRMWARE(FW_FILE_NAME_E1H); MODULE_FIRMWARE(FW_FILE_NAME_E2); int bnx2x_num_queues; -module_param_named(num_queues, bnx2x_num_queues, int, 0); +module_param_named(num_queues, bnx2x_num_queues, int, S_IRUGO); MODULE_PARM_DESC(num_queues, " Set number of queues (default is as a number of CPUs)"); static int disable_tpa; -module_param(disable_tpa, int, 0); +module_param(disable_tpa, int, S_IRUGO); MODULE_PARM_DESC(disable_tpa, " Disable the TPA (LRO) feature"); static int int_mode; -module_param(int_mode, int, 0); +module_param(int_mode, int, S_IRUGO); MODULE_PARM_DESC(int_mode, " Force interrupt mode other than MSI-X " "(1 INT#x; 2 MSI)"); static int dropless_fc; -module_param(dropless_fc, int, 0); +module_param(dropless_fc, int, S_IRUGO); MODULE_PARM_DESC(dropless_fc, " Pause on exhausted host ring"); static int mrrs = -1; -module_param(mrrs, int, 0); +module_param(mrrs, int, S_IRUGO); MODULE_PARM_DESC(mrrs, " Force Max Read Req Size (0..3) (for debug)"); static int debug; -module_param(debug, int, 0); +module_param(debug, int, S_IRUGO); MODULE_PARM_DESC(debug, " Default debug msglevel"); struct workqueue_struct *bnx2x_wq; From 63b5f152eb4a5bb79b9caf7ec37b4201d12f6e66 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Wed, 5 Feb 2014 08:38:25 +0100 Subject: [PATCH 329/596] ipv4: Fix runtime WARNING in rtmsg_ifa() On m68k/ARAnyM: WARNING: CPU: 0 PID: 407 at net/ipv4/devinet.c:1599 0x316a99() Modules linked in: CPU: 0 PID: 407 Comm: ifconfig Not tainted 3.13.0-atari-09263-g0c71d68014d1 #1378 Stack from 10c4fdf0: 10c4fdf0 002ffabb 000243e8 00000000 008ced6c 00024416 00316a99 0000063f 00316a99 00000009 00000000 002501b4 00316a99 0000063f c0a86117 00000080 c0a86117 00ad0c90 00250a5a 00000014 00ad0c90 00000000 00000000 00000001 00b02dd0 00356594 00000000 00356594 c0a86117 eff6c9e4 008ced6c 00000002 008ced60 0024f9b4 00250b52 00ad0c90 00000000 00000000 00252390 00ad0c90 eff6c9e4 0000004f 00000000 00000000 eff6c9e4 8000e25c eff6c9e4 80001020 Call Trace: [<000243e8>] warn_slowpath_common+0x52/0x6c [<00024416>] warn_slowpath_null+0x14/0x1a [<002501b4>] rtmsg_ifa+0xdc/0xf0 [<00250a5a>] __inet_insert_ifa+0xd6/0x1c2 [<0024f9b4>] inet_abc_len+0x0/0x42 [<00250b52>] inet_insert_ifa+0xc/0x12 [<00252390>] devinet_ioctl+0x2ae/0x5d6 Adding some debugging code reveals that net_fill_ifaddr() fails in put_cacheinfo(skb, ifa->ifa_cstamp, ifa->ifa_tstamp, preferred, valid)) nla_put complains: lib/nlattr.c:454: skb_tailroom(skb) = 12, nla_total_size(attrlen) = 20 Apparently commit 5c766d642bcaffd0c2a5b354db2068515b3846cf ("ipv4: introduce address lifetime") forgot to take into account the addition of struct ifa_cacheinfo in inet_nlmsg_size(). Hence add it, like is already done for ipv6. Suggested-by: Cong Wang Signed-off-by: Geert Uytterhoeven Signed-off-by: Cong Wang Signed-off-by: David S. Miller --- net/ipv4/devinet.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index ac2dff3c2c1c..bdbf68bb2e2d 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -1443,7 +1443,8 @@ static size_t inet_nlmsg_size(void) + nla_total_size(4) /* IFA_LOCAL */ + nla_total_size(4) /* IFA_BROADCAST */ + nla_total_size(IFNAMSIZ) /* IFA_LABEL */ - + nla_total_size(4); /* IFA_FLAGS */ + + nla_total_size(4) /* IFA_FLAGS */ + + nla_total_size(sizeof(struct ifa_cacheinfo)); /* IFA_CACHEINFO */ } static inline u32 cstamp_delta(unsigned long cstamp) From c6e27f2f3cf2631aae6f1f6fae1a7ab083fdb024 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Wed, 5 Feb 2014 16:29:21 +0300 Subject: [PATCH 330/596] tg3: cleanup an error path in tg3_phy_reset_5703_4_5() In the original code, if tg3_readphy() fails then it does an unnecessary check to verify "err" is still zero and then returns -EBUSY. My static checker complains about the unnecessary "if (!err)" check and anyway it is better to propagate the -EBUSY error code from tg3_readphy() instead of hard coding it here. And really the original code is confusing to look at. Signed-off-by: Dan Carpenter Acked-by: Nithin Nayak Sujir Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/tg3.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) diff --git a/drivers/net/ethernet/broadcom/tg3.c b/drivers/net/ethernet/broadcom/tg3.c index e2ca03e23dc1..c466e12b601e 100644 --- a/drivers/net/ethernet/broadcom/tg3.c +++ b/drivers/net/ethernet/broadcom/tg3.c @@ -2609,13 +2609,14 @@ static int tg3_phy_reset_5703_4_5(struct tg3 *tp) tg3_writephy(tp, MII_CTRL1000, phy9_orig); - if (!tg3_readphy(tp, MII_TG3_EXT_CTRL, ®32)) { - reg32 &= ~0x3000; - tg3_writephy(tp, MII_TG3_EXT_CTRL, reg32); - } else if (!err) - err = -EBUSY; + err = tg3_readphy(tp, MII_TG3_EXT_CTRL, ®32); + if (err) + return err; - return err; + reg32 &= ~0x3000; + tg3_writephy(tp, MII_TG3_EXT_CTRL, reg32); + + return 0; } static void tg3_carrier_off(struct tg3 *tp) From c6993dfd7db9b0c6b7ca7503a56fda9236a4710f Mon Sep 17 00:00:00 2001 From: Nithin Sujir Date: Thu, 6 Feb 2014 14:13:05 -0800 Subject: [PATCH 331/596] tg3: Fix deadlock in tg3_change_mtu() Quoting David Vrabel - "5780 cards cannot have jumbo frames and TSO enabled together. When jumbo frames are enabled by setting the MTU, the TSO feature must be cleared. This is done indirectly by calling netdev_update_features() which will call tg3_fix_features() to actually clear the flags. netdev_update_features() will also trigger a new netlink message for the feature change event which will result in a call to tg3_get_stats64() which deadlocks on the tg3 lock." tg3_set_mtu() does not need to be under the tg3 lock since converting the flags to use set_bit(). Move it out to after tg3_netif_stop(). Reported-by: David Vrabel Tested-by: David Vrabel Signed-off-by: Michael Chan Signed-off-by: Nithin Nayak Sujir Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/tg3.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/broadcom/tg3.c b/drivers/net/ethernet/broadcom/tg3.c index c466e12b601e..3167ed6593b0 100644 --- a/drivers/net/ethernet/broadcom/tg3.c +++ b/drivers/net/ethernet/broadcom/tg3.c @@ -14114,12 +14114,12 @@ static int tg3_change_mtu(struct net_device *dev, int new_mtu) tg3_netif_stop(tp); + tg3_set_mtu(dev, tp, new_mtu); + tg3_full_lock(tp, 1); tg3_halt(tp, RESET_KIND_SHUTDOWN, 1); - tg3_set_mtu(dev, tp, new_mtu); - /* Reset PHY, otherwise the read DMA engine will be in a mode that * breaks all requests to 256 bytes. */ From 656493d6e730690b76d6e74253d0208841cf4dd9 Mon Sep 17 00:00:00 2001 From: Yuval Mintz Date: Wed, 5 Feb 2014 16:07:12 +0200 Subject: [PATCH 332/596] bnx2x: Allow VF rss on higher PFs bnx2x driver uses incorrect PF identifier to configure (in HW) the VF interrupt scheme; As a result, in multi-function mode the configuration for PFs with a high index (4+) will overflow and the PF will erroneously configure a single ISR scheme for its VFs. As a result, if such a VF uses multiple queues, interrupt generation will stop after VF receives an Rx packet or sends a Tx packet on a queue other than queue[0]. Signed-off-by: Yuval Mintz Signed-off-by: Ariel Elior Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c index aec5ef2ed7ce..e42f48df6e94 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c @@ -1446,12 +1446,12 @@ static void bnx2x_vf_igu_reset(struct bnx2x *bp, struct bnx2x_virtf *vf) if (vf->cfg_flags & VF_CFG_INT_SIMD) val |= IGU_VF_CONF_SINGLE_ISR_EN; val &= ~IGU_VF_CONF_PARENT_MASK; - val |= BP_FUNC(bp) << IGU_VF_CONF_PARENT_SHIFT; /* parent PF */ + val |= (BP_ABS_FUNC(bp) >> 1) << IGU_VF_CONF_PARENT_SHIFT; REG_WR(bp, IGU_REG_VF_CONFIGURATION, val); DP(BNX2X_MSG_IOV, - "value in IGU_REG_VF_CONFIGURATION of vf %d after write %x\n", - vf->abs_vfid, REG_RD(bp, IGU_REG_VF_CONFIGURATION)); + "value in IGU_REG_VF_CONFIGURATION of vf %d after write is 0x%08x\n", + vf->abs_vfid, val); bnx2x_pretend_func(bp, BP_ABS_FUNC(bp)); From 3d55f44f563de2ab83209e8ead48d922b0d56a65 Mon Sep 17 00:00:00 2001 From: hayeswang Date: Thu, 6 Feb 2014 11:55:48 +0800 Subject: [PATCH 333/596] r8152: fix the submission of the interrupt transfer The submission of the interrupt transfer should be done after setting the bit of WORK_ENABLE, otherwise the callback function would have the opportunity to be returned directly. Clear the bit of WORK_ENABLE before killing the interrupt transfer. Signed-off-by: Hayes Wang Signed-off-by: David S. Miller --- drivers/net/usb/r8152.c | 19 +++++++++---------- 1 file changed, 9 insertions(+), 10 deletions(-) diff --git a/drivers/net/usb/r8152.c b/drivers/net/usb/r8152.c index e8fac732c6f1..d89dbe395ad2 100644 --- a/drivers/net/usb/r8152.c +++ b/drivers/net/usb/r8152.c @@ -2273,15 +2273,6 @@ static int rtl8152_open(struct net_device *netdev) struct r8152 *tp = netdev_priv(netdev); int res = 0; - res = usb_submit_urb(tp->intr_urb, GFP_KERNEL); - if (res) { - if (res == -ENODEV) - netif_device_detach(tp->netdev); - netif_warn(tp, ifup, netdev, "intr_urb submit failed: %d\n", - res); - return res; - } - rtl8152_set_speed(tp, AUTONEG_ENABLE, tp->mii.supports_gmii ? SPEED_1000 : SPEED_100, DUPLEX_FULL); @@ -2289,6 +2280,14 @@ static int rtl8152_open(struct net_device *netdev) netif_carrier_off(netdev); netif_start_queue(netdev); set_bit(WORK_ENABLE, &tp->flags); + res = usb_submit_urb(tp->intr_urb, GFP_KERNEL); + if (res) { + if (res == -ENODEV) + netif_device_detach(tp->netdev); + netif_warn(tp, ifup, netdev, "intr_urb submit failed: %d\n", + res); + } + return res; } @@ -2298,8 +2297,8 @@ static int rtl8152_close(struct net_device *netdev) struct r8152 *tp = netdev_priv(netdev); int res = 0; - usb_kill_urb(tp->intr_urb); clear_bit(WORK_ENABLE, &tp->flags); + usb_kill_urb(tp->intr_urb); cancel_delayed_work_sync(&tp->schedule); netif_stop_queue(netdev); tasklet_disable(&tp->tl); From 661dbf34129cecd20879cb7b1cbe936911da0f02 Mon Sep 17 00:00:00 2001 From: Matija Glavinic Pecotic Date: Thu, 6 Feb 2014 08:30:10 +0100 Subject: [PATCH 334/596] net: sctp: fix initialization of local source address on accepted ipv6 sockets commit efe4208f47f907b86f528788da711e8ab9dea44d: 'ipv6: make lookups simpler and faster' broke initialization of local source address on accepted ipv6 sockets. Before the mentioned commit receive address was copied along with the contents of ipv6_pinfo in sctp_v6_create_accept_sk. Now when it is moved, it has to be copied separately. This also fixes lksctp's ipv6 regression in a sense that test_getname_v6, TC5 - 'getsockname on a connected server socket' now passes. Signed-off-by: Matija Glavinic Pecotic Signed-off-by: David S. Miller --- net/sctp/ipv6.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c index 0f6259a6a932..2b1738ef9394 100644 --- a/net/sctp/ipv6.c +++ b/net/sctp/ipv6.c @@ -662,6 +662,8 @@ static struct sock *sctp_v6_create_accept_sk(struct sock *sk, */ sctp_v6_to_sk_daddr(&asoc->peer.primary_addr, newsk); + newsk->sk_v6_rcv_saddr = sk->sk_v6_rcv_saddr; + sk_refcnt_debug_inc(newsk); if (newsk->sk_prot->init(newsk)) { From cd139781546f22c41276045f6464c51074b50c40 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Thu, 6 Feb 2014 11:03:08 +0300 Subject: [PATCH 335/596] isdn/hisax: hex vs decimal typo in prfeatureind() This is a static checker fix, but judging from the context then I think hexidecimal 0x80 is intended here instead of decimal 80. Signed-off-by: Dan Carpenter Signed-off-by: David S. Miller --- drivers/isdn/hisax/q931.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/isdn/hisax/q931.c b/drivers/isdn/hisax/q931.c index af1b020a81f1..b420f8bd862e 100644 --- a/drivers/isdn/hisax/q931.c +++ b/drivers/isdn/hisax/q931.c @@ -810,7 +810,7 @@ prfeatureind(char *dest, u_char *p) dp += sprintf(dp, " octet 3 "); dp += prbits(dp, *p, 8, 8); *dp++ = '\n'; - if (!(*p++ & 80)) { + if (!(*p++ & 0x80)) { dp += sprintf(dp, " octet 4 "); dp += prbits(dp, *p++, 8, 8); *dp++ = '\n'; From ee262ad827f89e2dc7851ec2986953b5b125c6bc Mon Sep 17 00:00:00 2001 From: Jan Moskyto Matejka Date: Thu, 6 Feb 2014 12:10:00 +0100 Subject: [PATCH 336/596] inet: defines IPPROTO_* needed for module alias generation Commit cfd280c91253 ("net: sync some IP headers with glibc") changed a set of define's to an enum (with no explanation why) which introduced a bug in module mip6 where aliases are generated using the IPPROTO_* defines; mip6 doesn't load if require_module called with the aliases from xfrm_get_type(). Reverting this change back to define's to fix the aliases. modinfo mip6 (before this change) alias: xfrm-type-10-IPPROTO_DSTOPTS alias: xfrm-type-10-IPPROTO_ROUTING modinfo mip6 (after this change) alias: xfrm-type-10-43 alias: xfrm-type-10-60 Signed-off-by: Jan Moskyto Matejka Signed-off-by: David S. Miller --- include/uapi/linux/in6.h | 23 +++++++---------------- 1 file changed, 7 insertions(+), 16 deletions(-) diff --git a/include/uapi/linux/in6.h b/include/uapi/linux/in6.h index 633b93cac1ed..e9a1d2d973b6 100644 --- a/include/uapi/linux/in6.h +++ b/include/uapi/linux/in6.h @@ -128,22 +128,13 @@ struct in6_flowlabel_req { * IPV6 extension headers */ #if __UAPI_DEF_IPPROTO_V6 -enum { - IPPROTO_HOPOPTS = 0, /* IPv6 hop-by-hop options */ -#define IPPROTO_HOPOPTS IPPROTO_HOPOPTS - IPPROTO_ROUTING = 43, /* IPv6 routing header */ -#define IPPROTO_ROUTING IPPROTO_ROUTING - IPPROTO_FRAGMENT = 44, /* IPv6 fragmentation header */ -#define IPPROTO_FRAGMENT IPPROTO_FRAGMENT - IPPROTO_ICMPV6 = 58, /* ICMPv6 */ -#define IPPROTO_ICMPV6 IPPROTO_ICMPV6 - IPPROTO_NONE = 59, /* IPv6 no next header */ -#define IPPROTO_NONE IPPROTO_NONE - IPPROTO_DSTOPTS = 60, /* IPv6 destination options */ -#define IPPROTO_DSTOPTS IPPROTO_DSTOPTS - IPPROTO_MH = 135, /* IPv6 mobility header */ -#define IPPROTO_MH IPPROTO_MH -}; +#define IPPROTO_HOPOPTS 0 /* IPv6 hop-by-hop options */ +#define IPPROTO_ROUTING 43 /* IPv6 routing header */ +#define IPPROTO_FRAGMENT 44 /* IPv6 fragmentation header */ +#define IPPROTO_ICMPV6 58 /* ICMPv6 */ +#define IPPROTO_NONE 59 /* IPv6 no next header */ +#define IPPROTO_DSTOPTS 60 /* IPv6 destination options */ +#define IPPROTO_MH 135 /* IPv6 mobility header */ #endif /* __UAPI_DEF_IPPROTO_V6 */ /* From 4ccd0bb9bf136911f85a5c1eeb6bdcc259cde5e4 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Thu, 6 Feb 2014 15:53:19 +0300 Subject: [PATCH 337/596] hso: remove some dead code It seems like this function was intended to have special handling for urb statuses of -ENOENT and -ECONNRESET. But now it just prints some debugging and returns at the start of the function. I have removed the dead code, it's still in the git history if anyone wants to revive it. Signed-off-by: Dan Carpenter Signed-off-by: David S. Miller --- drivers/net/usb/hso.c | 32 +++++++++++--------------------- 1 file changed, 11 insertions(+), 21 deletions(-) diff --git a/drivers/net/usb/hso.c b/drivers/net/usb/hso.c index 1a482344b3f5..660bd5ea9fc0 100644 --- a/drivers/net/usb/hso.c +++ b/drivers/net/usb/hso.c @@ -1201,16 +1201,18 @@ static void hso_std_serial_read_bulk_callback(struct urb *urb) struct hso_serial *serial = urb->context; int status = urb->status; + D4("\n--- Got serial_read_bulk callback %02x ---", status); + /* sanity check */ if (!serial) { D1("serial == NULL"); return; - } else if (status) { + } + if (status) { handle_usb_error(status, __func__, serial->parent); return; } - D4("\n--- Got serial_read_bulk callback %02x ---", status); D1("Actual length = %d\n", urb->actual_length); DUMP1(urb->transfer_buffer, urb->actual_length); @@ -1218,25 +1220,13 @@ static void hso_std_serial_read_bulk_callback(struct urb *urb) if (serial->port.count == 0) return; - if (status == 0) { - if (serial->parent->port_spec & HSO_INFO_CRC_BUG) - fix_crc_bug(urb, serial->in_endp->wMaxPacketSize); - /* Valid data, handle RX data */ - spin_lock(&serial->serial_lock); - serial->rx_urb_filled[hso_urb_to_index(serial, urb)] = 1; - put_rxbuf_data_and_resubmit_bulk_urb(serial); - spin_unlock(&serial->serial_lock); - } else if (status == -ENOENT || status == -ECONNRESET) { - /* Unlinked - check for throttled port. */ - D2("Port %d, successfully unlinked urb", serial->minor); - spin_lock(&serial->serial_lock); - serial->rx_urb_filled[hso_urb_to_index(serial, urb)] = 0; - hso_resubmit_rx_bulk_urb(serial, urb); - spin_unlock(&serial->serial_lock); - } else { - D2("Port %d, status = %d for read urb", serial->minor, status); - return; - } + if (serial->parent->port_spec & HSO_INFO_CRC_BUG) + fix_crc_bug(urb, serial->in_endp->wMaxPacketSize); + /* Valid data, handle RX data */ + spin_lock(&serial->serial_lock); + serial->rx_urb_filled[hso_urb_to_index(serial, urb)] = 1; + put_rxbuf_data_and_resubmit_bulk_urb(serial); + spin_unlock(&serial->serial_lock); } /* From 440b87eac4ca21ccb3f56e7b1f126e0eebcc0f84 Mon Sep 17 00:00:00 2001 From: Paul Gortmaker Date: Thu, 6 Feb 2014 11:45:12 -0500 Subject: [PATCH 338/596] drivers/net: fix build warning in ethernet/sfc/tx.c Commit ee45fd92c739db5b7950163d91dfe5f016af6d24 ("sfc: Use TX PIO for sufficiently small packets") introduced the following warning: drivers/net/ethernet/sfc/tx.c: In function 'efx_enqueue_skb': drivers/net/ethernet/sfc/tx.c:432:1: warning: label 'finish_packet' defined but not used Stick the label inside the same #ifdef that the code which calls it uses. Note that this is only seen for arch that do not set ARCH_HAS_IOREMAP_WC, such as arm, mips, sparc, ..., as the others enable the write combining code and hence use the label. Cc: Jon Cooper Cc: Ben Hutchings Signed-off-by: Paul Gortmaker Acked-by: Shradha Shah Signed-off-by: David S. Miller --- drivers/net/ethernet/sfc/tx.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/ethernet/sfc/tx.c b/drivers/net/ethernet/sfc/tx.c index c49d1fb16965..75d11fa4eb0a 100644 --- a/drivers/net/ethernet/sfc/tx.c +++ b/drivers/net/ethernet/sfc/tx.c @@ -429,7 +429,9 @@ netdev_tx_t efx_enqueue_skb(struct efx_tx_queue *tx_queue, struct sk_buff *skb) } /* Transfer ownership of the skb to the final buffer */ +#ifdef EFX_USE_PIO finish_packet: +#endif buffer->skb = skb; buffer->flags = EFX_TX_BUF_SKB | dma_flags; From 00fe11b3c67dc670fe6391d22f1fe64e7c99a8ec Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Thu, 6 Feb 2014 18:34:12 +0100 Subject: [PATCH 339/596] netpoll: fix netconsole IPv6 setup Currently, to make netconsole start over IPv6, the source address needs to be specified. Without a source address, netpoll_parse_options assumes we're setting up over IPv4 and the destination IPv6 address is rejected. Check if the IP version has been forced by a source address before checking for a version mismatch when parsing the destination address. Signed-off-by: Sabrina Dubroca Acked-by: Cong Wang Signed-off-by: David S. Miller --- net/core/netpoll.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/net/core/netpoll.c b/net/core/netpoll.c index c03f3dec4763..a664f7829a6d 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -948,6 +948,7 @@ int netpoll_parse_options(struct netpoll *np, char *opt) { char *cur=opt, *delim; int ipv6; + bool ipversion_set = false; if (*cur != '@') { if ((delim = strchr(cur, '@')) == NULL) @@ -960,6 +961,7 @@ int netpoll_parse_options(struct netpoll *np, char *opt) cur++; if (*cur != '/') { + ipversion_set = true; if ((delim = strchr(cur, '/')) == NULL) goto parse_failed; *delim = 0; @@ -1002,7 +1004,7 @@ int netpoll_parse_options(struct netpoll *np, char *opt) ipv6 = netpoll_parse_ip_addr(cur, &np->remote_ip); if (ipv6 < 0) goto parse_failed; - else if (np->ipv6 != (bool)ipv6) + else if (ipversion_set && np->ipv6 != (bool)ipv6) goto parse_failed; else np->ipv6 = (bool)ipv6; From ed98df3361f059db42786c830ea96e2d18b8d4db Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 6 Feb 2014 10:42:42 -0800 Subject: [PATCH 340/596] net: use __GFP_NORETRY for high order allocations sock_alloc_send_pskb() & sk_page_frag_refill() have a loop trying high order allocations to prepare skb with low number of fragments as this increases performance. Problem is that under memory pressure/fragmentation, this can trigger OOM while the intent was only to try the high order allocations, then fallback to order-0 allocations. We had various reports from unexpected regressions. According to David, setting __GFP_NORETRY should be fine, as the asynchronous compaction is still enabled, and this will prevent OOM from kicking as in : CFSClientEventm invoked oom-killer: gfp_mask=0x42d0, order=3, oom_adj=0, oom_score_adj=0, oom_score_badness=2 (enabled),memcg_scoring=disabled CFSClientEventm Call Trace: [] dump_header+0xe1/0x23e [] oom_kill_process+0x6a/0x323 [] out_of_memory+0x4b3/0x50d [] __alloc_pages_may_oom+0xa2/0xc7 [] __alloc_pages_nodemask+0x1002/0x17f0 [] alloc_pages_current+0x103/0x2b0 [] sk_page_frag_refill+0x8f/0x160 [] tcp_sendmsg+0x560/0xee0 [] inet_sendmsg+0x67/0x100 [] __sock_sendmsg_nosec+0x6c/0x90 [] sock_sendmsg+0xc5/0xf0 [] __sys_sendmsg+0x136/0x430 [] sys_sendmsg+0x88/0x110 [] system_call_fastpath+0x16/0x1b Out of Memory: Kill process 2856 (bash) score 9999 or sacrifice child Signed-off-by: Eric Dumazet Acked-by: David Rientjes Acked-by: "Eric W. Biederman" Signed-off-by: David S. Miller --- net/core/sock.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/net/core/sock.c b/net/core/sock.c index 0c127dcdf6a8..5b6a9431b017 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -1775,7 +1775,9 @@ struct sk_buff *sock_alloc_send_pskb(struct sock *sk, unsigned long header_len, while (order) { if (npages >= 1 << order) { page = alloc_pages(sk->sk_allocation | - __GFP_COMP | __GFP_NOWARN, + __GFP_COMP | + __GFP_NOWARN | + __GFP_NORETRY, order); if (page) goto fill_page; @@ -1845,7 +1847,7 @@ bool skb_page_frag_refill(unsigned int sz, struct page_frag *pfrag, gfp_t prio) gfp_t gfp = prio; if (order) - gfp |= __GFP_COMP | __GFP_NOWARN; + gfp |= __GFP_COMP | __GFP_NOWARN | __GFP_NORETRY; pfrag->page = alloc_pages(gfp, order); if (likely(pfrag->page)) { pfrag->offset = 0; From dbe173079ab58a444e12dbebe96f5aec1e0bed1a Mon Sep 17 00:00:00 2001 From: Cong Wang Date: Thu, 6 Feb 2014 15:00:52 -0800 Subject: [PATCH 341/596] bridge: fix netconsole setup over bridge Commit 93d8bf9fb8f3 ("bridge: cleanup netpoll code") introduced a check in br_netpoll_enable(), but this check is incorrect for br_netpoll_setup(). This patch moves the code after the check into __br_netpoll_enable() and calls it in br_netpoll_setup(). For br_add_if(), the check is still needed. Fixes: 93d8bf9fb8f3 ("bridge: cleanup netpoll code") Cc: Toshiaki Makita Cc: Stephen Hemminger Cc: David S. Miller Signed-off-by: Cong Wang Signed-off-by: Cong Wang Acked-by: Toshiaki Makita Tested-by: Toshiaki Makita Signed-off-by: David S. Miller --- net/bridge/br_device.c | 59 +++++++++++++++++++++++------------------- 1 file changed, 32 insertions(+), 27 deletions(-) diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c index e4401a531afb..d9a9b0fc1795 100644 --- a/net/bridge/br_device.c +++ b/net/bridge/br_device.c @@ -226,37 +226,11 @@ static void br_netpoll_cleanup(struct net_device *dev) br_netpoll_disable(p); } -static int br_netpoll_setup(struct net_device *dev, struct netpoll_info *ni, - gfp_t gfp) -{ - struct net_bridge *br = netdev_priv(dev); - struct net_bridge_port *p; - int err = 0; - - list_for_each_entry(p, &br->port_list, list) { - if (!p->dev) - continue; - err = br_netpoll_enable(p, gfp); - if (err) - goto fail; - } - -out: - return err; - -fail: - br_netpoll_cleanup(dev); - goto out; -} - -int br_netpoll_enable(struct net_bridge_port *p, gfp_t gfp) +static int __br_netpoll_enable(struct net_bridge_port *p, gfp_t gfp) { struct netpoll *np; int err; - if (!p->br->dev->npinfo) - return 0; - np = kzalloc(sizeof(*p->np), gfp); if (!np) return -ENOMEM; @@ -271,6 +245,37 @@ int br_netpoll_enable(struct net_bridge_port *p, gfp_t gfp) return err; } +int br_netpoll_enable(struct net_bridge_port *p, gfp_t gfp) +{ + if (!p->br->dev->npinfo) + return 0; + + return __br_netpoll_enable(p, gfp); +} + +static int br_netpoll_setup(struct net_device *dev, struct netpoll_info *ni, + gfp_t gfp) +{ + struct net_bridge *br = netdev_priv(dev); + struct net_bridge_port *p; + int err = 0; + + list_for_each_entry(p, &br->port_list, list) { + if (!p->dev) + continue; + err = __br_netpoll_enable(p, gfp); + if (err) + goto fail; + } + +out: + return err; + +fail: + br_netpoll_cleanup(dev); + goto out; +} + void br_netpoll_disable(struct net_bridge_port *p) { struct netpoll *np = p->np; From 4a5ab4e224288403b0b4b6b8c4d339323150c312 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 6 Feb 2014 15:57:10 -0800 Subject: [PATCH 342/596] tcp: remove 1ms offset in srtt computation TCP pacing depends on an accurate srtt estimation. Current srtt estimation is using jiffie resolution, and has an artificial offset of at least 1 ms, which can produce slowdowns when FQ/pacing is used, especially in DC world, where typical rtt is below 1 ms. We are planning a switch to usec resolution for linux-3.15, but in the meantime, this patch removes the 1 ms offset. All we need is to have tp->srtt minimal value of 1 to differentiate the case of srtt being initialized or not, not 8. The problematic behavior was observed on a 40Gbit testbed, where 32 concurrent netperf were reaching 12Gbps of aggregate speed, instead of line speed. This patch also has the effect of reporting more accurate srtt and send rates to iproute2 ss command as in : $ ss -i dst cca2 Netid State Recv-Q Send-Q Local Address:Port Peer Address:Port tcp ESTAB 0 0 10.244.129.1:56984 10.244.129.2:12865 cubic wscale:6,6 rto:200 rtt:0.25/0.25 ato:40 mss:1448 cwnd:10 send 463.4Mbps rcv_rtt:1 rcv_space:29200 tcp ESTAB 0 390960 10.244.129.1:60247 10.244.129.2:50204 cubic wscale:6,6 rto:200 rtt:0.875/0.75 mss:1448 cwnd:73 ssthresh:51 send 966.4Mbps unacked:73 retrans:0/121 rcv_space:29200 Reported-by: Vytautas Valancius Signed-off-by: Eric Dumazet Cc: Yuchung Cheng Cc: Neal Cardwell Acked-by: Neal Cardwell Signed-off-by: David S. Miller --- net/ipv4/tcp_input.c | 18 ++++++++++-------- net/ipv4/tcp_output.c | 2 +- 2 files changed, 11 insertions(+), 9 deletions(-) diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 65cf90e063d5..227cba79fa6b 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -671,6 +671,7 @@ static void tcp_rtt_estimator(struct sock *sk, const __u32 mrtt) { struct tcp_sock *tp = tcp_sk(sk); long m = mrtt; /* RTT */ + u32 srtt = tp->srtt; /* The following amusing code comes from Jacobson's * article in SIGCOMM '88. Note that rtt and mdev @@ -688,11 +689,9 @@ static void tcp_rtt_estimator(struct sock *sk, const __u32 mrtt) * does not matter how to _calculate_ it. Seems, it was trap * that VJ failed to avoid. 8) */ - if (m == 0) - m = 1; - if (tp->srtt != 0) { - m -= (tp->srtt >> 3); /* m is now error in rtt est */ - tp->srtt += m; /* rtt = 7/8 rtt + 1/8 new */ + if (srtt != 0) { + m -= (srtt >> 3); /* m is now error in rtt est */ + srtt += m; /* rtt = 7/8 rtt + 1/8 new */ if (m < 0) { m = -m; /* m is now abs(error) */ m -= (tp->mdev >> 2); /* similar update on mdev */ @@ -723,11 +722,12 @@ static void tcp_rtt_estimator(struct sock *sk, const __u32 mrtt) } } else { /* no previous measure. */ - tp->srtt = m << 3; /* take the measured time to be rtt */ + srtt = m << 3; /* take the measured time to be rtt */ tp->mdev = m << 1; /* make sure rto = 3*rtt */ tp->mdev_max = tp->rttvar = max(tp->mdev, tcp_rto_min(sk)); tp->rtt_seq = tp->snd_nxt; } + tp->srtt = max(1U, srtt); } /* Set the sk_pacing_rate to allow proper sizing of TSO packets. @@ -746,8 +746,10 @@ static void tcp_update_pacing_rate(struct sock *sk) rate *= max(tp->snd_cwnd, tp->packets_out); - /* Correction for small srtt : minimum srtt being 8 (1 jiffy << 3), - * be conservative and assume srtt = 1 (125 us instead of 1.25 ms) + /* Correction for small srtt and scheduling constraints. + * For small rtt, consider noise is too high, and use + * the minimal value (srtt = 1 -> 125 us for HZ=1000) + * * We probably need usec resolution in the future. * Note: This also takes care of possible srtt=0 case, * when tcp_rtt_estimator() was not yet called. diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 03d26b85eab8..10435b3b9d0f 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -1977,7 +1977,7 @@ bool tcp_schedule_loss_probe(struct sock *sk) /* Schedule a loss probe in 2*RTT for SACK capable connections * in Open state, that are either limited by cwnd or application. */ - if (sysctl_tcp_early_retrans < 3 || !rtt || !tp->packets_out || + if (sysctl_tcp_early_retrans < 3 || !tp->srtt || !tp->packets_out || !tcp_is_sack(tp) || inet_csk(sk)->icsk_ca_state != TCP_CA_Open) return false; From 9e65dc371b5c8d7476c81353137efc13cc1bdabd Mon Sep 17 00:00:00 2001 From: Eli Cohen Date: Tue, 28 Jan 2014 14:52:47 +0200 Subject: [PATCH 343/596] IB/mlx5: Fix RC transport send queue overhead computation Fix the RC QPs send queue overhead computation to take into account two additional segments in the WQE which are needed for registration operations. Signed-off-by: Eli Cohen Signed-off-by: Roland Dreier --- drivers/infiniband/hw/mlx5/qp.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/infiniband/hw/mlx5/qp.c b/drivers/infiniband/hw/mlx5/qp.c index ae37fb9bf262..492dc330e907 100644 --- a/drivers/infiniband/hw/mlx5/qp.c +++ b/drivers/infiniband/hw/mlx5/qp.c @@ -216,7 +216,9 @@ static int sq_overhead(enum ib_qp_type qp_type) case IB_QPT_UC: size += sizeof(struct mlx5_wqe_ctrl_seg) + - sizeof(struct mlx5_wqe_raddr_seg); + sizeof(struct mlx5_wqe_raddr_seg) + + sizeof(struct mlx5_wqe_umr_ctrl_seg) + + sizeof(struct mlx5_mkey_seg); break; case IB_QPT_UD: From 78c0f98cc9dd46824fa66f35f14ea24ba733d145 Mon Sep 17 00:00:00 2001 From: Eli Cohen Date: Thu, 30 Jan 2014 13:49:48 +0200 Subject: [PATCH 344/596] IB/mlx5: Fix binary compatibility with libmlx5 Commit c1be5232d21d ("Fix micro UAR allocator") broke binary compatibility between libmlx5 and mlx5_ib since it defines a different value to the number of micro UARs per page, leading to wrong calculation in libmlx5. This patch defines struct mlx5_ib_alloc_ucontext_req_v2 as an extension to struct mlx5_ib_alloc_ucontext_req. The extended size is determined in mlx5_ib_alloc_ucontext() and in case of old library we use uuarn 0 which works fine -- this is acheived due to create_user_qp() falling back from high to medium then to low class where low class will return 0. For new libraries we use the more sophisticated allocation algorithm. Signed-off-by: Eli Cohen Reviewed-by: Yann Droneaud Signed-off-by: Roland Dreier --- drivers/infiniband/hw/mlx5/main.c | 19 +++++++++++++++++-- drivers/infiniband/hw/mlx5/qp.c | 10 ++++++++-- drivers/infiniband/hw/mlx5/user.h | 7 +++++++ include/linux/mlx5/driver.h | 1 + 4 files changed, 33 insertions(+), 4 deletions(-) diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c index 9660d093f8cf..f4ef4a24d410 100644 --- a/drivers/infiniband/hw/mlx5/main.c +++ b/drivers/infiniband/hw/mlx5/main.c @@ -536,24 +536,38 @@ static struct ib_ucontext *mlx5_ib_alloc_ucontext(struct ib_device *ibdev, struct ib_udata *udata) { struct mlx5_ib_dev *dev = to_mdev(ibdev); - struct mlx5_ib_alloc_ucontext_req req; + struct mlx5_ib_alloc_ucontext_req_v2 req; struct mlx5_ib_alloc_ucontext_resp resp; struct mlx5_ib_ucontext *context; struct mlx5_uuar_info *uuari; struct mlx5_uar *uars; int gross_uuars; int num_uars; + int ver; int uuarn; int err; int i; + int reqlen; if (!dev->ib_active) return ERR_PTR(-EAGAIN); - err = ib_copy_from_udata(&req, udata, sizeof(req)); + memset(&req, 0, sizeof(req)); + reqlen = udata->inlen - sizeof(struct ib_uverbs_cmd_hdr); + if (reqlen == sizeof(struct mlx5_ib_alloc_ucontext_req)) + ver = 0; + else if (reqlen == sizeof(struct mlx5_ib_alloc_ucontext_req_v2)) + ver = 2; + else + return ERR_PTR(-EINVAL); + + err = ib_copy_from_udata(&req, udata, reqlen); if (err) return ERR_PTR(err); + if (req.flags || req.reserved) + return ERR_PTR(-EINVAL); + if (req.total_num_uuars > MLX5_MAX_UUARS) return ERR_PTR(-ENOMEM); @@ -626,6 +640,7 @@ static struct ib_ucontext *mlx5_ib_alloc_ucontext(struct ib_device *ibdev, if (err) goto out_uars; + uuari->ver = ver; uuari->num_low_latency_uuars = req.num_low_latency_uuars; uuari->uars = uars; uuari->num_uars = num_uars; diff --git a/drivers/infiniband/hw/mlx5/qp.c b/drivers/infiniband/hw/mlx5/qp.c index 492dc330e907..091576a777e9 100644 --- a/drivers/infiniband/hw/mlx5/qp.c +++ b/drivers/infiniband/hw/mlx5/qp.c @@ -430,11 +430,17 @@ static int alloc_uuar(struct mlx5_uuar_info *uuari, break; case MLX5_IB_LATENCY_CLASS_MEDIUM: - uuarn = alloc_med_class_uuar(uuari); + if (uuari->ver < 2) + uuarn = -ENOMEM; + else + uuarn = alloc_med_class_uuar(uuari); break; case MLX5_IB_LATENCY_CLASS_HIGH: - uuarn = alloc_high_class_uuar(uuari); + if (uuari->ver < 2) + uuarn = -ENOMEM; + else + uuarn = alloc_high_class_uuar(uuari); break; case MLX5_IB_LATENCY_CLASS_FAST_PATH: diff --git a/drivers/infiniband/hw/mlx5/user.h b/drivers/infiniband/hw/mlx5/user.h index 32a2a5dfc523..0f4f8e42a17f 100644 --- a/drivers/infiniband/hw/mlx5/user.h +++ b/drivers/infiniband/hw/mlx5/user.h @@ -62,6 +62,13 @@ struct mlx5_ib_alloc_ucontext_req { __u32 num_low_latency_uuars; }; +struct mlx5_ib_alloc_ucontext_req_v2 { + __u32 total_num_uuars; + __u32 num_low_latency_uuars; + __u32 flags; + __u32 reserved; +}; + struct mlx5_ib_alloc_ucontext_resp { __u32 qp_tab_size; __u32 bf_reg_size; diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index 554548cd3dd4..32cb18c399c2 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -227,6 +227,7 @@ struct mlx5_uuar_info { * protect uuar allocation data structs */ struct mutex lock; + u32 ver; }; struct mlx5_bf { From 1a4c3a3dc5fdeef2a7bdf4ac7d81df58c3c0a51e Mon Sep 17 00:00:00 2001 From: Eli Cohen Date: Thu, 6 Feb 2014 17:41:25 +0200 Subject: [PATCH 345/596] IB/mlx5: Don't set "block multicast loopback" capability Currently Connect-IB does not support blocking multicast loopback, so don't set IB_DEVICE_BLOCK_MULTICAST_LOOPBACK in the device caps. Reported by: Or Gerlitz Signed-off-by: Eli Cohen Signed-off-by: Roland Dreier --- drivers/infiniband/hw/mlx5/main.c | 3 +-- drivers/infiniband/hw/mlx5/qp.c | 4 ++-- 2 files changed, 3 insertions(+), 4 deletions(-) diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c index f4ef4a24d410..aa03e732b6a8 100644 --- a/drivers/infiniband/hw/mlx5/main.c +++ b/drivers/infiniband/hw/mlx5/main.c @@ -261,8 +261,7 @@ static int mlx5_ib_query_device(struct ib_device *ibdev, props->device_cap_flags = IB_DEVICE_CHANGE_PHY_PORT | IB_DEVICE_PORT_ACTIVE_EVENT | IB_DEVICE_SYS_IMAGE_GUID | - IB_DEVICE_RC_RNR_NAK_GEN | - IB_DEVICE_BLOCK_MULTICAST_LOOPBACK; + IB_DEVICE_RC_RNR_NAK_GEN; flags = dev->mdev.caps.flags; if (flags & MLX5_DEV_CAP_FLAG_BAD_PKEY_CNTR) props->device_cap_flags |= IB_DEVICE_BAD_PKEY_CNTR; diff --git a/drivers/infiniband/hw/mlx5/qp.c b/drivers/infiniband/hw/mlx5/qp.c index 091576a777e9..7dfe8a1c84cf 100644 --- a/drivers/infiniband/hw/mlx5/qp.c +++ b/drivers/infiniband/hw/mlx5/qp.c @@ -665,8 +665,8 @@ static int create_kernel_qp(struct mlx5_ib_dev *dev, int err; uuari = &dev->mdev.priv.uuari; - if (init_attr->create_flags & IB_QP_CREATE_BLOCK_MULTICAST_LOOPBACK) - qp->flags |= MLX5_IB_QP_BLOCK_MULTICAST_LOOPBACK; + if (init_attr->create_flags) + return -EINVAL; if (init_attr->qp_type == MLX5_IB_QPT_REG_UMR) lc = MLX5_IB_LATENCY_CLASS_FAST_PATH; From f8f1becfa4ac3231da55de68698cf7facf089646 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Thu, 6 Feb 2014 18:14:03 +0100 Subject: [PATCH 346/596] ALSA: hda - Fix leftover ifdef checks after modularization Since the commit [595fe1b702c3: ALSA: hda - Make CONFIG_SND_HDA_CODEC_* tristate], the kconfig variables for the generic parser and codec drivers can be "m" instead of boolean, but some codes are left unchanged to check only #ifdef CONFIG_SND_HDA_CODEC_XXX, which is no longer true for modules. This patch fixes them by replacing with IS_ENABLED() macros. Fixes: 595fe1b702c3 ('ALSA: hda - Make CONFIG_SND_HDA_CODEC_* tristate') Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=70161 Signed-off-by: Takashi Iwai --- sound/pci/hda/hda_codec.c | 6 +++--- sound/pci/hda/hda_intel.c | 2 +- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/sound/pci/hda/hda_codec.c b/sound/pci/hda/hda_codec.c index ec4536c8d8d4..69acbf758bbb 100644 --- a/sound/pci/hda/hda_codec.c +++ b/sound/pci/hda/hda_codec.c @@ -932,7 +932,7 @@ int snd_hda_bus_new(struct snd_card *card, } EXPORT_SYMBOL_GPL(snd_hda_bus_new); -#ifdef CONFIG_SND_HDA_GENERIC +#if IS_ENABLED(CONFIG_SND_HDA_GENERIC) #define is_generic_config(codec) \ (codec->modelname && !strcmp(codec->modelname, "generic")) #else @@ -1570,7 +1570,7 @@ int snd_hda_codec_update_widgets(struct hda_codec *codec) EXPORT_SYMBOL_GPL(snd_hda_codec_update_widgets); -#ifdef CONFIG_SND_HDA_CODEC_HDMI +#if IS_ENABLED(CONFIG_SND_HDA_CODEC_HDMI) /* if all audio out widgets are digital, let's assume the codec as a HDMI/DP */ static bool is_likely_hdmi_codec(struct hda_codec *codec) { @@ -1622,7 +1622,7 @@ int snd_hda_codec_configure(struct hda_codec *codec) unload_parser(codec); /* to be sure */ if (is_likely_hdmi_codec(codec)) patch = load_parser(codec, snd_hda_parse_hdmi_codec); -#ifdef CONFIG_SND_HDA_GENERIC +#if IS_ENABLED(CONFIG_SND_HDA_GENERIC) if (!patch) patch = load_parser(codec, snd_hda_parse_generic_codec); #endif diff --git a/sound/pci/hda/hda_intel.c b/sound/pci/hda/hda_intel.c index fa2879a21a50..e354ab1ec20f 100644 --- a/sound/pci/hda/hda_intel.c +++ b/sound/pci/hda/hda_intel.c @@ -198,7 +198,7 @@ MODULE_DESCRIPTION("Intel HDA driver"); #endif #if defined(CONFIG_PM) && defined(CONFIG_VGA_SWITCHEROO) -#ifdef CONFIG_SND_HDA_CODEC_HDMI +#if IS_ENABLED(CONFIG_SND_HDA_CODEC_HDMI) #define SUPPORT_VGA_SWITCHEROO #endif #endif From 7fe307117db5bd7ec6efb93c563dcf44577b6d2b Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Thu, 30 Jan 2014 17:59:02 +0100 Subject: [PATCH 347/596] ALSA: hda - Fix inconsistent Mic mute LED The current code for controlling mic mute LED in patch_sigmatel.c blindly assumes that there is a single capture switch. But, there can be multiple multiple ones, and each of them flips the state, ended up in an inconsistent state. For fixing this problem, this patch adds kcontrol to be passed to the hook function so that the callee can check which switch is being accessed. In stac_capture_led_hook(), the state is checked as a bitmask, and turns on the LED when all capture switches are off. Signed-off-by: Takashi Iwai --- sound/pci/hda/hda_generic.c | 8 ++++---- sound/pci/hda/hda_generic.h | 1 + sound/pci/hda/patch_conexant.c | 3 ++- sound/pci/hda/patch_realtek.c | 9 ++++++--- sound/pci/hda/patch_sigmatel.c | 27 +++++++++++++++++---------- sound/pci/hda/thinkpad_helper.c | 1 + 6 files changed, 31 insertions(+), 18 deletions(-) diff --git a/sound/pci/hda/hda_generic.c b/sound/pci/hda/hda_generic.c index 8321a97d5c05..d9a09bdd09db 100644 --- a/sound/pci/hda/hda_generic.c +++ b/sound/pci/hda/hda_generic.c @@ -3269,7 +3269,7 @@ static int cap_put_caller(struct snd_kcontrol *kcontrol, mutex_unlock(&codec->control_mutex); snd_hda_codec_flush_cache(codec); /* flush the updates */ if (err >= 0 && spec->cap_sync_hook) - spec->cap_sync_hook(codec, ucontrol); + spec->cap_sync_hook(codec, kcontrol, ucontrol); return err; } @@ -3390,7 +3390,7 @@ static int cap_single_sw_put(struct snd_kcontrol *kcontrol, return ret; if (spec->cap_sync_hook) - spec->cap_sync_hook(codec, ucontrol); + spec->cap_sync_hook(codec, kcontrol, ucontrol); return ret; } @@ -3795,7 +3795,7 @@ static int mux_select(struct hda_codec *codec, unsigned int adc_idx, return 0; snd_hda_activate_path(codec, path, true, false); if (spec->cap_sync_hook) - spec->cap_sync_hook(codec, NULL); + spec->cap_sync_hook(codec, NULL, NULL); path_power_down_sync(codec, old_path); return 1; } @@ -5270,7 +5270,7 @@ static void init_input_src(struct hda_codec *codec) } if (spec->cap_sync_hook) - spec->cap_sync_hook(codec, NULL); + spec->cap_sync_hook(codec, NULL, NULL); } /* set right pin controls for digital I/O */ diff --git a/sound/pci/hda/hda_generic.h b/sound/pci/hda/hda_generic.h index 07f767231c9f..c908afbe4d94 100644 --- a/sound/pci/hda/hda_generic.h +++ b/sound/pci/hda/hda_generic.h @@ -274,6 +274,7 @@ struct hda_gen_spec { void (*init_hook)(struct hda_codec *codec); void (*automute_hook)(struct hda_codec *codec); void (*cap_sync_hook)(struct hda_codec *codec, + struct snd_kcontrol *kcontrol, struct snd_ctl_elem_value *ucontrol); /* PCM hooks */ diff --git a/sound/pci/hda/patch_conexant.c b/sound/pci/hda/patch_conexant.c index 4e0ec146553d..bcf91bea3317 100644 --- a/sound/pci/hda/patch_conexant.c +++ b/sound/pci/hda/patch_conexant.c @@ -3291,7 +3291,8 @@ static void cxt_update_headset_mode(struct hda_codec *codec) } static void cxt_update_headset_mode_hook(struct hda_codec *codec, - struct snd_ctl_elem_value *ucontrol) + struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_value *ucontrol) { cxt_update_headset_mode(codec); } diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index d9693ca9546f..f363238668b7 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -708,7 +708,8 @@ static void alc_inv_dmic_sync(struct hda_codec *codec, bool force) } static void alc_inv_dmic_hook(struct hda_codec *codec, - struct snd_ctl_elem_value *ucontrol) + struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_value *ucontrol) { alc_inv_dmic_sync(codec, false); } @@ -3218,7 +3219,8 @@ static void alc269_fixup_hp_gpio_mute_hook(void *private_data, int enabled) /* turn on/off mic-mute LED per capture hook */ static void alc269_fixup_hp_gpio_mic_mute_hook(struct hda_codec *codec, - struct snd_ctl_elem_value *ucontrol) + struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_value *ucontrol) { struct alc_spec *spec = codec->spec; unsigned int oldval = spec->gpio_led; @@ -3528,7 +3530,8 @@ static void alc_update_headset_mode(struct hda_codec *codec) } static void alc_update_headset_mode_hook(struct hda_codec *codec, - struct snd_ctl_elem_value *ucontrol) + struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_value *ucontrol) { alc_update_headset_mode(codec); } diff --git a/sound/pci/hda/patch_sigmatel.c b/sound/pci/hda/patch_sigmatel.c index 6998cf29b9bc..7311badf6a94 100644 --- a/sound/pci/hda/patch_sigmatel.c +++ b/sound/pci/hda/patch_sigmatel.c @@ -194,7 +194,7 @@ struct sigmatel_spec { int default_polarity; unsigned int mic_mute_led_gpio; /* capture mute LED GPIO */ - bool mic_mute_led_on; /* current mic mute state */ + unsigned int mic_enabled; /* current mic mute state (bitmask) */ /* stream */ unsigned int stream_delay; @@ -324,19 +324,26 @@ static void stac_gpio_set(struct hda_codec *codec, unsigned int mask, /* hook for controlling mic-mute LED GPIO */ static void stac_capture_led_hook(struct hda_codec *codec, - struct snd_ctl_elem_value *ucontrol) + struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_value *ucontrol) { struct sigmatel_spec *spec = codec->spec; - bool mute; + unsigned int mask; + bool cur_mute, prev_mute; - if (!ucontrol) + if (!kcontrol || !ucontrol) return; - mute = !(ucontrol->value.integer.value[0] || - ucontrol->value.integer.value[1]); - if (spec->mic_mute_led_on != mute) { - spec->mic_mute_led_on = mute; - if (mute) + mask = 1U << snd_ctl_get_ioffidx(kcontrol, &ucontrol->id); + prev_mute = !spec->mic_enabled; + if (ucontrol->value.integer.value[0] || + ucontrol->value.integer.value[1]) + spec->mic_enabled |= mask; + else + spec->mic_enabled &= ~mask; + cur_mute = !spec->mic_enabled; + if (cur_mute != prev_mute) { + if (cur_mute) spec->gpio_data |= spec->mic_mute_led_gpio; else spec->gpio_data &= ~spec->mic_mute_led_gpio; @@ -4462,7 +4469,7 @@ static void stac_setup_gpio(struct hda_codec *codec) if (spec->mic_mute_led_gpio) { spec->gpio_mask |= spec->mic_mute_led_gpio; spec->gpio_dir |= spec->mic_mute_led_gpio; - spec->mic_mute_led_on = true; + spec->mic_enabled = 0; spec->gpio_data |= spec->mic_mute_led_gpio; spec->gen.cap_sync_hook = stac_capture_led_hook; diff --git a/sound/pci/hda/thinkpad_helper.c b/sound/pci/hda/thinkpad_helper.c index 5799fbc24c28..8fe3b8c18ed4 100644 --- a/sound/pci/hda/thinkpad_helper.c +++ b/sound/pci/hda/thinkpad_helper.c @@ -39,6 +39,7 @@ static void update_tpacpi_mute_led(void *private_data, int enabled) } static void update_tpacpi_micmute_led(struct hda_codec *codec, + struct snd_kcontrol *kcontrol, struct snd_ctl_elem_value *ucontrol) { if (!ucontrol || !led_set_func) From f47e5dc464251f661da9495fcbf003a0d22c1360 Mon Sep 17 00:00:00 2001 From: David Henningsson Date: Fri, 7 Feb 2014 11:00:16 +0100 Subject: [PATCH 348/596] ALSA: hda - Add a headset quirk for Dell XPS 13 This quirk is needed for the headset microphone to work. Alsa-info at http://www.alsa-project.org/db/?f=8c7dfe857ceff462ca2de133e67023c0f68de9cb Cc: stable@vger.kernel.org (3.10+) Reported-by: Po-Hsu Lin Signed-off-by: David Henningsson Signed-off-by: Takashi Iwai --- sound/pci/hda/patch_realtek.c | 1 + 1 file changed, 1 insertion(+) diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index f363238668b7..1266f18ce352 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -5109,6 +5109,7 @@ static const struct snd_pci_quirk alc662_fixup_tbl[] = { SND_PCI_QUIRK(0x1025, 0x038b, "Acer Aspire 8943G", ALC662_FIXUP_ASPIRE), SND_PCI_QUIRK(0x1028, 0x05d8, "Dell", ALC668_FIXUP_DELL_MIC_NO_PRESENCE), SND_PCI_QUIRK(0x1028, 0x05db, "Dell", ALC668_FIXUP_DELL_MIC_NO_PRESENCE), + SND_PCI_QUIRK(0x1028, 0x060a, "Dell XPS 13", ALC668_FIXUP_DELL_MIC_NO_PRESENCE), SND_PCI_QUIRK(0x1028, 0x0623, "Dell", ALC668_FIXUP_AUTO_MUTE), SND_PCI_QUIRK(0x1028, 0x0624, "Dell", ALC668_FIXUP_AUTO_MUTE), SND_PCI_QUIRK(0x1028, 0x0625, "Dell", ALC668_FIXUP_DELL_MIC_NO_PRESENCE), From f88abaa0d0dc0d1f1a9ae21f8e822918e5aadfdf Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Fri, 7 Feb 2014 12:07:59 +0100 Subject: [PATCH 349/596] ALSA: hda - Fix mic capture on Sony VAIO Pro 11 The very same fixup is needed to make the mic on Sony VAIO Pro 11 working as well as VAIO Pro 13 model. Reported-and-tested-by: Hendrik-Jan Heins Cc: Signed-off-by: Takashi Iwai --- sound/pci/hda/patch_realtek.c | 1 + 1 file changed, 1 insertion(+) diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index 1266f18ce352..0ab0b9e05967 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -4332,6 +4332,7 @@ static const struct snd_pci_quirk alc269_fixup_tbl[] = { SND_PCI_QUIRK(0x1043, 0x8398, "ASUS P1005", ALC269_FIXUP_STEREO_DMIC), SND_PCI_QUIRK(0x1043, 0x83ce, "ASUS P1005", ALC269_FIXUP_STEREO_DMIC), SND_PCI_QUIRK(0x1043, 0x8516, "ASUS X101CH", ALC269_FIXUP_ASUS_X101), + SND_PCI_QUIRK(0x104d, 0x90b5, "Sony VAIO Pro 11", ALC286_FIXUP_SONY_MIC_NO_PRESENCE), SND_PCI_QUIRK(0x104d, 0x90b6, "Sony VAIO Pro 13", ALC286_FIXUP_SONY_MIC_NO_PRESENCE), SND_PCI_QUIRK(0x104d, 0x9073, "Sony VAIO", ALC275_FIXUP_SONY_VAIO_GPIO2), SND_PCI_QUIRK(0x104d, 0x907b, "Sony VAIO", ALC275_FIXUP_SONY_HWEQ), From bd7fc645dabab423ef362186db2917f3919321d3 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Fri, 7 Feb 2014 12:53:07 +0100 Subject: [PATCH 350/596] netfilter: nf_tables: do not allow NFT_SET_ELEM_INTERVAL_END flag and data This combination is not allowed since end interval elements cannot contain data. Signed-off-by: Pablo Neira Ayuso Acked-by: Patrick McHardy --- net/netfilter/nf_tables_api.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 3a2e4800b415..d0c790e3e495 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -2741,6 +2741,9 @@ static int nft_add_set_elem(const struct nft_ctx *ctx, struct nft_set *set, if (nla[NFTA_SET_ELEM_DATA] == NULL && !(elem.flags & NFT_SET_ELEM_INTERVAL_END)) return -EINVAL; + if (nla[NFTA_SET_ELEM_DATA] != NULL && + elem.flags & NFT_SET_ELEM_INTERVAL_END) + return -EINVAL; } else { if (nla[NFTA_SET_ELEM_DATA] != NULL) return -EINVAL; From 2fb91ddbf8e1dcb207e1e2085473aeaeff975102 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Thu, 6 Feb 2014 16:15:39 +0100 Subject: [PATCH 351/596] netfilter: nft_rbtree: fix data handling of end interval elements This patch fixes several things which related to the handling of end interval elements: * Chain use underflow with intervals and map: If you add a rule using intervals+map that introduces a loop, the error path of the rbtree set decrements the chain refcount for each side of the interval, leading to a chain use counter underflow. * Don't copy the data part of the end interval element since, this area is uninitialized and this confuses the loop detection code. * Don't allocate room for the data part of end interval elements since this is unused. So, after this patch the idea is that end interval elements don't have a data part. Signed-off-by: Pablo Neira Ayuso Acked-by: Patrick McHardy --- net/netfilter/nft_rbtree.c | 16 +++++++++++----- 1 file changed, 11 insertions(+), 5 deletions(-) diff --git a/net/netfilter/nft_rbtree.c b/net/netfilter/nft_rbtree.c index ca0c1b231bfe..e21d69d13506 100644 --- a/net/netfilter/nft_rbtree.c +++ b/net/netfilter/nft_rbtree.c @@ -69,8 +69,10 @@ static void nft_rbtree_elem_destroy(const struct nft_set *set, struct nft_rbtree_elem *rbe) { nft_data_uninit(&rbe->key, NFT_DATA_VALUE); - if (set->flags & NFT_SET_MAP) + if (set->flags & NFT_SET_MAP && + !(rbe->flags & NFT_SET_ELEM_INTERVAL_END)) nft_data_uninit(rbe->data, set->dtype); + kfree(rbe); } @@ -108,7 +110,8 @@ static int nft_rbtree_insert(const struct nft_set *set, int err; size = sizeof(*rbe); - if (set->flags & NFT_SET_MAP) + if (set->flags & NFT_SET_MAP && + !(elem->flags & NFT_SET_ELEM_INTERVAL_END)) size += sizeof(rbe->data[0]); rbe = kzalloc(size, GFP_KERNEL); @@ -117,7 +120,8 @@ static int nft_rbtree_insert(const struct nft_set *set, rbe->flags = elem->flags; nft_data_copy(&rbe->key, &elem->key); - if (set->flags & NFT_SET_MAP) + if (set->flags & NFT_SET_MAP && + !(rbe->flags & NFT_SET_ELEM_INTERVAL_END)) nft_data_copy(rbe->data, &elem->data); err = __nft_rbtree_insert(set, rbe); @@ -153,7 +157,8 @@ static int nft_rbtree_get(const struct nft_set *set, struct nft_set_elem *elem) parent = parent->rb_right; else { elem->cookie = rbe; - if (set->flags & NFT_SET_MAP) + if (set->flags & NFT_SET_MAP && + !(rbe->flags & NFT_SET_ELEM_INTERVAL_END)) nft_data_copy(&elem->data, rbe->data); elem->flags = rbe->flags; return 0; @@ -177,7 +182,8 @@ static void nft_rbtree_walk(const struct nft_ctx *ctx, rbe = rb_entry(node, struct nft_rbtree_elem, node); nft_data_copy(&elem.key, &rbe->key); - if (set->flags & NFT_SET_MAP) + if (set->flags & NFT_SET_MAP && + !(rbe->flags & NFT_SET_ELEM_INTERVAL_END)) nft_data_copy(&elem.data, rbe->data); elem.flags = rbe->flags; From c2703b13a63da335053a2f3cb560bb806bdda0ae Mon Sep 17 00:00:00 2001 From: Rob Clark Date: Thu, 6 Feb 2014 19:19:20 -0500 Subject: [PATCH 352/596] drm/msm: bigger synchronization hammer Because we use a list_head in the bo to track it's position in a submit, we need to serialize at a higher layer. Otherwise there are problems when multiple contexts are SUBMIT'ing in parallel cmdstreams referencing a shared bo. Signed-off-by: Rob Clark --- drivers/gpu/drm/msm/msm_gem_submit.c | 9 +++++---- drivers/gpu/drm/msm/msm_gpu.c | 3 --- 2 files changed, 5 insertions(+), 7 deletions(-) diff --git a/drivers/gpu/drm/msm/msm_gem_submit.c b/drivers/gpu/drm/msm/msm_gem_submit.c index 5281d4bc37f7..5423e914e491 100644 --- a/drivers/gpu/drm/msm/msm_gem_submit.c +++ b/drivers/gpu/drm/msm/msm_gem_submit.c @@ -163,7 +163,7 @@ static int submit_validate_objects(struct msm_gem_submit *submit) /* if locking succeeded, pin bo: */ - ret = msm_gem_get_iova(&msm_obj->base, + ret = msm_gem_get_iova_locked(&msm_obj->base, submit->gpu->id, &iova); /* this would break the logic in the fail path.. there is no @@ -247,7 +247,7 @@ static int submit_reloc(struct msm_gem_submit *submit, struct msm_gem_object *ob /* For now, just map the entire thing. Eventually we probably * to do it page-by-page, w/ kmap() if not vmap()d.. */ - ptr = msm_gem_vaddr(&obj->base); + ptr = msm_gem_vaddr_locked(&obj->base); if (IS_ERR(ptr)) { ret = PTR_ERR(ptr); @@ -307,14 +307,12 @@ static void submit_cleanup(struct msm_gem_submit *submit, bool fail) { unsigned i; - mutex_lock(&submit->dev->struct_mutex); for (i = 0; i < submit->nr_bos; i++) { struct msm_gem_object *msm_obj = submit->bos[i].obj; submit_unlock_unpin_bo(submit, i); list_del_init(&msm_obj->submit_entry); drm_gem_object_unreference(&msm_obj->base); } - mutex_unlock(&submit->dev->struct_mutex); ww_acquire_fini(&submit->ticket); kfree(submit); @@ -342,6 +340,8 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data, if (args->nr_cmds > MAX_CMDS) return -EINVAL; + mutex_lock(&dev->struct_mutex); + submit = submit_create(dev, gpu, args->nr_bos); if (!submit) { ret = -ENOMEM; @@ -410,5 +410,6 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data, out: if (submit) submit_cleanup(submit, !!ret); + mutex_unlock(&dev->struct_mutex); return ret; } diff --git a/drivers/gpu/drm/msm/msm_gpu.c b/drivers/gpu/drm/msm/msm_gpu.c index 4ebce8be489d..0cfe3f426ee4 100644 --- a/drivers/gpu/drm/msm/msm_gpu.c +++ b/drivers/gpu/drm/msm/msm_gpu.c @@ -298,8 +298,6 @@ int msm_gpu_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit, struct msm_drm_private *priv = dev->dev_private; int i, ret; - mutex_lock(&dev->struct_mutex); - submit->fence = ++priv->next_fence; gpu->submitted_fence = submit->fence; @@ -331,7 +329,6 @@ int msm_gpu_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit, msm_gem_move_to_active(&msm_obj->base, gpu, true, submit->fence); } hangcheck_timer_reset(gpu); - mutex_unlock(&dev->struct_mutex); return ret; } From 4e6b788c3f2388c519ab6ed7fe59f372c85d26e9 Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Fri, 7 Feb 2014 16:33:20 +0100 Subject: [PATCH 353/596] drm/i915: Disable dp aux irq on g4x Apparently it's broken in the exact same way as the gmbus irq. For reference of the full story see commit c12aba5aa0e60b7947bc8b6ea25ef55c4acf81a4 Author: Jiri Kosina Date: Tue Mar 19 09:56:57 2013 +0100 drm/i915: stop using GMBUS IRQs on Gen4 chips The effect is that we have a storm of unclaimed interrupts on the legacy irq line. If that one is used by a different device then the kernel will complain and rather quickly kill the irq source. Which breaks any device trying to actually use the legacy irq line. This regression has been introduced commit 4aeebd7443e36b0a40032e518a9338f48bd27efc Author: Daniel Vetter Date: Thu Oct 31 09:53:36 2013 +0100 drm/i915: dp aux irq support for g4x/vlv Note that disabling MSI works around the issue, but we can't do that since apparently then the hw will miss interrupts. At least if relevant comments in i915_irq.c are accurate. v2: Cross-reference dp aux and gmbus gen4 comments. v3: Consolidate harder into i915_drv.h as suggested by Chris. Cc: Jani Nikula Cc: Jiri Kosina Cc: Chris Wilson Reported-and-tested-by: Jiri Kosina Reviewed-by: Chris Wilson Signed-off-by: Daniel Vetter --- drivers/gpu/drm/i915/i915_drv.h | 8 ++++++++ drivers/gpu/drm/i915/intel_dp.c | 2 +- drivers/gpu/drm/i915/intel_i2c.c | 7 ------- 3 files changed, 9 insertions(+), 8 deletions(-) diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index 98322053eb2a..c0cba20ea6af 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -1831,6 +1831,14 @@ struct drm_i915_file_private { /* Early gen2 have a totally busted CS tlb and require pinned batches. */ #define HAS_BROKEN_CS_TLB(dev) (IS_I830(dev) || IS_845G(dev)) +/* + * dp aux and gmbus irq on gen4 seems to be able to generate legacy interrupts + * even when in MSI mode. This results in spurious interrupt warnings if the + * legacy irq no. is shared with another device. The kernel then disables that + * interrupt source and so prevents the other device from working properly. + */ +#define HAS_AUX_IRQ(dev) (INTEL_INFO(dev)->gen >= 5) +#define HAS_GMBUS_IRQ(dev) (INTEL_INFO(dev)->gen >= 5) /* With the 945 and later, Y tiling got adjusted so that it was 32 128-byte * rows, which changed the alignment requirements and fence programming. diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c index c00b6e352c2b..2f517b85b3f4 100644 --- a/drivers/gpu/drm/i915/intel_dp.c +++ b/drivers/gpu/drm/i915/intel_dp.c @@ -404,7 +404,7 @@ intel_dp_aux_ch(struct intel_dp *intel_dp, int i, ret, recv_bytes; uint32_t status; int try, precharge, clock = 0; - bool has_aux_irq = true; + bool has_aux_irq = HAS_AUX_IRQ(dev); uint32_t timeout; /* dp aux is extremely sensitive to irq latency, hence request the diff --git a/drivers/gpu/drm/i915/intel_i2c.c b/drivers/gpu/drm/i915/intel_i2c.c index b1dc33f47899..d33b61d0dd33 100644 --- a/drivers/gpu/drm/i915/intel_i2c.c +++ b/drivers/gpu/drm/i915/intel_i2c.c @@ -258,13 +258,6 @@ intel_gpio_setup(struct intel_gmbus *bus, u32 pin) algo->data = bus; } -/* - * gmbus on gen4 seems to be able to generate legacy interrupts even when in MSI - * mode. This results in spurious interrupt warnings if the legacy irq no. is - * shared with another device. The kernel then disables that interrupt source - * and so prevents the other device from working properly. - */ -#define HAS_GMBUS_IRQ(dev) (INTEL_INFO(dev)->gen >= 5) static int gmbus_wait_hw_status(struct drm_i915_private *dev_priv, u32 gmbus2_status, From 30c54df7cb9b15b222529a028390b9c9582dd65e Mon Sep 17 00:00:00 2001 From: Alexander Usyskin Date: Mon, 27 Jan 2014 22:27:23 +0200 Subject: [PATCH 354/596] mei: clear write cb from waiting list on reset Clear write callbacks sitting in write_waiting list on reset. Otherwise these callbacks are left dangling and cause memory leak. Signed-off-by: Alexander Usyskin Signed-off-by: Tomas Winkler Cc: stable Signed-off-by: Greg Kroah-Hartman --- drivers/misc/mei/client.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/drivers/misc/mei/client.c b/drivers/misc/mei/client.c index 1ee2b9492a82..ccdacb5fcd8b 100644 --- a/drivers/misc/mei/client.c +++ b/drivers/misc/mei/client.c @@ -942,8 +942,16 @@ void mei_cl_all_wakeup(struct mei_device *dev) void mei_cl_all_write_clear(struct mei_device *dev) { struct mei_cl_cb *cb, *next; + struct list_head *list; - list_for_each_entry_safe(cb, next, &dev->write_list.list, list) { + list = &dev->write_list.list; + list_for_each_entry_safe(cb, next, list, list) { + list_del(&cb->list); + mei_io_cb_free(cb); + } + + list = &dev->write_waiting_list.list; + list_for_each_entry_safe(cb, next, list, list) { list_del(&cb->list); mei_io_cb_free(cb); } From 5cb906c7035f03a3a44fecece9d3ff8fcc75d6e0 Mon Sep 17 00:00:00 2001 From: Alexander Usyskin Date: Mon, 27 Jan 2014 22:27:24 +0200 Subject: [PATCH 355/596] mei: don't unset read cb ptr on reset Don't set read callback to NULL during reset as this leads to memory leak of both cb and its buffer. The memory is correctly freed during mei_release. The memory leak is detectable by kmemleak if application has open read call while system is going through suspend/resume. unreferenced object 0xecead780 (size 64): comm "AsyncTask #1", pid 1018, jiffies 4294949621 (age 152.440s) hex dump (first 32 bytes): 00 01 10 00 00 02 20 00 00 bf 30 f1 00 00 00 00 ...... ...0..... 00 00 00 00 00 00 00 00 36 01 00 00 00 70 da e2 ........6....p.. backtrace: [] kmemleak_alloc+0x3c/0xa0 [] kmem_cache_alloc_trace+0xc6/0x190 [] mei_io_cb_init+0x29/0x50 [] mei_cl_read_start+0x102/0x360 [] mei_read+0x103/0x4e0 [] vfs_read+0x89/0x160 [] SyS_read+0x4f/0x80 [] syscall_call+0x7/0xb [] 0xffffffff unreferenced object 0xe2da7000 (size 512): comm "AsyncTask #1", pid 1018, jiffies 4294949621 (age 152.440s) hex dump (first 32 bytes): 00 6c da e2 7c 00 00 00 00 00 00 00 c0 eb 0c 59 .l..|..........Y 1b 00 00 00 01 00 00 00 02 10 00 00 01 00 00 00 ................ backtrace: [] kmemleak_alloc+0x3c/0xa0 [] __kmalloc+0xe7/0x1d0 [] mei_io_cb_alloc_resp_buf+0x2e/0x60 [] mei_cl_read_start+0x12c/0x360 [] mei_read+0x103/0x4e0 [] vfs_read+0x89/0x160 [] SyS_read+0x4f/0x80 [] syscall_call+0x7/0xb [] 0xffffffff Signed-off-by: Alexander Usyskin Signed-off-by: Tomas Winkler Cc: stable Signed-off-by: Greg Kroah-Hartman --- drivers/misc/mei/client.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/misc/mei/client.c b/drivers/misc/mei/client.c index ccdacb5fcd8b..9b809cfc2899 100644 --- a/drivers/misc/mei/client.c +++ b/drivers/misc/mei/client.c @@ -908,7 +908,6 @@ void mei_cl_all_disconnect(struct mei_device *dev) list_for_each_entry_safe(cl, next, &dev->file_list, link) { cl->state = MEI_FILE_DISCONNECTED; cl->mei_flow_ctrl_creds = 0; - cl->read_cb = NULL; cl->timer_count = 0; } } From f0342e66b397947ed8c3eef8c37b5ca2d5b1bb50 Mon Sep 17 00:00:00 2001 From: Martyn Welch Date: Fri, 7 Feb 2014 15:48:56 +0000 Subject: [PATCH 356/596] VME: Correct read/write alignment algorithm In order to ensure the correct width cycles on the VME bus, the VME bridge drivers implement an algorithm to utilise the largest possible width reads and writes whilst maintaining natural alignment constraints. The algorithm currently looks at the start address rather than the current read/write address when determining whether a 16-bit width cycle is required to get to 32-bit alignment. This results in incorrect alignment, Reported-by: Jim Strouth Tested-by: Jim Strouth Signed-off-by: Martyn Welch Cc: stable Signed-off-by: Greg Kroah-Hartman --- drivers/vme/bridges/vme_ca91cx42.c | 4 ++-- drivers/vme/bridges/vme_tsi148.c | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/vme/bridges/vme_ca91cx42.c b/drivers/vme/bridges/vme_ca91cx42.c index a06edbfa95ca..1b5d48c578e1 100644 --- a/drivers/vme/bridges/vme_ca91cx42.c +++ b/drivers/vme/bridges/vme_ca91cx42.c @@ -884,7 +884,7 @@ static ssize_t ca91cx42_master_read(struct vme_master_resource *image, if (done == count) goto out; } - if ((uintptr_t)addr & 0x2) { + if ((uintptr_t)(addr + done) & 0x2) { if ((count - done) < 2) { *(u8 *)(buf + done) = ioread8(addr + done); done += 1; @@ -938,7 +938,7 @@ static ssize_t ca91cx42_master_write(struct vme_master_resource *image, if (done == count) goto out; } - if ((uintptr_t)addr & 0x2) { + if ((uintptr_t)(addr + done) & 0x2) { if ((count - done) < 2) { iowrite8(*(u8 *)(buf + done), addr + done); done += 1; diff --git a/drivers/vme/bridges/vme_tsi148.c b/drivers/vme/bridges/vme_tsi148.c index 16830d8b777c..9911cd5fddb5 100644 --- a/drivers/vme/bridges/vme_tsi148.c +++ b/drivers/vme/bridges/vme_tsi148.c @@ -1289,7 +1289,7 @@ static ssize_t tsi148_master_read(struct vme_master_resource *image, void *buf, if (done == count) goto out; } - if ((uintptr_t)addr & 0x2) { + if ((uintptr_t)(addr + done) & 0x2) { if ((count - done) < 2) { *(u8 *)(buf + done) = ioread8(addr + done); done += 1; @@ -1371,7 +1371,7 @@ static ssize_t tsi148_master_write(struct vme_master_resource *image, void *buf, if (done == count) goto out; } - if ((uintptr_t)addr & 0x2) { + if ((uintptr_t)(addr + done) & 0x2) { if ((count - done) < 2) { iowrite8(*(u8 *)(buf + done), addr + done); done += 1; From 821003ba4b9ffbf020a12177f0ea3d27826463c3 Mon Sep 17 00:00:00 2001 From: Jean-Jacques Hiblot Date: Wed, 15 Jan 2014 11:24:46 +0100 Subject: [PATCH 357/596] ARM: at91/dt: sam9263: fix compatibility string for the I2C Signed-off-by: Jean-Jacques Hiblot Signed-off-by: Nicolas Ferre --- arch/arm/boot/dts/at91sam9263.dtsi | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/boot/dts/at91sam9263.dtsi b/arch/arm/boot/dts/at91sam9263.dtsi index 0042f73068b0..fece8665fb63 100644 --- a/arch/arm/boot/dts/at91sam9263.dtsi +++ b/arch/arm/boot/dts/at91sam9263.dtsi @@ -523,7 +523,7 @@ usb1: gadget@fff78000 { }; i2c0: i2c@fff88000 { - compatible = "atmel,at91sam9263-i2c"; + compatible = "atmel,at91sam9260-i2c"; reg = <0xfff88000 0x100>; interrupts = <13 IRQ_TYPE_LEVEL_HIGH 6>; #address-cells = <1>; From 5f877518775a681b516dde700b511641fb3e4f9b Mon Sep 17 00:00:00 2001 From: Boris BREZILLON Date: Thu, 16 Jan 2014 16:25:34 +0100 Subject: [PATCH 358/596] ARM: at91/dt: fix sama5d3 ohci hclk clock reference The hclk clock of the ohci node is referencing udphs_clk instead of uhphs_clk. Signed-off-by: Boris BREZILLON Signed-off-by: Nicolas Ferre --- arch/arm/boot/dts/sama5d3.dtsi | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/boot/dts/sama5d3.dtsi b/arch/arm/boot/dts/sama5d3.dtsi index 52447c17537a..3d5faf85f51b 100644 --- a/arch/arm/boot/dts/sama5d3.dtsi +++ b/arch/arm/boot/dts/sama5d3.dtsi @@ -1228,7 +1228,7 @@ usb1: ohci@00600000 { compatible = "atmel,at91rm9200-ohci", "usb-ohci"; reg = <0x00600000 0x100000>; interrupts = <32 IRQ_TYPE_LEVEL_HIGH 2>; - clocks = <&usb>, <&uhphs_clk>, <&udphs_clk>, + clocks = <&usb>, <&uhphs_clk>, <&uhphs_clk>, <&uhpck>; clock-names = "usb_clk", "ohci_clk", "hclk", "uhpck"; status = "disabled"; From d7e67ee8b2ce27f4481d3cf2ac6ba8febcc3d888 Mon Sep 17 00:00:00 2001 From: Bo Shen Date: Tue, 14 Jan 2014 14:34:44 +0800 Subject: [PATCH 359/596] ARM: at91: enable USB host on at91sam9n12ek board Enable USB host on at91sam9n12ek board. Signed-off-by: Bo Shen Signed-off-by: Nicolas Ferre --- arch/arm/boot/dts/at91sam9n12ek.dts | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/arch/arm/boot/dts/at91sam9n12ek.dts b/arch/arm/boot/dts/at91sam9n12ek.dts index e9487f6f0166..924a6a6ffd0f 100644 --- a/arch/arm/boot/dts/at91sam9n12ek.dts +++ b/arch/arm/boot/dts/at91sam9n12ek.dts @@ -124,6 +124,10 @@ nand0: nand@40000000 { nand-on-flash-bbt; status = "okay"; }; + + usb0: ohci@00500000 { + status = "okay"; + }; }; leds { From 425bb8d6eb8cf7b6aa0cddd3ee59a919b0b21a80 Mon Sep 17 00:00:00 2001 From: Boris BREZILLON Date: Tue, 17 Dec 2013 10:36:17 +0100 Subject: [PATCH 360/596] mmc: atmel-mci: document clock properties Document the clock properties required by the atmel-mci driver. Signed-off-by: Boris BREZILLON Acked-by: Ludovic Desroches Signed-off-by: Nicolas Ferre --- Documentation/devicetree/bindings/mmc/atmel-hsmci.txt | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/Documentation/devicetree/bindings/mmc/atmel-hsmci.txt b/Documentation/devicetree/bindings/mmc/atmel-hsmci.txt index 0a85c70cd30a..07ad02075a93 100644 --- a/Documentation/devicetree/bindings/mmc/atmel-hsmci.txt +++ b/Documentation/devicetree/bindings/mmc/atmel-hsmci.txt @@ -13,6 +13,9 @@ Required properties: - #address-cells: should be one. The cell is the slot id. - #size-cells: should be zero. - at least one slot node +- clock-names: tuple listing input clock names. + Required elements: "mci_clk" +- clocks: phandles to input clocks. The node contains child nodes for each slot that the platform uses @@ -24,6 +27,8 @@ mmc0: mmc@f0008000 { interrupts = <12 4>; #address-cells = <1>; #size-cells = <0>; + clock-names = "mci_clk"; + clocks = <&mci0_clk>; [ child node definitions...] }; From bdb90b6bb78ff51603ca85b2d353699238b54c8e Mon Sep 17 00:00:00 2001 From: Boris BREZILLON Date: Tue, 17 Dec 2013 12:01:34 +0100 Subject: [PATCH 361/596] spi/atmel: document clock properties Document the clock properties required by the spi-atmel driver. Signed-off-by: Boris BREZILLON Signed-off-by: Nicolas Ferre --- Documentation/devicetree/bindings/spi/spi_atmel.txt | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/Documentation/devicetree/bindings/spi/spi_atmel.txt b/Documentation/devicetree/bindings/spi/spi_atmel.txt index 07e04cdc0c9e..4f8184d069cb 100644 --- a/Documentation/devicetree/bindings/spi/spi_atmel.txt +++ b/Documentation/devicetree/bindings/spi/spi_atmel.txt @@ -5,6 +5,9 @@ Required properties: - reg: Address and length of the register set for the device - interrupts: Should contain spi interrupt - cs-gpios: chipselects +- clock-names: tuple listing input clock names. + Required elements: "spi_clk" +- clocks: phandles to input clocks. Example: @@ -14,6 +17,8 @@ spi1: spi@fffcc000 { interrupts = <13 4 5>; #address-cells = <1>; #size-cells = <0>; + clocks = <&spi1_clk>; + clock-names = "spi_clk"; cs-gpios = <&pioB 3 0>; status = "okay"; From b7c2b6157079811586180d13c44a1095b57c2d47 Mon Sep 17 00:00:00 2001 From: Nicolas Ferre Date: Tue, 4 Feb 2014 15:05:55 +0100 Subject: [PATCH 362/596] ARM: at91: add Atmel's SAMA5D3 Xplained board Add DT file for new SAMA5D3 Xplained board. This board is based on Atmel's SAMA5D36 Cortex-A5 SoC. Signed-off-by: Nicolas Ferre --- arch/arm/boot/dts/Makefile | 1 + arch/arm/boot/dts/at91-sama5d3_xplained.dts | 229 ++++++++++++++++++++ 2 files changed, 230 insertions(+) create mode 100644 arch/arm/boot/dts/at91-sama5d3_xplained.dts diff --git a/arch/arm/boot/dts/Makefile b/arch/arm/boot/dts/Makefile index b9d6a8b485e0..6d1e43d46187 100644 --- a/arch/arm/boot/dts/Makefile +++ b/arch/arm/boot/dts/Makefile @@ -38,6 +38,7 @@ dtb-$(CONFIG_ARCH_AT91) += at91sam9g35ek.dtb dtb-$(CONFIG_ARCH_AT91) += at91sam9x25ek.dtb dtb-$(CONFIG_ARCH_AT91) += at91sam9x35ek.dtb # sama5d3 +dtb-$(CONFIG_ARCH_AT91) += at91-sama5d3_xplained.dtb dtb-$(CONFIG_ARCH_AT91) += sama5d31ek.dtb dtb-$(CONFIG_ARCH_AT91) += sama5d33ek.dtb dtb-$(CONFIG_ARCH_AT91) += sama5d34ek.dtb diff --git a/arch/arm/boot/dts/at91-sama5d3_xplained.dts b/arch/arm/boot/dts/at91-sama5d3_xplained.dts new file mode 100644 index 000000000000..ce1375595e5f --- /dev/null +++ b/arch/arm/boot/dts/at91-sama5d3_xplained.dts @@ -0,0 +1,229 @@ +/* + * at91-sama5d3_xplained.dts - Device Tree file for the SAMA5D3 Xplained board + * + * Copyright (C) 2014 Atmel, + * 2014 Nicolas Ferre + * + * Licensed under GPLv2 or later. + */ +/dts-v1/; +#include "sama5d36.dtsi" + +/ { + model = "SAMA5D3 Xplained"; + compatible = "atmel,sama5d3-xplained", "atmel,sama5d3", "atmel,sama5"; + + chosen { + bootargs = "console=ttyS0,115200"; + }; + + memory { + reg = <0x20000000 0x10000000>; + }; + + ahb { + apb { + mmc0: mmc@f0000000 { + pinctrl-0 = <&pinctrl_mmc0_clk_cmd_dat0 &pinctrl_mmc0_dat1_3 &pinctrl_mmc0_dat4_7 &pinctrl_mmc0_cd>; + status = "okay"; + slot@0 { + reg = <0>; + bus-width = <8>; + cd-gpios = <&pioE 0 GPIO_ACTIVE_LOW>; + }; + }; + + spi0: spi@f0004000 { + cs-gpios = <&pioD 13 0>; + status = "okay"; + }; + + can0: can@f000c000 { + status = "okay"; + }; + + i2c0: i2c@f0014000 { + status = "okay"; + }; + + i2c1: i2c@f0018000 { + status = "okay"; + }; + + macb0: ethernet@f0028000 { + phy-mode = "rgmii"; + status = "okay"; + }; + + usart0: serial@f001c000 { + status = "okay"; + }; + + usart1: serial@f0020000 { + pinctrl-0 = <&pinctrl_usart1 &pinctrl_usart1_rts_cts>; + status = "okay"; + }; + + uart0: serial@f0024000 { + status = "okay"; + }; + + mmc1: mmc@f8000000 { + pinctrl-0 = <&pinctrl_mmc1_clk_cmd_dat0 &pinctrl_mmc1_dat1_3 &pinctrl_mmc1_cd>; + status = "okay"; + slot@0 { + reg = <0>; + bus-width = <4>; + cd-gpios = <&pioE 1 GPIO_ACTIVE_HIGH>; + }; + }; + + spi1: spi@f8008000 { + cs-gpios = <&pioC 25 0>, <0>, <0>, <&pioD 16 0>; + status = "okay"; + }; + + adc0: adc@f8018000 { + pinctrl-0 = < + &pinctrl_adc0_adtrg + &pinctrl_adc0_ad0 + &pinctrl_adc0_ad1 + &pinctrl_adc0_ad2 + &pinctrl_adc0_ad3 + &pinctrl_adc0_ad4 + &pinctrl_adc0_ad5 + &pinctrl_adc0_ad6 + &pinctrl_adc0_ad7 + &pinctrl_adc0_ad8 + &pinctrl_adc0_ad9 + >; + status = "okay"; + }; + + i2c2: i2c@f801c000 { + dmas = <0>, <0>; /* Do not use DMA for i2c2 */ + status = "okay"; + }; + + macb1: ethernet@f802c000 { + phy-mode = "rmii"; + status = "okay"; + }; + + dbgu: serial@ffffee00 { + status = "okay"; + }; + + pinctrl@fffff200 { + board { + pinctrl_mmc0_cd: mmc0_cd { + atmel,pins = + ; + }; + + pinctrl_mmc1_cd: mmc1_cd { + atmel,pins = + ; + }; + + pinctrl_usba_vbus: usba_vbus { + atmel,pins = + ; /* PE9, conflicts with A9 */ + }; + }; + }; + + pmc: pmc@fffffc00 { + main: mainck { + clock-frequency = <12000000>; + }; + }; + }; + + nand0: nand@60000000 { + nand-bus-width = <8>; + nand-ecc-mode = "hw"; + atmel,has-pmecc; + atmel,pmecc-cap = <4>; + atmel,pmecc-sector-size = <512>; + nand-on-flash-bbt; + status = "okay"; + + at91bootstrap@0 { + label = "at91bootstrap"; + reg = <0x0 0x40000>; + }; + + bootloader@40000 { + label = "bootloader"; + reg = <0x40000 0x80000>; + }; + + bootloaderenv@c0000 { + label = "bootloader env"; + reg = <0xc0000 0xc0000>; + }; + + dtb@180000 { + label = "device tree"; + reg = <0x180000 0x80000>; + }; + + kernel@200000 { + label = "kernel"; + reg = <0x200000 0x600000>; + }; + + rootfs@800000 { + label = "rootfs"; + reg = <0x800000 0x0f800000>; + }; + }; + + usb0: gadget@00500000 { + atmel,vbus-gpio = <&pioE 9 GPIO_ACTIVE_HIGH>; /* PE9, conflicts with A9 */ + pinctrl-names = "default"; + pinctrl-0 = <&pinctrl_usba_vbus>; + status = "okay"; + }; + + usb1: ohci@00600000 { + num-ports = <3>; + atmel,vbus-gpio = <0 + &pioE 3 GPIO_ACTIVE_LOW + &pioE 4 GPIO_ACTIVE_LOW + >; + status = "okay"; + }; + + usb2: ehci@00700000 { + status = "okay"; + }; + }; + + gpio_keys { + compatible = "gpio-keys"; + + bp3 { + label = "PB_USER"; + gpios = <&pioE 29 GPIO_ACTIVE_LOW>; + linux,code = <0x104>; + gpio-key,wakeup; + }; + }; + + leds { + compatible = "gpio-leds"; + + d2 { + label = "d2"; + gpios = <&pioE 23 GPIO_ACTIVE_LOW>; /* PE23, conflicts with A23, CTS2 */ + linux,default-trigger = "heartbeat"; + }; + + d3 { + label = "d3"; + gpios = <&pioE 24 GPIO_ACTIVE_HIGH>; + }; + }; +}; From 62f9c8b40d2db915f89a6aa47395412fb29f1cfc Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Fri, 7 Feb 2014 14:45:01 +0100 Subject: [PATCH 363/596] netfilter: nf_tables: fix loop checking with end interval elements Fix access to uninitialized data for end interval elements. The element data part is uninitialized in interval end elements. Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_tables_api.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index d0c790e3e495..adce01e8bb57 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -2998,6 +2998,9 @@ static int nf_tables_loop_check_setelem(const struct nft_ctx *ctx, const struct nft_set_iter *iter, const struct nft_set_elem *elem) { + if (elem->flags & NFT_SET_ELEM_INTERVAL_END) + return 0; + switch (elem->data.verdict) { case NFT_JUMP: case NFT_GOTO: From e28bab4828354583bb66ac09021ca69b341a7db4 Mon Sep 17 00:00:00 2001 From: "K. Y. Srinivasan" Date: Wed, 15 Jan 2014 17:12:58 -0800 Subject: [PATCH 364/596] Drivers: hv: vmbus: Specify the target CPU that should receive notification During the initial VMBUS connect phase, starting with WS2012 R2, we should specify the VPCU in the guest that should receive the notification. Fix this issue. This fix is required to properly connect to the host in the kexeced kernel. Signed-off-by: K. Y. Srinivasan Cc: [3.9+] Signed-off-by: Greg Kroah-Hartman --- drivers/hv/connection.c | 2 ++ include/linux/hyperv.h | 2 +- 2 files changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/hv/connection.c b/drivers/hv/connection.c index af6edf9b1936..855bbda9964d 100644 --- a/drivers/hv/connection.c +++ b/drivers/hv/connection.c @@ -78,6 +78,8 @@ static int vmbus_negotiate_version(struct vmbus_channel_msginfo *msginfo, msg->interrupt_page = virt_to_phys(vmbus_connection.int_page); msg->monitor_page1 = virt_to_phys(vmbus_connection.monitor_pages[0]); msg->monitor_page2 = virt_to_phys(vmbus_connection.monitor_pages[1]); + if (version == VERSION_WIN8) + msg->target_vcpu = hv_context.vp_index[smp_processor_id()]; /* * Add to list before we send the request since we may diff --git a/include/linux/hyperv.h b/include/linux/hyperv.h index 15da677478dd..344883dce584 100644 --- a/include/linux/hyperv.h +++ b/include/linux/hyperv.h @@ -875,7 +875,7 @@ struct vmbus_channel_relid_released { struct vmbus_channel_initiate_contact { struct vmbus_channel_message_header header; u32 vmbus_version_requested; - u32 padding2; + u32 target_vcpu; /* The VCPU the host should respond to */ u64 interrupt_page; u64 monitor_page1; u64 monitor_page2; From 269f979467cf49f2ea8132316c1f00f8c9678f7c Mon Sep 17 00:00:00 2001 From: "K. Y. Srinivasan" Date: Thu, 16 Jan 2014 11:59:58 -0800 Subject: [PATCH 365/596] Drivers: hv: vmbus: Don't timeout during the initial connection with host When the guest attempts to connect with the host when there may already be a connection with the host (as would be the case during the kdump/kexec path), it is difficult to guarantee timely response from the host. Starting with WS2012 R2, the host supports this ability to re-connect with the host (explicitly to support kexec). Prior to responding to the guest, the host needs to ensure that device states based on the previous connection to the host have been properly torn down. This may introduce unbounded delays. To deal with this issue, don't do a timed wait during the initial connect with the host. Signed-off-by: K. Y. Srinivasan Cc: Signed-off-by: Greg Kroah-Hartman --- drivers/hv/connection.c | 11 +---------- 1 file changed, 1 insertion(+), 10 deletions(-) diff --git a/drivers/hv/connection.c b/drivers/hv/connection.c index 855bbda9964d..f2d7bf90c9fe 100644 --- a/drivers/hv/connection.c +++ b/drivers/hv/connection.c @@ -67,7 +67,6 @@ static int vmbus_negotiate_version(struct vmbus_channel_msginfo *msginfo, int ret = 0; struct vmbus_channel_initiate_contact *msg; unsigned long flags; - int t; init_completion(&msginfo->waitevent); @@ -102,15 +101,7 @@ static int vmbus_negotiate_version(struct vmbus_channel_msginfo *msginfo, } /* Wait for the connection response */ - t = wait_for_completion_timeout(&msginfo->waitevent, 5*HZ); - if (t == 0) { - spin_lock_irqsave(&vmbus_connection.channelmsg_lock, - flags); - list_del(&msginfo->msglistentry); - spin_unlock_irqrestore(&vmbus_connection.channelmsg_lock, - flags); - return -ETIMEDOUT; - } + wait_for_completion(&msginfo->waitevent); spin_lock_irqsave(&vmbus_connection.channelmsg_lock, flags); list_del(&msginfo->msglistentry); From f17083c3affccfe955b0a419056784096c18fea8 Mon Sep 17 00:00:00 2001 From: Adam Thomson Date: Thu, 6 Feb 2014 18:03:21 +0000 Subject: [PATCH 366/596] regulator: da9055: Remove use of regmap_irq_get_virq() Signed-off-by: Adam Thomson Signed-off-by: Mark Brown --- drivers/regulator/da9055-regulator.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/regulator/da9055-regulator.c b/drivers/regulator/da9055-regulator.c index 7f340206d329..b14ebdad5dd2 100644 --- a/drivers/regulator/da9055-regulator.c +++ b/drivers/regulator/da9055-regulator.c @@ -576,7 +576,9 @@ static int da9055_regulator_probe(struct platform_device *pdev) /* Only LDO 5 and 6 has got the over current interrupt */ if (pdev->id == DA9055_ID_LDO5 || pdev->id == DA9055_ID_LDO6) { irq = platform_get_irq_byname(pdev, "REGULATOR"); - irq = regmap_irq_get_virq(da9055->irq_data, irq); + if (irq < 0) + return irq; + ret = devm_request_threaded_irq(&pdev->dev, irq, NULL, da9055_ldo5_6_oc_irq, IRQF_TRIGGER_HIGH | From 5bbb2ae3d6f896f8d2082d1eceb6131c2420b7cf Mon Sep 17 00:00:00 2001 From: Paul Bolle Date: Tue, 4 Feb 2014 23:23:12 +0100 Subject: [PATCH 367/596] raw: test against runtime value of max_raw_minors bind_get() checks the device number it is called with. It uses MAX_RAW_MINORS for the upper bound. But MAX_RAW_MINORS is set at compile time while the actual number of raw devices can be set at runtime. This means the test can either be too strict or too lenient. And if the test ends up being too lenient bind_get() might try to access memory beyond what was allocated for "raw_devices". So check against the runtime value (max_raw_minors) in this function. Signed-off-by: Paul Bolle Acked-by: Jan Kara Cc: stable Signed-off-by: Greg Kroah-Hartman --- drivers/char/raw.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/char/raw.c b/drivers/char/raw.c index f3223aac4df1..6e8d65e9b1d3 100644 --- a/drivers/char/raw.c +++ b/drivers/char/raw.c @@ -190,7 +190,7 @@ static int bind_get(int number, dev_t *dev) struct raw_device_data *rawdev; struct block_device *bdev; - if (number <= 0 || number >= MAX_RAW_MINORS) + if (number <= 0 || number >= max_raw_minors) return -EINVAL; rawdev = &raw_devices[number]; From 7143479a6a4f853881b19999310ada53b616d2ca Mon Sep 17 00:00:00 2001 From: Paul Bolle Date: Tue, 4 Feb 2014 23:23:40 +0100 Subject: [PATCH 368/596] raw: set range for MAX_RAW_DEVS The Kconfig symbol MAX_RAW_DEVS is meant to be between 1 and 65536. But those boundaries are not enforced by its Kconfig entry. Note that MAX_RAW_DEVS is used to set MAX_RAW_MINORS in drivers/char/raw.c. If one would accidentally set MAX_RAW_DEVS to an invalid value, that invalid value will actually end up being used in raw_init(). So add an appropriate range to this Kconfig entry. Signed-off-by: Paul Bolle Signed-off-by: Greg Kroah-Hartman --- drivers/char/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/char/Kconfig b/drivers/char/Kconfig index fa3243d71c76..1386749b48ff 100644 --- a/drivers/char/Kconfig +++ b/drivers/char/Kconfig @@ -499,6 +499,7 @@ config RAW_DRIVER config MAX_RAW_DEVS int "Maximum number of RAW devices to support (1-65536)" depends on RAW_DRIVER + range 1 65536 default "256" help The maximum number of RAW devices that are supported. From 4fedd0bf479558e50924c6d88f9197336742d20f Mon Sep 17 00:00:00 2001 From: Paul Gortmaker Date: Wed, 15 Jan 2014 16:35:43 -0500 Subject: [PATCH 369/596] drivers/tty/hvc: don't use module_init in non-modular hyp. console code The HVC_OPAL/RTAS/UDBG/XEN options are all bool, and hence their support is either present or absent. It will never be modular, so using module_init as an alias for __initcall is rather misleading. Fix this up now, so that we can relocate module_init from init.h into module.h in the future. If we don't do this, we'd have to add module.h to obviously non-modular code, and that would be a worse thing. Note that direct use of __initcall is discouraged, vs. one of the priority categorized subgroups. As __initcall gets mapped onto device_initcall, our use of device_initcall directly in this change means that the runtime impact is zero -- it will remain at level 6 in initcall ordering. Also the __exitcall functions have been outright deleted since they are only ever of interest to UML, and UML will never be using any of this code. Cc: Richard Weinberger Cc: Konrad Rzeszutek Wilk Cc: Boris Ostrovsky Signed-off-by: Paul Gortmaker Acked-by: David Vrabel Signed-off-by: Greg Kroah-Hartman --- drivers/tty/hvc/hvc_opal.c | 8 +------- drivers/tty/hvc/hvc_rtas.c | 12 +----------- drivers/tty/hvc/hvc_udbg.c | 9 +-------- drivers/tty/hvc/hvc_xen.c | 17 +---------------- 4 files changed, 4 insertions(+), 42 deletions(-) diff --git a/drivers/tty/hvc/hvc_opal.c b/drivers/tty/hvc/hvc_opal.c index 6496872e2e47..b01659bd4f7c 100644 --- a/drivers/tty/hvc/hvc_opal.c +++ b/drivers/tty/hvc/hvc_opal.c @@ -255,13 +255,7 @@ static int __init hvc_opal_init(void) /* Register as a vio device to receive callbacks */ return platform_driver_register(&hvc_opal_driver); } -module_init(hvc_opal_init); - -static void __exit hvc_opal_exit(void) -{ - platform_driver_unregister(&hvc_opal_driver); -} -module_exit(hvc_opal_exit); +device_initcall(hvc_opal_init); static void udbg_opal_putc(char c) { diff --git a/drivers/tty/hvc/hvc_rtas.c b/drivers/tty/hvc/hvc_rtas.c index 0069bb86ba49..08c87920b74a 100644 --- a/drivers/tty/hvc/hvc_rtas.c +++ b/drivers/tty/hvc/hvc_rtas.c @@ -102,17 +102,7 @@ static int __init hvc_rtas_init(void) return 0; } -module_init(hvc_rtas_init); - -/* This will tear down the tty portion of the driver */ -static void __exit hvc_rtas_exit(void) -{ - /* Really the fun isn't over until the worker thread breaks down and - * the tty cleans up */ - if (hvc_rtas_dev) - hvc_remove(hvc_rtas_dev); -} -module_exit(hvc_rtas_exit); +device_initcall(hvc_rtas_init); /* This will happen prior to module init. There is no tty at this time? */ static int __init hvc_rtas_console_init(void) diff --git a/drivers/tty/hvc/hvc_udbg.c b/drivers/tty/hvc/hvc_udbg.c index 72228276fe31..9cf573d06a29 100644 --- a/drivers/tty/hvc/hvc_udbg.c +++ b/drivers/tty/hvc/hvc_udbg.c @@ -80,14 +80,7 @@ static int __init hvc_udbg_init(void) return 0; } -module_init(hvc_udbg_init); - -static void __exit hvc_udbg_exit(void) -{ - if (hvc_udbg_dev) - hvc_remove(hvc_udbg_dev); -} -module_exit(hvc_udbg_exit); +device_initcall(hvc_udbg_init); static int __init hvc_udbg_console_init(void) { diff --git a/drivers/tty/hvc/hvc_xen.c b/drivers/tty/hvc/hvc_xen.c index 636c9baad7a5..2dc2831840ca 100644 --- a/drivers/tty/hvc/hvc_xen.c +++ b/drivers/tty/hvc/hvc_xen.c @@ -561,18 +561,7 @@ static int __init xen_hvc_init(void) #endif return r; } - -static void __exit xen_hvc_fini(void) -{ - struct xencons_info *entry, *next; - - if (list_empty(&xenconsoles)) - return; - - list_for_each_entry_safe(entry, next, &xenconsoles, list) { - xen_console_remove(entry); - } -} +device_initcall(xen_hvc_init); static int xen_cons_init(void) { @@ -598,10 +587,6 @@ static int xen_cons_init(void) hvc_instantiate(HVC_COOKIE, 0, ops); return 0; } - - -module_init(xen_hvc_init); -module_exit(xen_hvc_fini); console_initcall(xen_cons_init); #ifdef CONFIG_EARLY_PRINTK From 3ac06b905655b3ef2fd2196bab36e4587e1e4e4f Mon Sep 17 00:00:00 2001 From: Lars Poeschel Date: Tue, 7 Jan 2014 13:34:37 +0100 Subject: [PATCH 370/596] tty: n_gsm: Fix for modems with brk in modem status control 3GPP TS 07.10 states in section 5.4.6.3.7: "The length byte contains the value 2 or 3 ... depending on the break signal." The break byte is optional and if it is sent, the length is 3. In fact the driver was not able to work with modems that send this break byte in their modem status control message. If the modem just sends the break byte if it is really set, then weird things might happen. The code for deconding the modem status to the internal linux presentation in gsm_process_modem has already a big comment about this 2 or 3 byte length thing and it is already able to decode the brk, but the code calling the gsm_process_modem function in gsm_control_modem does not encode it and hand it over the right way. This patch fixes this. Without this fix if the modem sends the brk byte in it's modem status control message the driver will hang when opening a muxed channel. Signed-off-by: Lars Poeschel Cc: stable Signed-off-by: Greg Kroah-Hartman --- drivers/tty/n_gsm.c | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/drivers/tty/n_gsm.c b/drivers/tty/n_gsm.c index f34461c5f14e..2ebe47b78a3e 100644 --- a/drivers/tty/n_gsm.c +++ b/drivers/tty/n_gsm.c @@ -1090,6 +1090,7 @@ static void gsm_control_modem(struct gsm_mux *gsm, u8 *data, int clen) { unsigned int addr = 0; unsigned int modem = 0; + unsigned int brk = 0; struct gsm_dlci *dlci; int len = clen; u8 *dp = data; @@ -1116,6 +1117,16 @@ static void gsm_control_modem(struct gsm_mux *gsm, u8 *data, int clen) if (len == 0) return; } + len--; + if (len > 0) { + while (gsm_read_ea(&brk, *dp++) == 0) { + len--; + if (len == 0) + return; + } + modem <<= 7; + modem |= (brk & 0x7f); + } tty = tty_port_tty_get(&dlci->port); gsm_process_modem(tty, dlci, modem, clen); if (tty) { From d8a5dc3033af2fd6d16030d2ee4fbd073460fe54 Mon Sep 17 00:00:00 2001 From: Hannes Reinecke Date: Fri, 7 Feb 2014 11:38:30 +0100 Subject: [PATCH 371/596] tty: Set correct tty name in 'active' sysfs attribute The 'active' sysfs attribute should refer to the currently active tty devices the console is running on, not the currently active console. The console structure doesn't refer to any device in sysfs, only the tty the console is running on has. So we need to print out the tty names in 'active', not the console names. This resolves an issue on s390 platforms in determining the correct console device to use. Cc: Lennart Poettering Cc: Kay Sievers Cc: Jiri Slaby Cc: David Herrmann Signed-off-by: Werner Fink Signed-off-by: Hannes Reinecke Cc: stable Signed-off-by: Greg Kroah-Hartman --- Documentation/ABI/testing/sysfs-tty | 3 ++- drivers/tty/tty_io.c | 25 ++++++++++++++++++------- 2 files changed, 20 insertions(+), 8 deletions(-) diff --git a/Documentation/ABI/testing/sysfs-tty b/Documentation/ABI/testing/sysfs-tty index ad22fb0ee765..a2ccec35ffce 100644 --- a/Documentation/ABI/testing/sysfs-tty +++ b/Documentation/ABI/testing/sysfs-tty @@ -3,7 +3,8 @@ Date: Nov 2010 Contact: Kay Sievers Description: Shows the list of currently configured - console devices, like 'tty1 ttyS0'. + tty devices used for the console, + like 'tty1 ttyS0'. The last entry in the file is the active device connected to /dev/console. The file supports poll() to detect virtual diff --git a/drivers/tty/tty_io.c b/drivers/tty/tty_io.c index c74a00ad7add..bd2715a9d8e5 100644 --- a/drivers/tty/tty_io.c +++ b/drivers/tty/tty_io.c @@ -1267,16 +1267,17 @@ static void pty_line_name(struct tty_driver *driver, int index, char *p) * @p: output buffer of at least 7 bytes * * Generate a name from a driver reference and write it to the output - * buffer. + * buffer. Return the number of bytes written. * * Locking: None */ -static void tty_line_name(struct tty_driver *driver, int index, char *p) +static ssize_t tty_line_name(struct tty_driver *driver, int index, char *p) { if (driver->flags & TTY_DRIVER_UNNUMBERED_NODE) - strcpy(p, driver->name); + return sprintf(p, "%s", driver->name); else - sprintf(p, "%s%d", driver->name, index + driver->name_base); + return sprintf(p, "%s%d", driver->name, + index + driver->name_base); } /** @@ -3545,9 +3546,19 @@ static ssize_t show_cons_active(struct device *dev, if (i >= ARRAY_SIZE(cs)) break; } - while (i--) - count += sprintf(buf + count, "%s%d%c", - cs[i]->name, cs[i]->index, i ? ' ':'\n'); + while (i--) { + struct tty_driver *driver; + const char *name = cs[i]->name; + int index = cs[i]->index; + + driver = cs[i]->device(cs[i], &index); + if (driver) { + count += tty_line_name(driver, index, buf + count); + count += sprintf(buf + count, "%c", i ? ' ':'\n'); + } else + count += sprintf(buf + count, "%s%d%c", + name, index, i ? ' ':'\n'); + } console_unlock(); return count; From 8e86f0b409a44193f1587e87b69c5dcf8f65be67 Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Tue, 4 Feb 2014 12:29:12 +0000 Subject: [PATCH 372/596] arm64: atomics: fix use of acquire + release for full barrier semantics Linux requires a number of atomic operations to provide full barrier semantics, that is no memory accesses after the operation can be observed before any accesses up to and including the operation in program order. On arm64, these operations have been incorrectly implemented as follows: // A, B, C are independent memory locations // atomic_op (B) 1: ldaxr x0, [B] // Exclusive load with acquire stlxr w1, x0, [B] // Exclusive store with release cbnz w1, 1b The assumption here being that two half barriers are equivalent to a full barrier, so the only permitted ordering would be A -> B -> C (where B is the atomic operation involving both a load and a store). Unfortunately, this is not the case by the letter of the architecture and, in fact, the accesses to A and C are permitted to pass their nearest half barrier resulting in orderings such as Bl -> A -> C -> Bs or Bl -> C -> A -> Bs (where Bl is the load-acquire on B and Bs is the store-release on B). This is a clear violation of the full barrier requirement. The simple way to fix this is to implement the same algorithm as ARMv7 using explicit barriers: // atomic_op (B) dmb ish // Full barrier 1: ldxr x0, [B] // Exclusive load stxr w1, x0, [B] // Exclusive store cbnz w1, 1b dmb ish // Full barrier but this has the undesirable effect of introducing *two* full barrier instructions. A better approach is actually the following, non-intuitive sequence: // atomic_op (B) 1: ldxr x0, [B] // Exclusive load stlxr w1, x0, [B] // Exclusive store with release cbnz w1, 1b dmb ish // Full barrier The simple observations here are: - The dmb ensures that no subsequent accesses (e.g. the access to C) can enter or pass the atomic sequence. - The dmb also ensures that no prior accesses (e.g. the access to A) can pass the atomic sequence. - Therefore, no prior access can pass a subsequent access, or vice-versa (i.e. A is strictly ordered before C). - The stlxr ensures that no prior access can pass the store component of the atomic operation. The only tricky part remaining is the ordering between the ldxr and the access to A, since the absence of the first dmb means that we're now permitting re-ordering between the ldxr and any prior accesses. From an (arbitrary) observer's point of view, there are two scenarios: 1. We have observed the ldxr. This means that if we perform a store to [B], the ldxr will still return older data. If we can observe the ldxr, then we can potentially observe the permitted re-ordering with the access to A, which is clearly an issue when compared to the dmb variant of the code. Thankfully, the exclusive monitor will save us here since it will be cleared as a result of the store and the ldxr will retry. Notice that any use of a later memory observation to imply observation of the ldxr will also imply observation of the access to A, since the stlxr/dmb ensure strict ordering. 2. We have not observed the ldxr. This means we can perform a store and influence the later ldxr. However, that doesn't actually tell us anything about the access to [A], so we've not lost anything here either when compared to the dmb variant. This patch implements this solution for our barriered atomic operations, ensuring that we satisfy the full barrier requirements where they are needed. Cc: Cc: Peter Zijlstra Signed-off-by: Will Deacon Signed-off-by: Catalin Marinas --- arch/arm64/include/asm/atomic.h | 29 ++++++++++++++++++++--------- arch/arm64/include/asm/cmpxchg.h | 9 +++++---- arch/arm64/include/asm/futex.h | 6 ++++-- arch/arm64/kernel/kuser32.S | 6 ++++-- arch/arm64/lib/bitops.S | 3 ++- 5 files changed, 35 insertions(+), 18 deletions(-) diff --git a/arch/arm64/include/asm/atomic.h b/arch/arm64/include/asm/atomic.h index 01de5aaa3edc..e32893e005d4 100644 --- a/arch/arm64/include/asm/atomic.h +++ b/arch/arm64/include/asm/atomic.h @@ -64,7 +64,7 @@ static inline int atomic_add_return(int i, atomic_t *v) int result; asm volatile("// atomic_add_return\n" -"1: ldaxr %w0, %2\n" +"1: ldxr %w0, %2\n" " add %w0, %w0, %w3\n" " stlxr %w1, %w0, %2\n" " cbnz %w1, 1b" @@ -72,6 +72,7 @@ static inline int atomic_add_return(int i, atomic_t *v) : "Ir" (i) : "cc", "memory"); + smp_mb(); return result; } @@ -96,7 +97,7 @@ static inline int atomic_sub_return(int i, atomic_t *v) int result; asm volatile("// atomic_sub_return\n" -"1: ldaxr %w0, %2\n" +"1: ldxr %w0, %2\n" " sub %w0, %w0, %w3\n" " stlxr %w1, %w0, %2\n" " cbnz %w1, 1b" @@ -104,6 +105,7 @@ static inline int atomic_sub_return(int i, atomic_t *v) : "Ir" (i) : "cc", "memory"); + smp_mb(); return result; } @@ -112,17 +114,20 @@ static inline int atomic_cmpxchg(atomic_t *ptr, int old, int new) unsigned long tmp; int oldval; + smp_mb(); + asm volatile("// atomic_cmpxchg\n" -"1: ldaxr %w1, %2\n" +"1: ldxr %w1, %2\n" " cmp %w1, %w3\n" " b.ne 2f\n" -" stlxr %w0, %w4, %2\n" +" stxr %w0, %w4, %2\n" " cbnz %w0, 1b\n" "2:" : "=&r" (tmp), "=&r" (oldval), "+Q" (ptr->counter) : "Ir" (old), "r" (new) : "cc", "memory"); + smp_mb(); return oldval; } @@ -183,7 +188,7 @@ static inline long atomic64_add_return(long i, atomic64_t *v) unsigned long tmp; asm volatile("// atomic64_add_return\n" -"1: ldaxr %0, %2\n" +"1: ldxr %0, %2\n" " add %0, %0, %3\n" " stlxr %w1, %0, %2\n" " cbnz %w1, 1b" @@ -191,6 +196,7 @@ static inline long atomic64_add_return(long i, atomic64_t *v) : "Ir" (i) : "cc", "memory"); + smp_mb(); return result; } @@ -215,7 +221,7 @@ static inline long atomic64_sub_return(long i, atomic64_t *v) unsigned long tmp; asm volatile("// atomic64_sub_return\n" -"1: ldaxr %0, %2\n" +"1: ldxr %0, %2\n" " sub %0, %0, %3\n" " stlxr %w1, %0, %2\n" " cbnz %w1, 1b" @@ -223,6 +229,7 @@ static inline long atomic64_sub_return(long i, atomic64_t *v) : "Ir" (i) : "cc", "memory"); + smp_mb(); return result; } @@ -231,17 +238,20 @@ static inline long atomic64_cmpxchg(atomic64_t *ptr, long old, long new) long oldval; unsigned long res; + smp_mb(); + asm volatile("// atomic64_cmpxchg\n" -"1: ldaxr %1, %2\n" +"1: ldxr %1, %2\n" " cmp %1, %3\n" " b.ne 2f\n" -" stlxr %w0, %4, %2\n" +" stxr %w0, %4, %2\n" " cbnz %w0, 1b\n" "2:" : "=&r" (res), "=&r" (oldval), "+Q" (ptr->counter) : "Ir" (old), "r" (new) : "cc", "memory"); + smp_mb(); return oldval; } @@ -253,11 +263,12 @@ static inline long atomic64_dec_if_positive(atomic64_t *v) unsigned long tmp; asm volatile("// atomic64_dec_if_positive\n" -"1: ldaxr %0, %2\n" +"1: ldxr %0, %2\n" " subs %0, %0, #1\n" " b.mi 2f\n" " stlxr %w1, %0, %2\n" " cbnz %w1, 1b\n" +" dmb ish\n" "2:" : "=&r" (result), "=&r" (tmp), "+Q" (v->counter) : diff --git a/arch/arm64/include/asm/cmpxchg.h b/arch/arm64/include/asm/cmpxchg.h index 56166d7f4a25..189390ce8653 100644 --- a/arch/arm64/include/asm/cmpxchg.h +++ b/arch/arm64/include/asm/cmpxchg.h @@ -29,7 +29,7 @@ static inline unsigned long __xchg(unsigned long x, volatile void *ptr, int size switch (size) { case 1: asm volatile("// __xchg1\n" - "1: ldaxrb %w0, %2\n" + "1: ldxrb %w0, %2\n" " stlxrb %w1, %w3, %2\n" " cbnz %w1, 1b\n" : "=&r" (ret), "=&r" (tmp), "+Q" (*(u8 *)ptr) @@ -38,7 +38,7 @@ static inline unsigned long __xchg(unsigned long x, volatile void *ptr, int size break; case 2: asm volatile("// __xchg2\n" - "1: ldaxrh %w0, %2\n" + "1: ldxrh %w0, %2\n" " stlxrh %w1, %w3, %2\n" " cbnz %w1, 1b\n" : "=&r" (ret), "=&r" (tmp), "+Q" (*(u16 *)ptr) @@ -47,7 +47,7 @@ static inline unsigned long __xchg(unsigned long x, volatile void *ptr, int size break; case 4: asm volatile("// __xchg4\n" - "1: ldaxr %w0, %2\n" + "1: ldxr %w0, %2\n" " stlxr %w1, %w3, %2\n" " cbnz %w1, 1b\n" : "=&r" (ret), "=&r" (tmp), "+Q" (*(u32 *)ptr) @@ -56,7 +56,7 @@ static inline unsigned long __xchg(unsigned long x, volatile void *ptr, int size break; case 8: asm volatile("// __xchg8\n" - "1: ldaxr %0, %2\n" + "1: ldxr %0, %2\n" " stlxr %w1, %3, %2\n" " cbnz %w1, 1b\n" : "=&r" (ret), "=&r" (tmp), "+Q" (*(u64 *)ptr) @@ -67,6 +67,7 @@ static inline unsigned long __xchg(unsigned long x, volatile void *ptr, int size BUILD_BUG(); } + smp_mb(); return ret; } diff --git a/arch/arm64/include/asm/futex.h b/arch/arm64/include/asm/futex.h index 78cc3aba5d69..572193d0005d 100644 --- a/arch/arm64/include/asm/futex.h +++ b/arch/arm64/include/asm/futex.h @@ -24,10 +24,11 @@ #define __futex_atomic_op(insn, ret, oldval, uaddr, tmp, oparg) \ asm volatile( \ -"1: ldaxr %w1, %2\n" \ +"1: ldxr %w1, %2\n" \ insn "\n" \ "2: stlxr %w3, %w0, %2\n" \ " cbnz %w3, 1b\n" \ +" dmb ish\n" \ "3:\n" \ " .pushsection .fixup,\"ax\"\n" \ " .align 2\n" \ @@ -111,11 +112,12 @@ futex_atomic_cmpxchg_inatomic(u32 *uval, u32 __user *uaddr, return -EFAULT; asm volatile("// futex_atomic_cmpxchg_inatomic\n" -"1: ldaxr %w1, %2\n" +"1: ldxr %w1, %2\n" " sub %w3, %w1, %w4\n" " cbnz %w3, 3f\n" "2: stlxr %w3, %w5, %2\n" " cbnz %w3, 1b\n" +" dmb ish\n" "3:\n" " .pushsection .fixup,\"ax\"\n" "4: mov %w0, %w6\n" diff --git a/arch/arm64/kernel/kuser32.S b/arch/arm64/kernel/kuser32.S index 63c48ffdf230..7787208e8cc6 100644 --- a/arch/arm64/kernel/kuser32.S +++ b/arch/arm64/kernel/kuser32.S @@ -38,12 +38,13 @@ __kuser_cmpxchg64: // 0xffff0f60 .inst 0xe92d00f0 // push {r4, r5, r6, r7} .inst 0xe1c040d0 // ldrd r4, r5, [r0] .inst 0xe1c160d0 // ldrd r6, r7, [r1] - .inst 0xe1b20e9f // 1: ldaexd r0, r1, [r2] + .inst 0xe1b20f9f // 1: ldrexd r0, r1, [r2] .inst 0xe0303004 // eors r3, r0, r4 .inst 0x00313005 // eoreqs r3, r1, r5 .inst 0x01a23e96 // stlexdeq r3, r6, [r2] .inst 0x03330001 // teqeq r3, #1 .inst 0x0afffff9 // beq 1b + .inst 0xf57ff05b // dmb ish .inst 0xe2730000 // rsbs r0, r3, #0 .inst 0xe8bd00f0 // pop {r4, r5, r6, r7} .inst 0xe12fff1e // bx lr @@ -55,11 +56,12 @@ __kuser_memory_barrier: // 0xffff0fa0 .align 5 __kuser_cmpxchg: // 0xffff0fc0 - .inst 0xe1923e9f // 1: ldaex r3, [r2] + .inst 0xe1923f9f // 1: ldrex r3, [r2] .inst 0xe0533000 // subs r3, r3, r0 .inst 0x01823e91 // stlexeq r3, r1, [r2] .inst 0x03330001 // teqeq r3, #1 .inst 0x0afffffa // beq 1b + .inst 0xf57ff05b // dmb ish .inst 0xe2730000 // rsbs r0, r3, #0 .inst 0xe12fff1e // bx lr diff --git a/arch/arm64/lib/bitops.S b/arch/arm64/lib/bitops.S index e5db797790d3..7dac371cc9a2 100644 --- a/arch/arm64/lib/bitops.S +++ b/arch/arm64/lib/bitops.S @@ -46,11 +46,12 @@ ENTRY( \name ) mov x2, #1 add x1, x1, x0, lsr #3 // Get word offset lsl x4, x2, x3 // Create mask -1: ldaxr x2, [x1] +1: ldxr x2, [x1] lsr x0, x2, x3 // Save old value of bit \instr x2, x2, x4 // toggle bit stlxr w5, x2, [x1] cbnz w5, 1b + dmb ish and x0, x0, #1 3: ret ENDPROC(\name ) From 95c4189689f92fba7ecf9097173404d4928c6e9b Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Tue, 4 Feb 2014 12:29:13 +0000 Subject: [PATCH 373/596] arm64: asm: remove redundant "cc" clobbers cbnz/tbnz don't update the condition flags, so remove the "cc" clobbers from inline asm blocks that only use these instructions to implement conditional branches. Signed-off-by: Will Deacon Signed-off-by: Catalin Marinas --- arch/arm64/include/asm/atomic.h | 24 ++++++++++-------------- arch/arm64/include/asm/cmpxchg.h | 8 ++++---- arch/arm64/include/asm/futex.h | 4 ++-- arch/arm64/include/asm/spinlock.h | 10 +++++----- 4 files changed, 21 insertions(+), 25 deletions(-) diff --git a/arch/arm64/include/asm/atomic.h b/arch/arm64/include/asm/atomic.h index e32893e005d4..0237f0867e37 100644 --- a/arch/arm64/include/asm/atomic.h +++ b/arch/arm64/include/asm/atomic.h @@ -54,8 +54,7 @@ static inline void atomic_add(int i, atomic_t *v) " stxr %w1, %w0, %2\n" " cbnz %w1, 1b" : "=&r" (result), "=&r" (tmp), "+Q" (v->counter) - : "Ir" (i) - : "cc"); + : "Ir" (i)); } static inline int atomic_add_return(int i, atomic_t *v) @@ -70,7 +69,7 @@ static inline int atomic_add_return(int i, atomic_t *v) " cbnz %w1, 1b" : "=&r" (result), "=&r" (tmp), "+Q" (v->counter) : "Ir" (i) - : "cc", "memory"); + : "memory"); smp_mb(); return result; @@ -87,8 +86,7 @@ static inline void atomic_sub(int i, atomic_t *v) " stxr %w1, %w0, %2\n" " cbnz %w1, 1b" : "=&r" (result), "=&r" (tmp), "+Q" (v->counter) - : "Ir" (i) - : "cc"); + : "Ir" (i)); } static inline int atomic_sub_return(int i, atomic_t *v) @@ -103,7 +101,7 @@ static inline int atomic_sub_return(int i, atomic_t *v) " cbnz %w1, 1b" : "=&r" (result), "=&r" (tmp), "+Q" (v->counter) : "Ir" (i) - : "cc", "memory"); + : "memory"); smp_mb(); return result; @@ -125,7 +123,7 @@ static inline int atomic_cmpxchg(atomic_t *ptr, int old, int new) "2:" : "=&r" (tmp), "=&r" (oldval), "+Q" (ptr->counter) : "Ir" (old), "r" (new) - : "cc", "memory"); + : "cc"); smp_mb(); return oldval; @@ -178,8 +176,7 @@ static inline void atomic64_add(u64 i, atomic64_t *v) " stxr %w1, %0, %2\n" " cbnz %w1, 1b" : "=&r" (result), "=&r" (tmp), "+Q" (v->counter) - : "Ir" (i) - : "cc"); + : "Ir" (i)); } static inline long atomic64_add_return(long i, atomic64_t *v) @@ -194,7 +191,7 @@ static inline long atomic64_add_return(long i, atomic64_t *v) " cbnz %w1, 1b" : "=&r" (result), "=&r" (tmp), "+Q" (v->counter) : "Ir" (i) - : "cc", "memory"); + : "memory"); smp_mb(); return result; @@ -211,8 +208,7 @@ static inline void atomic64_sub(u64 i, atomic64_t *v) " stxr %w1, %0, %2\n" " cbnz %w1, 1b" : "=&r" (result), "=&r" (tmp), "+Q" (v->counter) - : "Ir" (i) - : "cc"); + : "Ir" (i)); } static inline long atomic64_sub_return(long i, atomic64_t *v) @@ -227,7 +223,7 @@ static inline long atomic64_sub_return(long i, atomic64_t *v) " cbnz %w1, 1b" : "=&r" (result), "=&r" (tmp), "+Q" (v->counter) : "Ir" (i) - : "cc", "memory"); + : "memory"); smp_mb(); return result; @@ -249,7 +245,7 @@ static inline long atomic64_cmpxchg(atomic64_t *ptr, long old, long new) "2:" : "=&r" (res), "=&r" (oldval), "+Q" (ptr->counter) : "Ir" (old), "r" (new) - : "cc", "memory"); + : "cc"); smp_mb(); return oldval; diff --git a/arch/arm64/include/asm/cmpxchg.h b/arch/arm64/include/asm/cmpxchg.h index 189390ce8653..57c0fa7bf711 100644 --- a/arch/arm64/include/asm/cmpxchg.h +++ b/arch/arm64/include/asm/cmpxchg.h @@ -34,7 +34,7 @@ static inline unsigned long __xchg(unsigned long x, volatile void *ptr, int size " cbnz %w1, 1b\n" : "=&r" (ret), "=&r" (tmp), "+Q" (*(u8 *)ptr) : "r" (x) - : "cc", "memory"); + : "memory"); break; case 2: asm volatile("// __xchg2\n" @@ -43,7 +43,7 @@ static inline unsigned long __xchg(unsigned long x, volatile void *ptr, int size " cbnz %w1, 1b\n" : "=&r" (ret), "=&r" (tmp), "+Q" (*(u16 *)ptr) : "r" (x) - : "cc", "memory"); + : "memory"); break; case 4: asm volatile("// __xchg4\n" @@ -52,7 +52,7 @@ static inline unsigned long __xchg(unsigned long x, volatile void *ptr, int size " cbnz %w1, 1b\n" : "=&r" (ret), "=&r" (tmp), "+Q" (*(u32 *)ptr) : "r" (x) - : "cc", "memory"); + : "memory"); break; case 8: asm volatile("// __xchg8\n" @@ -61,7 +61,7 @@ static inline unsigned long __xchg(unsigned long x, volatile void *ptr, int size " cbnz %w1, 1b\n" : "=&r" (ret), "=&r" (tmp), "+Q" (*(u64 *)ptr) : "r" (x) - : "cc", "memory"); + : "memory"); break; default: BUILD_BUG(); diff --git a/arch/arm64/include/asm/futex.h b/arch/arm64/include/asm/futex.h index 572193d0005d..5f750dc96e0f 100644 --- a/arch/arm64/include/asm/futex.h +++ b/arch/arm64/include/asm/futex.h @@ -41,7 +41,7 @@ " .popsection\n" \ : "=&r" (ret), "=&r" (oldval), "+Q" (*uaddr), "=&r" (tmp) \ : "r" (oparg), "Ir" (-EFAULT) \ - : "cc", "memory") + : "memory") static inline int futex_atomic_op_inuser (int encoded_op, u32 __user *uaddr) @@ -129,7 +129,7 @@ futex_atomic_cmpxchg_inatomic(u32 *uval, u32 __user *uaddr, " .popsection\n" : "+r" (ret), "=&r" (val), "+Q" (*uaddr), "=&r" (tmp) : "r" (oldval), "r" (newval), "Ir" (-EFAULT) - : "cc", "memory"); + : "memory"); *uval = val; return ret; diff --git a/arch/arm64/include/asm/spinlock.h b/arch/arm64/include/asm/spinlock.h index 3d5cf064d7a1..c45b7b1b7197 100644 --- a/arch/arm64/include/asm/spinlock.h +++ b/arch/arm64/include/asm/spinlock.h @@ -132,7 +132,7 @@ static inline void arch_write_lock(arch_rwlock_t *rw) " cbnz %w0, 2b\n" : "=&r" (tmp), "+Q" (rw->lock) : "r" (0x80000000) - : "cc", "memory"); + : "memory"); } static inline int arch_write_trylock(arch_rwlock_t *rw) @@ -146,7 +146,7 @@ static inline int arch_write_trylock(arch_rwlock_t *rw) "1:\n" : "=&r" (tmp), "+Q" (rw->lock) : "r" (0x80000000) - : "cc", "memory"); + : "memory"); return !tmp; } @@ -187,7 +187,7 @@ static inline void arch_read_lock(arch_rwlock_t *rw) " cbnz %w1, 2b\n" : "=&r" (tmp), "=&r" (tmp2), "+Q" (rw->lock) : - : "cc", "memory"); + : "memory"); } static inline void arch_read_unlock(arch_rwlock_t *rw) @@ -201,7 +201,7 @@ static inline void arch_read_unlock(arch_rwlock_t *rw) " cbnz %w1, 1b\n" : "=&r" (tmp), "=&r" (tmp2), "+Q" (rw->lock) : - : "cc", "memory"); + : "memory"); } static inline int arch_read_trylock(arch_rwlock_t *rw) @@ -216,7 +216,7 @@ static inline int arch_read_trylock(arch_rwlock_t *rw) "1:\n" : "=&r" (tmp), "+r" (tmp2), "+Q" (rw->lock) : - : "cc", "memory"); + : "memory"); return !tmp2; } From 6d8c00d58e9e484fdc41aaaf62e5d8364efe375a Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Thu, 9 Jan 2014 18:42:42 +0000 Subject: [PATCH 374/596] netfilter: nf_tables: unininline nft_trace_packet() It makes no sense to inline a rarely used function meant for debugging only that is called a total of five times in the main evaluation loop. Signed-off-by: Patrick McHardy Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_tables_core.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/net/netfilter/nf_tables_core.c b/net/netfilter/nf_tables_core.c index 0d879fcb8763..90998a6ff8b9 100644 --- a/net/netfilter/nf_tables_core.c +++ b/net/netfilter/nf_tables_core.c @@ -103,9 +103,9 @@ static struct nf_loginfo trace_loginfo = { }, }; -static inline void nft_trace_packet(const struct nft_pktinfo *pkt, - const struct nft_chain *chain, - int rulenum, enum nft_trace type) +static void nft_trace_packet(const struct nft_pktinfo *pkt, + const struct nft_chain *chain, + int rulenum, enum nft_trace type) { struct net *net = dev_net(pkt->in ? pkt->in : pkt->out); From d979f3b0a1f0b5499ab85e68cdf02b56852918b6 Mon Sep 17 00:00:00 2001 From: Steve French Date: Sat, 1 Feb 2014 23:27:18 -0600 Subject: [PATCH 375/596] Add protocol specific operation for CIFS xattrs Changeset 666753c3ef8fc88b0ddd5be4865d0aa66428ac35 added protocol operations for get/setxattr to avoid calling cifs operations on smb2/smb3 mounts for xattr operations and this changeset adds the calls to cifs specific protocol operations for xattrs (in order to reenable cifs support for xattrs which was temporarily disabled by the previous changeset. We do not have SMB2/SMB3 worker function for setting xattrs yet so this only enables it for cifs. CCing stable since without these two small changsets (its small coreq 666753c3ef8fc88b0ddd5be4865d0aa66428ac35 is also needed) calling getfattr/setfattr on smb2/smb3 mounts causes problems. Signed-off-by: Steve French Reviewed-by: Shirish Pargaonkar CC: Stable --- fs/cifs/inode.c | 13 +++++++++---- fs/cifs/smb1ops.c | 4 ++++ 2 files changed, 13 insertions(+), 4 deletions(-) diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c index 9cb9679d7357..be58b8fcdb3c 100644 --- a/fs/cifs/inode.c +++ b/fs/cifs/inode.c @@ -527,10 +527,15 @@ static int cifs_sfu_mode(struct cifs_fattr *fattr, const unsigned char *path, return PTR_ERR(tlink); tcon = tlink_tcon(tlink); - rc = CIFSSMBQAllEAs(xid, tcon, path, "SETFILEBITS", - ea_value, 4 /* size of buf */, cifs_sb->local_nls, - cifs_sb->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); + if (tcon->ses->server->ops->query_all_EAs == NULL) { + cifs_put_tlink(tlink); + return -EOPNOTSUPP; + } + + rc = tcon->ses->server->ops->query_all_EAs(xid, tcon, path, + "SETFILEBITS", ea_value, 4 /* size of buf */, + cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); cifs_put_tlink(tlink); if (rc < 0) return (int)rc; diff --git a/fs/cifs/smb1ops.c b/fs/cifs/smb1ops.c index 9ac5bfc9cc56..3e4ff79e3031 100644 --- a/fs/cifs/smb1ops.c +++ b/fs/cifs/smb1ops.c @@ -1067,6 +1067,10 @@ struct smb_version_operations smb1_operations = { .query_mf_symlink = cifs_query_mf_symlink, .create_mf_symlink = cifs_create_mf_symlink, .is_read_op = cifs_is_read_op, +#ifdef CONFIG_CIFS_XATTR + .query_all_EAs = CIFSSMBQAllEAs, + .set_EA = CIFSSMBSetEA, +#endif /* CIFS_XATTR */ }; struct smb_version_values smb1_values = { From 83e3bc23ef9ce7c03b7b4e5d3d790246ea59db3e Mon Sep 17 00:00:00 2001 From: Steve French Date: Sun, 2 Feb 2014 23:31:47 -0600 Subject: [PATCH 376/596] retrieving CIFS ACLs when mounted with SMB2 fails dropping session The get/set ACL xattr support for CIFS ACLs attempts to send old cifs dialect protocol requests even when mounted with SMB2 or later dialects. Sending cifs requests on an smb2 session causes problems - the server drops the session due to the illegal request. This patch makes CIFS ACL operations protocol specific to fix that. Attempting to query/set CIFS ACLs for SMB2 will now return EOPNOTSUPP (until we add worker routines for sending query ACL requests via SMB2) instead of sending invalid (cifs) requests. A separate followon patch will be needed to fix cifs_acl_to_fattr (which takes a cifs specific u16 fid so can't be abstracted to work with SMB2 until that is changed) and will be needed to fix mount problems when "cifsacl" is specified on mount with e.g. vers=2.1 Signed-off-by: Steve French Reviewed-by: Shirish Pargaonkar CC: Stable --- fs/cifs/cifsacl.c | 28 ++++++++++++++++++++++++---- fs/cifs/cifsglob.h | 4 ++++ fs/cifs/smb1ops.c | 4 ++++ fs/cifs/xattr.c | 15 +++++++++++---- 4 files changed, 43 insertions(+), 8 deletions(-) diff --git a/fs/cifs/cifsacl.c b/fs/cifs/cifsacl.c index 8f9b4f710d4a..c819b0bd491a 100644 --- a/fs/cifs/cifsacl.c +++ b/fs/cifs/cifsacl.c @@ -1043,15 +1043,30 @@ id_mode_to_cifs_acl(struct inode *inode, const char *path, __u64 nmode, __u32 secdesclen = 0; struct cifs_ntsd *pntsd = NULL; /* acl obtained from server */ struct cifs_ntsd *pnntsd = NULL; /* modified acl to be sent to server */ + struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); + struct tcon_link *tlink = cifs_sb_tlink(cifs_sb); + struct cifs_tcon *tcon; + + if (IS_ERR(tlink)) + return PTR_ERR(tlink); + tcon = tlink_tcon(tlink); cifs_dbg(NOISY, "set ACL from mode for %s\n", path); /* Get the security descriptor */ - pntsd = get_cifs_acl(CIFS_SB(inode->i_sb), inode, path, &secdesclen); + + if (tcon->ses->server->ops->get_acl == NULL) { + cifs_put_tlink(tlink); + return -EOPNOTSUPP; + } + + pntsd = tcon->ses->server->ops->get_acl(cifs_sb, inode, path, + &secdesclen); if (IS_ERR(pntsd)) { rc = PTR_ERR(pntsd); cifs_dbg(VFS, "%s: error %d getting sec desc\n", __func__, rc); - goto out; + cifs_put_tlink(tlink); + return rc; } /* @@ -1064,6 +1079,7 @@ id_mode_to_cifs_acl(struct inode *inode, const char *path, __u64 nmode, pnntsd = kmalloc(secdesclen, GFP_KERNEL); if (!pnntsd) { kfree(pntsd); + cifs_put_tlink(tlink); return -ENOMEM; } @@ -1072,14 +1088,18 @@ id_mode_to_cifs_acl(struct inode *inode, const char *path, __u64 nmode, cifs_dbg(NOISY, "build_sec_desc rc: %d\n", rc); + if (tcon->ses->server->ops->set_acl == NULL) + rc = -EOPNOTSUPP; + if (!rc) { /* Set the security descriptor */ - rc = set_cifs_acl(pnntsd, secdesclen, inode, path, aclflag); + rc = tcon->ses->server->ops->set_acl(pnntsd, secdesclen, inode, + path, aclflag); cifs_dbg(NOISY, "set_cifs_acl rc: %d\n", rc); } + cifs_put_tlink(tlink); kfree(pnntsd); kfree(pntsd); -out: return rc; } diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h index a245d1809ed8..615e35ab7ef6 100644 --- a/fs/cifs/cifsglob.h +++ b/fs/cifs/cifsglob.h @@ -395,6 +395,10 @@ struct smb_version_operations { int (*set_EA)(const unsigned int, struct cifs_tcon *, const char *, const char *, const void *, const __u16, const struct nls_table *, int); + struct cifs_ntsd * (*get_acl)(struct cifs_sb_info *, struct inode *, + const char *, u32 *); + int (*set_acl)(struct cifs_ntsd *, __u32, struct inode *, const char *, + int); }; struct smb_version_values { diff --git a/fs/cifs/smb1ops.c b/fs/cifs/smb1ops.c index 3e4ff79e3031..bfd66d84831e 100644 --- a/fs/cifs/smb1ops.c +++ b/fs/cifs/smb1ops.c @@ -1071,6 +1071,10 @@ struct smb_version_operations smb1_operations = { .query_all_EAs = CIFSSMBQAllEAs, .set_EA = CIFSSMBSetEA, #endif /* CIFS_XATTR */ +#ifdef CONFIG_CIFS_ACL + .get_acl = get_cifs_acl, + .set_acl = set_cifs_acl, +#endif /* CIFS_ACL */ }; struct smb_version_values smb1_values = { diff --git a/fs/cifs/xattr.c b/fs/cifs/xattr.c index 95c43bb20335..5ac836a86b18 100644 --- a/fs/cifs/xattr.c +++ b/fs/cifs/xattr.c @@ -176,8 +176,12 @@ int cifs_setxattr(struct dentry *direntry, const char *ea_name, rc = -ENOMEM; } else { memcpy(pacl, ea_value, value_size); - rc = set_cifs_acl(pacl, value_size, - direntry->d_inode, full_path, CIFS_ACL_DACL); + if (pTcon->ses->server->ops->set_acl) + rc = pTcon->ses->server->ops->set_acl(pacl, + value_size, direntry->d_inode, + full_path, CIFS_ACL_DACL); + else + rc = -EOPNOTSUPP; if (rc == 0) /* force revalidate of the inode */ CIFS_I(direntry->d_inode)->time = 0; kfree(pacl); @@ -323,8 +327,11 @@ ssize_t cifs_getxattr(struct dentry *direntry, const char *ea_name, u32 acllen; struct cifs_ntsd *pacl; - pacl = get_cifs_acl(cifs_sb, direntry->d_inode, - full_path, &acllen); + if (pTcon->ses->server->ops->get_acl == NULL) + goto get_ea_exit; /* rc already EOPNOTSUPP */ + + pacl = pTcon->ses->server->ops->get_acl(cifs_sb, + direntry->d_inode, full_path, &acllen); if (IS_ERR(pacl)) { rc = PTR_ERR(pacl); cifs_dbg(VFS, "%s: error %zd getting sec desc\n", From 55834a773fe343912b705bef8114ec93fd337188 Mon Sep 17 00:00:00 2001 From: Mark Rutland Date: Fri, 7 Feb 2014 17:12:45 +0000 Subject: [PATCH 377/596] arm64: defconfig: Expand default enabled features FPGA implementations of the Cortex-A57 and Cortex-A53 are now available in the form of the SMM-A57 and SMM-A53 Soft Macrocell Models (SMMs) for Versatile Express. As these attach to a Motherboard Express V2M-P1 it would be useful to have support for some V2M-P1 peripherals enabled by default. Additionally a couple of of features have been introduced since the last defconfig update (CMA, jump labels) that would be good to have enabled by default to ensure they are build and boot tested. This patch updates the arm64 defconfig to enable support for these devices and features. The arm64 Kconfig is modified to select HAVE_PATA_PLATFORM, which is required to enable support for the CompactFlash controller on the V2M-P1. A few options which don't need to appear in defconfig are trimmed: * BLK_DEV - selected by default * EXPERIMENTAL - otherwise gone from the kernel * MII - selected by drivers which require it * USB_SUPPORT - selected by default Signed-off-by: Mark Rutland Signed-off-by: Catalin Marinas --- arch/arm64/Kconfig | 1 + arch/arm64/configs/defconfig | 18 ++++++++++++++---- 2 files changed, 15 insertions(+), 4 deletions(-) diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig index dd4327f09ba4..27bbcfc7202a 100644 --- a/arch/arm64/Kconfig +++ b/arch/arm64/Kconfig @@ -36,6 +36,7 @@ config ARM64 select HAVE_GENERIC_DMA_COHERENT select HAVE_HW_BREAKPOINT if PERF_EVENTS select HAVE_MEMBLOCK + select HAVE_PATA_PLATFORM select HAVE_PERF_EVENTS select IRQ_DOMAIN select MODULES_USE_ELF_RELA diff --git a/arch/arm64/configs/defconfig b/arch/arm64/configs/defconfig index 84139be62ae6..7959dd0ca5d5 100644 --- a/arch/arm64/configs/defconfig +++ b/arch/arm64/configs/defconfig @@ -1,4 +1,3 @@ -CONFIG_EXPERIMENTAL=y # CONFIG_LOCALVERSION_AUTO is not set # CONFIG_SWAP is not set CONFIG_SYSVIPC=y @@ -19,6 +18,7 @@ CONFIG_BLK_DEV_INITRD=y CONFIG_KALLSYMS_ALL=y # CONFIG_COMPAT_BRK is not set CONFIG_PROFILING=y +CONFIG_JUMP_LABEL=y CONFIG_MODULES=y CONFIG_MODULE_UNLOAD=y # CONFIG_BLK_DEV_BSG is not set @@ -27,6 +27,7 @@ CONFIG_ARCH_VEXPRESS=y CONFIG_ARCH_XGENE=y CONFIG_SMP=y CONFIG_PREEMPT=y +CONFIG_CMA=y CONFIG_CMDLINE="console=ttyAMA0" # CONFIG_CORE_DUMP_DEFAULT_ELF_HEADERS is not set CONFIG_COMPAT=y @@ -42,14 +43,17 @@ CONFIG_IP_PNP_BOOTP=y # CONFIG_WIRELESS is not set CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug" CONFIG_DEVTMPFS=y -CONFIG_BLK_DEV=y +CONFIG_DMA_CMA=y CONFIG_SCSI=y # CONFIG_SCSI_PROC_FS is not set CONFIG_BLK_DEV_SD=y # CONFIG_SCSI_LOWLEVEL is not set +CONFIG_ATA=y +CONFIG_PATA_PLATFORM=y +CONFIG_PATA_OF_PLATFORM=y CONFIG_NETDEVICES=y -CONFIG_MII=y CONFIG_SMC91X=y +CONFIG_SMSC911X=y # CONFIG_WLAN is not set CONFIG_INPUT_EVDEV=y # CONFIG_SERIO_I8042 is not set @@ -62,13 +66,19 @@ CONFIG_SERIAL_AMBA_PL011=y CONFIG_SERIAL_AMBA_PL011_CONSOLE=y # CONFIG_HW_RANDOM is not set # CONFIG_HWMON is not set +CONFIG_REGULATOR=y +CONFIG_REGULATOR_FIXED_VOLTAGE=y CONFIG_FB=y # CONFIG_VGA_CONSOLE is not set CONFIG_FRAMEBUFFER_CONSOLE=y CONFIG_LOGO=y # CONFIG_LOGO_LINUX_MONO is not set # CONFIG_LOGO_LINUX_VGA16 is not set -# CONFIG_USB_SUPPORT is not set +CONFIG_USB=y +CONFIG_USB_ISP1760_HCD=y +CONFIG_USB_STORAGE=y +CONFIG_MMC=y +CONFIG_MMC_ARMMMCI=y # CONFIG_IOMMU_SUPPORT is not set CONFIG_EXT2_FS=y CONFIG_EXT3_FS=y From 2ed22e3c3bec5b92b9aba4afdef0cc5e6d859a11 Mon Sep 17 00:00:00 2001 From: Matt Rushton Date: Tue, 4 Feb 2014 11:26:12 +0100 Subject: [PATCH 378/596] xen-blkback: fix memory leak when persistent grants are used MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Currently shrink_free_pagepool() is called before the pages used for persistent grants are released via free_persistent_gnts(). This results in a memory leak when a VBD that uses persistent grants is torn down. Cc: Konrad Rzeszutek Wilk Cc: "Roger Pau Monné" Cc: Ian Campbell Reviewed-by: David Vrabel Cc: linux-kernel@vger.kernel.org Cc: xen-devel@lists.xen.org Cc: Anthony Liguori Signed-off-by: Matt Rushton Signed-off-by: Matt Wilson Signed-off-by: Konrad Rzeszutek Wilk --- drivers/block/xen-blkback/blkback.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/block/xen-blkback/blkback.c b/drivers/block/xen-blkback/blkback.c index 6620b73d0490..30ef7b390df5 100644 --- a/drivers/block/xen-blkback/blkback.c +++ b/drivers/block/xen-blkback/blkback.c @@ -625,9 +625,6 @@ int xen_blkif_schedule(void *arg) print_stats(blkif); } - /* Since we are shutting down remove all pages from the buffer */ - shrink_free_pagepool(blkif, 0 /* All */); - /* Free all persistent grant pages */ if (!RB_EMPTY_ROOT(&blkif->persistent_gnts)) free_persistent_gnts(blkif, &blkif->persistent_gnts, @@ -636,6 +633,9 @@ int xen_blkif_schedule(void *arg) BUG_ON(!RB_EMPTY_ROOT(&blkif->persistent_gnts)); blkif->persistent_gnt_c = 0; + /* Since we are shutting down remove all pages from the buffer */ + shrink_free_pagepool(blkif, 0 /* All */); + if (log_stats) print_stats(blkif); From ef753411339eae46b9a3151906901f8bfd12b0f1 Mon Sep 17 00:00:00 2001 From: Roger Pau Monne Date: Tue, 4 Feb 2014 11:26:13 +0100 Subject: [PATCH 379/596] xen-blkback: fix memory leaks MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit I've at least identified two possible memory leaks in blkback, both related to the shutdown path of a VBD: - blkback doesn't wait for any pending purge work to finish before cleaning the list of free_pages. The purge work will call put_free_pages and thus we might end up with pages being added to the free_pages list after we have emptied it. Fix this by making sure there's no pending purge work before exiting xen_blkif_schedule, and moving the free_page cleanup code to xen_blkif_free. - blkback doesn't wait for pending requests to end before cleaning persistent grants and the list of free_pages. Again this can add pages to the free_pages list or persistent grants to the persistent_gnts red-black tree. Fixed by moving the persistent grants and free_pages cleanup code to xen_blkif_free. Also, add some checks in xen_blkif_free to make sure we are cleaning everything. Signed-off-by: Roger Pau Monné Cc: Konrad Rzeszutek Wilk Reviewed-by: David Vrabel Cc: Boris Ostrovsky Tested-by: Matt Rushton Reviewed-by: Matt Rushton Cc: Matt Wilson Cc: Ian Campbell Signed-off-by: Konrad Rzeszutek Wilk --- drivers/block/xen-blkback/blkback.c | 27 ++++++++++++++++++--------- drivers/block/xen-blkback/common.h | 1 + drivers/block/xen-blkback/xenbus.c | 12 ++++++++++++ 3 files changed, 31 insertions(+), 9 deletions(-) diff --git a/drivers/block/xen-blkback/blkback.c b/drivers/block/xen-blkback/blkback.c index 30ef7b390df5..dcfe49fd3cb4 100644 --- a/drivers/block/xen-blkback/blkback.c +++ b/drivers/block/xen-blkback/blkback.c @@ -375,7 +375,7 @@ static void purge_persistent_gnt(struct xen_blkif *blkif) pr_debug(DRV_PFX "Going to purge %u persistent grants\n", num_clean); - INIT_LIST_HEAD(&blkif->persistent_purge_list); + BUG_ON(!list_empty(&blkif->persistent_purge_list)); root = &blkif->persistent_gnts; purge_list: foreach_grant_safe(persistent_gnt, n, root, node) { @@ -625,6 +625,23 @@ int xen_blkif_schedule(void *arg) print_stats(blkif); } + /* Drain pending purge work */ + flush_work(&blkif->persistent_purge_work); + + if (log_stats) + print_stats(blkif); + + blkif->xenblkd = NULL; + xen_blkif_put(blkif); + + return 0; +} + +/* + * Remove persistent grants and empty the pool of free pages + */ +void xen_blkbk_free_caches(struct xen_blkif *blkif) +{ /* Free all persistent grant pages */ if (!RB_EMPTY_ROOT(&blkif->persistent_gnts)) free_persistent_gnts(blkif, &blkif->persistent_gnts, @@ -635,14 +652,6 @@ int xen_blkif_schedule(void *arg) /* Since we are shutting down remove all pages from the buffer */ shrink_free_pagepool(blkif, 0 /* All */); - - if (log_stats) - print_stats(blkif); - - blkif->xenblkd = NULL; - xen_blkif_put(blkif); - - return 0; } /* diff --git a/drivers/block/xen-blkback/common.h b/drivers/block/xen-blkback/common.h index 8d8807563d99..f733d7627120 100644 --- a/drivers/block/xen-blkback/common.h +++ b/drivers/block/xen-blkback/common.h @@ -376,6 +376,7 @@ int xen_blkif_xenbus_init(void); irqreturn_t xen_blkif_be_int(int irq, void *dev_id); int xen_blkif_schedule(void *arg); int xen_blkif_purge_persistent(void *arg); +void xen_blkbk_free_caches(struct xen_blkif *blkif); int xen_blkbk_flush_diskcache(struct xenbus_transaction xbt, struct backend_info *be, int state); diff --git a/drivers/block/xen-blkback/xenbus.c b/drivers/block/xen-blkback/xenbus.c index c2014a0aa206..8afef67fecdd 100644 --- a/drivers/block/xen-blkback/xenbus.c +++ b/drivers/block/xen-blkback/xenbus.c @@ -125,6 +125,7 @@ static struct xen_blkif *xen_blkif_alloc(domid_t domid) blkif->persistent_gnts.rb_node = NULL; spin_lock_init(&blkif->free_pages_lock); INIT_LIST_HEAD(&blkif->free_pages); + INIT_LIST_HEAD(&blkif->persistent_purge_list); blkif->free_pages_num = 0; atomic_set(&blkif->persistent_gnt_in_use, 0); @@ -259,6 +260,17 @@ static void xen_blkif_free(struct xen_blkif *blkif) if (!atomic_dec_and_test(&blkif->refcnt)) BUG(); + /* Remove all persistent grants and the cache of ballooned pages. */ + xen_blkbk_free_caches(blkif); + + /* Make sure everything is drained before shutting down */ + BUG_ON(blkif->persistent_gnt_c != 0); + BUG_ON(atomic_read(&blkif->persistent_gnt_in_use) != 0); + BUG_ON(blkif->free_pages_num != 0); + BUG_ON(!list_empty(&blkif->persistent_purge_list)); + BUG_ON(!list_empty(&blkif->free_pages)); + BUG_ON(!RB_EMPTY_ROOT(&blkif->persistent_gnts)); + /* Check that there is no request in use */ list_for_each_entry_safe(req, n, &blkif->pending_free, free_list) { list_del(&req->free_list); From c05f3e3c85df1d89673e00cee7ece5ae4eb4c6ec Mon Sep 17 00:00:00 2001 From: Roger Pau Monne Date: Tue, 4 Feb 2014 11:26:14 +0100 Subject: [PATCH 380/596] xen-blkback: fix shutdown race MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Introduce a new variable to keep track of the number of in-flight requests. We need to make sure that when xen_blkif_put is called the request has already been freed and we can safely free xen_blkif, which was not the case before. Signed-off-by: Roger Pau Monné Cc: Konrad Rzeszutek Wilk Cc: David Vrabel Reviewed-by: Boris Ostrovsky Tested-by: Matt Rushton Reviewed-by: Matt Rushton Cc: Matt Wilson Cc: Ian Campbell Signed-off-by: Konrad Rzeszutek Wilk --- drivers/block/xen-blkback/blkback.c | 32 ++++++++++++++++++++--------- drivers/block/xen-blkback/common.h | 1 + drivers/block/xen-blkback/xenbus.c | 1 + 3 files changed, 24 insertions(+), 10 deletions(-) diff --git a/drivers/block/xen-blkback/blkback.c b/drivers/block/xen-blkback/blkback.c index dcfe49fd3cb4..394fa2eabf87 100644 --- a/drivers/block/xen-blkback/blkback.c +++ b/drivers/block/xen-blkback/blkback.c @@ -943,9 +943,7 @@ static void xen_blk_drain_io(struct xen_blkif *blkif) { atomic_set(&blkif->drain, 1); do { - /* The initial value is one, and one refcnt taken at the - * start of the xen_blkif_schedule thread. */ - if (atomic_read(&blkif->refcnt) <= 2) + if (atomic_read(&blkif->inflight) == 0) break; wait_for_completion_interruptible_timeout( &blkif->drain_complete, HZ); @@ -985,17 +983,30 @@ static void __end_block_io_op(struct pending_req *pending_req, int error) * the proper response on the ring. */ if (atomic_dec_and_test(&pending_req->pendcnt)) { - xen_blkbk_unmap(pending_req->blkif, + struct xen_blkif *blkif = pending_req->blkif; + + xen_blkbk_unmap(blkif, pending_req->segments, pending_req->nr_pages); - make_response(pending_req->blkif, pending_req->id, + make_response(blkif, pending_req->id, pending_req->operation, pending_req->status); - xen_blkif_put(pending_req->blkif); - if (atomic_read(&pending_req->blkif->refcnt) <= 2) { - if (atomic_read(&pending_req->blkif->drain)) - complete(&pending_req->blkif->drain_complete); + free_req(blkif, pending_req); + /* + * Make sure the request is freed before releasing blkif, + * or there could be a race between free_req and the + * cleanup done in xen_blkif_free during shutdown. + * + * NB: The fact that we might try to wake up pending_free_wq + * before drain_complete (in case there's a drain going on) + * it's not a problem with our current implementation + * because we can assure there's no thread waiting on + * pending_free_wq if there's a drain going on, but it has + * to be taken into account if the current model is changed. + */ + if (atomic_dec_and_test(&blkif->inflight) && atomic_read(&blkif->drain)) { + complete(&blkif->drain_complete); } - free_req(pending_req->blkif, pending_req); + xen_blkif_put(blkif); } } @@ -1249,6 +1260,7 @@ static int dispatch_rw_block_io(struct xen_blkif *blkif, * below (in "!bio") if we are handling a BLKIF_OP_DISCARD. */ xen_blkif_get(blkif); + atomic_inc(&blkif->inflight); for (i = 0; i < nseg; i++) { while ((bio == NULL) || diff --git a/drivers/block/xen-blkback/common.h b/drivers/block/xen-blkback/common.h index f733d7627120..e40326a7f707 100644 --- a/drivers/block/xen-blkback/common.h +++ b/drivers/block/xen-blkback/common.h @@ -278,6 +278,7 @@ struct xen_blkif { /* for barrier (drain) requests */ struct completion drain_complete; atomic_t drain; + atomic_t inflight; /* One thread per one blkif. */ struct task_struct *xenblkd; unsigned int waiting_reqs; diff --git a/drivers/block/xen-blkback/xenbus.c b/drivers/block/xen-blkback/xenbus.c index 8afef67fecdd..84973c6a856a 100644 --- a/drivers/block/xen-blkback/xenbus.c +++ b/drivers/block/xen-blkback/xenbus.c @@ -128,6 +128,7 @@ static struct xen_blkif *xen_blkif_alloc(domid_t domid) INIT_LIST_HEAD(&blkif->persistent_purge_list); blkif->free_pages_num = 0; atomic_set(&blkif->persistent_gnt_in_use, 0); + atomic_set(&blkif->inflight, 0); INIT_LIST_HEAD(&blkif->pending_free); From 80bfa2f6e2e81049fc6cd3bfaeedcb64db3a9ba6 Mon Sep 17 00:00:00 2001 From: Roger Pau Monne Date: Tue, 4 Feb 2014 11:26:15 +0100 Subject: [PATCH 381/596] xen-blkif: drop struct blkif_request_segment_aligned MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This was wrongly introduced in commit 402b27f9, the only difference between blkif_request_segment_aligned and blkif_request_segment is that the former has a named padding, while both share the same memory layout. Also correct a few minor glitches in the description, including for it to no longer assume PAGE_SIZE == 4096. Signed-off-by: Roger Pau Monné [Description fix by Jan Beulich] Signed-off-by: Jan Beulich Reported-by: Jan Beulich Cc: Konrad Rzeszutek Wilk Cc: David Vrabel Cc: Boris Ostrovsky Tested-by: Matt Rushton Cc: Matt Wilson Signed-off-by: Konrad Rzeszutek Wilk --- drivers/block/xen-blkback/blkback.c | 2 +- drivers/block/xen-blkback/common.h | 2 +- drivers/block/xen-blkfront.c | 6 ++--- include/xen/interface/io/blkif.h | 34 ++++++++++++----------------- 4 files changed, 19 insertions(+), 25 deletions(-) diff --git a/drivers/block/xen-blkback/blkback.c b/drivers/block/xen-blkback/blkback.c index 394fa2eabf87..e612627ae981 100644 --- a/drivers/block/xen-blkback/blkback.c +++ b/drivers/block/xen-blkback/blkback.c @@ -847,7 +847,7 @@ static int xen_blkbk_parse_indirect(struct blkif_request *req, struct grant_page **pages = pending_req->indirect_pages; struct xen_blkif *blkif = pending_req->blkif; int indirect_grefs, rc, n, nseg, i; - struct blkif_request_segment_aligned *segments = NULL; + struct blkif_request_segment *segments = NULL; nseg = pending_req->nr_pages; indirect_grefs = INDIRECT_PAGES(nseg); diff --git a/drivers/block/xen-blkback/common.h b/drivers/block/xen-blkback/common.h index e40326a7f707..9eb34e24b4fe 100644 --- a/drivers/block/xen-blkback/common.h +++ b/drivers/block/xen-blkback/common.h @@ -57,7 +57,7 @@ #define MAX_INDIRECT_SEGMENTS 256 #define SEGS_PER_INDIRECT_FRAME \ - (PAGE_SIZE/sizeof(struct blkif_request_segment_aligned)) + (PAGE_SIZE/sizeof(struct blkif_request_segment)) #define MAX_INDIRECT_PAGES \ ((MAX_INDIRECT_SEGMENTS + SEGS_PER_INDIRECT_FRAME - 1)/SEGS_PER_INDIRECT_FRAME) #define INDIRECT_PAGES(_segs) \ diff --git a/drivers/block/xen-blkfront.c b/drivers/block/xen-blkfront.c index c4a4c9006288..7d09dfc9735c 100644 --- a/drivers/block/xen-blkfront.c +++ b/drivers/block/xen-blkfront.c @@ -162,7 +162,7 @@ static DEFINE_SPINLOCK(minor_lock); #define DEV_NAME "xvd" /* name in /dev */ #define SEGS_PER_INDIRECT_FRAME \ - (PAGE_SIZE/sizeof(struct blkif_request_segment_aligned)) + (PAGE_SIZE/sizeof(struct blkif_request_segment)) #define INDIRECT_GREFS(_segs) \ ((_segs + SEGS_PER_INDIRECT_FRAME - 1)/SEGS_PER_INDIRECT_FRAME) @@ -393,7 +393,7 @@ static int blkif_queue_request(struct request *req) unsigned long id; unsigned int fsect, lsect; int i, ref, n; - struct blkif_request_segment_aligned *segments = NULL; + struct blkif_request_segment *segments = NULL; /* * Used to store if we are able to queue the request by just using @@ -550,7 +550,7 @@ static int blkif_queue_request(struct request *req) } else { n = i % SEGS_PER_INDIRECT_FRAME; segments[n] = - (struct blkif_request_segment_aligned) { + (struct blkif_request_segment) { .gref = ref, .first_sect = fsect, .last_sect = lsect }; diff --git a/include/xen/interface/io/blkif.h b/include/xen/interface/io/blkif.h index ae665ac59c36..32ec05a6572f 100644 --- a/include/xen/interface/io/blkif.h +++ b/include/xen/interface/io/blkif.h @@ -113,13 +113,13 @@ typedef uint64_t blkif_sector_t; * it's less than the number provided by the backend. The indirect_grefs field * in blkif_request_indirect should be filled by the frontend with the * grant references of the pages that are holding the indirect segments. - * This pages are filled with an array of blkif_request_segment_aligned - * that hold the information about the segments. The number of indirect - * pages to use is determined by the maximum number of segments - * a indirect request contains. Every indirect page can contain a maximum - * of 512 segments (PAGE_SIZE/sizeof(blkif_request_segment_aligned)), - * so to calculate the number of indirect pages to use we have to do - * ceil(indirect_segments/512). + * These pages are filled with an array of blkif_request_segment that hold the + * information about the segments. The number of indirect pages to use is + * determined by the number of segments an indirect request contains. Every + * indirect page can contain a maximum of + * (PAGE_SIZE / sizeof(struct blkif_request_segment)) segments, so to + * calculate the number of indirect pages to use we have to do + * ceil(indirect_segments / (PAGE_SIZE / sizeof(struct blkif_request_segment))). * * If a backend does not recognize BLKIF_OP_INDIRECT, it should *not* * create the "feature-max-indirect-segments" node! @@ -135,13 +135,12 @@ typedef uint64_t blkif_sector_t; #define BLKIF_MAX_INDIRECT_PAGES_PER_REQUEST 8 -struct blkif_request_segment_aligned { - grant_ref_t gref; /* reference to I/O buffer frame */ - /* @first_sect: first sector in frame to transfer (inclusive). */ - /* @last_sect: last sector in frame to transfer (inclusive). */ - uint8_t first_sect, last_sect; - uint16_t _pad; /* padding to make it 8 bytes, so it's cache-aligned */ -} __attribute__((__packed__)); +struct blkif_request_segment { + grant_ref_t gref; /* reference to I/O buffer frame */ + /* @first_sect: first sector in frame to transfer (inclusive). */ + /* @last_sect: last sector in frame to transfer (inclusive). */ + uint8_t first_sect, last_sect; +}; struct blkif_request_rw { uint8_t nr_segments; /* number of segments */ @@ -151,12 +150,7 @@ struct blkif_request_rw { #endif uint64_t id; /* private guest value, echoed in resp */ blkif_sector_t sector_number;/* start sector idx on disk (r/w only) */ - struct blkif_request_segment { - grant_ref_t gref; /* reference to I/O buffer frame */ - /* @first_sect: first sector in frame to transfer (inclusive). */ - /* @last_sect: last sector in frame to transfer (inclusive). */ - uint8_t first_sect, last_sect; - } seg[BLKIF_MAX_SEGMENTS_PER_REQUEST]; + struct blkif_request_segment seg[BLKIF_MAX_SEGMENTS_PER_REQUEST]; } __attribute__((__packed__)); struct blkif_request_discard { From 3661371701e714f0cea4120f6a365340858fb4e4 Mon Sep 17 00:00:00 2001 From: David Vrabel Date: Tue, 4 Feb 2014 18:53:56 +0000 Subject: [PATCH 382/596] xen-blkfront: handle backend CLOSED without CLOSING Backend drivers shouldn't transistion to CLOSED unless the frontend is CLOSED. If a backend does transition to CLOSED too soon then the frontend may not see the CLOSING state and will not properly shutdown. So, treat an unexpected backend CLOSED state the same as CLOSING. Signed-off-by: David Vrabel Acked-by: Konrad Rzeszutek Wilk Cc: stable@vger.kernel.org Signed-off-by: Konrad Rzeszutek Wilk --- drivers/block/xen-blkfront.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/block/xen-blkfront.c b/drivers/block/xen-blkfront.c index 7d09dfc9735c..76e5c0b265b1 100644 --- a/drivers/block/xen-blkfront.c +++ b/drivers/block/xen-blkfront.c @@ -1904,13 +1904,16 @@ static void blkback_changed(struct xenbus_device *dev, case XenbusStateReconfiguring: case XenbusStateReconfigured: case XenbusStateUnknown: - case XenbusStateClosed: break; case XenbusStateConnected: blkfront_connect(info); break; + case XenbusStateClosed: + if (dev->state == XenbusStateClosed) + break; + /* Missed the backend's Closing state -- fallthrough */ case XenbusStateClosing: blkfront_closing(info); break; From 0bbfdfe8d25fcc1d5c2edb6b060fb0c5cf66aff9 Mon Sep 17 00:00:00 2001 From: Ilya Dryomov Date: Fri, 31 Jan 2014 19:33:39 +0200 Subject: [PATCH 383/596] libceph: factor out logic from ceph_osdc_start_request() Factor out logic from ceph_osdc_start_request() into a new helper, __ceph_osdc_start_request(). ceph_osdc_start_request() now amounts to taking locks and calling __ceph_osdc_start_request(). Signed-off-by: Ilya Dryomov Reviewed-by: Sage Weil --- net/ceph/osd_client.c | 62 +++++++++++++++++++++++++++---------------- 1 file changed, 39 insertions(+), 23 deletions(-) diff --git a/net/ceph/osd_client.c b/net/ceph/osd_client.c index 166d4c7ba065..2aa82b6bb305 100644 --- a/net/ceph/osd_client.c +++ b/net/ceph/osd_client.c @@ -1426,6 +1426,40 @@ static void __send_queued(struct ceph_osd_client *osdc) __send_request(osdc, req); } +/* + * Caller should hold map_sem for read and request_mutex. + */ +static int __ceph_osdc_start_request(struct ceph_osd_client *osdc, + struct ceph_osd_request *req, + bool nofail) +{ + int rc; + + __register_request(osdc, req); + req->r_sent = 0; + req->r_got_reply = 0; + rc = __map_request(osdc, req, 0); + if (rc < 0) { + if (nofail) { + dout("osdc_start_request failed map, " + " will retry %lld\n", req->r_tid); + rc = 0; + } else { + __unregister_request(osdc, req); + } + return rc; + } + + if (req->r_osd == NULL) { + dout("send_request %p no up osds in pg\n", req); + ceph_monc_request_next_osdmap(&osdc->client->monc); + } else { + __send_queued(osdc); + } + + return 0; +} + /* * Timeout callback, called every N seconds when 1 or more osd * requests has been active for more than N seconds. When this @@ -2351,34 +2385,16 @@ int ceph_osdc_start_request(struct ceph_osd_client *osdc, struct ceph_osd_request *req, bool nofail) { - int rc = 0; + int rc; down_read(&osdc->map_sem); mutex_lock(&osdc->request_mutex); - __register_request(osdc, req); - req->r_sent = 0; - req->r_got_reply = 0; - rc = __map_request(osdc, req, 0); - if (rc < 0) { - if (nofail) { - dout("osdc_start_request failed map, " - " will retry %lld\n", req->r_tid); - rc = 0; - } else { - __unregister_request(osdc, req); - } - goto out_unlock; - } - if (req->r_osd == NULL) { - dout("send_request %p no up osds in pg\n", req); - ceph_monc_request_next_osdmap(&osdc->client->monc); - } else { - __send_queued(osdc); - } - rc = 0; -out_unlock: + + rc = __ceph_osdc_start_request(osdc, req, nofail); + mutex_unlock(&osdc->request_mutex); up_read(&osdc->map_sem); + return rc; } EXPORT_SYMBOL(ceph_osdc_start_request); From ff513ace9b772e75e337f8e058cc7f12816843fe Mon Sep 17 00:00:00 2001 From: Ilya Dryomov Date: Mon, 3 Feb 2014 13:56:33 +0200 Subject: [PATCH 384/596] libceph: take map_sem for read in handle_reply() Handling redirect replies requires both map_sem and request_mutex. Taking map_sem unconditionally near the top of handle_reply() avoids possible race conditions that arise from releasing request_mutex to be able to acquire map_sem in redirect reply case. (Lock ordering is: map_sem, request_mutex, crush_mutex.) Signed-off-by: Ilya Dryomov Reviewed-by: Sage Weil --- net/ceph/osd_client.c | 17 +++++++++++------ 1 file changed, 11 insertions(+), 6 deletions(-) diff --git a/net/ceph/osd_client.c b/net/ceph/osd_client.c index 2aa82b6bb305..0676f2b199d6 100644 --- a/net/ceph/osd_client.c +++ b/net/ceph/osd_client.c @@ -1687,6 +1687,7 @@ static void handle_reply(struct ceph_osd_client *osdc, struct ceph_msg *msg, osdmap_epoch = ceph_decode_32(&p); /* lookup */ + down_read(&osdc->map_sem); mutex_lock(&osdc->request_mutex); req = __lookup_request(osdc, tid); if (req == NULL) { @@ -1743,7 +1744,6 @@ static void handle_reply(struct ceph_osd_client *osdc, struct ceph_msg *msg, dout("redirect pool %lld\n", redir.oloc.pool); __unregister_request(osdc, req); - mutex_unlock(&osdc->request_mutex); req->r_target_oloc = redir.oloc; /* struct */ @@ -1755,10 +1755,10 @@ static void handle_reply(struct ceph_osd_client *osdc, struct ceph_msg *msg, * successfully. In the future we might want to follow * original request's nofail setting here. */ - err = ceph_osdc_start_request(osdc, req, true); + err = __ceph_osdc_start_request(osdc, req, true); BUG_ON(err); - goto done; + goto out_unlock; } already_completed = req->r_got_reply; @@ -1776,8 +1776,7 @@ static void handle_reply(struct ceph_osd_client *osdc, struct ceph_msg *msg, req->r_got_reply = 1; } else if ((flags & CEPH_OSD_FLAG_ONDISK) == 0) { dout("handle_reply tid %llu dup ack\n", tid); - mutex_unlock(&osdc->request_mutex); - goto done; + goto out_unlock; } dout("handle_reply tid %llu flags %d\n", tid, flags); @@ -1792,6 +1791,7 @@ static void handle_reply(struct ceph_osd_client *osdc, struct ceph_msg *msg, __unregister_request(osdc, req); mutex_unlock(&osdc->request_mutex); + up_read(&osdc->map_sem); if (!already_completed) { if (req->r_unsafe_callback && @@ -1809,10 +1809,14 @@ static void handle_reply(struct ceph_osd_client *osdc, struct ceph_msg *msg, complete_request(req); } -done: +out: dout("req=%p req->r_linger=%d\n", req, req->r_linger); ceph_osdc_put_request(req); return; +out_unlock: + mutex_unlock(&osdc->request_mutex); + up_read(&osdc->map_sem); + goto out; bad_put: req->r_result = -EIO; @@ -1825,6 +1829,7 @@ static void handle_reply(struct ceph_osd_client *osdc, struct ceph_msg *msg, ceph_osdc_put_request(req); bad_mutex: mutex_unlock(&osdc->request_mutex); + up_read(&osdc->map_sem); bad: pr_err("corrupt osd_op_reply got %d %d\n", (int)msg->front.iov_len, le32_to_cpu(msg->hdr.front_len)); From 72a0a36e2854a6eadb4cf2561858f613f9cd4639 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Fri, 7 Feb 2014 10:22:36 -0800 Subject: [PATCH 385/596] blk-mq: support at_head inserations for blk_execute_rq This is neede for proper SG_IO operation as well as various uses of blk_execute_rq from the SCSI midlayer. Signed-off-by: Christoph Hellwig Signed-off-by: Jens Axboe --- block/blk-exec.c | 2 +- block/blk-mq.c | 17 ++++++++++------- include/linux/blk-mq.h | 3 ++- 3 files changed, 13 insertions(+), 9 deletions(-) diff --git a/block/blk-exec.c b/block/blk-exec.c index bbfc072a79c2..c68613bb4c79 100644 --- a/block/blk-exec.c +++ b/block/blk-exec.c @@ -65,7 +65,7 @@ void blk_execute_rq_nowait(struct request_queue *q, struct gendisk *bd_disk, * be resued after dying flag is set */ if (q->mq_ops) { - blk_mq_insert_request(q, rq, true); + blk_mq_insert_request(q, rq, at_head, true); return; } diff --git a/block/blk-mq.c b/block/blk-mq.c index 9072d0ab184f..c9306e3403fe 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -714,13 +714,16 @@ static void blk_mq_work_fn(struct work_struct *work) } static void __blk_mq_insert_request(struct blk_mq_hw_ctx *hctx, - struct request *rq) + struct request *rq, bool at_head) { struct blk_mq_ctx *ctx = rq->mq_ctx; trace_block_rq_insert(hctx->queue, rq); - list_add_tail(&rq->queuelist, &ctx->rq_list); + if (at_head) + list_add(&rq->queuelist, &ctx->rq_list); + else + list_add_tail(&rq->queuelist, &ctx->rq_list); blk_mq_hctx_mark_pending(hctx, ctx); /* @@ -730,7 +733,7 @@ static void __blk_mq_insert_request(struct blk_mq_hw_ctx *hctx, } void blk_mq_insert_request(struct request_queue *q, struct request *rq, - bool run_queue) + bool at_head, bool run_queue) { struct blk_mq_hw_ctx *hctx; struct blk_mq_ctx *ctx, *current_ctx; @@ -749,7 +752,7 @@ void blk_mq_insert_request(struct request_queue *q, struct request *rq, rq->mq_ctx = ctx; } spin_lock(&ctx->lock); - __blk_mq_insert_request(hctx, rq); + __blk_mq_insert_request(hctx, rq, at_head); spin_unlock(&ctx->lock); blk_mq_put_ctx(current_ctx); @@ -781,7 +784,7 @@ void blk_mq_run_request(struct request *rq, bool run_queue, bool async) /* ctx->cpu might be offline */ spin_lock(&ctx->lock); - __blk_mq_insert_request(hctx, rq); + __blk_mq_insert_request(hctx, rq, false); spin_unlock(&ctx->lock); blk_mq_put_ctx(current_ctx); @@ -819,7 +822,7 @@ static void blk_mq_insert_requests(struct request_queue *q, rq = list_first_entry(list, struct request, queuelist); list_del_init(&rq->queuelist); rq->mq_ctx = ctx; - __blk_mq_insert_request(hctx, rq); + __blk_mq_insert_request(hctx, rq, false); } spin_unlock(&ctx->lock); @@ -971,7 +974,7 @@ static void blk_mq_make_request(struct request_queue *q, struct bio *bio) __blk_mq_free_request(hctx, ctx, rq); else { blk_mq_bio_to_request(rq, bio); - __blk_mq_insert_request(hctx, rq); + __blk_mq_insert_request(hctx, rq, false); } spin_unlock(&ctx->lock); diff --git a/include/linux/blk-mq.h b/include/linux/blk-mq.h index 1e8f16f65af4..b7638be58599 100644 --- a/include/linux/blk-mq.h +++ b/include/linux/blk-mq.h @@ -122,7 +122,8 @@ void blk_mq_init_commands(struct request_queue *, void (*init)(void *data, struc void blk_mq_flush_plug_list(struct blk_plug *plug, bool from_schedule); -void blk_mq_insert_request(struct request_queue *, struct request *, bool); +void blk_mq_insert_request(struct request_queue *, struct request *, + bool, bool); void blk_mq_run_queues(struct request_queue *q, bool async); void blk_mq_free_request(struct request *rq); bool blk_mq_can_queue(struct blk_mq_hw_ctx *); From 6f5ba581c0d3ba0a76fe138123c1c2817ffcbeb1 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Fri, 7 Feb 2014 10:22:37 -0800 Subject: [PATCH 386/596] blk-mq: divert __blk_put_request for MQ ops __blk_put_request needs to call into the blk-mq code just like blk_put_request. As we don't have the queue lock in this case both end up calling the same function. Signed-off-by: Christoph Hellwig Signed-off-by: Jens Axboe --- block/blk-core.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/block/blk-core.c b/block/blk-core.c index c00e0bdeab4a..06636f3ad424 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -1278,6 +1278,11 @@ void __blk_put_request(struct request_queue *q, struct request *req) if (unlikely(!q)) return; + if (q->mq_ops) { + blk_mq_free_request(req); + return; + } + blk_pm_put_request(req); elv_completed_request(q, req); From 4f7f418c4835d3ce1b66d00502df41f324d13ec0 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Fri, 7 Feb 2014 10:22:38 -0800 Subject: [PATCH 387/596] blk-mq: handle dma_drain_size Make blk-mq handle the dma_drain_size field the same way as the old request path. Signed-off-by: Christoph Hellwig Signed-off-by: Jens Axboe --- block/blk-mq.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/block/blk-mq.c b/block/blk-mq.c index c9306e3403fe..a99bea455452 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -582,6 +582,16 @@ static void __blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx) list_del_init(&rq->queuelist); blk_mq_start_request(rq); + if (q->dma_drain_size && blk_rq_bytes(rq)) { + /* + * make sure space for the drain appears we + * know we can do this because max_hw_segments + * has been adjusted to be one fewer than the + * device can handle + */ + rq->nr_phys_segments++; + } + /* * Last request in the series. Flag it as such, this * enables drivers to know when IO should be kicked off, From 1be036e9464032362def6b3c13f57bfceefe2dab Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Fri, 7 Feb 2014 10:22:39 -0800 Subject: [PATCH 388/596] blk-mq: initialize sg_reserved_size To behave the same way as the old request path. Signed-off-by: Christoph Hellwig Signed-off-by: Jens Axboe --- block/blk-mq.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/block/blk-mq.c b/block/blk-mq.c index a99bea455452..f1e63c2ece71 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -1387,6 +1387,8 @@ struct request_queue *blk_mq_init_queue(struct blk_mq_reg *reg, q->mq_ops = reg->ops; q->queue_flags |= QUEUE_FLAG_MQ_DEFAULT; + q->sg_reserved_size = INT_MAX; + blk_queue_make_request(q, blk_mq_make_request); blk_queue_rq_timed_out(q, reg->ops->timeout); if (reg->timeout) From 03b56329f9bb5a1cb73d7dc659d529a9a9bf3acc Mon Sep 17 00:00:00 2001 From: Jan Moskyto Matejka Date: Fri, 7 Feb 2014 19:15:11 +0100 Subject: [PATCH 389/596] Modpost: fixed USB alias generation for ranges including 0x9 and 0xA Commit afe2dab4f6 ("USB: add hex/bcd detection to usb modalias generation") changed the routine that generates alias ranges. Before that change, only digits 0-9 were supported; the commit tried to fix the case when the range includes higher values than 0x9. Unfortunately, the commit didn't fix the case when the range includes both 0x9 and 0xA, meaning that the final range must look like [x-9A-y] where x <= 0x9 and y >= 0xA -- instead the [x-9A-x] range was produced. Modprobe doesn't complain as it sees no difference between no-match and bad-pattern results of fnmatch(). Fixing this simple bug to fix the aliases. Also changing the hardcoded beginning of the range to uppercase as all the other letters are also uppercase in the device version numbers. Fortunately, this affects only the dvb-usb-dib0700 module, AFAIK. Signed-off-by: Jan Moskyto Matejka Cc: stable Signed-off-by: Greg Kroah-Hartman --- scripts/mod/file2alias.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/scripts/mod/file2alias.c b/scripts/mod/file2alias.c index 23708636b05c..25e5cb0aaef6 100644 --- a/scripts/mod/file2alias.c +++ b/scripts/mod/file2alias.c @@ -210,8 +210,8 @@ static void do_usb_entry(void *symval, range_lo < 0x9 ? "[%X-9" : "[%X", range_lo); sprintf(alias + strlen(alias), - range_hi > 0xA ? "a-%X]" : "%X]", - range_lo); + range_hi > 0xA ? "A-%X]" : "%X]", + range_hi); } } if (bcdDevice_initial_digits < (sizeof(bcdDevice_lo) * 2 - 1)) From 0ec1d15ec6ed513ab2cc86c67d94761d71228a32 Mon Sep 17 00:00:00 2001 From: Ilya Dryomov Date: Wed, 5 Feb 2014 15:19:55 +0200 Subject: [PATCH 390/596] libceph: do not dereference a NULL bio pointer Commit f38a5181d9f3 ("ceph: Convert to immutable biovecs") introduced a NULL pointer dereference, which broke rbd in -rc1. Fix it. Cc: Kent Overstreet Signed-off-by: Ilya Dryomov Reviewed-by: Sage Weil --- net/ceph/messenger.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/net/ceph/messenger.c b/net/ceph/messenger.c index 0e478a0f4204..30efc5c18622 100644 --- a/net/ceph/messenger.c +++ b/net/ceph/messenger.c @@ -840,9 +840,13 @@ static bool ceph_msg_data_bio_advance(struct ceph_msg_data_cursor *cursor, if (!cursor->bvec_iter.bi_size) { bio = bio->bi_next; - cursor->bvec_iter = bio->bi_iter; + cursor->bio = bio; + if (bio) + cursor->bvec_iter = bio->bi_iter; + else + memset(&cursor->bvec_iter, 0, + sizeof(cursor->bvec_iter)); } - cursor->bio = bio; if (!cursor->last_piece) { BUG_ON(!cursor->resid); From 14ec77f352cb00ab8425ec2af03bd7e529eefe24 Mon Sep 17 00:00:00 2001 From: Nicholas Bellinger Date: Fri, 7 Feb 2014 13:45:39 -0700 Subject: [PATCH 391/596] blk-mq: Add bio_integrity setup to blk_mq_make_request This patch adds the missing bio_integrity_enabled() + bio_integrity_prep() setup into blk_mq_make_request() in order to use DIF protection with scsi-mq. Cc: Martin K. Petersen Signed-off-by: Nicholas Bellinger Signed-off-by: Jens Axboe --- block/blk-mq.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/block/blk-mq.c b/block/blk-mq.c index f1e63c2ece71..cee96234bf58 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -922,6 +922,11 @@ static void blk_mq_make_request(struct request_queue *q, struct bio *bio) blk_queue_bounce(q, &bio); + if (bio_integrity_enabled(bio) && bio_integrity_prep(bio)) { + bio_endio(bio, -EIO); + return; + } + if (use_plug && blk_attempt_plug_merge(q, bio, &request_count)) return; From 087787959ce851d7bbb19f10f6e9241b7f85a3ca Mon Sep 17 00:00:00 2001 From: "Martin K. Petersen" Date: Thu, 6 Feb 2014 15:14:13 -0500 Subject: [PATCH 392/596] block: Fix nr_vecs for inline integrity vectors Commit 9f060e2231ca changed the way we handle allocations for the integrity vectors. When the vectors are inline there is no associated slab and consequently bvec_nr_vecs() returns 0. Ensure that we check against BIP_INLINE_VECS in that case. Reported-by: David Milburn Tested-by: David Milburn Cc: stable@vger.kernel.org # v3.10+ Signed-off-by: Martin K. Petersen Signed-off-by: Jens Axboe --- fs/bio-integrity.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/fs/bio-integrity.c b/fs/bio-integrity.c index 0bad24ddc2e7..bcb4a4d533a2 100644 --- a/fs/bio-integrity.c +++ b/fs/bio-integrity.c @@ -114,6 +114,14 @@ void bio_integrity_free(struct bio *bio) } EXPORT_SYMBOL(bio_integrity_free); +static inline unsigned int bip_integrity_vecs(struct bio_integrity_payload *bip) +{ + if (bip->bip_slab == BIO_POOL_NONE) + return BIP_INLINE_VECS; + + return bvec_nr_vecs(bip->bip_slab); +} + /** * bio_integrity_add_page - Attach integrity metadata * @bio: bio to update @@ -129,7 +137,7 @@ int bio_integrity_add_page(struct bio *bio, struct page *page, struct bio_integrity_payload *bip = bio->bi_integrity; struct bio_vec *iv; - if (bip->bip_vcnt >= bvec_nr_vecs(bip->bip_slab)) { + if (bip->bip_vcnt >= bip_integrity_vecs(bip)) { printk(KERN_ERR "%s: bip_vec full\n", __func__); return 0; } From 5cb8850c9c4a7605f74f5c9c7ecadd0b02e87a25 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 7 Feb 2014 13:53:46 -0700 Subject: [PATCH 393/596] block: Explicitly handle discard/write same segments Immutable biovecs changed the way biovecs are interpreted - drivers no longer use bi_vcnt, they have to go by bi_iter.bi_size (to allow for using part of an existing segment without modifying it). This breaks with discards and write_same bios, since for those bi_size has nothing to do with segments in the biovec. So for now, we need a fairly gross hack - we fortunately know that there will never be more than one segment for the entire request, so we can special case discard/write_same. Signed-off-by: Kent Overstreet Tested-by: Hugh Dickins Signed-off-by: Jens Axboe --- block/blk-merge.c | 91 ++++++++++++++++++++++++++++++++--------------- 1 file changed, 62 insertions(+), 29 deletions(-) diff --git a/block/blk-merge.c b/block/blk-merge.c index 8f8adaa95466..6c583f9c5b65 100644 --- a/block/blk-merge.c +++ b/block/blk-merge.c @@ -21,6 +21,16 @@ static unsigned int __blk_recalc_rq_segments(struct request_queue *q, if (!bio) return 0; + /* + * This should probably be returning 0, but blk_add_request_payload() + * (Christoph!!!!) + */ + if (bio->bi_rw & REQ_DISCARD) + return 1; + + if (bio->bi_rw & REQ_WRITE_SAME) + return 1; + fbio = bio; cluster = blk_queue_cluster(q); seg_size = 0; @@ -161,6 +171,48 @@ __blk_segment_map_sg(struct request_queue *q, struct bio_vec *bvec, *bvprv = *bvec; } +static int __blk_bios_map_sg(struct request_queue *q, struct bio *bio, + struct scatterlist *sglist, + struct scatterlist **sg) +{ + struct bio_vec bvec, bvprv = { NULL }; + struct bvec_iter iter; + int nsegs, cluster; + + nsegs = 0; + cluster = blk_queue_cluster(q); + + if (bio->bi_rw & REQ_DISCARD) { + /* + * This is a hack - drivers should be neither modifying the + * biovec, nor relying on bi_vcnt - but because of + * blk_add_request_payload(), a discard bio may or may not have + * a payload we need to set up here (thank you Christoph) and + * bi_vcnt is really the only way of telling if we need to. + */ + + if (bio->bi_vcnt) + goto single_segment; + + return 0; + } + + if (bio->bi_rw & REQ_WRITE_SAME) { +single_segment: + *sg = sglist; + bvec = bio_iovec(bio); + sg_set_page(*sg, bvec.bv_page, bvec.bv_len, bvec.bv_offset); + return 1; + } + + for_each_bio(bio) + bio_for_each_segment(bvec, bio, iter) + __blk_segment_map_sg(q, &bvec, sglist, &bvprv, sg, + &nsegs, &cluster); + + return nsegs; +} + /* * map a request to scatterlist, return number of sg entries setup. Caller * must make sure sg can hold rq->nr_phys_segments entries @@ -168,23 +220,11 @@ __blk_segment_map_sg(struct request_queue *q, struct bio_vec *bvec, int blk_rq_map_sg(struct request_queue *q, struct request *rq, struct scatterlist *sglist) { - struct bio_vec bvec, bvprv = { NULL }; - struct req_iterator iter; - struct scatterlist *sg; - int nsegs, cluster; - - nsegs = 0; - cluster = blk_queue_cluster(q); - - /* - * for each bio in rq - */ - sg = NULL; - rq_for_each_segment(bvec, rq, iter) { - __blk_segment_map_sg(q, &bvec, sglist, &bvprv, &sg, - &nsegs, &cluster); - } /* segments in rq */ + struct scatterlist *sg = NULL; + int nsegs = 0; + if (rq->bio) + nsegs = __blk_bios_map_sg(q, rq->bio, sglist, &sg); if (unlikely(rq->cmd_flags & REQ_COPY_USER) && (blk_rq_bytes(rq) & q->dma_pad_mask)) { @@ -230,20 +270,13 @@ EXPORT_SYMBOL(blk_rq_map_sg); int blk_bio_map_sg(struct request_queue *q, struct bio *bio, struct scatterlist *sglist) { - struct bio_vec bvec, bvprv = { NULL }; - struct scatterlist *sg; - int nsegs, cluster; - struct bvec_iter iter; - - nsegs = 0; - cluster = blk_queue_cluster(q); - - sg = NULL; - bio_for_each_segment(bvec, bio, iter) { - __blk_segment_map_sg(q, &bvec, sglist, &bvprv, &sg, - &nsegs, &cluster); - } /* segments in bio */ + struct scatterlist *sg = NULL; + int nsegs; + struct bio *next = bio->bi_next; + bio->bi_next = NULL; + nsegs = __blk_bios_map_sg(q, bio, sglist, &sg); + bio->bi_next = next; if (sg) sg_mark_end(sg); From d7790b928d42597b7da21a4e43080774903e3b5c Mon Sep 17 00:00:00 2001 From: Shlomo Pongratz Date: Thu, 6 Feb 2014 18:33:17 +0200 Subject: [PATCH 394/596] block/null_blk: Fix completion processing from LIFO to FIFO The completion queue is implemented using lockless list. The llist_add is adds the events to the list head which is a push operation. The processing of the completion elements is done by disconnecting all the pushed elements and iterating over the disconnected list. The problem is that the processing is done in reverse order w.r.t order of the insertion i.e. LIFO processing. By reversing the disconnected list which is done in linear time the desired FIFO processing is achieved. Signed-off-by: Shlomo Pongratz Signed-off-by: Jens Axboe --- drivers/block/null_blk.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/block/null_blk.c b/drivers/block/null_blk.c index 3107282a9741..cc801cde5cfa 100644 --- a/drivers/block/null_blk.c +++ b/drivers/block/null_blk.c @@ -195,6 +195,7 @@ static enum hrtimer_restart null_cmd_timer_expired(struct hrtimer *timer) cq = &per_cpu(completion_queues, smp_processor_id()); while ((entry = llist_del_all(&cq->list)) != NULL) { + entry = llist_reverse_order(entry); do { cmd = container_of(entry, struct nullb_cmd, ll_list); end_cmd(cmd); @@ -235,6 +236,7 @@ static void null_ipi_cmd_end_io(void *data) cq = &per_cpu(completion_queues, smp_processor_id()); entry = llist_del_all(&cq->list); + entry = llist_reverse_order(entry); while (entry) { next = entry->next; From 247bf557273dd775505fb9240d2d152f4f20d304 Mon Sep 17 00:00:00 2001 From: Sarah Sharp Date: Fri, 31 Jan 2014 11:26:25 -0800 Subject: [PATCH 395/596] xhci 1.0: Limit arbitrarily-aligned scatter gather. MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit xHCI 1.0 hosts have a set of requirements on how to align transfer buffers on the endpoint rings called "TD fragment" rules. When the ax88179_178a driver added support for scatter gather in 3.12, with commit 804fad45411b48233b48003e33a78f290d227c8 "USBNET: ax88179_178a: enable tso if usb host supports sg dma", it broke the device under xHCI 1.0 hosts. Under certain network loads, the device would see an unexpected short packet from the host, which would cause the device to stop sending ethernet packets, even through USB packets would still be sent. Commit 35773dac5f86 "usb: xhci: Link TRB must not occur within a USB payload burst" attempted to fix this. It was a quick hack to partially implement the TD fragment rules. However, it caused regressions in the usb-storage layer and userspace USB drivers using libusb. The patches to attempt to fix this are too far reaching into the USB core, and we really need to implement the TD fragment rules correctly in the xHCI driver, instead of continuing to wallpaper over the issues. Disable arbitrarily-aligned scatter-gather in the xHCI driver for 1.0 hosts. Only the ax88179_178a driver checks the no_sg_constraint flag, so don't set it for 1.0 hosts. This should not impact usb-storage or usbfs behavior, since they pass down max packet sized aligned sg-list entries (512 for USB 2.0 and 1024 for USB 3.0). Signed-off-by: Sarah Sharp Tested-by: Mark Lord Cc: David Laight Cc: Bjørn Mork Cc: Freddy Xin Cc: Ming Lei Cc: stable@vger.kernel.org # 3.12 --- drivers/usb/host/xhci.c | 14 +++++++++++--- 1 file changed, 11 insertions(+), 3 deletions(-) diff --git a/drivers/usb/host/xhci.c b/drivers/usb/host/xhci.c index 3712359d18ba..86a2e834fdff 100644 --- a/drivers/usb/host/xhci.c +++ b/drivers/usb/host/xhci.c @@ -4733,9 +4733,6 @@ int xhci_gen_setup(struct usb_hcd *hcd, xhci_get_quirks_t get_quirks) /* Limit the block layer scatter-gather lists to half a segment. */ hcd->self.sg_tablesize = TRBS_PER_SEGMENT / 2; - /* support to build packet from discontinuous buffers */ - hcd->self.no_sg_constraint = 1; - /* XHCI controllers don't stop the ep queue on short packets :| */ hcd->self.no_stop_on_short = 1; @@ -4760,6 +4757,14 @@ int xhci_gen_setup(struct usb_hcd *hcd, xhci_get_quirks_t get_quirks) /* xHCI private pointer was set in xhci_pci_probe for the second * registered roothub. */ + xhci = hcd_to_xhci(hcd); + /* + * Support arbitrarily aligned sg-list entries on hosts without + * TD fragment rules (which are currently unsupported). + */ + if (xhci->hci_version < 0x100) + hcd->self.no_sg_constraint = 1; + return 0; } @@ -4788,6 +4793,9 @@ int xhci_gen_setup(struct usb_hcd *hcd, xhci_get_quirks_t get_quirks) if (xhci->hci_version > 0x96) xhci->quirks |= XHCI_SPURIOUS_SUCCESS; + if (xhci->hci_version < 0x100) + hcd->self.no_sg_constraint = 1; + /* Make sure the HC is halted. */ retval = xhci_halt(xhci); if (retval) From 1386ff75797a187df324062fb4e929152392da88 Mon Sep 17 00:00:00 2001 From: Sarah Sharp Date: Fri, 31 Jan 2014 11:45:02 -0800 Subject: [PATCH 396/596] Revert "xhci: Set scatter-gather limit to avoid failed block writes." This reverts commit f2d9b991c549f159dc9ae81f77d8206c790cbfee. We are ripping out commit 35773dac5f862cb1c82ea151eba3e2f6de51ec3e "usb: xhci: Link TRB must not occur within a USB payload burst" because it's a hack that caused regressions in the usb-storage and userspace USB drivers that use usbfs and libusb. This commit attempted to fix the issues with that patch. Signed-off-by: Sarah Sharp Cc: stable@vger.kernel.org #3.12 --- drivers/usb/host/xhci.c | 4 ++-- drivers/usb/host/xhci.h | 2 +- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/usb/host/xhci.c b/drivers/usb/host/xhci.c index 86a2e834fdff..6fe577d46fa2 100644 --- a/drivers/usb/host/xhci.c +++ b/drivers/usb/host/xhci.c @@ -4730,8 +4730,8 @@ int xhci_gen_setup(struct usb_hcd *hcd, xhci_get_quirks_t get_quirks) struct device *dev = hcd->self.controller; int retval; - /* Limit the block layer scatter-gather lists to half a segment. */ - hcd->self.sg_tablesize = TRBS_PER_SEGMENT / 2; + /* Accept arbitrarily long scatter-gather lists */ + hcd->self.sg_tablesize = ~0; /* XHCI controllers don't stop the ep queue on short packets :| */ hcd->self.no_stop_on_short = 1; diff --git a/drivers/usb/host/xhci.h b/drivers/usb/host/xhci.h index 9154fd6cf24c..58ed9d088e63 100644 --- a/drivers/usb/host/xhci.h +++ b/drivers/usb/host/xhci.h @@ -1268,7 +1268,7 @@ union xhci_trb { * since the command ring is 64-byte aligned. * It must also be greater than 16. */ -#define TRBS_PER_SEGMENT 256 +#define TRBS_PER_SEGMENT 64 /* Allow two commands + a link TRB, along with any reserved command TRBs */ #define MAX_RSVD_CMD_TRBS (TRBS_PER_SEGMENT - 3) #define TRB_SEGMENT_SIZE (TRBS_PER_SEGMENT*16) From 9cf00d91708221ff2d8a11143315f7ebab8d5da8 Mon Sep 17 00:00:00 2001 From: Sarah Sharp Date: Fri, 31 Jan 2014 11:51:59 -0800 Subject: [PATCH 397/596] Revert "xhci: Avoid infinite loop when sg urb requires too many trbs" This reverts commit d6c9ea9069af684358efedcaf2f2f687f51c58ee. We are ripping out commit 35773dac5f862cb1c82ea151eba3e2f6de51ec3e "usb: xhci: Link TRB must not occur within a USB payload burst" because it's a hack that caused regressions in the usb-storage and userspace USB drivers that use usbfs and libusb. This commit attempted to fix the issues with that patch. Signed-off-by: Sarah Sharp Cc: stable@vger.kernel.org # 3.12 --- drivers/usb/host/xhci-ring.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/usb/host/xhci-ring.c b/drivers/usb/host/xhci-ring.c index 909b32a4412f..0272450a1667 100644 --- a/drivers/usb/host/xhci-ring.c +++ b/drivers/usb/host/xhci-ring.c @@ -3002,7 +3002,7 @@ static int prepare_ring(struct xhci_hcd *xhci, struct xhci_ring *ep_ring, if (num_trbs >= TRBS_PER_SEGMENT) { xhci_err(xhci, "Too many fragments %d, max %d\n", num_trbs, TRBS_PER_SEGMENT - 1); - return -EINVAL; + return -ENOMEM; } nop_cmd = cpu_to_le32(TRB_TYPE(TRB_TR_NOOP) | From 3d4b81eda2211f32886e2978daf6f39885042fc4 Mon Sep 17 00:00:00 2001 From: Sarah Sharp Date: Fri, 31 Jan 2014 11:52:57 -0800 Subject: [PATCH 398/596] Revert "usb: xhci: Link TRB must not occur within a USB payload burst" This reverts commit 35773dac5f862cb1c82ea151eba3e2f6de51ec3e. It's a hack that caused regressions in the usb-storage and userspace USB drivers that use usbfs and libusb. Commit 70cabb7d992f "xhci 1.0: Limit arbitrarily-aligned scatter gather." should fix the issues seen with the ax88179_178a driver on xHCI 1.0 hosts, without causing regressions. Signed-off-by: Sarah Sharp Cc: stable@vger.kernel.org # 3.12 --- drivers/usb/host/xhci-ring.c | 54 ++---------------------------------- include/linux/usb.h | 2 -- 2 files changed, 2 insertions(+), 54 deletions(-) diff --git a/drivers/usb/host/xhci-ring.c b/drivers/usb/host/xhci-ring.c index 0272450a1667..0ed64eb68e48 100644 --- a/drivers/usb/host/xhci-ring.c +++ b/drivers/usb/host/xhci-ring.c @@ -2967,58 +2967,8 @@ static int prepare_ring(struct xhci_hcd *xhci, struct xhci_ring *ep_ring, } while (1) { - if (room_on_ring(xhci, ep_ring, num_trbs)) { - union xhci_trb *trb = ep_ring->enqueue; - unsigned int usable = ep_ring->enq_seg->trbs + - TRBS_PER_SEGMENT - 1 - trb; - u32 nop_cmd; - - /* - * Section 4.11.7.1 TD Fragments states that a link - * TRB must only occur at the boundary between - * data bursts (eg 512 bytes for 480M). - * While it is possible to split a large fragment - * we don't know the size yet. - * Simplest solution is to fill the trb before the - * LINK with nop commands. - */ - if (num_trbs == 1 || num_trbs <= usable || usable == 0) - break; - - if (ep_ring->type != TYPE_BULK) - /* - * While isoc transfers might have a buffer that - * crosses a 64k boundary it is unlikely. - * Since we can't add NOPs without generating - * gaps in the traffic just hope it never - * happens at the end of the ring. - * This could be fixed by writing a LINK TRB - * instead of the first NOP - however the - * TRB_TYPE_LINK_LE32() calls would all need - * changing to check the ring length. - */ - break; - - if (num_trbs >= TRBS_PER_SEGMENT) { - xhci_err(xhci, "Too many fragments %d, max %d\n", - num_trbs, TRBS_PER_SEGMENT - 1); - return -ENOMEM; - } - - nop_cmd = cpu_to_le32(TRB_TYPE(TRB_TR_NOOP) | - ep_ring->cycle_state); - ep_ring->num_trbs_free -= usable; - do { - trb->generic.field[0] = 0; - trb->generic.field[1] = 0; - trb->generic.field[2] = 0; - trb->generic.field[3] = nop_cmd; - trb++; - } while (--usable); - ep_ring->enqueue = trb; - if (room_on_ring(xhci, ep_ring, num_trbs)) - break; - } + if (room_on_ring(xhci, ep_ring, num_trbs)) + break; if (ep_ring == xhci->cmd_ring) { xhci_err(xhci, "Do not support expand command ring\n"); diff --git a/include/linux/usb.h b/include/linux/usb.h index c716da18c668..7f6eb859873e 100644 --- a/include/linux/usb.h +++ b/include/linux/usb.h @@ -1265,8 +1265,6 @@ typedef void (*usb_complete_t)(struct urb *); * @sg: scatter gather buffer list, the buffer size of each element in * the list (except the last) must be divisible by the endpoint's * max packet size if no_sg_constraint isn't set in 'struct usb_bus' - * (FIXME: scatter-gather under xHCI is broken for periodic transfers. - * Do not use urb->sg for interrupt endpoints for now, only bulk.) * @num_mapped_sgs: (internal) number of mapped sg entries * @num_sgs: number of entries in the sg list * @transfer_buffer_length: How big is transfer_buffer. The transfer may From fa69c87232cd6dac4d7e33e376467724697c43b9 Mon Sep 17 00:00:00 2001 From: Fu Wei Date: Tue, 28 Jan 2014 20:36:44 +0800 Subject: [PATCH 399/596] Documentation:Chinese translation of Documentation/arm64/tagged-pointers.txt This is a Chinese translated version of Documentation/arm64/tagged-pointers.txt It is based on the modifications of Documentation/arm64/tagged-pointers.txt in submission: "d50240a5", "374ed9d1". Signed-off-by: Fu Wei Acked-by: Harry Wei Signed-off-by: Greg Kroah-Hartman --- Documentation/zh_CN/arm64/tagged-pointers.txt | 52 +++++++++++++++++++ 1 file changed, 52 insertions(+) create mode 100644 Documentation/zh_CN/arm64/tagged-pointers.txt diff --git a/Documentation/zh_CN/arm64/tagged-pointers.txt b/Documentation/zh_CN/arm64/tagged-pointers.txt new file mode 100644 index 000000000000..2664d1bd5a1c --- /dev/null +++ b/Documentation/zh_CN/arm64/tagged-pointers.txt @@ -0,0 +1,52 @@ +Chinese translated version of Documentation/arm64/tagged-pointers.txt + +If you have any comment or update to the content, please contact the +original document maintainer directly. However, if you have a problem +communicating in English you can also ask the Chinese maintainer for +help. Contact the Chinese maintainer if this translation is outdated +or if there is a problem with the translation. + +Maintainer: Will Deacon +Chinese maintainer: Fu Wei +--------------------------------------------------------------------- +Documentation/arm64/tagged-pointers.txt 的中文翻译 + +如果想评论或更新本文的内容，请直接联系原文档的维护者。如果你使用英文 +交流有困难的话，也可以向中文版维护者求助。如果本翻译更新不及时或者翻 +译存在问题，请联系中文版维护者。 + +英文版维护者： Will Deacon +中文版维护者： 傅炜 Fu Wei +中文版翻译者： 傅炜 Fu Wei +中文版校译者： 傅炜 Fu Wei + +以下为正文 +--------------------------------------------------------------------- + Linux 在 AArch64 中带标记的虚拟地址 + ================================= + +作者: Will Deacon +日期: 2013 年 06 月 12 日 + +本文档简述了在 AArch64 地址转换系统中提供的带标记的虚拟地址及其在 +AArch64 Linux 中的潜在用途。 + +内核提供的地址转换表配置使通过 TTBR0 完成的虚拟地址转换（即用户空间 +映射），其虚拟地址的最高 8 位（63:56）会被转换硬件所忽略。这种机制 +让这些位可供应用程序自由使用，其注意事项如下： + + (1) 内核要求所有传递到 EL1 的用户空间地址带有 0x00 标记。 + 这意味着任何携带用户空间虚拟地址的系统调用（syscall） + 参数 *必须* 在陷入内核前使它们的最高字节被清零。 + + (2) 非零标记在传递信号时不被保存。这意味着在应用程序中利用了 + 标记的信号处理函数无法依赖 siginfo_t 的用户空间虚拟 + 地址所携带的包含其内部域信息的标记。此规则的一个例外是 + 当信号是在调试观察点的异常处理程序中产生的，此时标记的 + 信息将被保存。 + + (3) 当使用带标记的指针时需特别留心，因为仅对两个虚拟地址 + 的高字节，C 编译器很可能无法判断它们是不同的。 + +此构架会阻止对带标记的 PC 指针的利用，因此在异常返回时，其高字节 +将被设置成一个为 “55” 的扩展符。 From b0ab8a9030b655e02c5532d1f78b9ba9d9ff4420 Mon Sep 17 00:00:00 2001 From: Fu Wei Date: Tue, 28 Jan 2014 20:28:29 +0800 Subject: [PATCH 400/596] Documentation:Update Documentation/zh_CN/arm64/booting.txt This is a update of Chinese documentation: Documentation/zh_CN/arm64/booting.txt It is based on the modifications of Documentation/arm64/booting.txt in submission: "4d5e0b15", "4370eec0", "4fcd6e14". Update maintainer's Email address. Signed-off-by: Fu Wei Acked-by: Harry Wei Signed-off-by: Greg Kroah-Hartman --- Documentation/zh_CN/arm64/booting.txt | 65 +++++++++++++++++++-------- 1 file changed, 46 insertions(+), 19 deletions(-) diff --git a/Documentation/zh_CN/arm64/booting.txt b/Documentation/zh_CN/arm64/booting.txt index 28fa325b7461..6f6d956ac1c9 100644 --- a/Documentation/zh_CN/arm64/booting.txt +++ b/Documentation/zh_CN/arm64/booting.txt @@ -7,7 +7,7 @@ help. Contact the Chinese maintainer if this translation is outdated or if there is a problem with the translation. Maintainer: Will Deacon -Chinese maintainer: Fu Wei +Chinese maintainer: Fu Wei --------------------------------------------------------------------- Documentation/arm64/booting.txt 的中文翻译 @@ -16,9 +16,9 @@ Documentation/arm64/booting.txt 的中文翻译 译存在问题，请联系中文版维护者。 英文版维护者： Will Deacon -中文版维护者： 傅炜 Fu Wei -中文版翻译者： 傅炜 Fu Wei -中文版校译者： 傅炜 Fu Wei +中文版维护者： 傅炜 Fu Wei +中文版翻译者： 傅炜 Fu Wei +中文版校译者： 傅炜 Fu Wei 以下为正文 --------------------------------------------------------------------- @@ -64,8 +64,8 @@ RAM，或可能使用对这个设备已知的 RAM 信息，还可能使用任何 必要性: 强制 -设备树数据块（dtb）大小必须不大于 2 MB，且位于从内核映像起始算起第一个 -512MB 内的 2MB 边界上。这使得内核可以通过初始页表中的单个节描述符来 +设备树数据块（dtb）必须 8 字节对齐，并位于从内核映像起始算起第一个 512MB +内，且不得跨越 2MB 对齐边界。这使得内核可以通过初始页表中的单个节描述符来 映射此数据块。 @@ -84,13 +84,23 @@ AArch64 内核当前没有提供自解压代码，因此如果使用了压缩内 必要性: 强制 -已解压的内核映像包含一个 32 字节的头，内容如下： +已解压的内核映像包含一个 64 字节的头，内容如下： - u32 magic = 0x14000008; /* 跳转到 stext, 小端 */ - u32 res0 = 0; /* 保留 */ + u32 code0; /* 可执行代码 */ + u32 code1; /* 可执行代码 */ u64 text_offset; /* 映像装载偏移 */ + u64 res0 = 0; /* 保留 */ u64 res1 = 0; /* 保留 */ u64 res2 = 0; /* 保留 */ + u64 res3 = 0; /* 保留 */ + u64 res4 = 0; /* 保留 */ + u32 magic = 0x644d5241; /* 魔数, 小端, "ARM\x64" */ + u32 res5 = 0; /* 保留 */ + + +映像头注释： + +- code0/code1 负责跳转到 stext. 映像必须位于系统 RAM 起始处的特定偏移（当前是 0x80000）。系统 RAM 的起始地址必须是以 2MB 对齐的。 @@ -118,9 +128,9 @@ AArch64 内核当前没有提供自解压代码，因此如果使用了压缩内 外部高速缓存（如果存在）必须配置并禁用。 - 架构计时器 - CNTFRQ 必须设定为计时器的频率。 - 如果在 EL1 模式下进入内核，则 CNTHCTL_EL2 中的 EL1PCTEN (bit 0) - 必须置位。 + CNTFRQ 必须设定为计时器的频率，且 CNTVOFF 必须设定为对所有 CPU + 都一致的值。如果在 EL1 模式下进入内核，则 CNTHCTL_EL2 中的 + EL1PCTEN (bit 0) 必须置位。 - 一致性 通过内核启动的所有 CPU 在内核入口地址上必须处于相同的一致性域中。 @@ -131,23 +141,40 @@ AArch64 内核当前没有提供自解压代码，因此如果使用了压缩内 在进入内核映像的异常级中，所有构架中可写的系统寄存器必须通过软件 在一个更高的异常级别下初始化，以防止在 未知 状态下运行。 +以上对于 CPU 模式、高速缓存、MMU、架构计时器、一致性、系统寄存器的 +必要条件描述适用于所有 CPU。所有 CPU 必须在同一异常级别跳入内核。 + 引导装载程序必须在每个 CPU 处于以下状态时跳入内核入口： - 主 CPU 必须直接跳入内核映像的第一条指令。通过此 CPU 传递的设备树 - 数据块必须在每个 CPU 节点中包含以下内容： - - 1、‘enable-method’属性。目前，此字段支持的值仅为字符串“spin-table”。 - - 2、‘cpu-release-addr’标识一个 64-bit、初始化为零的内存位置。 + 数据块必须在每个 CPU 节点中包含一个 ‘enable-method’ 属性，所 + 支持的 enable-method 请见下文。 引导装载程序必须生成这些设备树属性，并在跳入内核入口之前将其插入 数据块。 -- 任何辅助 CPU 必须在内存保留区（通过设备树中的 /memreserve/ 域传递 +- enable-method 为 “spin-table” 的 CPU 必须在它们的 CPU + 节点中包含一个 ‘cpu-release-addr’ 属性。这个属性标识了一个 + 64 位自然对齐且初始化为零的内存位置。 + + 这些 CPU 必须在内存保留区（通过设备树中的 /memreserve/ 域传递 给内核）中自旋于内核之外，轮询它们的 cpu-release-addr 位置（必须 包含在保留区中）。可通过插入 wfe 指令来降低忙循环开销，而主 CPU 将 发出 sev 指令。当对 cpu-release-addr 所指位置的读取操作返回非零值 - 时，CPU 必须直接跳入此值所指向的地址。 + 时，CPU 必须跳入此值所指向的地址。此值为一个单独的 64 位小端值， + 因此 CPU 须在跳转前将所读取的值转换为其本身的端模式。 + +- enable-method 为 “psci” 的 CPU 保持在内核外（比如，在 + memory 节点中描述为内核空间的内存区外，或在通过设备树 /memreserve/ + 域中描述为内核保留区的空间中）。内核将会发起在 ARM 文档（编号 + ARM DEN 0022A：用于 ARM 上的电源状态协调接口系统软件）中描述的 + CPU_ON 调用来将 CPU 带入内核。 + + *译者注:到文档翻译时，此文档已更新为 ARM DEN 0022B。 + + 设备树必须包含一个 ‘psci’ 节点，请参考以下文档： + Documentation/devicetree/bindings/arm/psci.txt + - 辅助 CPU 通用寄存器设置 x0 = 0 (保留，将来可能使用) From d0f2a808b2f0fa6e115cc6f6c84bb4f29aa8af49 Mon Sep 17 00:00:00 2001 From: Fu Wei Date: Tue, 28 Jan 2014 20:14:29 +0800 Subject: [PATCH 401/596] Documentation:Update Documentation/zh_CN/arm64/memory.txt This is a update of Chinese documentation:Documentation/zh_CN/arm64/memory.txt It is based on the modifications of Documentation/arm64/memory.txt in submission: "2475ff9d", "847264fb", "aa4a73a0", "e29a074b". Update maintainer's Email address. Signed-off-by: Fu Wei Acked-by: Harry Wei Signed-off-by: Greg Kroah-Hartman --- Documentation/zh_CN/arm64/memory.txt | 46 ++++++++++++++++++++++++---- 1 file changed, 40 insertions(+), 6 deletions(-) diff --git a/Documentation/zh_CN/arm64/memory.txt b/Documentation/zh_CN/arm64/memory.txt index a5f6283829f9..a782704c1cb5 100644 --- a/Documentation/zh_CN/arm64/memory.txt +++ b/Documentation/zh_CN/arm64/memory.txt @@ -7,7 +7,7 @@ help. Contact the Chinese maintainer if this translation is outdated or if there is a problem with the translation. Maintainer: Catalin Marinas -Chinese maintainer: Fu Wei +Chinese maintainer: Fu Wei --------------------------------------------------------------------- Documentation/arm64/memory.txt 的中文翻译 @@ -16,9 +16,9 @@ Documentation/arm64/memory.txt 的中文翻译 译存在问题，请联系中文版维护者。 英文版维护者： Catalin Marinas -中文版维护者： 傅炜 Fu Wei -中文版翻译者： 傅炜 Fu Wei -中文版校译者： 傅炜 Fu Wei +中文版维护者： 傅炜 Fu Wei +中文版翻译者： 傅炜 Fu Wei +中文版校译者： 傅炜 Fu Wei 以下为正文 --------------------------------------------------------------------- @@ -41,7 +41,7 @@ AArch64 Linux 使用页大小为 4KB 的 3 级转换表配置，对于用户和 TTBR1 中，且从不写入 TTBR0。 -AArch64 Linux 内存布局： +AArch64 Linux 在页大小为 4KB 时的内存布局： 起始地址 结束地址 大小 用途 ----------------------------------------------------------------------- @@ -55,15 +55,42 @@ ffffffbc00000000 ffffffbdffffffff 8GB vmemmap ffffffbe00000000 ffffffbffbbfffff ~8GB [防护页，未来用于 vmmemap] +ffffffbffbc00000 ffffffbffbdfffff 2MB earlyprintk 设备 + ffffffbffbe00000 ffffffbffbe0ffff 64KB PCI I/O 空间 -ffffffbbffff0000 ffffffbcffffffff ~2MB [防护页] +ffffffbffbe10000 ffffffbcffffffff ~2MB [防护页] ffffffbffc000000 ffffffbfffffffff 64MB 模块 ffffffc000000000 ffffffffffffffff 256GB 内核逻辑内存映射 +AArch64 Linux 在页大小为 64KB 时的内存布局： + +起始地址 结束地址 大小 用途 +----------------------------------------------------------------------- +0000000000000000 000003ffffffffff 4TB 用户空间 + +fffffc0000000000 fffffdfbfffeffff ~2TB vmalloc + +fffffdfbffff0000 fffffdfbffffffff 64KB [防护页] + +fffffdfc00000000 fffffdfdffffffff 8GB vmemmap + +fffffdfe00000000 fffffdfffbbfffff ~8GB [防护页，未来用于 vmmemap] + +fffffdfffbc00000 fffffdfffbdfffff 2MB earlyprintk 设备 + +fffffdfffbe00000 fffffdfffbe0ffff 64KB PCI I/O 空间 + +fffffdfffbe10000 fffffdfffbffffff ~2MB [防护页] + +fffffdfffc000000 fffffdffffffffff 64MB 模块 + +fffffe0000000000 ffffffffffffffff 2TB 内核逻辑内存映射 + + 4KB 页大小的转换表查找： +--------+--------+--------+--------+--------+--------+--------+--------+ @@ -91,3 +118,10 @@ ffffffc000000000 ffffffffffffffff 256GB 内核逻辑内存映射 | | +--------------------------> [41:29] L2 索引 (仅使用 38:29 ) | +-------------------------------> [47:42] L1 索引 (未使用) +-------------------------------------------------> [63] TTBR0/1 + +当使用 KVM 时, 管理程序（hypervisor）在 EL2 中通过相对内核虚拟地址的 +一个固定偏移来映射内核页（内核虚拟地址的高 24 位设为零）: + +起始地址 结束地址 大小 用途 +----------------------------------------------------------------------- +0000004000000000 0000007fffffffff 256GB 在 HYP 中映射的内核对象 From d913c7439add288b50752186b306634df2ae21e7 Mon Sep 17 00:00:00 2001 From: Christian Engelmayer Date: Sat, 25 Jan 2014 22:47:44 +0100 Subject: [PATCH 402/596] misc: genwqe: Fix potential memory leak when pinning memory Fix a memory leak in the genwqe_pin_mem() error path as called by ioctl GENWQE_PIN_MEM. In case there is an error encountered when mapping memory, the already allocated dma_mapping struct needs to be freed correctly. Detected by Coverity: CID 1162606. Signed-off-by: Christian Engelmayer Acked-by: Frank Haverkamp Signed-off-by: Greg Kroah-Hartman --- drivers/misc/genwqe/card_dev.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/misc/genwqe/card_dev.c b/drivers/misc/genwqe/card_dev.c index 8f8a6b327cdb..2c2c9cc75231 100644 --- a/drivers/misc/genwqe/card_dev.c +++ b/drivers/misc/genwqe/card_dev.c @@ -787,6 +787,7 @@ static int genwqe_pin_mem(struct genwqe_file *cfile, struct genwqe_mem *m) if (rc != 0) { dev_err(&pci_dev->dev, "[%s] genwqe_user_vmap rc=%d\n", __func__, rc); + kfree(dma_map); return rc; } From f0de8e04a7201a2000f3c6d09732c11e7f35d42d Mon Sep 17 00:00:00 2001 From: Maxime Ripard Date: Sun, 2 Feb 2014 14:52:05 +0100 Subject: [PATCH 403/596] misc: eeprom: sunxi: Add new compatibles The Allwinner A10 compatibles were following a slightly different compatible patterns than the rest of the SoCs for historical reasons. Add compatibles matching the other pattern to the SID driver for consistency, and keep the older one for backward compatibility. Signed-off-by: Maxime Ripard Signed-off-by: Greg Kroah-Hartman --- .../devicetree/bindings/misc/allwinner,sunxi-sid.txt | 5 +++-- drivers/misc/eeprom/sunxi_sid.c | 5 ++++- 2 files changed, 7 insertions(+), 3 deletions(-) diff --git a/Documentation/devicetree/bindings/misc/allwinner,sunxi-sid.txt b/Documentation/devicetree/bindings/misc/allwinner,sunxi-sid.txt index 68ba37295565..c10a61ad5f95 100644 --- a/Documentation/devicetree/bindings/misc/allwinner,sunxi-sid.txt +++ b/Documentation/devicetree/bindings/misc/allwinner,sunxi-sid.txt @@ -1,12 +1,13 @@ Allwinner sunxi-sid Required properties: -- compatible: "allwinner,sun4i-sid" or "allwinner,sun7i-a20-sid". +- compatible: "allwinner,sun4i-a10-sid" or "allwinner,sun7i-a20-sid" + (Deprecated: "allwinner,sun4i-sid"). - reg: Should contain registers location and length Example for sun4i: sid@01c23800 { - compatible = "allwinner,sun4i-sid"; + compatible = "allwinner,sun4i-a10-sid"; reg = <0x01c23800 0x10> }; diff --git a/drivers/misc/eeprom/sunxi_sid.c b/drivers/misc/eeprom/sunxi_sid.c index 9c34e5704304..e137e75a89e5 100644 --- a/drivers/misc/eeprom/sunxi_sid.c +++ b/drivers/misc/eeprom/sunxi_sid.c @@ -96,8 +96,11 @@ static int sunxi_sid_remove(struct platform_device *pdev) } static const struct of_device_id sunxi_sid_of_match[] = { - { .compatible = "allwinner,sun4i-sid", .data = (void *)16}, + { .compatible = "allwinner,sun4i-a10-sid", .data = (void *)16}, { .compatible = "allwinner,sun7i-a20-sid", .data = (void *)512}, + + /* Deprecated */ + { .compatible = "allwinner,sun4i-sid", .data = (void *)16}, {/* sentinel */}, }; MODULE_DEVICE_TABLE(of, sunxi_sid_of_match); From 01ab1167cd2d861d20195eda08505652c536df97 Mon Sep 17 00:00:00 2001 From: Maxime Ripard Date: Sun, 2 Feb 2014 14:52:06 +0100 Subject: [PATCH 404/596] ARM: sunxi: dt: Convert to the new SID compatibles Switch the device tree to the new compatibles introduced in the SID drivers to have a common pattern accross all Allwinner SoCs. Signed-off-by: Maxime Ripard Signed-off-by: Greg Kroah-Hartman --- arch/arm/boot/dts/sun4i-a10.dtsi | 2 +- arch/arm/boot/dts/sun5i-a10s.dtsi | 2 +- arch/arm/boot/dts/sun5i-a13.dtsi | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) diff --git a/arch/arm/boot/dts/sun4i-a10.dtsi b/arch/arm/boot/dts/sun4i-a10.dtsi index 040bb0eba152..7b10f85ef988 100644 --- a/arch/arm/boot/dts/sun4i-a10.dtsi +++ b/arch/arm/boot/dts/sun4i-a10.dtsi @@ -421,7 +421,7 @@ rtc: rtc@01c20d00 { }; sid: eeprom@01c23800 { - compatible = "allwinner,sun4i-sid"; + compatible = "allwinner,sun4i-a10-sid"; reg = <0x01c23800 0x10>; }; diff --git a/arch/arm/boot/dts/sun5i-a10s.dtsi b/arch/arm/boot/dts/sun5i-a10s.dtsi index ea16054857a4..6d287a0dac5d 100644 --- a/arch/arm/boot/dts/sun5i-a10s.dtsi +++ b/arch/arm/boot/dts/sun5i-a10s.dtsi @@ -378,7 +378,7 @@ wdt: watchdog@01c20c90 { }; sid: eeprom@01c23800 { - compatible = "allwinner,sun4i-sid"; + compatible = "allwinner,sun4i-a10-sid"; reg = <0x01c23800 0x10>; }; diff --git a/arch/arm/boot/dts/sun5i-a13.dtsi b/arch/arm/boot/dts/sun5i-a13.dtsi index 320335abfccd..f49eb13c33c3 100644 --- a/arch/arm/boot/dts/sun5i-a13.dtsi +++ b/arch/arm/boot/dts/sun5i-a13.dtsi @@ -341,7 +341,7 @@ wdt: watchdog@01c20c90 { }; sid: eeprom@01c23800 { - compatible = "allwinner,sun4i-sid"; + compatible = "allwinner,sun4i-a10-sid"; reg = <0x01c23800 0x10>; }; From 3b1cc9b9622a022208ec95b1259b05bbdf712eb7 Mon Sep 17 00:00:00 2001 From: Sudeep Dutt Date: Mon, 3 Feb 2014 14:53:19 -0800 Subject: [PATCH 405/596] misc: mic: fix possible signed underflow (undefined behavior) in userspace API iovcnt is declared as a signed integer in both the userspace API and as a local variable in mic_virtio.c. The while() loop in mic_virtio.c iterates until the local variable iovcnt reaches the value 0. If userspace passes e.g. INT_MIN as iovcnt field, this loop then appears to depend on an undefined behavior (signed underflow) to complete. The fix is to use unsigned integers in both the userspace API and the local variable. This issue was reported @ https://lkml.org/lkml/2014/1/10/10 Reported-by: Mathieu Desnoyers Reviewed-by: Ashutosh Dixit Signed-off-by: Sudeep Dutt Cc: stable Signed-off-by: Greg Kroah-Hartman --- drivers/misc/mic/host/mic_virtio.c | 3 ++- include/uapi/linux/mic_ioctl.h | 2 +- 2 files changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/misc/mic/host/mic_virtio.c b/drivers/misc/mic/host/mic_virtio.c index 752ff873f891..7e1ef0ebbb80 100644 --- a/drivers/misc/mic/host/mic_virtio.c +++ b/drivers/misc/mic/host/mic_virtio.c @@ -156,7 +156,8 @@ static int mic_vringh_copy(struct mic_vdev *mvdev, struct vringh_kiov *iov, static int _mic_virtio_copy(struct mic_vdev *mvdev, struct mic_copy_desc *copy) { - int ret = 0, iovcnt = copy->iovcnt; + int ret = 0; + u32 iovcnt = copy->iovcnt; struct iovec iov; struct iovec __user *u_iov = copy->iov; void __user *ubuf = NULL; diff --git a/include/uapi/linux/mic_ioctl.h b/include/uapi/linux/mic_ioctl.h index 7fabba5059cf..feb0b4c0814c 100644 --- a/include/uapi/linux/mic_ioctl.h +++ b/include/uapi/linux/mic_ioctl.h @@ -39,7 +39,7 @@ struct mic_copy_desc { #else struct iovec *iov; #endif - int iovcnt; + __u32 iovcnt; __u8 vr_idx; __u8 update_used; __u32 out_len; From 9e1ccb4a7700fcb3ddfe2767e0bbb6131f8ab54e Mon Sep 17 00:00:00 2001 From: Russell King Date: Fri, 7 Feb 2014 20:09:27 +0000 Subject: [PATCH 406/596] drivers/base: fix devres handling for master device We weren't handling the devres issues for the master device failing a bind, or being unbound properly. Add a devres group to contain these, and release the resources at the appropriate points. Signed-off-by: Russell King Signed-off-by: Greg Kroah-Hartman --- drivers/base/component.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/base/component.c b/drivers/base/component.c index c53efe6c6d8e..c4778995cd72 100644 --- a/drivers/base/component.c +++ b/drivers/base/component.c @@ -133,9 +133,16 @@ static int try_to_bring_up_master(struct master *master, goto out; } + if (!devres_open_group(master->dev, NULL, GFP_KERNEL)) { + ret = -ENOMEM; + goto out; + } + /* Found all components */ ret = master->ops->bind(master->dev); if (ret < 0) { + devres_release_group(master->dev, NULL); + dev_info(master->dev, "master bind failed: %d\n", ret); master_remove_components(master); goto out; } @@ -166,6 +173,7 @@ static void take_down_master(struct master *master) { if (master->bound) { master->ops->unbind(master->dev); + devres_release_group(master->dev, NULL); master->bound = false; } From 26c8f0d601f5d4c0d9f4bc8c5151539aae5dc26a Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Fri, 7 Feb 2014 11:04:04 -0500 Subject: [PATCH 407/596] cifs: use a flexarray in cifs_writedata The cifs_writedata code uses a single element trailing array, which just adds unneeded complexity. Use a flexarray instead. Signed-off-by: Jeff Layton Reviewed-by: Pavel Shilovsky Signed-off-by: Steve French --- fs/cifs/cifsglob.h | 2 +- fs/cifs/cifssmb.c | 8 +------- 2 files changed, 2 insertions(+), 8 deletions(-) diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h index 615e35ab7ef6..d6a031ed391b 100644 --- a/fs/cifs/cifsglob.h +++ b/fs/cifs/cifsglob.h @@ -1068,7 +1068,7 @@ struct cifs_writedata { unsigned int pagesz; unsigned int tailsz; unsigned int nr_pages; - struct page *pages[1]; + struct page *pages[]; }; /* diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index 4d881c35eeca..0cd742ccb019 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -1962,15 +1962,9 @@ cifs_writedata_alloc(unsigned int nr_pages, work_func_t complete) { struct cifs_writedata *wdata; - /* this would overflow */ - if (nr_pages == 0) { - cifs_dbg(VFS, "%s: called with nr_pages == 0!\n", __func__); - return NULL; - } - /* writedata + number of page pointers */ wdata = kzalloc(sizeof(*wdata) + - sizeof(struct page *) * (nr_pages - 1), GFP_NOFS); + sizeof(struct page *) * nr_pages, GFP_NOFS); if (wdata != NULL) { kref_init(&wdata->refcount); INIT_LIST_HEAD(&wdata->list); From 4a5c80d7b5615be8098f9d5da97d166afc318abc Mon Sep 17 00:00:00 2001 From: Steve French Date: Fri, 7 Feb 2014 20:45:12 -0600 Subject: [PATCH 408/596] [CIFS] clean up page array when uncached write send fails In the event that a send fails in an uncached write, or we end up needing to reissue it (-EAGAIN case), we'll kfree the wdata but the pages currently leak. Fix this by adding a new kref release routine for uncached writedata that releases the pages, and have the uncached codepaths use that. [original patch by Jeff modified to fix minor formatting problems] Signed-off-by: Jeff Layton Reviewed-by: Pavel Shilovsky Signed-off-by: Steve French --- fs/cifs/cifsglob.h | 3 ++- fs/cifs/cifsproto.h | 3 ++- fs/cifs/cifssmb.c | 7 ++++--- fs/cifs/file.c | 31 ++++++++++++++++++++----------- fs/cifs/smb2pdu.c | 5 +++-- fs/cifs/smb2proto.h | 3 ++- 6 files changed, 33 insertions(+), 19 deletions(-) diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h index d6a031ed391b..86dc28c7aa5c 100644 --- a/fs/cifs/cifsglob.h +++ b/fs/cifs/cifsglob.h @@ -323,7 +323,8 @@ struct smb_version_operations { /* async read from the server */ int (*async_readv)(struct cifs_readdata *); /* async write to the server */ - int (*async_writev)(struct cifs_writedata *); + int (*async_writev)(struct cifs_writedata *, + void (*release)(struct kref *)); /* sync read from the server */ int (*sync_read)(const unsigned int, struct cifsFileInfo *, struct cifs_io_parms *, unsigned int *, char **, diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h index 79e6e9a93a8c..d00e09dfc452 100644 --- a/fs/cifs/cifsproto.h +++ b/fs/cifs/cifsproto.h @@ -488,7 +488,8 @@ void cifs_readdata_release(struct kref *refcount); int cifs_async_readv(struct cifs_readdata *rdata); int cifs_readv_receive(struct TCP_Server_Info *server, struct mid_q_entry *mid); -int cifs_async_writev(struct cifs_writedata *wdata); +int cifs_async_writev(struct cifs_writedata *wdata, + void (*release)(struct kref *kref)); void cifs_writev_complete(struct work_struct *work); struct cifs_writedata *cifs_writedata_alloc(unsigned int nr_pages, work_func_t complete); diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index 0cd742ccb019..f3264bd7a83d 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -1910,7 +1910,7 @@ cifs_writev_requeue(struct cifs_writedata *wdata) do { server = tlink_tcon(wdata->cfile->tlink)->ses->server; - rc = server->ops->async_writev(wdata); + rc = server->ops->async_writev(wdata, cifs_writedata_release); } while (rc == -EAGAIN); for (i = 0; i < wdata->nr_pages; i++) { @@ -2025,7 +2025,8 @@ cifs_writev_callback(struct mid_q_entry *mid) /* cifs_async_writev - send an async write, and set up mid to handle result */ int -cifs_async_writev(struct cifs_writedata *wdata) +cifs_async_writev(struct cifs_writedata *wdata, + void (*release)(struct kref *kref)) { int rc = -EACCES; WRITE_REQ *smb = NULL; @@ -2099,7 +2100,7 @@ cifs_async_writev(struct cifs_writedata *wdata) if (rc == 0) cifs_stats_inc(&tcon->stats.cifs_stats.num_writes); else - kref_put(&wdata->refcount, cifs_writedata_release); + kref_put(&wdata->refcount, release); async_writev_out: cifs_small_buf_release(smb); diff --git a/fs/cifs/file.c b/fs/cifs/file.c index 853d6d1cc822..a301edbdad4a 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -2043,7 +2043,8 @@ static int cifs_writepages(struct address_space *mapping, } wdata->pid = wdata->cfile->pid; server = tlink_tcon(wdata->cfile->tlink)->ses->server; - rc = server->ops->async_writev(wdata); + rc = server->ops->async_writev(wdata, + cifs_writedata_release); } while (wbc->sync_mode == WB_SYNC_ALL && rc == -EAGAIN); for (i = 0; i < nr_pages; ++i) @@ -2331,9 +2332,20 @@ size_t get_numpages(const size_t wsize, const size_t len, size_t *cur_len) } static void -cifs_uncached_writev_complete(struct work_struct *work) +cifs_uncached_writedata_release(struct kref *refcount) { int i; + struct cifs_writedata *wdata = container_of(refcount, + struct cifs_writedata, refcount); + + for (i = 0; i < wdata->nr_pages; i++) + put_page(wdata->pages[i]); + cifs_writedata_release(refcount); +} + +static void +cifs_uncached_writev_complete(struct work_struct *work) +{ struct cifs_writedata *wdata = container_of(work, struct cifs_writedata, work); struct inode *inode = wdata->cfile->dentry->d_inode; @@ -2347,12 +2359,7 @@ cifs_uncached_writev_complete(struct work_struct *work) complete(&wdata->done); - if (wdata->result != -EAGAIN) { - for (i = 0; i < wdata->nr_pages; i++) - put_page(wdata->pages[i]); - } - - kref_put(&wdata->refcount, cifs_writedata_release); + kref_put(&wdata->refcount, cifs_uncached_writedata_release); } /* attempt to send write to server, retry on any -EAGAIN errors */ @@ -2370,7 +2377,8 @@ cifs_uncached_retry_writev(struct cifs_writedata *wdata) if (rc != 0) continue; } - rc = server->ops->async_writev(wdata); + rc = server->ops->async_writev(wdata, + cifs_uncached_writedata_release); } while (rc == -EAGAIN); return rc; @@ -2454,7 +2462,8 @@ cifs_iovec_write(struct file *file, const struct iovec *iov, wdata->tailsz = cur_len - ((nr_pages - 1) * PAGE_SIZE); rc = cifs_uncached_retry_writev(wdata); if (rc) { - kref_put(&wdata->refcount, cifs_writedata_release); + kref_put(&wdata->refcount, + cifs_uncached_writedata_release); break; } @@ -2496,7 +2505,7 @@ cifs_iovec_write(struct file *file, const struct iovec *iov, } } list_del_init(&wdata->list); - kref_put(&wdata->refcount, cifs_writedata_release); + kref_put(&wdata->refcount, cifs_uncached_writedata_release); } if (total_written > 0) diff --git a/fs/cifs/smb2pdu.c b/fs/cifs/smb2pdu.c index 2013234b73ad..a3f7a9c3cc69 100644 --- a/fs/cifs/smb2pdu.c +++ b/fs/cifs/smb2pdu.c @@ -1890,7 +1890,8 @@ smb2_writev_callback(struct mid_q_entry *mid) /* smb2_async_writev - send an async write, and set up mid to handle result */ int -smb2_async_writev(struct cifs_writedata *wdata) +smb2_async_writev(struct cifs_writedata *wdata, + void (*release)(struct kref *kref)) { int rc = -EACCES; struct smb2_write_req *req = NULL; @@ -1938,7 +1939,7 @@ smb2_async_writev(struct cifs_writedata *wdata) smb2_writev_callback, wdata, 0); if (rc) { - kref_put(&wdata->refcount, cifs_writedata_release); + kref_put(&wdata->refcount, release); cifs_stats_fail_inc(tcon, SMB2_WRITE_HE); } diff --git a/fs/cifs/smb2proto.h b/fs/cifs/smb2proto.h index 93adc64666f3..0ce48db20a65 100644 --- a/fs/cifs/smb2proto.h +++ b/fs/cifs/smb2proto.h @@ -123,7 +123,8 @@ extern int SMB2_get_srv_num(const unsigned int xid, struct cifs_tcon *tcon, extern int smb2_async_readv(struct cifs_readdata *rdata); extern int SMB2_read(const unsigned int xid, struct cifs_io_parms *io_parms, unsigned int *nbytes, char **buf, int *buf_type); -extern int smb2_async_writev(struct cifs_writedata *wdata); +extern int smb2_async_writev(struct cifs_writedata *wdata, + void (*release)(struct kref *kref)); extern int SMB2_write(const unsigned int xid, struct cifs_io_parms *io_parms, unsigned int *nbytes, struct kvec *iov, int n_vec); extern int SMB2_echo(struct TCP_Server_Info *server); From 1ccfe6f982674e4b7bd832b904bafcb3db890252 Mon Sep 17 00:00:00 2001 From: Richard Weinberger Date: Fri, 31 Jan 2014 13:47:34 +0100 Subject: [PATCH 409/596] watchdog: dw_wdt: Add dependency on HAS_IOMEM On archs like S390 or um this driver cannot build nor work. Make it depend on HAS_IOMEM to bypass build failures. drivers/built-in.o: In function `dw_wdt_drv_probe': drivers/watchdog/dw_wdt.c:302: undefined reference to `devm_ioremap_resource' Signed-off-by: Richard Weinberger Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/watchdog/Kconfig b/drivers/watchdog/Kconfig index 4c4c566c52a3..79d25894343a 100644 --- a/drivers/watchdog/Kconfig +++ b/drivers/watchdog/Kconfig @@ -223,6 +223,7 @@ config SA1100_WATCHDOG config DW_WATCHDOG tristate "Synopsys DesignWare watchdog" + depends on HAS_IOMEM help Say Y here if to include support for the Synopsys DesignWare watchdog timer found in many chips. From c18f7b51200c3c8b76c63e391f9995b65ace9c83 Mon Sep 17 00:00:00 2001 From: Dave Kleikamp Date: Fri, 7 Feb 2014 14:36:10 -0600 Subject: [PATCH 410/596] jfs: fix generic posix ACL regression I missed a couple errors in reviewing the patches converting jfs to use the generic posix ACL function. Setting ACL's currently fails with -EOPNOTSUPP. Signed-off-by: Dave Kleikamp Reported-by: Michael L. Semon Reviewed-by: Christoph Hellwig --- fs/jfs/xattr.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/fs/jfs/xattr.c b/fs/jfs/xattr.c index 3bd5ee45f7b3..46325d5c34fc 100644 --- a/fs/jfs/xattr.c +++ b/fs/jfs/xattr.c @@ -854,9 +854,6 @@ int jfs_setxattr(struct dentry *dentry, const char *name, const void *value, int rc; tid_t tid; - if ((rc = can_set_xattr(inode, name, value, value_len))) - return rc; - /* * If this is a request for a synthetic attribute in the system.* * namespace use the generic infrastructure to resolve a handler @@ -865,6 +862,9 @@ int jfs_setxattr(struct dentry *dentry, const char *name, const void *value, if (!strncmp(name, XATTR_SYSTEM_PREFIX, XATTR_SYSTEM_PREFIX_LEN)) return generic_setxattr(dentry, name, value, value_len, flags); + if ((rc = can_set_xattr(inode, name, value, value_len))) + return rc; + if (value == NULL) { /* empty EA, do not remove */ value = ""; value_len = 0; @@ -1034,9 +1034,6 @@ int jfs_removexattr(struct dentry *dentry, const char *name) int rc; tid_t tid; - if ((rc = can_set_xattr(inode, name, NULL, 0))) - return rc; - /* * If this is a request for a synthetic attribute in the system.* * namespace use the generic infrastructure to resolve a handler @@ -1045,6 +1042,9 @@ int jfs_removexattr(struct dentry *dentry, const char *name) if (!strncmp(name, XATTR_SYSTEM_PREFIX, XATTR_SYSTEM_PREFIX_LEN)) return generic_removexattr(dentry, name); + if ((rc = can_set_xattr(inode, name, NULL, 0))) + return rc; + tid = txBegin(inode->i_sb, 0); mutex_lock(&ji->commit_mutex); rc = __jfs_setxattr(tid, dentry->d_inode, name, NULL, 0, XATTR_REPLACE); @@ -1061,7 +1061,7 @@ int jfs_removexattr(struct dentry *dentry, const char *name) * attributes are handled directly. */ const struct xattr_handler *jfs_xattr_handlers[] = { -#ifdef JFS_POSIX_ACL +#ifdef CONFIG_JFS_POSIX_ACL &posix_acl_access_xattr_handler, &posix_acl_default_xattr_handler, #endif From 6cc98d90f8d14f8ebce2391323929024d7eef39f Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Wed, 5 Feb 2014 16:19:21 -0500 Subject: [PATCH 411/596] Btrfs: fix assert screwup for the pending move stuff Wang noticed that he was failing btrfs/030 even though me and Filipe couldn't reproduce. Turns out this is because Wang didn't have CONFIG_BTRFS_ASSERT set, which meant that a key part of Filipe's original patch was not being built in. This appears to be a mess up with merging Filipe's patch as it does not exist in his original patch. Fix this by changing how we make sure del_waiting_dir_move asserts that it did not error and take the function out of the ifdef check. This makes btrfs/030 pass with the assert on or off. Thanks, Signed-off-by: Josef Bacik Reviewed-by: Filipe Manana Signed-off-by: Chris Mason --- fs/btrfs/send.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c index cf9107a64204..9c8d1a3fdc3a 100644 --- a/fs/btrfs/send.c +++ b/fs/btrfs/send.c @@ -2774,8 +2774,6 @@ static int add_waiting_dir_move(struct send_ctx *sctx, u64 ino) return 0; } -#ifdef CONFIG_BTRFS_ASSERT - static int del_waiting_dir_move(struct send_ctx *sctx, u64 ino) { struct rb_node *n = sctx->waiting_dir_moves.rb_node; @@ -2796,8 +2794,6 @@ static int del_waiting_dir_move(struct send_ctx *sctx, u64 ino) return -ENOENT; } -#endif - static int add_pending_dir_move(struct send_ctx *sctx, u64 parent_ino) { struct rb_node **p = &sctx->pending_dir_moves.rb_node; @@ -2902,7 +2898,9 @@ static int apply_dir_move(struct send_ctx *sctx, struct pending_dir_move *pm) } sctx->send_progress = sctx->cur_ino + 1; - ASSERT(del_waiting_dir_move(sctx, pm->ino) == 0); + ret = del_waiting_dir_move(sctx, pm->ino); + ASSERT(ret == 0); + ret = get_cur_path(sctx, pm->ino, pm->gen, to_path); if (ret < 0) goto out; From d0270aca88966641eb15306e9bd0c7ad15321440 Mon Sep 17 00:00:00 2001 From: Jeff Mahoney Date: Fri, 7 Feb 2014 14:33:57 +0100 Subject: [PATCH 412/596] btrfs: commit transaction after setting label and features The set_fslabel ioctl uses btrfs_end_transaction, which means it's possible that the change will be lost if the system crashes, same for the newly set features. Let's use btrfs_commit_transaction instead. Signed-off-by: Jeff Mahoney Signed-off-by: David Sterba Signed-off-by: Chris Mason --- fs/btrfs/ioctl.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index 34772cbcc7aa..5bbf6b7216c3 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -4547,7 +4547,7 @@ static int btrfs_ioctl_set_fslabel(struct file *file, void __user *arg) spin_lock(&root->fs_info->super_lock); strcpy(super_block->label, label); spin_unlock(&root->fs_info->super_lock); - ret = btrfs_end_transaction(trans, root); + ret = btrfs_commit_transaction(trans, root); out_unlock: mnt_drop_write_file(file); @@ -4711,7 +4711,7 @@ static int btrfs_ioctl_set_features(struct file *file, void __user *arg) btrfs_set_super_incompat_flags(super_block, newflags); spin_unlock(&root->fs_info->super_lock); - return btrfs_end_transaction(trans, root); + return btrfs_commit_transaction(trans, root); } long btrfs_ioctl(struct file *file, unsigned int From 8051aa1a3d5aaa7bd4c062cad94d09c3d567ef2e Mon Sep 17 00:00:00 2001 From: David Sterba Date: Fri, 7 Feb 2014 14:34:04 +0100 Subject: [PATCH 413/596] btrfs: reserve no transaction units in btrfs_ioctl_set_features Added in patch "btrfs: add ioctls to query/change feature bits online" modifications to superblock don't need to reserve metadata blocks when starting a transaction. Signed-off-by: David Sterba Signed-off-by: Chris Mason --- fs/btrfs/ioctl.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index 5bbf6b7216c3..ebdd866d4cfd 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -4690,7 +4690,7 @@ static int btrfs_ioctl_set_features(struct file *file, void __user *arg) if (ret) return ret; - trans = btrfs_start_transaction(root, 1); + trans = btrfs_start_transaction(root, 0); if (IS_ERR(trans)) return PTR_ERR(trans); From 27a377db745ed4d11b3b9b340756857cb8dde07f Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Fri, 7 Feb 2014 13:57:59 -0500 Subject: [PATCH 414/596] Btrfs: don't loop forever if we can't run because of the tree mod log A user reported a 100% cpu hang with my new delayed ref code. Turns out I forgot to increase the count check when we can't run a delayed ref because of the tree mod log. If we can't run any delayed refs during this there is no point in continuing to look, and we need to break out. Thanks, Signed-off-by: Josef Bacik Signed-off-by: Chris Mason --- fs/btrfs/extent-tree.c | 1 + 1 file changed, 1 insertion(+) diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 9c9ecc93ae2c..32312e09f0f5 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -2385,6 +2385,7 @@ static noinline int __btrfs_run_delayed_refs(struct btrfs_trans_handle *trans, spin_unlock(&delayed_refs->lock); locked_ref = NULL; cond_resched(); + count++; continue; } From a2aa75e18a21b21952dc6daa9bac7c9f4426f81f Mon Sep 17 00:00:00 2001 From: Filipe David Borba Manana Date: Sat, 8 Feb 2014 15:47:46 +0000 Subject: [PATCH 415/596] Btrfs: fix data corruption when reading/updating compressed extents When using a mix of compressed file extents and prealloc extents, it is possible to fill a page of a file with random, garbage data from some unrelated previous use of the page, instead of a sequence of zeroes. A simple sequence of steps to get into such case, taken from the test case I made for xfstests, is: _scratch_mkfs _scratch_mount "-o compress-force=lzo" $XFS_IO_PROG -f -c "pwrite -S 0x06 -b 18670 266978 18670" $SCRATCH_MNT/foobar $XFS_IO_PROG -c "falloc 26450 665194" $SCRATCH_MNT/foobar $XFS_IO_PROG -c "truncate 542872" $SCRATCH_MNT/foobar $XFS_IO_PROG -c "fsync" $SCRATCH_MNT/foobar This results in the following file items in the fs tree: item 4 key (257 INODE_ITEM 0) itemoff 15879 itemsize 160 inode generation 6 transid 6 size 542872 block group 0 mode 100600 item 5 key (257 INODE_REF 256) itemoff 15863 itemsize 16 inode ref index 2 namelen 6 name: foobar item 6 key (257 EXTENT_DATA 0) itemoff 15810 itemsize 53 extent data disk byte 0 nr 0 gen 6 extent data offset 0 nr 24576 ram 266240 extent compression 0 item 7 key (257 EXTENT_DATA 24576) itemoff 15757 itemsize 53 prealloc data disk byte 12849152 nr 241664 gen 6 prealloc data offset 0 nr 241664 item 8 key (257 EXTENT_DATA 266240) itemoff 15704 itemsize 53 extent data disk byte 12845056 nr 4096 gen 6 extent data offset 0 nr 20480 ram 20480 extent compression 2 item 9 key (257 EXTENT_DATA 286720) itemoff 15651 itemsize 53 prealloc data disk byte 13090816 nr 405504 gen 6 prealloc data offset 0 nr 258048 The on disk extent at offset 266240 (which corresponds to 1 single disk block), contains 5 compressed chunks of file data. Each of the first 4 compress 4096 bytes of file data, while the last one only compresses 3024 bytes of file data. Therefore a read into the file region [285648 ; 286720[ (length = 4096 - 3024 = 1072 bytes) should always return zeroes (our next extent is a prealloc one). The solution here is the compression code path to zero the remaining (untouched) bytes of the last page it uncompressed data into, as the information about how much space the file data consumes in the last page is not known in the upper layer fs/btrfs/extent_io.c:__do_readpage(). In __do_readpage we were correctly zeroing the remainder of the page but only if it corresponds to the last page of the inode and if the inode's size is not a multiple of the page size. This would cause not only returning random data on reads, but also permanently storing random data when updating parts of the region that should be zeroed. For the example above, it means updating a single byte in the region [285648 ; 286720[ would store that byte correctly but also store random data on disk. A test case for xfstests follows soon. Signed-off-by: Filipe David Borba Manana Signed-off-by: Chris Mason --- fs/btrfs/compression.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/fs/btrfs/compression.c b/fs/btrfs/compression.c index af815eb8f970..ed1ff1cb1017 100644 --- a/fs/btrfs/compression.c +++ b/fs/btrfs/compression.c @@ -1011,6 +1011,8 @@ int btrfs_decompress_buf2page(char *buf, unsigned long buf_start, bytes = min(bytes, working_bytes); kaddr = kmap_atomic(page_out); memcpy(kaddr + *pg_offset, buf + buf_offset, bytes); + if (*pg_index == (vcnt - 1) && *pg_offset == 0) + memset(kaddr + bytes, 0, PAGE_CACHE_SIZE - bytes); kunmap_atomic(kaddr); flush_dcache_page(page_out); From 569d6557ab957d6ae7e97a46ae669174be4189e6 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Tue, 4 Feb 2014 14:13:15 -0500 Subject: [PATCH 416/596] x86: Use preempt_disable_notrace() in cycles_2_ns() When debug preempt is enabled, preempt_disable() can be traced by function and function graph tracing. There's a place in the function graph tracer that calls trace_clock() which eventually calls cycles_2_ns() outside of the recursion protection. When cycles_2_ns() calls preempt_disable() it gets traced and the graph tracer will go into a recursive loop causing a crash or worse, a triple fault. Simple fix is to use preempt_disable_notrace() in cycles_2_ns, which makes sense because the preempt_disable() tracing may use that code too, and it tracing it, even with recursion protection is rather pointless. Signed-off-by: Steven Rostedt Acked-by: Thomas Gleixner Signed-off-by: Peter Zijlstra Cc: Linus Torvalds Link: http://lkml.kernel.org/r/20140204141315.2a968a72@gandalf.local.home Signed-off-by: Ingo Molnar --- arch/x86/kernel/tsc.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c index 19e5adb49a27..acb3b606613e 100644 --- a/arch/x86/kernel/tsc.c +++ b/arch/x86/kernel/tsc.c @@ -209,7 +209,7 @@ static inline unsigned long long cycles_2_ns(unsigned long long cyc) * dance when its actually needed. */ - preempt_disable(); + preempt_disable_notrace(); data = this_cpu_read(cyc2ns.head); tail = this_cpu_read(cyc2ns.tail); @@ -229,7 +229,7 @@ static inline unsigned long long cycles_2_ns(unsigned long long cyc) if (!--data->__count) this_cpu_write(cyc2ns.tail, data); } - preempt_enable(); + preempt_enable_notrace(); return ns; } From 0668d3065128d39449c097e62dbdb5707820137d Mon Sep 17 00:00:00 2001 From: Stephen Boyd Date: Thu, 2 Jan 2014 16:37:32 -0800 Subject: [PATCH 417/596] genirq: Add devm_request_any_context_irq() Some drivers use request_any_context_irq() but there isn't a devm_* function for it. Add one so that these drivers don't need to explicitly free the irq on driver detach. Signed-off-by: Stephen Boyd Cc: linux-arm-kernel@lists.infradead.org Cc: Dmitry Torokhov Link: http://lkml.kernel.org/r/1388709460-19222-3-git-send-email-sboyd@codeaurora.org Signed-off-by: Thomas Gleixner --- include/linux/interrupt.h | 5 +++++ kernel/irq/devres.c | 45 +++++++++++++++++++++++++++++++++++++++ 2 files changed, 50 insertions(+) diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h index 0053adde0ed9..a2678d35b5a2 100644 --- a/include/linux/interrupt.h +++ b/include/linux/interrupt.h @@ -158,6 +158,11 @@ devm_request_irq(struct device *dev, unsigned int irq, irq_handler_t handler, devname, dev_id); } +extern int __must_check +devm_request_any_context_irq(struct device *dev, unsigned int irq, + irq_handler_t handler, unsigned long irqflags, + const char *devname, void *dev_id); + extern void devm_free_irq(struct device *dev, unsigned int irq, void *dev_id); /* diff --git a/kernel/irq/devres.c b/kernel/irq/devres.c index bd8e788d71e0..1ef0606797c9 100644 --- a/kernel/irq/devres.c +++ b/kernel/irq/devres.c @@ -72,6 +72,51 @@ int devm_request_threaded_irq(struct device *dev, unsigned int irq, } EXPORT_SYMBOL(devm_request_threaded_irq); +/** + * devm_request_any_context_irq - allocate an interrupt line for a managed device + * @dev: device to request interrupt for + * @irq: Interrupt line to allocate + * @handler: Function to be called when the IRQ occurs + * @thread_fn: function to be called in a threaded interrupt context. NULL + * for devices which handle everything in @handler + * @irqflags: Interrupt type flags + * @devname: An ascii name for the claiming device + * @dev_id: A cookie passed back to the handler function + * + * Except for the extra @dev argument, this function takes the + * same arguments and performs the same function as + * request_any_context_irq(). IRQs requested with this function will be + * automatically freed on driver detach. + * + * If an IRQ allocated with this function needs to be freed + * separately, devm_free_irq() must be used. + */ +int devm_request_any_context_irq(struct device *dev, unsigned int irq, + irq_handler_t handler, unsigned long irqflags, + const char *devname, void *dev_id) +{ + struct irq_devres *dr; + int rc; + + dr = devres_alloc(devm_irq_release, sizeof(struct irq_devres), + GFP_KERNEL); + if (!dr) + return -ENOMEM; + + rc = request_any_context_irq(irq, handler, irqflags, devname, dev_id); + if (rc) { + devres_free(dr); + return rc; + } + + dr->irq = irq; + dr->dev_id = dev_id; + devres_add(dev, dr); + + return 0; +} +EXPORT_SYMBOL(devm_request_any_context_irq); + /** * devm_free_irq - free an interrupt * @dev: device to free interrupt for From d311d79de305f1ada47cadd672e6ed1b28a949eb Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 9 Feb 2014 15:18:09 -0500 Subject: [PATCH 418/596] fix O_SYNC|O_APPEND syncing the wrong range on write() It actually goes back to 2004 ([PATCH] Concurrent O_SYNC write support) when sync_page_range() had been introduced; generic_file_write{,v}() correctly synced pos_after_write - written .. pos_after_write - 1 but generic_file_aio_write() synced pos_before_write .. pos_before_write + written - 1 instead. Which is not the same thing with O_APPEND, obviously. A couple of years later correct variant had been killed off when everything switched to use of generic_file_aio_write(). All users of generic_file_aio_write() are affected, and the same bug has been copied into other instances of ->aio_write(). The fix is trivial; the only subtle point is that generic_write_sync() ought to be inlined to avoid calculations useless for the majority of calls. Signed-off-by: Al Viro --- fs/cifs/file.c | 4 ++-- fs/ext4/file.c | 2 +- fs/ntfs/file.c | 2 +- fs/sync.c | 17 ----------------- fs/xfs/xfs_file.c | 2 +- include/linux/fs.h | 8 +++++++- mm/filemap.c | 4 ++-- 7 files changed, 14 insertions(+), 25 deletions(-) diff --git a/fs/cifs/file.c b/fs/cifs/file.c index 853d6d1cc822..a7eda8ebfacc 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -2559,8 +2559,8 @@ cifs_writev(struct kiocb *iocb, const struct iovec *iov, if (rc > 0) { ssize_t err; - err = generic_write_sync(file, pos, rc); - if (err < 0 && rc > 0) + err = generic_write_sync(file, iocb->ki_pos - rc, rc); + if (err < 0) rc = err; } diff --git a/fs/ext4/file.c b/fs/ext4/file.c index 43e64f6022eb..1a5073959f32 100644 --- a/fs/ext4/file.c +++ b/fs/ext4/file.c @@ -152,7 +152,7 @@ ext4_file_dio_write(struct kiocb *iocb, const struct iovec *iov, if (ret > 0) { ssize_t err; - err = generic_write_sync(file, pos, ret); + err = generic_write_sync(file, iocb->ki_pos - ret, ret); if (err < 0 && ret > 0) ret = err; } diff --git a/fs/ntfs/file.c b/fs/ntfs/file.c index ea4ba9daeb47..db9bd8a31725 100644 --- a/fs/ntfs/file.c +++ b/fs/ntfs/file.c @@ -2134,7 +2134,7 @@ static ssize_t ntfs_file_aio_write(struct kiocb *iocb, const struct iovec *iov, ret = ntfs_file_aio_write_nolock(iocb, iov, nr_segs, &iocb->ki_pos); mutex_unlock(&inode->i_mutex); if (ret > 0) { - int err = generic_write_sync(file, pos, ret); + int err = generic_write_sync(file, iocb->ki_pos - ret, ret); if (err < 0) ret = err; } diff --git a/fs/sync.c b/fs/sync.c index f15537452231..e8ba024a055b 100644 --- a/fs/sync.c +++ b/fs/sync.c @@ -222,23 +222,6 @@ SYSCALL_DEFINE1(fdatasync, unsigned int, fd) return do_fsync(fd, 1); } -/** - * generic_write_sync - perform syncing after a write if file / inode is sync - * @file: file to which the write happened - * @pos: offset where the write started - * @count: length of the write - * - * This is just a simple wrapper about our general syncing function. - */ -int generic_write_sync(struct file *file, loff_t pos, loff_t count) -{ - if (!(file->f_flags & O_DSYNC) && !IS_SYNC(file->f_mapping->host)) - return 0; - return vfs_fsync_range(file, pos, pos + count - 1, - (file->f_flags & __O_SYNC) ? 0 : 1); -} -EXPORT_SYMBOL(generic_write_sync); - /* * sys_sync_file_range() permits finely controlled syncing over a segment of * a file in the range offset .. (offset+nbytes-1) inclusive. If nbytes is diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c index 2e7989e3a2d6..64b48eade91d 100644 --- a/fs/xfs/xfs_file.c +++ b/fs/xfs/xfs_file.c @@ -799,7 +799,7 @@ xfs_file_aio_write( XFS_STATS_ADD(xs_write_bytes, ret); /* Handle various SYNC-type writes */ - err = generic_write_sync(file, pos, ret); + err = generic_write_sync(file, iocb->ki_pos - ret, ret); if (err < 0) ret = err; } diff --git a/include/linux/fs.h b/include/linux/fs.h index 09f553c59813..75ff961be051 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -2273,7 +2273,13 @@ extern int filemap_fdatawrite_range(struct address_space *mapping, extern int vfs_fsync_range(struct file *file, loff_t start, loff_t end, int datasync); extern int vfs_fsync(struct file *file, int datasync); -extern int generic_write_sync(struct file *file, loff_t pos, loff_t count); +static inline int generic_write_sync(struct file *file, loff_t pos, loff_t count) +{ + if (!(file->f_flags & O_DSYNC) && !IS_SYNC(file->f_mapping->host)) + return 0; + return vfs_fsync_range(file, pos, pos + count - 1, + (file->f_flags & __O_SYNC) ? 0 : 1); +} extern void emergency_sync(void); extern void emergency_remount(void); #ifdef CONFIG_BLOCK diff --git a/mm/filemap.c b/mm/filemap.c index d56d3c145b9f..7a13f6ac5421 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -2553,8 +2553,8 @@ ssize_t generic_file_aio_write(struct kiocb *iocb, const struct iovec *iov, if (ret > 0) { ssize_t err; - err = generic_write_sync(file, pos, ret); - if (err < 0 && ret > 0) + err = generic_write_sync(file, iocb->ki_pos - ret, ret); + if (err < 0) ret = err; } return ret; From c9efe51165fa0aff57be54e3cb0201ac87f68980 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 2 Feb 2014 07:05:05 -0500 Subject: [PATCH 419/596] fix a kmap leak in virtio_console While we are at it, don't do kmap() under kmap_atomic(), *especially* for a page we'd allocated with GFP_KERNEL. It's spelled "page_address", and had that been more than that, we'd have a real trouble - kmap_high() can block, and doing that while holding kmap_atomic() is a Bad Idea(tm). Signed-off-by: Al Viro --- drivers/char/virtio_console.c | 9 +++------ 1 file changed, 3 insertions(+), 6 deletions(-) diff --git a/drivers/char/virtio_console.c b/drivers/char/virtio_console.c index feea87cc6b8f..6928d094451d 100644 --- a/drivers/char/virtio_console.c +++ b/drivers/char/virtio_console.c @@ -890,12 +890,10 @@ static int pipe_to_sg(struct pipe_inode_info *pipe, struct pipe_buffer *buf, } else { /* Failback to copying a page */ struct page *page = alloc_page(GFP_KERNEL); - char *src = buf->ops->map(pipe, buf, 1); - char *dst; + char *src; if (!page) return -ENOMEM; - dst = kmap(page); offset = sd->pos & ~PAGE_MASK; @@ -903,9 +901,8 @@ static int pipe_to_sg(struct pipe_inode_info *pipe, struct pipe_buffer *buf, if (len + offset > PAGE_SIZE) len = PAGE_SIZE - offset; - memcpy(dst + offset, src + buf->offset, len); - - kunmap(page); + src = buf->ops->map(pipe, buf, 1); + memcpy(page_address(page) + offset, src + buf->offset, len); buf->ops->unmap(pipe, buf, src); sg_set_page(&(sgl->sg[sgl->n]), page, len, offset); From 0b4ef8de090a60e04560ed69aa35f439b8d5145f Mon Sep 17 00:00:00 2001 From: Rashika Kheria Date: Sun, 9 Feb 2014 18:22:53 +0530 Subject: [PATCH 420/596] fs: Mark function as static in fs/bio-integrity.c MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Mark functions as static in bio-integrity.c because it is not used outside this file. This eliminates the following warnings in bio-integrity.c: fs/bio-integrity.c:224:5: warning: no previous prototype for ‘bio_integrity_tag’ [-Wmissing-prototypes] Signed-off-by: Rashika Kheria Reviewed-by: Josh Triplett Signed-off-by: Jens Axboe --- fs/bio-integrity.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/fs/bio-integrity.c b/fs/bio-integrity.c index bcb4a4d533a2..0129b78a6908 100644 --- a/fs/bio-integrity.c +++ b/fs/bio-integrity.c @@ -234,7 +234,8 @@ unsigned int bio_integrity_tag_size(struct bio *bio) } EXPORT_SYMBOL(bio_integrity_tag_size); -int bio_integrity_tag(struct bio *bio, void *tag_buf, unsigned int len, int set) +static int bio_integrity_tag(struct bio *bio, void *tag_buf, unsigned int len, + int set) { struct bio_integrity_payload *bip = bio->bi_integrity; struct blk_integrity *bi = bdev_get_integrity(bio->bi_bdev); From c4540a7d8c1e595560e53acedf88901daf15a2b5 Mon Sep 17 00:00:00 2001 From: Rashika Kheria Date: Sun, 9 Feb 2014 18:30:39 +0530 Subject: [PATCH 421/596] fs: Add prototype declaration to appropriate header file include/linux/bio.h MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Add prototype declaration to header file include/linux/bio.h because it is used by more than one file. This eliminates the following warning in bio-integrity.c: fs/bio-integrity.c:214:14: warning: no previous prototype for ‘bio_integrity_tag_size’ [-Wmissing-prototypes] Signed-off-by: Rashika Kheria Reviewed-by: Josh Triplett Signed-off-by: Jens Axboe --- include/linux/bio.h | 1 + 1 file changed, 1 insertion(+) diff --git a/include/linux/bio.h b/include/linux/bio.h index 70654521dab6..d6791bba8264 100644 --- a/include/linux/bio.h +++ b/include/linux/bio.h @@ -332,6 +332,7 @@ extern struct bio *bio_clone_fast(struct bio *, gfp_t, struct bio_set *); extern struct bio *bio_clone_bioset(struct bio *, gfp_t, struct bio_set *bs); extern struct bio_set *fs_bio_set; +unsigned int bio_integrity_tag_size(struct bio *bio); static inline struct bio *bio_alloc(gfp_t gfp_mask, unsigned int nr_iovecs) { From fd944bded5ce32247941ce54eb8f65646549886b Mon Sep 17 00:00:00 2001 From: Rashika Kheria Date: Sun, 9 Feb 2014 19:57:33 +0530 Subject: [PATCH 422/596] net: Mark function as static in 9p/client.c MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Mark function as static in net/9p/client.c because it is not used outside this file. This eliminates the following warning in net/9p/client.c: net/9p/client.c:207:18: warning: no previous prototype for ‘p9_fcall_alloc’ [-Wmissing-prototypes] Signed-off-by: Rashika Kheria Reviewed-by: Josh Triplett Signed-off-by: David S. Miller --- net/9p/client.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/9p/client.c b/net/9p/client.c index a5e4d2dcb03e..9186550d77a6 100644 --- a/net/9p/client.c +++ b/net/9p/client.c @@ -204,7 +204,7 @@ static int parse_opts(char *opts, struct p9_client *clnt) return ret; } -struct p9_fcall *p9_fcall_alloc(int alloc_msize) +static struct p9_fcall *p9_fcall_alloc(int alloc_msize) { struct p9_fcall *fc; fc = kmalloc(sizeof(struct p9_fcall) + alloc_msize, GFP_NOFS); From 8203274e1541392e8a85d4bcbcda55d62fe62469 Mon Sep 17 00:00:00 2001 From: Rashika Kheria Date: Sun, 9 Feb 2014 19:59:04 +0530 Subject: [PATCH 423/596] net: Include appropriate header file in caif/caif_dev.c MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Include appropriate header file net/caif/caif_dev.h in caif/caif_dev.c because it has prototype declarations of function defined in caif/caif_dev.c. This eliminates the following file in caif/caif_dev.c: net/caif/caif_dev.c:303:6: warning: no previous prototype for ‘caif_enroll_dev’ [-Wmissing-prototypes] Signed-off-by: Rashika Kheria Reviewed-by: Josh Triplett Signed-off-by: David S. Miller --- net/caif/caif_dev.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/caif/caif_dev.c b/net/caif/caif_dev.c index 4dca159435cf..edbca468fa73 100644 --- a/net/caif/caif_dev.c +++ b/net/caif/caif_dev.c @@ -22,6 +22,7 @@ #include #include #include +#include #include #include #include From 02fe72c9edc58d6b5d85c08327d7ef05cd3e97dc Mon Sep 17 00:00:00 2001 From: Rashika Kheria Date: Sun, 9 Feb 2014 20:02:16 +0530 Subject: [PATCH 424/596] net: Include appropriate header file in caif/cfsrvl.c MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Include appropriate header file net/caif/caif_dev.h in caif/cfsrvl.c because it has prototype declaration of functions defined in caif/cfsrvl.c. This eliminates the following warning in caif/cfsrvl.c: net/caif/cfsrvl.c:198:6: warning: no previous prototype for ‘caif_free_client’ [-Wmissing-prototypes] net/caif/cfsrvl.c:208:6: warning: no previous prototype for ‘caif_client_register_refcnt’ [-Wmissing-prototypes] Signed-off-by: Rashika Kheria Reviewed-by: Josh Triplett Signed-off-by: David S. Miller --- net/caif/cfsrvl.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/caif/cfsrvl.c b/net/caif/cfsrvl.c index 353f793d1b3b..a6e115463052 100644 --- a/net/caif/cfsrvl.c +++ b/net/caif/cfsrvl.c @@ -15,6 +15,7 @@ #include #include #include +#include #define SRVL_CTRL_PKT_SIZE 1 #define SRVL_FLOW_OFF 0x81 From 0a59f3a9fd7e2801a445682465ea0522ea497183 Mon Sep 17 00:00:00 2001 From: Rashika Kheria Date: Sun, 9 Feb 2014 20:26:25 +0530 Subject: [PATCH 425/596] net: Mark functions as static in core/dev.c MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Mark functions as static in core/dev.c because they are not used outside this file. This eliminates the following warning in core/dev.c: net/core/dev.c:2806:5: warning: no previous prototype for ‘__dev_queue_xmit’ [-Wmissing-prototypes] net/core/dev.c:4640:5: warning: no previous prototype for ‘netdev_adjacent_sysfs_add’ [-Wmissing-prototypes] net/core/dev.c:4650:6: warning: no previous prototype for ‘netdev_adjacent_sysfs_del’ [-Wmissing-prototypes] Signed-off-by: Rashika Kheria Reviewed-by: Josh Triplett Reviewed-by: Veaceslav Falico Signed-off-by: David S. Miller --- net/core/dev.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/net/core/dev.c b/net/core/dev.c index 3721db716350..4ad1b78c9c77 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2803,7 +2803,7 @@ EXPORT_SYMBOL(dev_loopback_xmit); * the BH enable code must have IRQs enabled so that it will not deadlock. * --BLG */ -int __dev_queue_xmit(struct sk_buff *skb, void *accel_priv) +static int __dev_queue_xmit(struct sk_buff *skb, void *accel_priv) { struct net_device *dev = skb->dev; struct netdev_queue *txq; @@ -4637,7 +4637,7 @@ struct net_device *netdev_master_upper_dev_get_rcu(struct net_device *dev) } EXPORT_SYMBOL(netdev_master_upper_dev_get_rcu); -int netdev_adjacent_sysfs_add(struct net_device *dev, +static int netdev_adjacent_sysfs_add(struct net_device *dev, struct net_device *adj_dev, struct list_head *dev_list) { @@ -4647,7 +4647,7 @@ int netdev_adjacent_sysfs_add(struct net_device *dev, return sysfs_create_link(&(dev->dev.kobj), &(adj_dev->dev.kobj), linkname); } -void netdev_adjacent_sysfs_del(struct net_device *dev, +static void netdev_adjacent_sysfs_del(struct net_device *dev, char *name, struct list_head *dev_list) { From f56b8bf6e445b22bb6207dbcf30b2c7f5ddfdf96 Mon Sep 17 00:00:00 2001 From: Rashika Kheria Date: Sun, 9 Feb 2014 22:20:09 +0530 Subject: [PATCH 426/596] net: Move prototype declaration to appropriate header file from decnet/af_decnet.c MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Move prototype declaration of functions to header file include/net/dn_route.h from net/decnet/af_decnet.c because it is used by more than one file. This eliminates the following warning in net/decnet/dn_route.c: net/decnet/dn_route.c:629:5: warning: no previous prototype for ‘dn_route_rcv’ [-Wmissing-prototypes] Signed-off-by: Rashika Kheria Reviewed-by: Josh Triplett Signed-off-by: David S. Miller --- include/net/dn_route.h | 2 ++ net/decnet/af_decnet.c | 2 -- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/include/net/dn_route.h b/include/net/dn_route.h index b409ad6b8d7a..55df9939bca2 100644 --- a/include/net/dn_route.h +++ b/include/net/dn_route.h @@ -20,6 +20,8 @@ int dn_route_output_sock(struct dst_entry __rcu **pprt, struct flowidn *, struct sock *sk, int flags); int dn_cache_dump(struct sk_buff *skb, struct netlink_callback *cb); void dn_rt_cache_flush(int delay); +int dn_route_rcv(struct sk_buff *skb, struct net_device *dev, + struct packet_type *pt, struct net_device *orig_dev); /* Masks for flags field */ #define DN_RT_F_PID 0x07 /* Mask for packet type */ diff --git a/net/decnet/af_decnet.c b/net/decnet/af_decnet.c index 2954dcbca832..24d9193240e3 100644 --- a/net/decnet/af_decnet.c +++ b/net/decnet/af_decnet.c @@ -2104,8 +2104,6 @@ static struct notifier_block dn_dev_notifier = { .notifier_call = dn_device_event, }; -extern int dn_route_rcv(struct sk_buff *, struct net_device *, struct packet_type *, struct net_device *); - static struct packet_type dn_dix_packet_type __read_mostly = { .type = cpu_to_be16(ETH_P_DNA_RT), .func = dn_route_rcv, From ab3301bd96c024ec9c2e1c35c90327dc5c8012a4 Mon Sep 17 00:00:00 2001 From: Rashika Kheria Date: Sun, 9 Feb 2014 22:22:53 +0530 Subject: [PATCH 427/596] net: Move prototype declaration to header file include/net/dn.h from net/decnet/af_decnet.c MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Move prototype declaration of functions to header file include/net/dn.h from net/decnet/af_decnet.c because they are used by more than one file. This eliminates the following warning in net/decnet/af_decnet.c: net/decnet/sysctl_net_decnet.c:354:6: warning: no previous prototype for ‘dn_register_sysctl’ [-Wmissing-prototypes] net/decnet/sysctl_net_decnet.c:359:6: warning: no previous prototype for ‘dn_unregister_sysctl’ [-Wmissing-prototypes] Signed-off-by: Rashika Kheria Reviewed-by: Josh Triplett Signed-off-by: David S. Miller --- include/net/dn.h | 2 ++ net/decnet/af_decnet.c | 3 --- 2 files changed, 2 insertions(+), 3 deletions(-) diff --git a/include/net/dn.h b/include/net/dn.h index ccc15588d108..913b73d239f5 100644 --- a/include/net/dn.h +++ b/include/net/dn.h @@ -200,6 +200,8 @@ static inline void dn_sk_ports_copy(struct flowidn *fld, struct dn_scp *scp) } unsigned int dn_mss_from_pmtu(struct net_device *dev, int mtu); +void dn_register_sysctl(void); +void dn_unregister_sysctl(void); #define DN_MENUVER_ACC 0x01 #define DN_MENUVER_USR 0x02 diff --git a/net/decnet/af_decnet.c b/net/decnet/af_decnet.c index 24d9193240e3..4c04848953bd 100644 --- a/net/decnet/af_decnet.c +++ b/net/decnet/af_decnet.c @@ -2351,9 +2351,6 @@ static const struct proto_ops dn_proto_ops = { .sendpage = sock_no_sendpage, }; -void dn_register_sysctl(void); -void dn_unregister_sysctl(void); - MODULE_DESCRIPTION("The Linux DECnet Network Protocol"); MODULE_AUTHOR("Linux DECnet Project Team"); MODULE_LICENSE("GPL"); From 493cc5e5ba450356560cc12a8f71ff6a8574c91c Mon Sep 17 00:00:00 2001 From: Rashika Kheria Date: Sun, 9 Feb 2014 22:24:33 +0530 Subject: [PATCH 428/596] net: Move prototype declaration to include/net/ipx.h from net/ipx/ipx_route.c MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Move prototype definition of function to header file include/net/ipx.h from net/ipx/ipx_route.c because they are used by more than one file. This eliminates the following warning from net/ipx/af_ipx.c: net/ipx/af_ipx.c:193:23: warning: no previous prototype for ‘ipxitf_find_using_net’ [-Wmissing-prototypes] net/ipx/af_ipx.c:577:5: warning: no previous prototype for ‘ipxitf_send’ [-Wmissing-prototypes] net/ipx/af_ipx.c:1219:8: warning: no previous prototype for ‘ipx_cksum’ [-Wmissing-prototypes] Signed-off-by: Rashika Kheria Reviewed-by: Josh Triplett Signed-off-by: David S. Miller --- include/net/ipx.h | 3 +++ net/ipx/ipx_route.c | 4 ---- 2 files changed, 3 insertions(+), 4 deletions(-) diff --git a/include/net/ipx.h b/include/net/ipx.h index 9e9e35465baf..75466acdce21 100644 --- a/include/net/ipx.h +++ b/include/net/ipx.h @@ -140,6 +140,9 @@ static __inline__ void ipxitf_hold(struct ipx_interface *intrfc) } void ipxitf_down(struct ipx_interface *intrfc); +struct ipx_interface *ipxitf_find_using_net(__be32 net); +int ipxitf_send(struct ipx_interface *intrfc, struct sk_buff *skb, char *node); +__be16 ipx_cksum(struct ipxhdr *packet, int length); static __inline__ void ipxitf_put(struct ipx_interface *intrfc) { diff --git a/net/ipx/ipx_route.c b/net/ipx/ipx_route.c index 30f4519b092f..c1f03185c5e1 100644 --- a/net/ipx/ipx_route.c +++ b/net/ipx/ipx_route.c @@ -20,15 +20,11 @@ DEFINE_RWLOCK(ipx_routes_lock); extern struct ipx_interface *ipx_internal_net; -extern __be16 ipx_cksum(struct ipxhdr *packet, int length); extern struct ipx_interface *ipxitf_find_using_net(__be32 net); extern int ipxitf_demux_socket(struct ipx_interface *intrfc, struct sk_buff *skb, int copy); extern int ipxitf_demux_socket(struct ipx_interface *intrfc, struct sk_buff *skb, int copy); -extern int ipxitf_send(struct ipx_interface *intrfc, struct sk_buff *skb, - char *node); -extern struct ipx_interface *ipxitf_find_using_net(__be32 net); struct ipx_route *ipxrtr_lookup(__be32 net) { From 578efbc19f468686ad7eb1e505bab56d04e92784 Mon Sep 17 00:00:00 2001 From: Rashika Kheria Date: Sun, 9 Feb 2014 22:26:32 +0530 Subject: [PATCH 429/596] net: Move prototype declaration to header file include/net/ipx.h from net/ipx/af_ipx.c MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Move prototype declaration of functions to header file include/net/ipx.h from net/ipx/af_ipx.c because they are used by more than one file. This eliminates the following warning in net/ipx/ipx_route.c:33:19: warning: no previous prototype for ‘ipxrtr_lookup’ [-Wmissing-prototypes] net/ipx/ipx_route.c:52:5: warning: no previous prototype for ‘ipxrtr_add_route’ [-Wmissing-prototypes] net/ipx/ipx_route.c:94:6: warning: no previous prototype for ‘ipxrtr_del_routes’ [-Wmissing-prototypes] net/ipx/ipx_route.c:149:5: warning: no previous prototype for ‘ipxrtr_route_skb’ [-Wmissing-prototypes] net/ipx/ipx_route.c:171:5: warning: no previous prototype for ‘ipxrtr_route_packet’ [-Wmissing-prototypes] net/ipx/ipx_route.c:261:5: warning: no previous prototype for ‘ipxrtr_ioctl’ [-Wmissing-prototypes] Signed-off-by: Rashika Kheria Reviewed-by: Josh Triplett Signed-off-by: David S. Miller --- include/net/ipx.h | 8 ++++++++ net/ipx/af_ipx.c | 9 --------- 2 files changed, 8 insertions(+), 9 deletions(-) diff --git a/include/net/ipx.h b/include/net/ipx.h index 75466acdce21..0143180fecc9 100644 --- a/include/net/ipx.h +++ b/include/net/ipx.h @@ -143,6 +143,14 @@ void ipxitf_down(struct ipx_interface *intrfc); struct ipx_interface *ipxitf_find_using_net(__be32 net); int ipxitf_send(struct ipx_interface *intrfc, struct sk_buff *skb, char *node); __be16 ipx_cksum(struct ipxhdr *packet, int length); +int ipxrtr_add_route(__be32 network, struct ipx_interface *intrfc, + unsigned char *node); +void ipxrtr_del_routes(struct ipx_interface *intrfc); +int ipxrtr_route_packet(struct sock *sk, struct sockaddr_ipx *usipx, + struct iovec *iov, size_t len, int noblock); +int ipxrtr_route_skb(struct sk_buff *skb); +struct ipx_route *ipxrtr_lookup(__be32 net); +int ipxrtr_ioctl(unsigned int cmd, void __user *arg); static __inline__ void ipxitf_put(struct ipx_interface *intrfc) { diff --git a/net/ipx/af_ipx.c b/net/ipx/af_ipx.c index 994e28bfb32e..e5a00a9b52f3 100644 --- a/net/ipx/af_ipx.c +++ b/net/ipx/af_ipx.c @@ -84,15 +84,6 @@ DEFINE_SPINLOCK(ipx_interfaces_lock); struct ipx_interface *ipx_primary_net; struct ipx_interface *ipx_internal_net; -extern int ipxrtr_add_route(__be32 network, struct ipx_interface *intrfc, - unsigned char *node); -extern void ipxrtr_del_routes(struct ipx_interface *intrfc); -extern int ipxrtr_route_packet(struct sock *sk, struct sockaddr_ipx *usipx, - struct iovec *iov, size_t len, int noblock); -extern int ipxrtr_route_skb(struct sk_buff *skb); -extern struct ipx_route *ipxrtr_lookup(__be32 net); -extern int ipxrtr_ioctl(unsigned int cmd, void __user *arg); - struct ipx_interface *ipx_interfaces_head(void) { struct ipx_interface *rc = NULL; From 7780d8ae4ae1960ef7c0570de0a1ecd7b60c8152 Mon Sep 17 00:00:00 2001 From: Rashika Kheria Date: Sun, 9 Feb 2014 22:27:41 +0530 Subject: [PATCH 430/596] net: Move prototype declaration to header file include/net/datalink.h from net/ipx/af_ipx.c MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Move prototype declarations of function to header file include/net/datalink.h from net/ipx/af_ipx.c because they are used by more than one file. This eliminates the following warning in net/ipx/pe2.c: net/ipx/pe2.c:20:24: warning: no previous prototype for ‘make_EII_client’ [-Wmissing-prototypes] net/ipx/pe2.c:32:6: warning: no previous prototype for ‘destroy_EII_client’ [-Wmissing-prototypes] Signed-off-by: Rashika Kheria Reviewed-by: Josh Triplett Signed-off-by: David S. Miller --- include/net/datalink.h | 2 ++ net/ipx/af_ipx.c | 4 +--- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/include/net/datalink.h b/include/net/datalink.h index deb7ca75db48..93cb18f729b5 100644 --- a/include/net/datalink.h +++ b/include/net/datalink.h @@ -15,4 +15,6 @@ struct datalink_proto { struct list_head node; }; +struct datalink_proto *make_EII_client(void); +void destroy_EII_client(struct datalink_proto *dl); #endif diff --git a/net/ipx/af_ipx.c b/net/ipx/af_ipx.c index e5a00a9b52f3..224d05856b85 100644 --- a/net/ipx/af_ipx.c +++ b/net/ipx/af_ipx.c @@ -52,6 +52,7 @@ #include #include #include +#include #include #include @@ -1977,9 +1978,6 @@ static struct notifier_block ipx_dev_notifier = { .notifier_call = ipxitf_device_event, }; -extern struct datalink_proto *make_EII_client(void); -extern void destroy_EII_client(struct datalink_proto *); - static const unsigned char ipx_8022_type = 0xE0; static const unsigned char ipx_snap_id[5] = { 0x0, 0x0, 0x0, 0x81, 0x37 }; static const char ipx_EII_err_msg[] __initconst = From 535d3ae9c808e6a5248f0524dbc7a9e997cf3288 Mon Sep 17 00:00:00 2001 From: Rashika Kheria Date: Sun, 9 Feb 2014 22:29:14 +0530 Subject: [PATCH 431/596] net: Move prototype declaration to header file include/net/net_namespace.h from net/ipx/af_ipx.c MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Move prototype declaration of function to header file include/net/net_namespace.h from net/ipx/af_ipx.c because they are used by more than one file. This eliminates the following warning in net/ipx/sysctl_net_ipx.c: net/ipx/sysctl_net_ipx.c:33:6: warning: no previous prototype for ‘ipx_register_sysctl’ [-Wmissing-prototypes] net/ipx/sysctl_net_ipx.c:38:6: warning: no previous prototype for ‘ipx_unregister_sysctl’ [-Wmissing-prototypes] Signed-off-by: Rashika Kheria Signed-off-by: David S. Miller --- include/net/net_namespace.h | 8 ++++++++ net/ipx/af_ipx.c | 9 +-------- 2 files changed, 9 insertions(+), 8 deletions(-) diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h index da68c9a90ac5..991dcd94cbbf 100644 --- a/include/net/net_namespace.h +++ b/include/net/net_namespace.h @@ -162,6 +162,14 @@ extern struct list_head net_namespace_list; struct net *get_net_ns_by_pid(pid_t pid); struct net *get_net_ns_by_fd(int pid); +#ifdef CONFIG_SYSCTL +void ipx_register_sysctl(void); +void ipx_unregister_sysctl(void); +#else +#define ipx_register_sysctl() +#define ipx_unregister_sysctl() +#endif + #ifdef CONFIG_NET_NS void __put_net(struct net *net); diff --git a/net/ipx/af_ipx.c b/net/ipx/af_ipx.c index 224d05856b85..00b2a6d1c009 100644 --- a/net/ipx/af_ipx.c +++ b/net/ipx/af_ipx.c @@ -54,17 +54,10 @@ #include #include #include +#include #include -#ifdef CONFIG_SYSCTL -extern void ipx_register_sysctl(void); -extern void ipx_unregister_sysctl(void); -#else -#define ipx_register_sysctl() -#define ipx_unregister_sysctl() -#endif - /* Configuration Variables */ static unsigned char ipxcfg_max_hops = 16; static char ipxcfg_auto_select_primary; From bd76ed36bac50b5402fd09c3fc2f368ef324ffe2 Mon Sep 17 00:00:00 2001 From: Rashika Kheria Date: Sun, 9 Feb 2014 22:31:42 +0530 Subject: [PATCH 432/596] net: Include appropriate header file in netfilter/nft_lookup.c MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Include appropriate header file net/netfilter/nf_tables_core.h in net/netfilter/nft_lookup.c because it has prototype declaration of functions defined in net/netfilter/nft_lookup.c. This eliminates the following warning in net/netfilter/nft_lookup.c: net/netfilter/nft_lookup.c:133:12: warning: no previous prototype for ‘nft_lookup_module_init’ [-Wmissing-prototypes] net/netfilter/nft_lookup.c:138:6: warning: no previous prototype for ‘nft_lookup_module_exit’ [-Wmissing-prototypes] Signed-off-by: Rashika Kheria Reviewed-by: Josh Triplett Signed-off-by: David S. Miller --- net/netfilter/nft_lookup.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/netfilter/nft_lookup.c b/net/netfilter/nft_lookup.c index 8a6116b75b5a..bb4ef4cccb6e 100644 --- a/net/netfilter/nft_lookup.c +++ b/net/netfilter/nft_lookup.c @@ -16,6 +16,7 @@ #include #include #include +#include struct nft_lookup { struct nft_set *set; From e1d83ee673eff5407255ba9e884312219f6832d7 Mon Sep 17 00:00:00 2001 From: Rashika Kheria Date: Sun, 9 Feb 2014 22:33:03 +0530 Subject: [PATCH 433/596] net: Mark functions as static in net/sunrpc/svc_xprt.c MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Mark functions as static in net/sunrpc/svc_xprt.c because they are not used outside this file. This eliminates the following warning in net/sunrpc/svc_xprt.c: net/sunrpc/svc_xprt.c:574:5: warning: no previous prototype for ‘svc_alloc_arg’ [-Wmissing-prototypes] net/sunrpc/svc_xprt.c:615:18: warning: no previous prototype for ‘svc_get_next_xprt’ [-Wmissing-prototypes] net/sunrpc/svc_xprt.c:694:6: warning: no previous prototype for ‘svc_add_new_temp_xprt’ [-Wmissing-prototypes] Signed-off-by: Rashika Kheria Reviewed-by: Josh Triplett Signed-off-by: David S. Miller --- net/sunrpc/svc_xprt.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c index 80a6640f329b..06c6ff0cb911 100644 --- a/net/sunrpc/svc_xprt.c +++ b/net/sunrpc/svc_xprt.c @@ -571,7 +571,7 @@ static void svc_check_conn_limits(struct svc_serv *serv) } } -int svc_alloc_arg(struct svc_rqst *rqstp) +static int svc_alloc_arg(struct svc_rqst *rqstp) { struct svc_serv *serv = rqstp->rq_server; struct xdr_buf *arg; @@ -612,7 +612,7 @@ int svc_alloc_arg(struct svc_rqst *rqstp) return 0; } -struct svc_xprt *svc_get_next_xprt(struct svc_rqst *rqstp, long timeout) +static struct svc_xprt *svc_get_next_xprt(struct svc_rqst *rqstp, long timeout) { struct svc_xprt *xprt; struct svc_pool *pool = rqstp->rq_pool; @@ -691,7 +691,7 @@ struct svc_xprt *svc_get_next_xprt(struct svc_rqst *rqstp, long timeout) return xprt; } -void svc_add_new_temp_xprt(struct svc_serv *serv, struct svc_xprt *newxpt) +static void svc_add_new_temp_xprt(struct svc_serv *serv, struct svc_xprt *newxpt) { spin_lock_bh(&serv->sv_lock); set_bit(XPT_TEMP, &newxpt->xpt_flags); From d94c1f92bbf2c42c5febd68bbea51fffeac90834 Mon Sep 17 00:00:00 2001 From: FX Le Bail Date: Fri, 7 Feb 2014 11:22:37 +0100 Subject: [PATCH 434/596] ipv6: icmp6_send: fix Oops when pinging a not set up IPv6 peer on a sit tunnel The patch 446fab59333dea91e54688f033dd8d788d0486fb ("ipv6: enable anycast addresses as source addresses in ICMPv6 error messages") causes an Oops when pinging a not set up IPv6 peer on a sit tunnel. The problem is that ipv6_anycast_destination() uses unconditionally skb_dst(skb), which is NULL in this case. The solution is to use instead the ipv6_chk_acast_addr_src() function. Here are the steps to reproduce it: modprobe sit ip link add sit1 type sit remote 10.16.0.121 local 10.16.0.249 ip l s sit1 up ip -6 a a dev sit1 2001:1234::123 remote 2001:1234::121 ping6 2001:1234::121 Reported-by: Nicolas Dichtel Tested-by: Nicolas Dichtel Signed-off-by: Francois-Xavier Le Bail Signed-off-by: David S. Miller --- net/ipv6/icmp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index f81f59686f21..f2610e157660 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -414,7 +414,7 @@ static void icmp6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info) addr_type = ipv6_addr_type(&hdr->daddr); if (ipv6_chk_addr(net, &hdr->daddr, skb->dev, 0) || - ipv6_anycast_destination(skb)) + ipv6_chk_acast_addr_src(net, skb->dev, &hdr->daddr)) saddr = &hdr->daddr; /* From b28a960c42fcd9cfc987441fa6d1c1a471f0f9ed Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 9 Feb 2014 18:15:47 -0800 Subject: [PATCH 435/596] Linux 3.14-rc2 --- Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Makefile b/Makefile index 606ef7c4a544..933e1def6baf 100644 --- a/Makefile +++ b/Makefile @@ -1,7 +1,7 @@ VERSION = 3 PATCHLEVEL = 14 SUBLEVEL = 0 -EXTRAVERSION = -rc1 +EXTRAVERSION = -rc2 NAME = Shuffling Zombie Juror # *DOCUMENTATION* From 310f6fd0ca9de8c1fc6399af494855e15157aa35 Mon Sep 17 00:00:00 2001 From: Christian Engelmayer Date: Fri, 7 Feb 2014 22:58:38 +0100 Subject: [PATCH 436/596] 6lowpan: Remove unused pointer in lowpan_header_create() Commit 8df8c56a (6lowpan: Moving generic compression code into 6lowpan_iphc.c) left pointer 'hdr' unused - remove it. Detected by Coverity: CID 1164868. Signed-off-by: Christian Engelmayer Signed-off-by: David S. Miller --- net/ieee802154/6lowpan.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/net/ieee802154/6lowpan.c b/net/ieee802154/6lowpan.c index 48b25c0af4d0..8bfb40153fe7 100644 --- a/net/ieee802154/6lowpan.c +++ b/net/ieee802154/6lowpan.c @@ -106,7 +106,6 @@ static int lowpan_header_create(struct sk_buff *skb, unsigned short type, const void *_daddr, const void *_saddr, unsigned int len) { - struct ipv6hdr *hdr; const u8 *saddr = _saddr; const u8 *daddr = _daddr; struct ieee802154_addr sa, da; @@ -117,8 +116,6 @@ static int lowpan_header_create(struct sk_buff *skb, if (type != ETH_P_IPV6) return 0; - hdr = ipv6_hdr(skb); - if (!saddr) saddr = dev->dev_addr; From a88f92b1283a485b6096b6fba3f84e266df026fc Mon Sep 17 00:00:00 2001 From: Christian Engelmayer Date: Sat, 8 Feb 2014 00:21:10 +0100 Subject: [PATCH 437/596] wan: dlci: Remove unused netdev_priv pointer Remove occurrences of unused pointer to network device private data in functions dlci_header() and dlci_receive(). Detected by Coverity: CID 139844, CID 139845. Signed-off-by: Christian Engelmayer Signed-off-by: David S. Miller --- drivers/net/wan/dlci.c | 5 ----- 1 file changed, 5 deletions(-) diff --git a/drivers/net/wan/dlci.c b/drivers/net/wan/dlci.c index 0d1c7592efa0..19f7cb2cdef3 100644 --- a/drivers/net/wan/dlci.c +++ b/drivers/net/wan/dlci.c @@ -71,12 +71,9 @@ static int dlci_header(struct sk_buff *skb, struct net_device *dev, const void *saddr, unsigned len) { struct frhdr hdr; - struct dlci_local *dlp; unsigned int hlen; char *dest; - dlp = netdev_priv(dev); - hdr.control = FRAD_I_UI; switch (type) { @@ -107,11 +104,9 @@ static int dlci_header(struct sk_buff *skb, struct net_device *dev, static void dlci_receive(struct sk_buff *skb, struct net_device *dev) { - struct dlci_local *dlp; struct frhdr *hdr; int process, header; - dlp = netdev_priv(dev); if (!pskb_may_pull(skb, sizeof(*hdr))) { netdev_notice(dev, "invalid data no header\n"); dev->stats.rx_errors++; From 946c032e5a53992ea45e062ecb08670ba39b99e3 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Maciej=20=C5=BBenczykowski?= Date: Fri, 7 Feb 2014 16:23:48 -0800 Subject: [PATCH 438/596] net: fix 'ip rule' iif/oif device rename MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit ip rules with iif/oif references do not update: (detach/attach) across interface renames. Signed-off-by: Maciej Żenczykowski CC: Willem de Bruijn CC: Eric Dumazet CC: Chris Davis CC: Carlo Contavalli Google-Bug-Id: 12936021 Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/fib_rules.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c index f409e0bd35c0..185c341fafbd 100644 --- a/net/core/fib_rules.c +++ b/net/core/fib_rules.c @@ -745,6 +745,13 @@ static int fib_rules_event(struct notifier_block *this, unsigned long event, attach_rules(&ops->rules_list, dev); break; + case NETDEV_CHANGENAME: + list_for_each_entry(ops, &net->rules_ops, list) { + detach_rules(&ops->rules_list, dev); + attach_rules(&ops->rules_list, dev); + } + break; + case NETDEV_UNREGISTER: list_for_each_entry(ops, &net->rules_ops, list) detach_rules(&ops->rules_list, dev); From 4e6ecd442db707cd29cae1774dc951855d82df5e Mon Sep 17 00:00:00 2001 From: Christian Engelmayer Date: Sat, 8 Feb 2014 18:11:17 +0100 Subject: [PATCH 439/596] 3c59x: Remove unused pointer in vortex_eisa_cleanup() Remove unused network device private data pointer 'vp' in function vortex_eisa_cleanup(). Detected by Coverity: CID 139826. Signed-off-by: Christian Engelmayer Signed-off-by: David S. Miller --- drivers/net/ethernet/3com/3c59x.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/drivers/net/ethernet/3com/3c59x.c b/drivers/net/ethernet/3com/3c59x.c index 0f4241c6e97e..238ccea965c8 100644 --- a/drivers/net/ethernet/3com/3c59x.c +++ b/drivers/net/ethernet/3com/3c59x.c @@ -3294,7 +3294,6 @@ static int __init vortex_init(void) static void __exit vortex_eisa_cleanup(void) { - struct vortex_private *vp; void __iomem *ioaddr; #ifdef CONFIG_EISA @@ -3303,7 +3302,6 @@ static void __exit vortex_eisa_cleanup(void) #endif if (compaq_net_device) { - vp = netdev_priv(compaq_net_device); ioaddr = ioport_map(compaq_net_device->base_addr, VORTEX_TOTAL_SIZE); From 7653aabfbdc73c1567e29a9790701f5898ba1420 Mon Sep 17 00:00:00 2001 From: Raymond Wanyoike Date: Sun, 9 Feb 2014 00:01:02 +0300 Subject: [PATCH 440/596] net: qmi_wwan: add ZTE MF667 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The driver description files give these descriptions to the vendor specific ports on this modem: VID_19D2&PID_1270&MI_00: "ZTE MF667 Diagnostics Port" VID_19D2&PID_1270&MI_01: "ZTE MF667 AT Port" VID_19D2&PID_1270&MI_02: "ZTE MF667 ATExt2 Port" VID_19D2&PID_1270&MI_03: "ZTE MF667 ATExt Port" VID_19D2&PID_1270&MI_04: "ZTE MF667 USB Modem" VID_19D2&PID_1270&MI_05: "ZTE MF667 Network Adapter" Signed-off-by: Raymond Wanyoike Acked-by: Bjørn Mork Signed-off-by: David S. Miller --- drivers/net/usb/qmi_wwan.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/usb/qmi_wwan.c b/drivers/net/usb/qmi_wwan.c index ef4a45aea5aa..ff5c87128ffe 100644 --- a/drivers/net/usb/qmi_wwan.c +++ b/drivers/net/usb/qmi_wwan.c @@ -712,6 +712,7 @@ static const struct usb_device_id products[] = { {QMI_FIXED_INTF(0x19d2, 0x1255, 3)}, {QMI_FIXED_INTF(0x19d2, 0x1255, 4)}, {QMI_FIXED_INTF(0x19d2, 0x1256, 4)}, + {QMI_FIXED_INTF(0x19d2, 0x1270, 5)}, /* ZTE MF667 */ {QMI_FIXED_INTF(0x19d2, 0x1401, 2)}, {QMI_FIXED_INTF(0x19d2, 0x1402, 2)}, /* ZTE MF60 */ {QMI_FIXED_INTF(0x19d2, 0x1424, 2)}, From 0fbb8297b5468eb7d269dc868b80df1fb166a4ab Mon Sep 17 00:00:00 2001 From: Christian Engelmayer Date: Sun, 9 Feb 2014 00:16:17 +0100 Subject: [PATCH 441/596] net: vxge: Remove unused device pointer Remove occurrences of unused struct __vxge_hw_device pointer in functions vxge_learn_mac() and vxge_rem_isr(). Detected by Coverity: CID 139839, CID 139842. Signed-off-by: Christian Engelmayer Reviewed-by: Jingoo Han Signed-off-by: David S. Miller --- drivers/net/ethernet/neterion/vxge/vxge-main.c | 6 ------ 1 file changed, 6 deletions(-) diff --git a/drivers/net/ethernet/neterion/vxge/vxge-main.c b/drivers/net/ethernet/neterion/vxge/vxge-main.c index 1ded50ca1600..e46e8698e630 100644 --- a/drivers/net/ethernet/neterion/vxge/vxge-main.c +++ b/drivers/net/ethernet/neterion/vxge/vxge-main.c @@ -726,9 +726,6 @@ static int vxge_learn_mac(struct vxgedev *vdev, u8 *mac_header) int vpath_idx = 0; enum vxge_hw_status status = VXGE_HW_OK; struct vxge_vpath *vpath = NULL; - struct __vxge_hw_device *hldev; - - hldev = pci_get_drvdata(vdev->pdev); mac_address = (u8 *)&mac_addr; memcpy(mac_address, mac_header, ETH_ALEN); @@ -2443,9 +2440,6 @@ static void vxge_rem_msix_isr(struct vxgedev *vdev) static void vxge_rem_isr(struct vxgedev *vdev) { - struct __vxge_hw_device *hldev; - hldev = pci_get_drvdata(vdev->pdev); - #ifdef CONFIG_PCI_MSI if (vdev->config.intr_type == MSI_X) { vxge_rem_msix_isr(vdev); From b10bd54c05a6c3c96549f4642d7de23c99b9853b Mon Sep 17 00:00:00 2001 From: Jesper Juhl Date: Sun, 9 Feb 2014 23:30:32 +0100 Subject: [PATCH 442/596] tcp: correct code comment stating 3 min timeout for FIN_WAIT2, we only do 1 min As far as I can tell we have used a default of 60 seconds for FIN_WAIT2 timeout for ages (since 2.x times??). In any case, the timeout these days is 60 seconds, so the 3 min comment is wrong (and cost me a few minutes of my life when I was debugging a FIN_WAIT2 related problem in a userspace application and checked the kernel source for details). Signed-off-by: Jesper Juhl Signed-off-by: David S. Miller --- net/ipv4/tcp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 4475b3bb494d..9f3a2db9109e 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -2229,7 +2229,7 @@ void tcp_close(struct sock *sk, long timeout) /* This is a (useful) BSD violating of the RFC. There is a * problem with TCP as specified in that the other end could * keep a socket open forever with no application left this end. - * We use a 3 minute timeout (about the same as BSD) then kill + * We use a 1 minute timeout (about the same as BSD) then kill * our end. If they send after that then tough - BUT: long enough * that we won't make the old 4*rto = almost no time - whoops * reset mistake. From ac1566ece497b2a37667cc5d4258edb0bf3b6836 Mon Sep 17 00:00:00 2001 From: Michal Simek Date: Thu, 30 Jan 2014 15:25:37 +0100 Subject: [PATCH 443/596] microblaze: Fix missing HZ macro Add missing param.h header because of HZ macro. It is causing compilation failure: In file included from include/linux/delay.h:14:0, from drivers/clk/qcom/reset.c:18: drivers/clk/qcom/reset.c: In function 'qcom_reset': arch/microblaze/include/asm/delay.h:39:35: error: 'HZ' undeclared (first use in this function) Signed-off-by: Michal Simek --- arch/microblaze/include/asm/delay.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/microblaze/include/asm/delay.h b/arch/microblaze/include/asm/delay.h index 05b7d39e4391..66fc24c24238 100644 --- a/arch/microblaze/include/asm/delay.h +++ b/arch/microblaze/include/asm/delay.h @@ -13,6 +13,8 @@ #ifndef _ASM_MICROBLAZE_DELAY_H #define _ASM_MICROBLAZE_DELAY_H +#include + extern inline void __delay(unsigned long loops) { asm volatile ("# __delay \n\t" \ From 4f3cbd79afd99cc5143b2beb126e7425af7f1ff9 Mon Sep 17 00:00:00 2001 From: Michal Simek Date: Thu, 30 Jan 2014 15:27:26 +0100 Subject: [PATCH 444/596] microblaze: Define readq and writeq IO helper function Some drivers check if readq/writeq functions are defined. If not internal driver functions are used which cause compilation failures. Signed-off-by: Michal Simek --- arch/microblaze/include/asm/io.h | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/arch/microblaze/include/asm/io.h b/arch/microblaze/include/asm/io.h index a2cea7206077..3fbb7f1db3bc 100644 --- a/arch/microblaze/include/asm/io.h +++ b/arch/microblaze/include/asm/io.h @@ -89,6 +89,11 @@ static inline unsigned int readl(const volatile void __iomem *addr) { return le32_to_cpu(*(volatile unsigned int __force *)addr); } +#define readq readq +static inline u64 readq(const volatile void __iomem *addr) +{ + return le64_to_cpu(__raw_readq(addr)); +} static inline void writeb(unsigned char v, volatile void __iomem *addr) { *(volatile unsigned char __force *)addr = v; @@ -101,6 +106,7 @@ static inline void writel(unsigned int v, volatile void __iomem *addr) { *(volatile unsigned int __force *)addr = cpu_to_le32(v); } +#define writeq(b, addr) __raw_writeq(cpu_to_le64(b), addr) /* ioread and iowrite variants. thease are for now same as __raw_ * variants of accessors. we might check for endianess in the feature From a49f56eec54d864ba0fda838e4c8bf5c72f3eb08 Mon Sep 17 00:00:00 2001 From: "Edgar E. Iglesias" Date: Sat, 8 Feb 2014 02:31:31 +0000 Subject: [PATCH 445/596] microblaze: Fix a typo when disabling stack protection Correct a typo causing the stack protector to be left enabled. 0xFFFFFFF -> 0xFFFFFFFF Signed-off-by: Edgar E. Iglesias Tested-by: Guenter Roeck Signed-off-by: Michal Simek --- arch/microblaze/kernel/head.S | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/microblaze/kernel/head.S b/arch/microblaze/kernel/head.S index b7fb0438458c..17645b2e2f07 100644 --- a/arch/microblaze/kernel/head.S +++ b/arch/microblaze/kernel/head.S @@ -66,7 +66,7 @@ real_start: mts rmsr, r0 /* Disable stack protection from bootloader */ mts rslr, r0 - addi r8, r0, 0xFFFFFFF + addi r8, r0, 0xFFFFFFFF mts rshr, r8 /* * According to Xilinx, msrclr instruction behaves like 'mfs rX,rpc' From ef8e39b51f24c8ac907e1f1995910ee3083375e0 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Mon, 10 Feb 2014 11:43:41 +0100 Subject: [PATCH 446/596] ALSA: hda - Fix undefined symbol due to builtin/module mixup Even after the fix for leftover kconfig handling (commit f8f1becf), the current code still doesn't handle properly the builtin/module mixup case between the core snd-hda-codec and other codec drivers. For example, when CONFIG_SND_HDA_INTEL=y and CONFIG_SND_HDA_CODEC_HDMI=m, it'll end up with an unresolved symbol snd_hda_parse_hdmi_codec. This patch fixes the issue. Now codec->parser points to the parser object *only* when a module (either generic or HDMI parser) is loaded and bound. When a builtin symbol is used, codec->parser still points to NULL. This is the difference from the previous versions. Fixes: f8f1becfa4ac ('ALSA: hda - Fix leftover ifdef checks after modularization') Reported-by: Fengguang Wu Signed-off-by: Takashi Iwai --- sound/pci/hda/hda_codec.c | 32 ++++++++++++++++---------------- 1 file changed, 16 insertions(+), 16 deletions(-) diff --git a/sound/pci/hda/hda_codec.c b/sound/pci/hda/hda_codec.c index 69acbf758bbb..dafcf82139e2 100644 --- a/sound/pci/hda/hda_codec.c +++ b/sound/pci/hda/hda_codec.c @@ -1339,23 +1339,15 @@ get_hda_cvt_setup(struct hda_codec *codec, hda_nid_t nid) /* * Dynamic symbol binding for the codec parsers */ -#ifdef MODULE -#define load_parser_sym(sym) ((int (*)(struct hda_codec *))symbol_request(sym)) -#define unload_parser_addr(addr) symbol_put_addr(addr) -#else -#define load_parser_sym(sym) (sym) -#define unload_parser_addr(addr) do {} while (0) -#endif #define load_parser(codec, sym) \ - ((codec)->parser = load_parser_sym(sym)) + ((codec)->parser = (int (*)(struct hda_codec *))symbol_request(sym)) static void unload_parser(struct hda_codec *codec) { - if (codec->parser) { - unload_parser_addr(codec->parser); - codec->parser = NULL; - } + if (codec->parser) + symbol_put_addr(codec->parser); + codec->parser = NULL; } /* @@ -1620,12 +1612,20 @@ int snd_hda_codec_configure(struct hda_codec *codec) patch = codec->preset->patch; if (!patch) { unload_parser(codec); /* to be sure */ - if (is_likely_hdmi_codec(codec)) + if (is_likely_hdmi_codec(codec)) { +#if IS_MODULE(CONFIG_SND_HDA_CODEC_HDMI) patch = load_parser(codec, snd_hda_parse_hdmi_codec); -#if IS_ENABLED(CONFIG_SND_HDA_GENERIC) - if (!patch) - patch = load_parser(codec, snd_hda_parse_generic_codec); +#elif IS_BUILTIN(CONFIG_SND_HDA_CODEC_HDMI) + patch = snd_hda_parse_hdmi_codec; #endif + } + if (!patch) { +#if IS_MODULE(CONFIG_SND_HDA_GENERIC) + patch = load_parser(codec, snd_hda_parse_generic_codec); +#elif IS_BUILTIN(CONFIG_SND_HDA_GENERIC) + patch = snd_hda_parse_generic_codec; +#endif + } if (!patch) { printk(KERN_ERR "hda-codec: No codec parser is available\n"); return -ENODEV; From 0a8e5c3d5f0f4929761e6a5bef5358f0ccd8810c Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Mon, 10 Feb 2014 13:44:20 +0100 Subject: [PATCH 447/596] ACPI / dock: Use acpi_device_enumerated() to check if dock is present After commit 202317a573b2 (ACPI / scan: Add acpi_device objects for all device nodes in the namespace) acpi_bus_get_device() will always return 0 for dock devices in dock_notify(), so the dock station docking code under ACPI_NOTIFY_DEVICE_CHECK will never be executed and docking will not work as a result of that. Fix the problem by making dock_notify() use acpi_device_enumerated() to check the presence of the device instead of checking the return value of acpi_bus_get_device(). Fixes: 202317a573b2 (ACPI / scan: Add acpi_device objects for all device nodes in the namespace) Signed-off-by: Rafael J. Wysocki --- drivers/acpi/dock.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/acpi/dock.c b/drivers/acpi/dock.c index c431c88faaff..e9b3081c4fe9 100644 --- a/drivers/acpi/dock.c +++ b/drivers/acpi/dock.c @@ -609,7 +609,7 @@ static int handle_eject_request(struct dock_station *ds, u32 event) static void dock_notify(struct dock_station *ds, u32 event) { acpi_handle handle = ds->handle; - struct acpi_device *ad; + struct acpi_device *adev = NULL; int surprise_removal = 0; /* @@ -632,7 +632,8 @@ static void dock_notify(struct dock_station *ds, u32 event) switch (event) { case ACPI_NOTIFY_BUS_CHECK: case ACPI_NOTIFY_DEVICE_CHECK: - if (!dock_in_progress(ds) && acpi_bus_get_device(handle, &ad)) { + acpi_bus_get_device(handle, &adev); + if (!dock_in_progress(ds) && !acpi_device_enumerated(adev)) { begin_dock(ds); dock(ds); if (!dock_present(ds)) { From 30a91cb4ef385fe1b260df204ef314d86fff2850 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 10 Feb 2014 03:24:38 -0800 Subject: [PATCH 448/596] blk-mq: rework I/O completions Rework I/O completions to work more like the old code path. blk_mq_end_io now stays out of the business of deferring completions to others CPUs and calling blk_mark_rq_complete. The latter is very important to allow completing requests that have timed out and thus are already marked completed, the former allows using the IPI callout even for driver specific completions instead of having to reimplement them. Signed-off-by: Christoph Hellwig Signed-off-by: Jens Axboe --- block/blk-mq.c | 52 +++++++++++++++++++++++++----------------- block/blk-mq.h | 3 +-- block/blk-timeout.c | 2 +- include/linux/blk-mq.h | 4 ++++ 4 files changed, 37 insertions(+), 24 deletions(-) diff --git a/block/blk-mq.c b/block/blk-mq.c index cee96234bf58..14c8f35946e1 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -326,7 +326,7 @@ static void blk_mq_bio_endio(struct request *rq, struct bio *bio, int error) bio_endio(bio, error); } -void blk_mq_complete_request(struct request *rq, int error) +void blk_mq_end_io(struct request *rq, int error) { struct bio *bio = rq->bio; unsigned int bytes = 0; @@ -351,46 +351,53 @@ void blk_mq_complete_request(struct request *rq, int error) else blk_mq_free_request(rq); } +EXPORT_SYMBOL(blk_mq_end_io); -void __blk_mq_end_io(struct request *rq, int error) -{ - if (!blk_mark_rq_complete(rq)) - blk_mq_complete_request(rq, error); -} - -static void blk_mq_end_io_remote(void *data) +static void __blk_mq_complete_request_remote(void *data) { struct request *rq = data; - __blk_mq_end_io(rq, rq->errors); + rq->q->softirq_done_fn(rq); } -/* - * End IO on this request on a multiqueue enabled driver. We'll either do - * it directly inline, or punt to a local IPI handler on the matching - * remote CPU. - */ -void blk_mq_end_io(struct request *rq, int error) +void __blk_mq_complete_request(struct request *rq) { struct blk_mq_ctx *ctx = rq->mq_ctx; int cpu; - if (!ctx->ipi_redirect) - return __blk_mq_end_io(rq, error); + if (!ctx->ipi_redirect) { + rq->q->softirq_done_fn(rq); + return; + } cpu = get_cpu(); if (cpu != ctx->cpu && cpu_online(ctx->cpu)) { - rq->errors = error; - rq->csd.func = blk_mq_end_io_remote; + rq->csd.func = __blk_mq_complete_request_remote; rq->csd.info = rq; rq->csd.flags = 0; __smp_call_function_single(ctx->cpu, &rq->csd, 0); } else { - __blk_mq_end_io(rq, error); + rq->q->softirq_done_fn(rq); } put_cpu(); } -EXPORT_SYMBOL(blk_mq_end_io); + +/** + * blk_mq_complete_request - end I/O on a request + * @rq: the request being processed + * + * Description: + * Ends all I/O on a request. It does not handle partial completions. + * The actual completion happens out-of-order, through a IPI handler. + **/ +void blk_mq_complete_request(struct request *rq) +{ + if (unlikely(blk_should_fake_timeout(rq->q))) + return; + if (!blk_mark_rq_complete(rq)) + __blk_mq_complete_request(rq); +} +EXPORT_SYMBOL(blk_mq_complete_request); static void blk_mq_start_request(struct request *rq) { @@ -1399,6 +1406,9 @@ struct request_queue *blk_mq_init_queue(struct blk_mq_reg *reg, if (reg->timeout) blk_queue_rq_timeout(q, reg->timeout); + if (reg->ops->complete) + blk_queue_softirq_done(q, reg->ops->complete); + blk_mq_init_flush(q); blk_mq_init_cpu_queues(q, reg->nr_hw_queues); diff --git a/block/blk-mq.h b/block/blk-mq.h index 5c3917984b00..f29b645f0e1c 100644 --- a/block/blk-mq.h +++ b/block/blk-mq.h @@ -22,8 +22,7 @@ struct blk_mq_ctx { struct kobject kobj; }; -void __blk_mq_end_io(struct request *rq, int error); -void blk_mq_complete_request(struct request *rq, int error); +void __blk_mq_complete_request(struct request *rq); void blk_mq_run_request(struct request *rq, bool run_queue, bool async); void blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx, bool async); void blk_mq_init_flush(struct request_queue *q); diff --git a/block/blk-timeout.c b/block/blk-timeout.c index bba81c9348e1..d96f7061c6fd 100644 --- a/block/blk-timeout.c +++ b/block/blk-timeout.c @@ -91,7 +91,7 @@ static void blk_rq_timed_out(struct request *req) case BLK_EH_HANDLED: /* Can we use req->errors here? */ if (q->mq_ops) - blk_mq_complete_request(req, req->errors); + __blk_mq_complete_request(req); else __blk_complete_request(req); break; diff --git a/include/linux/blk-mq.h b/include/linux/blk-mq.h index b7638be58599..468be242db90 100644 --- a/include/linux/blk-mq.h +++ b/include/linux/blk-mq.h @@ -86,6 +86,8 @@ struct blk_mq_ops { */ rq_timed_out_fn *timeout; + softirq_done_fn *complete; + /* * Override for hctx allocations (should probably go) */ @@ -137,6 +139,8 @@ void blk_mq_free_single_hw_queue(struct blk_mq_hw_ctx *, unsigned int); void blk_mq_end_io(struct request *rq, int error); +void blk_mq_complete_request(struct request *rq); + void blk_mq_stop_hw_queue(struct blk_mq_hw_ctx *hctx); void blk_mq_start_hw_queue(struct blk_mq_hw_ctx *hctx); void blk_mq_stop_hw_queues(struct request_queue *q); From 5124c285797aa33d5fdb909fbe808a61bcc5eb9d Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 10 Feb 2014 03:24:39 -0800 Subject: [PATCH 449/596] virtio_blk: use blk_mq_complete_request Make sure to complete requests on the submitting CPU. Previously this was done in blk_mq_end_io, but the responsibility shifted to the drivers. Signed-off-by: Christoph Hellwig Signed-off-by: Jens Axboe --- drivers/block/virtio_blk.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/block/virtio_blk.c b/drivers/block/virtio_blk.c index 6a680d4de7f1..b1cb3f4c4db4 100644 --- a/drivers/block/virtio_blk.c +++ b/drivers/block/virtio_blk.c @@ -110,9 +110,9 @@ static int __virtblk_add_req(struct virtqueue *vq, return virtqueue_add_sgs(vq, sgs, num_out, num_in, vbr, GFP_ATOMIC); } -static inline void virtblk_request_done(struct virtblk_req *vbr) +static inline void virtblk_request_done(struct request *req) { - struct request *req = vbr->req; + struct virtblk_req *vbr = req->special; int error = virtblk_result(vbr); if (req->cmd_type == REQ_TYPE_BLOCK_PC) { @@ -138,7 +138,7 @@ static void virtblk_done(struct virtqueue *vq) do { virtqueue_disable_cb(vq); while ((vbr = virtqueue_get_buf(vblk->vq, &len)) != NULL) { - virtblk_request_done(vbr); + blk_mq_complete_request(vbr->req); req_done = true; } if (unlikely(virtqueue_is_broken(vq))) @@ -479,6 +479,7 @@ static struct blk_mq_ops virtio_mq_ops = { .map_queue = blk_mq_map_queue, .alloc_hctx = blk_mq_alloc_single_hw_queue, .free_hctx = blk_mq_free_single_hw_queue, + .complete = virtblk_request_done, }; static struct blk_mq_reg virtio_mq_reg = { From ce2c350b2cfe5b5ca5023a6b1ec4d21821d39add Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 10 Feb 2014 03:24:40 -0800 Subject: [PATCH 450/596] null_blk: use blk_complete_request and blk_mq_complete_request Use the block layer helpers for CPU-local completions instead of reimplementing them locally. Signed-off-by: Christoph Hellwig Signed-off-by: Jens Axboe --- drivers/block/null_blk.c | 97 +++++++++++++--------------------------- 1 file changed, 32 insertions(+), 65 deletions(-) diff --git a/drivers/block/null_blk.c b/drivers/block/null_blk.c index cc801cde5cfa..091b9ea14feb 100644 --- a/drivers/block/null_blk.c +++ b/drivers/block/null_blk.c @@ -60,7 +60,9 @@ enum { NULL_IRQ_NONE = 0, NULL_IRQ_SOFTIRQ = 1, NULL_IRQ_TIMER = 2, +}; +enum { NULL_Q_BIO = 0, NULL_Q_RQ = 1, NULL_Q_MQ = 2, @@ -172,18 +174,20 @@ static struct nullb_cmd *alloc_cmd(struct nullb_queue *nq, int can_wait) static void end_cmd(struct nullb_cmd *cmd) { - if (cmd->rq) { - if (queue_mode == NULL_Q_MQ) - blk_mq_end_io(cmd->rq, 0); - else { - INIT_LIST_HEAD(&cmd->rq->queuelist); - blk_end_request_all(cmd->rq, 0); - } - } else if (cmd->bio) + switch (queue_mode) { + case NULL_Q_MQ: + blk_mq_end_io(cmd->rq, 0); + return; + case NULL_Q_RQ: + INIT_LIST_HEAD(&cmd->rq->queuelist); + blk_end_request_all(cmd->rq, 0); + break; + case NULL_Q_BIO: bio_endio(cmd->bio, 0); + break; + } - if (queue_mode != NULL_Q_MQ) - free_cmd(cmd); + free_cmd(cmd); } static enum hrtimer_restart null_cmd_timer_expired(struct hrtimer *timer) @@ -222,63 +226,32 @@ static void null_cmd_end_timer(struct nullb_cmd *cmd) static void null_softirq_done_fn(struct request *rq) { - blk_end_request_all(rq, 0); + end_cmd(rq->special); } -#ifdef CONFIG_SMP - -static void null_ipi_cmd_end_io(void *data) -{ - struct completion_queue *cq; - struct llist_node *entry, *next; - struct nullb_cmd *cmd; - - cq = &per_cpu(completion_queues, smp_processor_id()); - - entry = llist_del_all(&cq->list); - entry = llist_reverse_order(entry); - - while (entry) { - next = entry->next; - cmd = llist_entry(entry, struct nullb_cmd, ll_list); - end_cmd(cmd); - entry = next; - } -} - -static void null_cmd_end_ipi(struct nullb_cmd *cmd) -{ - struct call_single_data *data = &cmd->csd; - int cpu = get_cpu(); - struct completion_queue *cq = &per_cpu(completion_queues, cpu); - - cmd->ll_list.next = NULL; - - if (llist_add(&cmd->ll_list, &cq->list)) { - data->func = null_ipi_cmd_end_io; - data->flags = 0; - __smp_call_function_single(cpu, data, 0); - } - - put_cpu(); -} - -#endif /* CONFIG_SMP */ - static inline void null_handle_cmd(struct nullb_cmd *cmd) { /* Complete IO by inline, softirq or timer */ switch (irqmode) { + case NULL_IRQ_SOFTIRQ: + switch (queue_mode) { + case NULL_Q_MQ: + blk_mq_complete_request(cmd->rq); + break; + case NULL_Q_RQ: + blk_complete_request(cmd->rq); + break; + case NULL_Q_BIO: + /* + * XXX: no proper submitting cpu information available. + */ + end_cmd(cmd); + break; + } + break; case NULL_IRQ_NONE: end_cmd(cmd); break; - case NULL_IRQ_SOFTIRQ: -#ifdef CONFIG_SMP - null_cmd_end_ipi(cmd); -#else - end_cmd(cmd); -#endif - break; case NULL_IRQ_TIMER: null_cmd_end_timer(cmd); break; @@ -413,6 +386,7 @@ static struct blk_mq_ops null_mq_ops = { .queue_rq = null_queue_rq, .map_queue = blk_mq_map_queue, .init_hctx = null_init_hctx, + .complete = null_softirq_done_fn, }; static struct blk_mq_reg null_mq_reg = { @@ -611,13 +585,6 @@ static int __init null_init(void) { unsigned int i; -#if !defined(CONFIG_SMP) - if (irqmode == NULL_IRQ_SOFTIRQ) { - pr_warn("null_blk: softirq completions not available.\n"); - pr_warn("null_blk: using direct completions.\n"); - irqmode = NULL_IRQ_NONE; - } -#endif if (bs > PAGE_SIZE) { pr_warn("null_blk: invalid block size\n"); pr_warn("null_blk: defaults block size to %lu\n", PAGE_SIZE); From 18741986a4b1dc4b1f171634c4191abc3b0fa023 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 10 Feb 2014 09:29:00 -0700 Subject: [PATCH 451/596] blk-mq: rework flush sequencing logic Witch to using a preallocated flush_rq for blk-mq similar to what's done with the old request path. This allows us to set up the request properly with a tag from the actually allowed range and ->rq_disk as needed by some drivers. To make life easier we also switch to dynamic allocation of ->flush_rq for the old path. This effectively reverts most of "blk-mq: fix for flush deadlock" and "blk-mq: Don't reserve a tag for flush request" Signed-off-by: Christoph Hellwig Signed-off-by: Jens Axboe --- block/blk-core.c | 15 ++++-- block/blk-flush.c | 105 +++++++++++++++-------------------------- block/blk-mq.c | 54 ++++++++------------- block/blk-mq.h | 1 + block/blk-sysfs.c | 2 + include/linux/blk-mq.h | 5 +- include/linux/blkdev.h | 11 ++--- 7 files changed, 76 insertions(+), 117 deletions(-) diff --git a/block/blk-core.c b/block/blk-core.c index 06636f3ad424..853f92749202 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -693,11 +693,20 @@ blk_init_queue_node(request_fn_proc *rfn, spinlock_t *lock, int node_id) if (!uninit_q) return NULL; + uninit_q->flush_rq = kzalloc(sizeof(struct request), GFP_KERNEL); + if (!uninit_q->flush_rq) + goto out_cleanup_queue; + q = blk_init_allocated_queue(uninit_q, rfn, lock); if (!q) - blk_cleanup_queue(uninit_q); - + goto out_free_flush_rq; return q; + +out_free_flush_rq: + kfree(uninit_q->flush_rq); +out_cleanup_queue: + blk_cleanup_queue(uninit_q); + return NULL; } EXPORT_SYMBOL(blk_init_queue_node); @@ -1127,7 +1136,7 @@ static struct request *blk_old_get_request(struct request_queue *q, int rw, struct request *blk_get_request(struct request_queue *q, int rw, gfp_t gfp_mask) { if (q->mq_ops) - return blk_mq_alloc_request(q, rw, gfp_mask, false); + return blk_mq_alloc_request(q, rw, gfp_mask); else return blk_old_get_request(q, rw, gfp_mask); } diff --git a/block/blk-flush.c b/block/blk-flush.c index 9143e85226c7..66e2b697f5db 100644 --- a/block/blk-flush.c +++ b/block/blk-flush.c @@ -130,20 +130,26 @@ static void blk_flush_restore_request(struct request *rq) blk_clear_rq_complete(rq); } -static void mq_flush_data_run(struct work_struct *work) +static void mq_flush_run(struct work_struct *work) { struct request *rq; - rq = container_of(work, struct request, mq_flush_data); + rq = container_of(work, struct request, mq_flush_work); memset(&rq->csd, 0, sizeof(rq->csd)); blk_mq_run_request(rq, true, false); } -static void blk_mq_flush_data_insert(struct request *rq) +static bool blk_flush_queue_rq(struct request *rq) { - INIT_WORK(&rq->mq_flush_data, mq_flush_data_run); - kblockd_schedule_work(rq->q, &rq->mq_flush_data); + if (rq->q->mq_ops) { + INIT_WORK(&rq->mq_flush_work, mq_flush_run); + kblockd_schedule_work(rq->q, &rq->mq_flush_work); + return false; + } else { + list_add_tail(&rq->queuelist, &rq->q->queue_head); + return true; + } } /** @@ -187,12 +193,7 @@ static bool blk_flush_complete_seq(struct request *rq, unsigned int seq, case REQ_FSEQ_DATA: list_move_tail(&rq->flush.list, &q->flush_data_in_flight); - if (q->mq_ops) - blk_mq_flush_data_insert(rq); - else { - list_add(&rq->queuelist, &q->queue_head); - queued = true; - } + queued = blk_flush_queue_rq(rq); break; case REQ_FSEQ_DONE: @@ -216,9 +217,6 @@ static bool blk_flush_complete_seq(struct request *rq, unsigned int seq, } kicked = blk_kick_flush(q); - /* blk_mq_run_flush will run queue */ - if (q->mq_ops) - return queued; return kicked | queued; } @@ -230,10 +228,9 @@ static void flush_end_io(struct request *flush_rq, int error) struct request *rq, *n; unsigned long flags = 0; - if (q->mq_ops) { - blk_mq_free_request(flush_rq); + if (q->mq_ops) spin_lock_irqsave(&q->mq_flush_lock, flags); - } + running = &q->flush_queue[q->flush_running_idx]; BUG_ON(q->flush_pending_idx == q->flush_running_idx); @@ -263,48 +260,14 @@ static void flush_end_io(struct request *flush_rq, int error) * kblockd. */ if (queued || q->flush_queue_delayed) { - if (!q->mq_ops) - blk_run_queue_async(q); - else - /* - * This can be optimized to only run queues with requests - * queued if necessary. - */ - blk_mq_run_queues(q, true); + WARN_ON(q->mq_ops); + blk_run_queue_async(q); } q->flush_queue_delayed = 0; if (q->mq_ops) spin_unlock_irqrestore(&q->mq_flush_lock, flags); } -static void mq_flush_work(struct work_struct *work) -{ - struct request_queue *q; - struct request *rq; - - q = container_of(work, struct request_queue, mq_flush_work); - - rq = blk_mq_alloc_request(q, WRITE_FLUSH|REQ_FLUSH_SEQ, - __GFP_WAIT|GFP_ATOMIC, false); - rq->cmd_type = REQ_TYPE_FS; - rq->end_io = flush_end_io; - - blk_mq_run_request(rq, true, false); -} - -/* - * We can't directly use q->flush_rq, because it doesn't have tag and is not in - * hctx->rqs[]. so we must allocate a new request, since we can't sleep here, - * so offload the work to workqueue. - * - * Note: we assume a flush request finished in any hardware queue will flush - * the whole disk cache. - */ -static void mq_run_flush(struct request_queue *q) -{ - kblockd_schedule_work(q, &q->mq_flush_work); -} - /** * blk_kick_flush - consider issuing flush request * @q: request_queue being kicked @@ -339,19 +302,31 @@ static bool blk_kick_flush(struct request_queue *q) * different from running_idx, which means flush is in flight. */ q->flush_pending_idx ^= 1; + if (q->mq_ops) { - mq_run_flush(q); - return true; + struct blk_mq_ctx *ctx = first_rq->mq_ctx; + struct blk_mq_hw_ctx *hctx = q->mq_ops->map_queue(q, ctx->cpu); + + blk_mq_rq_init(hctx, q->flush_rq); + q->flush_rq->mq_ctx = ctx; + + /* + * Reuse the tag value from the fist waiting request, + * with blk-mq the tag is generated during request + * allocation and drivers can rely on it being inside + * the range they asked for. + */ + q->flush_rq->tag = first_rq->tag; + } else { + blk_rq_init(q, q->flush_rq); } - blk_rq_init(q, &q->flush_rq); - q->flush_rq.cmd_type = REQ_TYPE_FS; - q->flush_rq.cmd_flags = WRITE_FLUSH | REQ_FLUSH_SEQ; - q->flush_rq.rq_disk = first_rq->rq_disk; - q->flush_rq.end_io = flush_end_io; + q->flush_rq->cmd_type = REQ_TYPE_FS; + q->flush_rq->cmd_flags = WRITE_FLUSH | REQ_FLUSH_SEQ; + q->flush_rq->rq_disk = first_rq->rq_disk; + q->flush_rq->end_io = flush_end_io; - list_add_tail(&q->flush_rq.queuelist, &q->queue_head); - return true; + return blk_flush_queue_rq(q->flush_rq); } static void flush_data_end_io(struct request *rq, int error) @@ -407,11 +382,8 @@ void blk_insert_flush(struct request *rq) /* * @policy now records what operations need to be done. Adjust * REQ_FLUSH and FUA for the driver. - * We keep REQ_FLUSH for mq to track flush requests. For !FUA, - * we never dispatch the request directly. */ - if (rq->cmd_flags & REQ_FUA) - rq->cmd_flags &= ~REQ_FLUSH; + rq->cmd_flags &= ~REQ_FLUSH; if (!(fflags & REQ_FUA)) rq->cmd_flags &= ~REQ_FUA; @@ -560,5 +532,4 @@ EXPORT_SYMBOL(blkdev_issue_flush); void blk_mq_init_flush(struct request_queue *q) { spin_lock_init(&q->mq_flush_lock); - INIT_WORK(&q->mq_flush_work, mq_flush_work); } diff --git a/block/blk-mq.c b/block/blk-mq.c index 14c8f35946e1..a59b0565e940 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -194,27 +194,9 @@ static void blk_mq_rq_ctx_init(struct request_queue *q, struct blk_mq_ctx *ctx, } static struct request *__blk_mq_alloc_request(struct blk_mq_hw_ctx *hctx, - gfp_t gfp, bool reserved, - int rw) + gfp_t gfp, bool reserved) { - struct request *req; - bool is_flush = false; - /* - * flush need allocate a request, leave at least one request for - * non-flush IO to avoid deadlock - */ - if ((rw & REQ_FLUSH) && !(rw & REQ_FLUSH_SEQ)) { - if (atomic_inc_return(&hctx->pending_flush) >= - hctx->queue_depth - hctx->reserved_tags - 1) { - atomic_dec(&hctx->pending_flush); - return NULL; - } - is_flush = true; - } - req = blk_mq_alloc_rq(hctx, gfp, reserved); - if (!req && is_flush) - atomic_dec(&hctx->pending_flush); - return req; + return blk_mq_alloc_rq(hctx, gfp, reserved); } static struct request *blk_mq_alloc_request_pinned(struct request_queue *q, @@ -227,7 +209,7 @@ static struct request *blk_mq_alloc_request_pinned(struct request_queue *q, struct blk_mq_ctx *ctx = blk_mq_get_ctx(q); struct blk_mq_hw_ctx *hctx = q->mq_ops->map_queue(q, ctx->cpu); - rq = __blk_mq_alloc_request(hctx, gfp & ~__GFP_WAIT, reserved, rw); + rq = __blk_mq_alloc_request(hctx, gfp & ~__GFP_WAIT, reserved); if (rq) { blk_mq_rq_ctx_init(q, ctx, rq, rw); break; @@ -244,15 +226,14 @@ static struct request *blk_mq_alloc_request_pinned(struct request_queue *q, return rq; } -struct request *blk_mq_alloc_request(struct request_queue *q, int rw, - gfp_t gfp, bool reserved) +struct request *blk_mq_alloc_request(struct request_queue *q, int rw, gfp_t gfp) { struct request *rq; if (blk_mq_queue_enter(q)) return NULL; - rq = blk_mq_alloc_request_pinned(q, rw, gfp, reserved); + rq = blk_mq_alloc_request_pinned(q, rw, gfp, false); if (rq) blk_mq_put_ctx(rq->mq_ctx); return rq; @@ -276,7 +257,7 @@ EXPORT_SYMBOL(blk_mq_alloc_reserved_request); /* * Re-init and set pdu, if we have it */ -static void blk_mq_rq_init(struct blk_mq_hw_ctx *hctx, struct request *rq) +void blk_mq_rq_init(struct blk_mq_hw_ctx *hctx, struct request *rq) { blk_rq_init(hctx->queue, rq); @@ -290,9 +271,6 @@ static void __blk_mq_free_request(struct blk_mq_hw_ctx *hctx, const int tag = rq->tag; struct request_queue *q = rq->q; - if ((rq->cmd_flags & REQ_FLUSH) && !(rq->cmd_flags & REQ_FLUSH_SEQ)) - atomic_dec(&hctx->pending_flush); - blk_mq_rq_init(hctx, rq); blk_mq_put_tag(hctx->tags, tag); @@ -946,14 +924,14 @@ static void blk_mq_make_request(struct request_queue *q, struct bio *bio) hctx = q->mq_ops->map_queue(q, ctx->cpu); trace_block_getrq(q, bio, rw); - rq = __blk_mq_alloc_request(hctx, GFP_ATOMIC, false, bio->bi_rw); + rq = __blk_mq_alloc_request(hctx, GFP_ATOMIC, false); if (likely(rq)) - blk_mq_rq_ctx_init(q, ctx, rq, bio->bi_rw); + blk_mq_rq_ctx_init(q, ctx, rq, rw); else { blk_mq_put_ctx(ctx); trace_block_sleeprq(q, bio, rw); - rq = blk_mq_alloc_request_pinned(q, bio->bi_rw, - __GFP_WAIT|GFP_ATOMIC, false); + rq = blk_mq_alloc_request_pinned(q, rw, __GFP_WAIT|GFP_ATOMIC, + false); ctx = rq->mq_ctx; hctx = q->mq_ops->map_queue(q, ctx->cpu); } @@ -1230,9 +1208,7 @@ static int blk_mq_init_hw_queues(struct request_queue *q, hctx->queue_num = i; hctx->flags = reg->flags; hctx->queue_depth = reg->queue_depth; - hctx->reserved_tags = reg->reserved_tags; hctx->cmd_size = reg->cmd_size; - atomic_set(&hctx->pending_flush, 0); blk_mq_init_cpu_notifier(&hctx->cpu_notifier, blk_mq_hctx_notify, hctx); @@ -1412,9 +1388,14 @@ struct request_queue *blk_mq_init_queue(struct blk_mq_reg *reg, blk_mq_init_flush(q); blk_mq_init_cpu_queues(q, reg->nr_hw_queues); - if (blk_mq_init_hw_queues(q, reg, driver_data)) + q->flush_rq = kzalloc(round_up(sizeof(struct request) + reg->cmd_size, + cache_line_size()), GFP_KERNEL); + if (!q->flush_rq) goto err_hw; + if (blk_mq_init_hw_queues(q, reg, driver_data)) + goto err_flush_rq; + blk_mq_map_swqueue(q); mutex_lock(&all_q_mutex); @@ -1422,6 +1403,9 @@ struct request_queue *blk_mq_init_queue(struct blk_mq_reg *reg, mutex_unlock(&all_q_mutex); return q; + +err_flush_rq: + kfree(q->flush_rq); err_hw: kfree(q->mq_map); err_map: diff --git a/block/blk-mq.h b/block/blk-mq.h index f29b645f0e1c..ed0035cd458e 100644 --- a/block/blk-mq.h +++ b/block/blk-mq.h @@ -28,6 +28,7 @@ void blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx, bool async); void blk_mq_init_flush(struct request_queue *q); void blk_mq_drain_queue(struct request_queue *q); void blk_mq_free_queue(struct request_queue *q); +void blk_mq_rq_init(struct blk_mq_hw_ctx *hctx, struct request *rq); /* * CPU hotplug helpers diff --git a/block/blk-sysfs.c b/block/blk-sysfs.c index 8095c4a21fc0..7500f876dae4 100644 --- a/block/blk-sysfs.c +++ b/block/blk-sysfs.c @@ -549,6 +549,8 @@ static void blk_release_queue(struct kobject *kobj) if (q->mq_ops) blk_mq_free_queue(q); + kfree(q->flush_rq); + blk_trace_shutdown(q); bdi_destroy(&q->backing_dev_info); diff --git a/include/linux/blk-mq.h b/include/linux/blk-mq.h index 468be242db90..18ba8a627f46 100644 --- a/include/linux/blk-mq.h +++ b/include/linux/blk-mq.h @@ -36,15 +36,12 @@ struct blk_mq_hw_ctx { struct list_head page_list; struct blk_mq_tags *tags; - atomic_t pending_flush; - unsigned long queued; unsigned long run; #define BLK_MQ_MAX_DISPATCH_ORDER 10 unsigned long dispatched[BLK_MQ_MAX_DISPATCH_ORDER]; unsigned int queue_depth; - unsigned int reserved_tags; unsigned int numa_node; unsigned int cmd_size; /* per-request extra data */ @@ -129,7 +126,7 @@ void blk_mq_insert_request(struct request_queue *, struct request *, void blk_mq_run_queues(struct request_queue *q, bool async); void blk_mq_free_request(struct request *rq); bool blk_mq_can_queue(struct blk_mq_hw_ctx *); -struct request *blk_mq_alloc_request(struct request_queue *q, int rw, gfp_t gfp, bool reserved); +struct request *blk_mq_alloc_request(struct request_queue *q, int rw, gfp_t gfp); struct request *blk_mq_alloc_reserved_request(struct request_queue *q, int rw, gfp_t gfp); struct request *blk_mq_rq_from_tag(struct request_queue *q, unsigned int tag); diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 0375654adb28..b2d25ecbcbc1 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -101,7 +101,7 @@ struct request { }; union { struct call_single_data csd; - struct work_struct mq_flush_data; + struct work_struct mq_flush_work; }; struct request_queue *q; @@ -451,13 +451,8 @@ struct request_queue { unsigned long flush_pending_since; struct list_head flush_queue[2]; struct list_head flush_data_in_flight; - union { - struct request flush_rq; - struct { - spinlock_t mq_flush_lock; - struct work_struct mq_flush_work; - }; - }; + struct request *flush_rq; + spinlock_t mq_flush_lock; struct mutex sysfs_lock; From 11c94444074f40b479a05f6657d935204e992f2e Mon Sep 17 00:00:00 2001 From: Masanari Iida Date: Mon, 10 Feb 2014 10:39:18 -0700 Subject: [PATCH 452/596] block: Fix type mismatch in ssize_t_blk_mq_tag_sysfs_show cppcheck detected following format string mismatch. [blk-mq-tag.c:201]: (warning) %u in format string (no. 1) requires 'unsigned int' but the argument type is 'int'. Change "cpu" from int to unsigned int, because the cpu never become minus value. Signed-off-by: Masanari Iida Signed-off-by: Jens Axboe --- block/blk-mq-tag.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/block/blk-mq-tag.c b/block/blk-mq-tag.c index d64a02fb1f73..4025050320b9 100644 --- a/block/blk-mq-tag.c +++ b/block/blk-mq-tag.c @@ -182,7 +182,7 @@ void blk_mq_free_tags(struct blk_mq_tags *tags) ssize_t blk_mq_tag_sysfs_show(struct blk_mq_tags *tags, char *page) { char *orig_page = page; - int cpu; + unsigned int cpu; if (!tags) return 0; From 29ffa48fa64fcdfc71d80593c8ae79248bc27677 Mon Sep 17 00:00:00 2001 From: Linus Walleij Date: Tue, 4 Feb 2014 13:30:05 +0100 Subject: [PATCH 453/596] ARM: pxa: fix compilation problem on AM300EPD board MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This board fails compilation like this: arch/arm/mach-pxa/am300epd.c: In function ‘am300_cleanup’: arch/arm/mach-pxa/am300epd.c:179:2: error: implicit declaration of function ‘PXA_GPIO_TO_IRQ’ [-Werror=implicit-function-declaration] free_irq(PXA_GPIO_TO_IRQ(RDY_GPIO_PIN), par); This was caused by commit 88f718e3fa4d67f3a8dbe79a2f97d722323e4051 "ARM: pxa: delete the custom GPIO header" This is because it was previously getting the macro PXA_GPIO_TO_IRQ implicitly from which in turn implicitly included which in turn included . Add the missing include so that the board compiles again. Reported-by: Arnd Bergmann Signed-off-by: Linus Walleij Signed-off-by: Kevin Hilman --- arch/arm/mach-pxa/am300epd.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/arm/mach-pxa/am300epd.c b/arch/arm/mach-pxa/am300epd.c index c9f309ae88c5..8b90c4f2d430 100644 --- a/arch/arm/mach-pxa/am300epd.c +++ b/arch/arm/mach-pxa/am300epd.c @@ -30,6 +30,7 @@ #include #include +#include #include #include "generic.h" From 9705e74671f0e4f994d86b00cecf441917c64a66 Mon Sep 17 00:00:00 2001 From: Linus Walleij Date: Tue, 4 Feb 2014 13:53:07 +0100 Subject: [PATCH 454/596] ARM: pxa: fix various compilation problems MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Due to commit 88f718e3fa4d67f3a8dbe79a2f97d722323e4051 "ARM: pxa: delete the custom GPIO header" some drivers fail compilation, for example like this: In file included from sound/soc/pxa/spitz.c:28:0: sound/soc/pxa/spitz.c: In function ‘spitz_ext_control’: arch/arm/mach-pxa/include/mach/spitz.h:111:30: error: ‘PXA_NR_BUILTIN_GPIO’ undeclared (first use in this function) #define SPITZ_SCP_GPIO_BASE (PXA_NR_BUILTIN_GPIO) (etc.) This is caused by implicit inclusion of from various board-specific headers under in the PXA platform. So we take a sweep over these, and for every such header that uses PXA_NR_BUILTIN_GPIO or PXA_GPIO_TO_IRQ() we explicitly #include "irqs.h" so that we satisfy the dependency in the board include file alone. Reported-by: Arnd Bergmann Signed-off-by: Linus Walleij Cc: stable@vger.kernel.org # v3.13+ Signed-off-by: Kevin Hilman --- arch/arm/mach-pxa/include/mach/balloon3.h | 2 ++ arch/arm/mach-pxa/include/mach/corgi.h | 1 + arch/arm/mach-pxa/include/mach/csb726.h | 2 ++ arch/arm/mach-pxa/include/mach/gumstix.h | 1 + arch/arm/mach-pxa/include/mach/idp.h | 1 + arch/arm/mach-pxa/include/mach/palmld.h | 2 ++ arch/arm/mach-pxa/include/mach/palmt5.h | 2 ++ arch/arm/mach-pxa/include/mach/palmtc.h | 2 ++ arch/arm/mach-pxa/include/mach/palmtx.h | 2 ++ arch/arm/mach-pxa/include/mach/pcm027.h | 2 ++ arch/arm/mach-pxa/include/mach/pcm990_baseboard.h | 1 + arch/arm/mach-pxa/include/mach/poodle.h | 2 ++ arch/arm/mach-pxa/include/mach/spitz.h | 2 +- arch/arm/mach-pxa/include/mach/tosa.h | 2 ++ arch/arm/mach-pxa/include/mach/trizeps4.h | 2 ++ 15 files changed, 25 insertions(+), 1 deletion(-) diff --git a/arch/arm/mach-pxa/include/mach/balloon3.h b/arch/arm/mach-pxa/include/mach/balloon3.h index 954641e6c8b1..1b0825911e62 100644 --- a/arch/arm/mach-pxa/include/mach/balloon3.h +++ b/arch/arm/mach-pxa/include/mach/balloon3.h @@ -14,6 +14,8 @@ #ifndef ASM_ARCH_BALLOON3_H #define ASM_ARCH_BALLOON3_H +#include "irqs.h" /* PXA_NR_BUILTIN_GPIO */ + enum balloon3_features { BALLOON3_FEATURE_OHCI, BALLOON3_FEATURE_MMC, diff --git a/arch/arm/mach-pxa/include/mach/corgi.h b/arch/arm/mach-pxa/include/mach/corgi.h index f3c3493b468d..c030d955bbd7 100644 --- a/arch/arm/mach-pxa/include/mach/corgi.h +++ b/arch/arm/mach-pxa/include/mach/corgi.h @@ -13,6 +13,7 @@ #ifndef __ASM_ARCH_CORGI_H #define __ASM_ARCH_CORGI_H 1 +#include "irqs.h" /* PXA_NR_BUILTIN_GPIO */ /* * Corgi (Non Standard) GPIO Definitions diff --git a/arch/arm/mach-pxa/include/mach/csb726.h b/arch/arm/mach-pxa/include/mach/csb726.h index 2628e7b72116..00cfbbbf73f7 100644 --- a/arch/arm/mach-pxa/include/mach/csb726.h +++ b/arch/arm/mach-pxa/include/mach/csb726.h @@ -11,6 +11,8 @@ #ifndef CSB726_H #define CSB726_H +#include "irqs.h" /* PXA_GPIO_TO_IRQ */ + #define CSB726_GPIO_IRQ_LAN 52 #define CSB726_GPIO_IRQ_SM501 53 #define CSB726_GPIO_MMC_DETECT 100 diff --git a/arch/arm/mach-pxa/include/mach/gumstix.h b/arch/arm/mach-pxa/include/mach/gumstix.h index dba14b6503ad..f7df27bbb42e 100644 --- a/arch/arm/mach-pxa/include/mach/gumstix.h +++ b/arch/arm/mach-pxa/include/mach/gumstix.h @@ -6,6 +6,7 @@ * published by the Free Software Foundation. */ +#include "irqs.h" /* PXA_GPIO_TO_IRQ */ /* BTRESET - Reset line to Bluetooth module, active low signal. */ #define GPIO_GUMSTIX_BTRESET 7 diff --git a/arch/arm/mach-pxa/include/mach/idp.h b/arch/arm/mach-pxa/include/mach/idp.h index 22a96f87232b..7e63f4680271 100644 --- a/arch/arm/mach-pxa/include/mach/idp.h +++ b/arch/arm/mach-pxa/include/mach/idp.h @@ -23,6 +23,7 @@ * IDP hardware. */ +#include "irqs.h" /* PXA_GPIO_TO_IRQ */ #define IDP_FLASH_PHYS (PXA_CS0_PHYS) #define IDP_ALT_FLASH_PHYS (PXA_CS1_PHYS) diff --git a/arch/arm/mach-pxa/include/mach/palmld.h b/arch/arm/mach-pxa/include/mach/palmld.h index 2c4471336570..b184f296023b 100644 --- a/arch/arm/mach-pxa/include/mach/palmld.h +++ b/arch/arm/mach-pxa/include/mach/palmld.h @@ -13,6 +13,8 @@ #ifndef _INCLUDE_PALMLD_H_ #define _INCLUDE_PALMLD_H_ +#include "irqs.h" /* PXA_GPIO_TO_IRQ */ + /** HERE ARE GPIOs **/ /* GPIOs */ diff --git a/arch/arm/mach-pxa/include/mach/palmt5.h b/arch/arm/mach-pxa/include/mach/palmt5.h index 0bd4f036c72f..e342c5921405 100644 --- a/arch/arm/mach-pxa/include/mach/palmt5.h +++ b/arch/arm/mach-pxa/include/mach/palmt5.h @@ -15,6 +15,8 @@ #ifndef _INCLUDE_PALMT5_H_ #define _INCLUDE_PALMT5_H_ +#include "irqs.h" /* PXA_GPIO_TO_IRQ */ + /** HERE ARE GPIOs **/ /* GPIOs */ diff --git a/arch/arm/mach-pxa/include/mach/palmtc.h b/arch/arm/mach-pxa/include/mach/palmtc.h index c383a21680b6..81c727b3cfd2 100644 --- a/arch/arm/mach-pxa/include/mach/palmtc.h +++ b/arch/arm/mach-pxa/include/mach/palmtc.h @@ -16,6 +16,8 @@ #ifndef _INCLUDE_PALMTC_H_ #define _INCLUDE_PALMTC_H_ +#include "irqs.h" /* PXA_GPIO_TO_IRQ */ + /** HERE ARE GPIOs **/ /* GPIOs */ diff --git a/arch/arm/mach-pxa/include/mach/palmtx.h b/arch/arm/mach-pxa/include/mach/palmtx.h index f2e530380253..92bc1f05300d 100644 --- a/arch/arm/mach-pxa/include/mach/palmtx.h +++ b/arch/arm/mach-pxa/include/mach/palmtx.h @@ -16,6 +16,8 @@ #ifndef _INCLUDE_PALMTX_H_ #define _INCLUDE_PALMTX_H_ +#include "irqs.h" /* PXA_GPIO_TO_IRQ */ + /** HERE ARE GPIOs **/ /* GPIOs */ diff --git a/arch/arm/mach-pxa/include/mach/pcm027.h b/arch/arm/mach-pxa/include/mach/pcm027.h index 6bf28de228bd..86ebd7b6c960 100644 --- a/arch/arm/mach-pxa/include/mach/pcm027.h +++ b/arch/arm/mach-pxa/include/mach/pcm027.h @@ -23,6 +23,8 @@ * Definitions of CPU card resources only */ +#include "irqs.h" /* PXA_GPIO_TO_IRQ */ + /* phyCORE-PXA270 (PCM027) Interrupts */ #define PCM027_IRQ(x) (IRQ_BOARD_START + (x)) #define PCM027_BTDET_IRQ PCM027_IRQ(0) diff --git a/arch/arm/mach-pxa/include/mach/pcm990_baseboard.h b/arch/arm/mach-pxa/include/mach/pcm990_baseboard.h index 0260aaa2fc17..7e544c14967e 100644 --- a/arch/arm/mach-pxa/include/mach/pcm990_baseboard.h +++ b/arch/arm/mach-pxa/include/mach/pcm990_baseboard.h @@ -20,6 +20,7 @@ */ #include +#include "irqs.h" /* PXA_GPIO_TO_IRQ */ /* * definitions relevant only when the PCM-990 diff --git a/arch/arm/mach-pxa/include/mach/poodle.h b/arch/arm/mach-pxa/include/mach/poodle.h index f32ff75dcca8..b56b19351a03 100644 --- a/arch/arm/mach-pxa/include/mach/poodle.h +++ b/arch/arm/mach-pxa/include/mach/poodle.h @@ -15,6 +15,8 @@ #ifndef __ASM_ARCH_POODLE_H #define __ASM_ARCH_POODLE_H 1 +#include "irqs.h" /* PXA_GPIO_TO_IRQ */ + /* * GPIOs */ diff --git a/arch/arm/mach-pxa/include/mach/spitz.h b/arch/arm/mach-pxa/include/mach/spitz.h index 0bfe6507c95d..25c9f62e46aa 100644 --- a/arch/arm/mach-pxa/include/mach/spitz.h +++ b/arch/arm/mach-pxa/include/mach/spitz.h @@ -15,8 +15,8 @@ #define __ASM_ARCH_SPITZ_H 1 #endif +#include "irqs.h" /* PXA_NR_BUILTIN_GPIO, PXA_GPIO_TO_IRQ */ #include -#include /* Spitz/Akita GPIOs */ diff --git a/arch/arm/mach-pxa/include/mach/tosa.h b/arch/arm/mach-pxa/include/mach/tosa.h index 2bb0e862598c..0497d95cef25 100644 --- a/arch/arm/mach-pxa/include/mach/tosa.h +++ b/arch/arm/mach-pxa/include/mach/tosa.h @@ -13,6 +13,8 @@ #ifndef _ASM_ARCH_TOSA_H_ #define _ASM_ARCH_TOSA_H_ 1 +#include "irqs.h" /* PXA_NR_BUILTIN_GPIO */ + /* TOSA Chip selects */ #define TOSA_LCDC_PHYS PXA_CS4_PHYS /* Internel Scoop */ diff --git a/arch/arm/mach-pxa/include/mach/trizeps4.h b/arch/arm/mach-pxa/include/mach/trizeps4.h index d2ca01053f69..ae3ca013afab 100644 --- a/arch/arm/mach-pxa/include/mach/trizeps4.h +++ b/arch/arm/mach-pxa/include/mach/trizeps4.h @@ -10,6 +10,8 @@ #ifndef _TRIPEPS4_H_ #define _TRIPEPS4_H_ +#include "irqs.h" /* PXA_GPIO_TO_IRQ */ + /* physical memory regions */ #define TRIZEPS4_FLASH_PHYS (PXA_CS0_PHYS) /* Flash region */ #define TRIZEPS4_DISK_PHYS (PXA_CS1_PHYS) /* Disk On Chip region */ From e7c57ecd6019cc6392223605aed18cce257c3eff Mon Sep 17 00:00:00 2001 From: Philipp Zabel Date: Wed, 29 Jan 2014 17:10:04 +0100 Subject: [PATCH 455/596] ARM: imx6: Initialize low-power mode early again Since commit 9e8147bb5ec5d1dda2141da70f96b98985a306cb "ARM: imx6q: move low-power code out of clock driver" the kernel fails to boot on i.MX6Q/D if preemption is enabled (CONFIG_PREEMPT=y). The kernel just hangs before the console comes up. The above commit moved the initalization of the low-power mode setting (enabling clocked WAIT states), which was introduced in commit 83ae20981ae924c37d02a42c829155fc3851260c "ARM: imx: correct low-power mode setting", from imx6q_clks_init to imx6q_pm_init. Now it is called much later, after all cores are enabled. This patch moves the low-power mode initialization back to imx6q_clks_init again (and to imx6sl_clks_init). Signed-off-by: Philipp Zabel Signed-off-by: Shawn Guo Signed-off-by: Kevin Hilman --- arch/arm/mach-imx/clk-imx6q.c | 3 +++ arch/arm/mach-imx/clk-imx6sl.c | 3 +++ arch/arm/mach-imx/pm-imx6q.c | 2 -- 3 files changed, 6 insertions(+), 2 deletions(-) diff --git a/arch/arm/mach-imx/clk-imx6q.c b/arch/arm/mach-imx/clk-imx6q.c index af2e582d2b74..4d677f442539 100644 --- a/arch/arm/mach-imx/clk-imx6q.c +++ b/arch/arm/mach-imx/clk-imx6q.c @@ -482,6 +482,9 @@ static void __init imx6q_clocks_init(struct device_node *ccm_node) if (IS_ENABLED(CONFIG_PCI_IMX6)) clk_set_parent(clk[lvds1_sel], clk[sata_ref]); + /* Set initial power mode */ + imx6q_set_lpm(WAIT_CLOCKED); + np = of_find_compatible_node(NULL, NULL, "fsl,imx6q-gpt"); base = of_iomap(np, 0); WARN_ON(!base); diff --git a/arch/arm/mach-imx/clk-imx6sl.c b/arch/arm/mach-imx/clk-imx6sl.c index 3781a1853998..4c86f3035205 100644 --- a/arch/arm/mach-imx/clk-imx6sl.c +++ b/arch/arm/mach-imx/clk-imx6sl.c @@ -266,6 +266,9 @@ static void __init imx6sl_clocks_init(struct device_node *ccm_node) /* Audio-related clocks configuration */ clk_set_parent(clks[IMX6SL_CLK_SPDIF0_SEL], clks[IMX6SL_CLK_PLL3_PFD3]); + /* Set initial power mode */ + imx6q_set_lpm(WAIT_CLOCKED); + np = of_find_compatible_node(NULL, NULL, "fsl,imx6sl-gpt"); base = of_iomap(np, 0); WARN_ON(!base); diff --git a/arch/arm/mach-imx/pm-imx6q.c b/arch/arm/mach-imx/pm-imx6q.c index 9d47adc078aa..7a9b98589db7 100644 --- a/arch/arm/mach-imx/pm-imx6q.c +++ b/arch/arm/mach-imx/pm-imx6q.c @@ -236,8 +236,6 @@ void __init imx6q_pm_init(void) regmap_update_bits(gpr, IOMUXC_GPR1, IMX6Q_GPR1_GINT, IMX6Q_GPR1_GINT); - /* Set initial power mode */ - imx6q_set_lpm(WAIT_CLOCKED); suspend_set_ops(&imx6q_pm_ops); } From 5db1dabc562b27c356376fbb527a746dcf7c9572 Mon Sep 17 00:00:00 2001 From: Nishanth Menon Date: Thu, 6 Feb 2014 09:29:31 -0600 Subject: [PATCH 456/596] ARM: multi_v7_defconfig: Select CONFIG_SOC_DRA7XX Select CONFIG_SOC_DRA7XX so that we can boot dra7-evm. DRA7 family are A15 based processors that supports LPAE and an evolutionary update to the OMAP5 generation of processors. Signed-off-by: Nishanth Menon Signed-off-by: Kevin Hilman --- arch/arm/configs/multi_v7_defconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/arm/configs/multi_v7_defconfig b/arch/arm/configs/multi_v7_defconfig index 845bc745706b..ee6982976d66 100644 --- a/arch/arm/configs/multi_v7_defconfig +++ b/arch/arm/configs/multi_v7_defconfig @@ -29,6 +29,7 @@ CONFIG_ARCH_OMAP3=y CONFIG_ARCH_OMAP4=y CONFIG_SOC_OMAP5=y CONFIG_SOC_AM33XX=y +CONFIG_SOC_DRA7XX=y CONFIG_SOC_AM43XX=y CONFIG_ARCH_ROCKCHIP=y CONFIG_ARCH_SOCFPGA=y From 46f5b96085b03f9e71a0134de77efc6a1f9d84c3 Mon Sep 17 00:00:00 2001 From: Michal Simek Date: Fri, 31 Jan 2014 12:55:06 +0100 Subject: [PATCH 457/596] ARM: zynq: Reserve not DMAable space in front of the kernel Reserve space from 0x0 - __pa(swapper_pg_dir), if kernel is loaded from 0, which is not DMAable. It is causing problem with MMC driver and others which want to add dma buffers to this space. Signed-off-by: Michal Simek Signed-off-by: Kevin Hilman --- arch/arm/mach-zynq/common.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+) diff --git a/arch/arm/mach-zynq/common.c b/arch/arm/mach-zynq/common.c index 1db2a5ca9ab8..8c09a8393fb6 100644 --- a/arch/arm/mach-zynq/common.c +++ b/arch/arm/mach-zynq/common.c @@ -25,6 +25,7 @@ #include #include #include +#include #include #include @@ -41,6 +42,18 @@ void __iomem *zynq_scu_base; +/** + * zynq_memory_init - Initialize special memory + * + * We need to stop things allocating the low memory as DMA can't work in + * the 1st 512K of memory. + */ +static void __init zynq_memory_init(void) +{ + if (!__pa(PAGE_OFFSET)) + memblock_reserve(__pa(PAGE_OFFSET), __pa(swapper_pg_dir)); +} + static struct platform_device zynq_cpuidle_device = { .name = "cpuidle-zynq", }; @@ -117,5 +130,6 @@ DT_MACHINE_START(XILINX_EP107, "Xilinx Zynq Platform") .init_machine = zynq_init_machine, .init_time = zynq_timer_init, .dt_compat = zynq_dt_match, + .reserve = zynq_memory_init, .restart = zynq_system_reset, MACHINE_END From a5642ab4744bc8c5a8c7ce7c6e30c01bd6bbc691 Mon Sep 17 00:00:00 2001 From: Toshiaki Makita Date: Fri, 7 Feb 2014 16:48:18 +0900 Subject: [PATCH 458/596] bridge: Fix the way to find old local fdb entries in br_fdb_changeaddr br_fdb_changeaddr() assumes that there is at most one local entry per port per vlan. It used to be true, but since commit 36fd2b63e3b4 ("bridge: allow creating/deleting fdb entries via netlink"), it has not been so. Therefore, the function might fail to search a correct previous address to be deleted and delete an arbitrary local entry if user has added local entries manually. Example of problematic case: ip link set eth0 address ee:ff:12:34:56:78 brctl addif br0 eth0 bridge fdb add 12:34:56:78:90:ab dev eth0 master ip link set eth0 address aa:bb:cc:dd:ee:ff Then, the address 12:34:56:78:90:ab might be deleted instead of ee:ff:12:34:56:78, the original mac address of eth0. Address this issue by introducing a new flag, added_by_user, to struct net_bridge_fdb_entry. Note that br_fdb_delete_by_port() has to set added_by_user to 0 in cases like: ip link set eth0 address 12:34:56:78:90:ab ip link set eth1 address aa:bb:cc:dd:ee:ff brctl addif br0 eth0 bridge fdb add aa:bb:cc:dd:ee:ff dev eth0 master brctl addif br0 eth1 brctl delif br0 eth0 In this case, kernel should delete the user-added entry aa:bb:cc:dd:ee:ff, but it also should have been added by "brctl addif br0 eth1" originally, so we don't delete it and treat it a new kernel-created entry. Signed-off-by: Toshiaki Makita Signed-off-by: David S. Miller --- net/bridge/br_fdb.c | 16 ++++++++++++---- net/bridge/br_input.c | 4 ++-- net/bridge/br_private.h | 3 ++- 3 files changed, 16 insertions(+), 7 deletions(-) diff --git a/net/bridge/br_fdb.c b/net/bridge/br_fdb.c index c5f5a4a933f4..ce5411995a63 100644 --- a/net/bridge/br_fdb.c +++ b/net/bridge/br_fdb.c @@ -104,7 +104,7 @@ void br_fdb_changeaddr(struct net_bridge_port *p, const unsigned char *newaddr) struct net_bridge_fdb_entry *f; f = hlist_entry(h, struct net_bridge_fdb_entry, hlist); - if (f->dst == p && f->is_local) { + if (f->dst == p && f->is_local && !f->added_by_user) { /* maybe another port has same hw addr? */ struct net_bridge_port *op; u16 vid = f->vlan_id; @@ -247,6 +247,7 @@ void br_fdb_delete_by_port(struct net_bridge *br, ether_addr_equal(op->dev->dev_addr, f->addr.addr)) { f->dst = op; + f->added_by_user = 0; goto skip_delete; } } @@ -397,6 +398,7 @@ static struct net_bridge_fdb_entry *fdb_create(struct hlist_head *head, fdb->vlan_id = vid; fdb->is_local = 0; fdb->is_static = 0; + fdb->added_by_user = 0; fdb->updated = fdb->used = jiffies; hlist_add_head_rcu(&fdb->hlist, head); } @@ -447,7 +449,7 @@ int br_fdb_insert(struct net_bridge *br, struct net_bridge_port *source, } void br_fdb_update(struct net_bridge *br, struct net_bridge_port *source, - const unsigned char *addr, u16 vid) + const unsigned char *addr, u16 vid, bool added_by_user) { struct hlist_head *head = &br->hash[br_mac_hash(addr, vid)]; struct net_bridge_fdb_entry *fdb; @@ -473,13 +475,18 @@ void br_fdb_update(struct net_bridge *br, struct net_bridge_port *source, /* fastpath: update of existing entry */ fdb->dst = source; fdb->updated = jiffies; + if (unlikely(added_by_user)) + fdb->added_by_user = 1; } } else { spin_lock(&br->hash_lock); if (likely(!fdb_find(head, addr, vid))) { fdb = fdb_create(head, source, addr, vid); - if (fdb) + if (fdb) { + if (unlikely(added_by_user)) + fdb->added_by_user = 1; fdb_notify(br, fdb, RTM_NEWNEIGH); + } } /* else we lose race and someone else inserts * it first, don't bother updating @@ -647,6 +654,7 @@ static int fdb_add_entry(struct net_bridge_port *source, const __u8 *addr, modified = true; } + fdb->added_by_user = 1; fdb->used = jiffies; if (modified) { @@ -664,7 +672,7 @@ static int __br_fdb_add(struct ndmsg *ndm, struct net_bridge_port *p, if (ndm->ndm_flags & NTF_USE) { rcu_read_lock(); - br_fdb_update(p->br, p, addr, vid); + br_fdb_update(p->br, p, addr, vid, true); rcu_read_unlock(); } else { spin_lock_bh(&p->br->hash_lock); diff --git a/net/bridge/br_input.c b/net/bridge/br_input.c index bf8dc7d308d6..28d544627422 100644 --- a/net/bridge/br_input.c +++ b/net/bridge/br_input.c @@ -77,7 +77,7 @@ int br_handle_frame_finish(struct sk_buff *skb) /* insert into forwarding database after filtering to avoid spoofing */ br = p->br; if (p->flags & BR_LEARNING) - br_fdb_update(br, p, eth_hdr(skb)->h_source, vid); + br_fdb_update(br, p, eth_hdr(skb)->h_source, vid, false); if (!is_broadcast_ether_addr(dest) && is_multicast_ether_addr(dest) && br_multicast_rcv(br, p, skb, vid)) @@ -148,7 +148,7 @@ static int br_handle_local_finish(struct sk_buff *skb) br_vlan_get_tag(skb, &vid); if (p->flags & BR_LEARNING) - br_fdb_update(p->br, p, eth_hdr(skb)->h_source, vid); + br_fdb_update(p->br, p, eth_hdr(skb)->h_source, vid, false); return 0; /* process further */ } diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h index fcd12333c59b..939a59e15036 100644 --- a/net/bridge/br_private.h +++ b/net/bridge/br_private.h @@ -104,6 +104,7 @@ struct net_bridge_fdb_entry mac_addr addr; unsigned char is_local; unsigned char is_static; + unsigned char added_by_user; __u16 vlan_id; }; @@ -383,7 +384,7 @@ int br_fdb_fillbuf(struct net_bridge *br, void *buf, unsigned long count, int br_fdb_insert(struct net_bridge *br, struct net_bridge_port *source, const unsigned char *addr, u16 vid); void br_fdb_update(struct net_bridge *br, struct net_bridge_port *source, - const unsigned char *addr, u16 vid); + const unsigned char *addr, u16 vid, bool added_by_user); int fdb_delete_by_addr(struct net_bridge *br, const u8 *addr, u16 vid); int br_fdb_delete(struct ndmsg *ndm, struct nlattr *tb[], From 2836882fe07718fe3263745b1aa07284ec71871c Mon Sep 17 00:00:00 2001 From: Toshiaki Makita Date: Fri, 7 Feb 2014 16:48:19 +0900 Subject: [PATCH 459/596] bridge: Fix the way to insert new local fdb entries in br_fdb_changeaddr Since commit bc9a25d21ef8 ("bridge: Add vlan support for local fdb entries"), br_fdb_changeaddr() has inserted a new local fdb entry only if it can find old one. But if we have two ports where they have the same address or user has deleted a local entry, there will be no entry for one of the ports. Example of problematic case: ip link set eth0 address aa:bb:cc:dd:ee:ff ip link set eth1 address aa:bb:cc:dd:ee:ff brctl addif br0 eth0 brctl addif br0 eth1 # eth1 will not have a local entry due to dup. ip link set eth1 address 12:34:56:78:90:ab Then, the new entry for the address 12:34:56:78:90:ab will not be created, and the bridge device will not be able to communicate. Insert new entries regardless of whether we can find old entries or not. Signed-off-by: Toshiaki Makita Acked-by: Vlad Yasevich Signed-off-by: David S. Miller --- net/bridge/br_fdb.c | 29 ++++++++++++++++++++--------- 1 file changed, 20 insertions(+), 9 deletions(-) diff --git a/net/bridge/br_fdb.c b/net/bridge/br_fdb.c index ce5411995a63..96ab1d1748d0 100644 --- a/net/bridge/br_fdb.c +++ b/net/bridge/br_fdb.c @@ -92,8 +92,10 @@ static void fdb_delete(struct net_bridge *br, struct net_bridge_fdb_entry *f) void br_fdb_changeaddr(struct net_bridge_port *p, const unsigned char *newaddr) { struct net_bridge *br = p->br; - bool no_vlan = (nbp_get_vlan_info(p) == NULL) ? true : false; + struct net_port_vlans *pv = nbp_get_vlan_info(p); + bool no_vlan = !pv; int i; + u16 vid; spin_lock_bh(&br->hash_lock); @@ -114,28 +116,37 @@ void br_fdb_changeaddr(struct net_bridge_port *p, const unsigned char *newaddr) f->addr.addr) && nbp_vlan_find(op, vid)) { f->dst = op; - goto insert; + goto skip_delete; } } /* delete old one */ fdb_delete(br, f); -insert: - /* insert new address, may fail if invalid - * address or dup. - */ - fdb_insert(br, p, newaddr, vid); - +skip_delete: /* if this port has no vlan information * configured, we can safely be done at * this point. */ if (no_vlan) - goto done; + goto insert; } } } +insert: + /* insert new address, may fail if invalid address or dup. */ + fdb_insert(br, p, newaddr, 0); + + if (no_vlan) + goto done; + + /* Now add entries for every VLAN configured on the port. + * This function runs under RTNL so the bitmap will not change + * from under us. + */ + for_each_set_bit(vid, pv->vlan_bitmap, VLAN_N_VID) + fdb_insert(br, p, newaddr, vid); + done: spin_unlock_bh(&br->hash_lock); } From a3ebb7efe7033d903d86c9d664f07a9cc21747b3 Mon Sep 17 00:00:00 2001 From: Toshiaki Makita Date: Fri, 7 Feb 2014 16:48:20 +0900 Subject: [PATCH 460/596] bridge: Fix the way to find old local fdb entries in br_fdb_change_mac_address We have been always failed to delete the old entry at br_fdb_change_mac_address() because br_set_mac_address() updates dev->dev_addr before calling br_fdb_change_mac_address() and br_fdb_change_mac_address() uses dev->dev_addr to find the old entry. That update of dev_addr is completely unnecessary because the same work is done in br_stp_change_bridge_id() which is called right away after calling br_fdb_change_mac_address(). Signed-off-by: Toshiaki Makita Acked-by: Vlad Yasevich Signed-off-by: David S. Miller --- net/bridge/br_device.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c index d9a9b0fc1795..6f5cbd1a2f38 100644 --- a/net/bridge/br_device.c +++ b/net/bridge/br_device.c @@ -187,8 +187,8 @@ static int br_set_mac_address(struct net_device *dev, void *p) spin_lock_bh(&br->lock); if (!ether_addr_equal(dev->dev_addr, addr->sa_data)) { - memcpy(dev->dev_addr, addr->sa_data, ETH_ALEN); br_fdb_change_mac_address(br, addr->sa_data); + /* Mac address will be changed in br_stp_change_bridge_id(). */ br_stp_change_bridge_id(br, addr->sa_data); } spin_unlock_bh(&br->lock); From a4b816d8ba1c1917842dc3de97cbf8ef116e043e Mon Sep 17 00:00:00 2001 From: Toshiaki Makita Date: Fri, 7 Feb 2014 16:48:21 +0900 Subject: [PATCH 461/596] bridge: Change local fdb entries whenever mac address of bridge device changes Vlan code may need fdb change when changing mac address of bridge device even if it is caused by the mac address changing of a bridge port. Example configuration: ip link set eth0 address 12:34:56:78:90:ab ip link set eth1 address aa:bb:cc:dd:ee:ff brctl addif br0 eth0 brctl addif br0 eth1 # br0 will have mac address 12:34:56:78:90:ab bridge vlan add dev br0 vid 10 self bridge vlan add dev eth0 vid 10 We will have fdb entry such that f->dst == NULL, f->vlan_id == 10 and f->addr == 12:34:56:78:90:ab at this time. Next, change the mac address of eth0 to greater value. ip link set eth0 address ee:ff:12:34:56:78 Then, mac address of br0 will be recalculated and set to aa:bb:cc:dd:ee:ff. However, an entry aa:bb:cc:dd:ee:ff will not be created and we will be not able to communicate using br0 on vlan 10. Address this issue by deleting and adding local entries whenever changing the mac address of the bridge device. If there already exists an entry that has the same address, for example, in case that br_fdb_changeaddr() has already inserted it, br_fdb_change_mac_address() will simply fail to insert it and no duplicated entry will be made, as it was. This approach also needs br_add_if() to call br_fdb_insert() before br_stp_recalculate_bridge_id() so that we don't create an entry whose dst == NULL in this function to preserve previous behavior. Note that this is a slight change in behavior where the bridge device can receive the traffic to the new address before calling br_stp_recalculate_bridge_id() in br_add_if(). However, it is not a problem because we have already the address on the new port and such a way to insert new one before recalculating bridge id is taken in br_device_event() as well. Signed-off-by: Toshiaki Makita Acked-by: Vlad Yasevich Signed-off-by: David S. Miller --- net/bridge/br_device.c | 1 - net/bridge/br_if.c | 6 +++--- net/bridge/br_stp_if.c | 2 ++ 3 files changed, 5 insertions(+), 4 deletions(-) diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c index 6f5cbd1a2f38..63f0455c0bc3 100644 --- a/net/bridge/br_device.c +++ b/net/bridge/br_device.c @@ -187,7 +187,6 @@ static int br_set_mac_address(struct net_device *dev, void *p) spin_lock_bh(&br->lock); if (!ether_addr_equal(dev->dev_addr, addr->sa_data)) { - br_fdb_change_mac_address(br, addr->sa_data); /* Mac address will be changed in br_stp_change_bridge_id(). */ br_stp_change_bridge_id(br, addr->sa_data); } diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c index cffe1d666ba1..54d207d3a31c 100644 --- a/net/bridge/br_if.c +++ b/net/bridge/br_if.c @@ -389,6 +389,9 @@ int br_add_if(struct net_bridge *br, struct net_device *dev) if (br->dev->needed_headroom < dev->needed_headroom) br->dev->needed_headroom = dev->needed_headroom; + if (br_fdb_insert(br, p, dev->dev_addr, 0)) + netdev_err(dev, "failed insert local address bridge forwarding table\n"); + spin_lock_bh(&br->lock); changed_addr = br_stp_recalculate_bridge_id(br); @@ -404,9 +407,6 @@ int br_add_if(struct net_bridge *br, struct net_device *dev) dev_set_mtu(br->dev, br_min_mtu(br)); - if (br_fdb_insert(br, p, dev->dev_addr, 0)) - netdev_err(dev, "failed insert local address bridge forwarding table\n"); - kobject_uevent(&p->kobj, KOBJ_ADD); return 0; diff --git a/net/bridge/br_stp_if.c b/net/bridge/br_stp_if.c index 656a6f3e40de..189ba1e7d851 100644 --- a/net/bridge/br_stp_if.c +++ b/net/bridge/br_stp_if.c @@ -194,6 +194,8 @@ void br_stp_change_bridge_id(struct net_bridge *br, const unsigned char *addr) wasroot = br_is_root_bridge(br); + br_fdb_change_mac_address(br, addr); + memcpy(oldaddr, br->bridge_id.addr, ETH_ALEN); memcpy(br->bridge_id.addr, addr, ETH_ALEN); memcpy(br->dev->dev_addr, addr, ETH_ALEN); From 2b292fb4a57dc233e298a84196d33be0bc3828e4 Mon Sep 17 00:00:00 2001 From: Toshiaki Makita Date: Fri, 7 Feb 2014 16:48:22 +0900 Subject: [PATCH 462/596] bridge: Fix the way to check if a local fdb entry can be deleted We should take into account the followings when deleting a local fdb entry. - nbp_vlan_find() can be used only when vid != 0 to check if an entry is deletable, because a fdb entry with vid 0 can exist at any time while nbp_vlan_find() always return false with vid 0. Example of problematic case: ip link set eth0 address 12:34:56:78:90:ab ip link set eth1 address 12:34:56:78:90:ab brctl addif br0 eth0 brctl addif br0 eth1 ip link set eth0 address aa:bb:cc:dd:ee:ff Then, the fdb entry 12:34:56:78:90:ab will be deleted even though the bridge port eth1 still has that address. - The port to which the bridge device is attached might needs a local entry if its mac address is set manually. Example of problematic case: ip link set eth0 address 12:34:56:78:90:ab brctl addif br0 eth0 ip link set br0 address 12:34:56:78:90:ab ip link set eth0 address aa:bb:cc:dd:ee:ff Then, the fdb still must have the entry 12:34:56:78:90:ab, but it will be deleted. We can use br->dev->addr_assign_type to check if the address is manually set or not, but I propose another approach. Since we delete and insert local entries whenever changing mac address of the bridge device, we can change dst of the entry to NULL regardless of addr_assign_type when deleting an entry associated with a certain port, and if it is found to be unnecessary later, then delete it. That is, if changing mac address of a port, the entry might be changed to its dst being NULL first, but is eventually deleted when recalculating and changing bridge id. This approach is especially useful when we want to share the code with deleting vlan in which the bridge device might want such an entry regardless of addr_assign_type, and makes things easy because we don't have to consider if mac address of the bridge device will be changed or not at the time we delete a local entry of a port, which means fdb code will not be bothered even if the bridge id calculating logic is changed in the future. Also, this change reduces inconsistent state, where frames whose dst is the mac address of the bridge, can't reach the bridge because of premature fdb entry deletion. This change reduces the possibility that the bridge device replies unreachable mac address to arp requests, which could occur during the short window between calling del_nbp() and br_stp_recalculate_bridge_id() in br_del_if(). This will effective after br_fdb_delete_by_port() starts to use the same code by following patch. Signed-off-by: Toshiaki Makita Acked-by: Vlad Yasevich Signed-off-by: David S. Miller --- net/bridge/br_fdb.c | 10 +++++++++- net/bridge/br_private.h | 6 ++++++ net/bridge/br_vlan.c | 19 +++++++++++++++++++ 3 files changed, 34 insertions(+), 1 deletion(-) diff --git a/net/bridge/br_fdb.c b/net/bridge/br_fdb.c index 96ab1d1748d0..b4005f5b28f4 100644 --- a/net/bridge/br_fdb.c +++ b/net/bridge/br_fdb.c @@ -114,12 +114,20 @@ void br_fdb_changeaddr(struct net_bridge_port *p, const unsigned char *newaddr) if (op != p && ether_addr_equal(op->dev->dev_addr, f->addr.addr) && - nbp_vlan_find(op, vid)) { + (!vid || nbp_vlan_find(op, vid))) { f->dst = op; goto skip_delete; } } + /* maybe bridge device has same hw addr? */ + if (ether_addr_equal(br->dev->dev_addr, + f->addr.addr) && + (!vid || br_vlan_find(br, vid))) { + f->dst = NULL; + goto skip_delete; + } + /* delete old one */ fdb_delete(br, f); skip_delete: diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h index 939a59e15036..f91e1d9c8e92 100644 --- a/net/bridge/br_private.h +++ b/net/bridge/br_private.h @@ -585,6 +585,7 @@ struct sk_buff *br_handle_vlan(struct net_bridge *br, int br_vlan_add(struct net_bridge *br, u16 vid, u16 flags); int br_vlan_delete(struct net_bridge *br, u16 vid); void br_vlan_flush(struct net_bridge *br); +bool br_vlan_find(struct net_bridge *br, u16 vid); int br_vlan_filter_toggle(struct net_bridge *br, unsigned long val); int nbp_vlan_add(struct net_bridge_port *port, u16 vid, u16 flags); int nbp_vlan_delete(struct net_bridge_port *port, u16 vid); @@ -666,6 +667,11 @@ static inline void br_vlan_flush(struct net_bridge *br) { } +static inline bool br_vlan_find(struct net_bridge *br, u16 vid) +{ + return false; +} + static inline int nbp_vlan_add(struct net_bridge_port *port, u16 vid, u16 flags) { return -EOPNOTSUPP; diff --git a/net/bridge/br_vlan.c b/net/bridge/br_vlan.c index 4ca4d0a0151c..233ec1c6e9db 100644 --- a/net/bridge/br_vlan.c +++ b/net/bridge/br_vlan.c @@ -295,6 +295,25 @@ void br_vlan_flush(struct net_bridge *br) __vlan_flush(pv); } +bool br_vlan_find(struct net_bridge *br, u16 vid) +{ + struct net_port_vlans *pv; + bool found = false; + + rcu_read_lock(); + pv = rcu_dereference(br->vlan_info); + + if (!pv) + goto out; + + if (test_bit(vid, pv->vlan_bitmap)) + found = true; + +out: + rcu_read_unlock(); + return found; +} + int br_vlan_filter_toggle(struct net_bridge *br, unsigned long val) { if (!rtnl_trylock()) From 960b589f86c74ce582922fcb996103271081f4de Mon Sep 17 00:00:00 2001 From: Toshiaki Makita Date: Fri, 7 Feb 2014 16:48:23 +0900 Subject: [PATCH 463/596] bridge: Properly check if local fdb entry can be deleted in br_fdb_change_mac_address br_fdb_change_mac_address() doesn't check if the local entry has the same address as any of bridge ports. Although I'm not sure when it is beneficial, current implementation allow the bridge device to receive any mac address of its ports. To preserve this behavior, we have to check if the mac address of the entry being deleted is identical to that of any port. As this check is almost the same as that in br_fdb_changeaddr(), create a common function fdb_delete_local() and call it from br_fdb_changeadddr() and br_fdb_change_mac_address(). Signed-off-by: Toshiaki Makita Acked-by: Vlad Yasevich Signed-off-by: David S. Miller --- net/bridge/br_fdb.c | 57 +++++++++++++++++++++++++-------------------- 1 file changed, 32 insertions(+), 25 deletions(-) diff --git a/net/bridge/br_fdb.c b/net/bridge/br_fdb.c index b4005f5b28f4..15bf13d0b543 100644 --- a/net/bridge/br_fdb.c +++ b/net/bridge/br_fdb.c @@ -89,6 +89,34 @@ static void fdb_delete(struct net_bridge *br, struct net_bridge_fdb_entry *f) call_rcu(&f->rcu, fdb_rcu_free); } +/* Delete a local entry if no other port had the same address. */ +static void fdb_delete_local(struct net_bridge *br, + const struct net_bridge_port *p, + struct net_bridge_fdb_entry *f) +{ + const unsigned char *addr = f->addr.addr; + u16 vid = f->vlan_id; + struct net_bridge_port *op; + + /* Maybe another port has same hw addr? */ + list_for_each_entry(op, &br->port_list, list) { + if (op != p && ether_addr_equal(op->dev->dev_addr, addr) && + (!vid || nbp_vlan_find(op, vid))) { + f->dst = op; + return; + } + } + + /* Maybe bridge device has same hw addr? */ + if (p && ether_addr_equal(br->dev->dev_addr, addr) && + (!vid || br_vlan_find(br, vid))) { + f->dst = NULL; + return; + } + + fdb_delete(br, f); +} + void br_fdb_changeaddr(struct net_bridge_port *p, const unsigned char *newaddr) { struct net_bridge *br = p->br; @@ -107,30 +135,9 @@ void br_fdb_changeaddr(struct net_bridge_port *p, const unsigned char *newaddr) f = hlist_entry(h, struct net_bridge_fdb_entry, hlist); if (f->dst == p && f->is_local && !f->added_by_user) { - /* maybe another port has same hw addr? */ - struct net_bridge_port *op; - u16 vid = f->vlan_id; - list_for_each_entry(op, &br->port_list, list) { - if (op != p && - ether_addr_equal(op->dev->dev_addr, - f->addr.addr) && - (!vid || nbp_vlan_find(op, vid))) { - f->dst = op; - goto skip_delete; - } - } - - /* maybe bridge device has same hw addr? */ - if (ether_addr_equal(br->dev->dev_addr, - f->addr.addr) && - (!vid || br_vlan_find(br, vid))) { - f->dst = NULL; - goto skip_delete; - } - /* delete old one */ - fdb_delete(br, f); -skip_delete: + fdb_delete_local(br, p, f); + /* if this port has no vlan information * configured, we can safely be done at * this point. @@ -168,7 +175,7 @@ void br_fdb_change_mac_address(struct net_bridge *br, const u8 *newaddr) /* If old entry was unassociated with any port, then delete it. */ f = __br_fdb_get(br, br->dev->dev_addr, 0); if (f && f->is_local && !f->dst) - fdb_delete(br, f); + fdb_delete_local(br, NULL, f); fdb_insert(br, NULL, newaddr, 0); @@ -183,7 +190,7 @@ void br_fdb_change_mac_address(struct net_bridge *br, const u8 *newaddr) for_each_set_bit_from(vid, pv->vlan_bitmap, VLAN_N_VID) { f = __br_fdb_get(br, br->dev->dev_addr, vid); if (f && f->is_local && !f->dst) - fdb_delete(br, f); + fdb_delete_local(br, NULL, f); fdb_insert(br, NULL, newaddr, vid); } } From a778e6d1a51faaafa6a3a3cef9bee11c3bd47f9f Mon Sep 17 00:00:00 2001 From: Toshiaki Makita Date: Fri, 7 Feb 2014 16:48:24 +0900 Subject: [PATCH 464/596] bridge: Properly check if local fdb entry can be deleted in br_fdb_delete_by_port br_fdb_delete_by_port() doesn't care about vlan and mac address of the bridge device. As the check is almost the same as mac address changing, slightly modify fdb_delete_local() and use it. Note that we can always set added_by_user to 0 in fdb_delete_local() because - br_fdb_delete_by_port() calls fdb_delete_local() for local entries regardless of its added_by_user. In this case, we have to check if another port has the same address and vlan, and if found, we have to create the entry (by changing dst). This is kernel-added entry, not user-added. - br_fdb_changeaddr() doesn't call fdb_delete_local() for user-added entry. Signed-off-by: Toshiaki Makita Acked-by: Vlad Yasevich Signed-off-by: David S. Miller --- net/bridge/br_fdb.c | 25 ++++++------------------- 1 file changed, 6 insertions(+), 19 deletions(-) diff --git a/net/bridge/br_fdb.c b/net/bridge/br_fdb.c index 15bf13d0b543..43e54d1ae956 100644 --- a/net/bridge/br_fdb.c +++ b/net/bridge/br_fdb.c @@ -103,6 +103,7 @@ static void fdb_delete_local(struct net_bridge *br, if (op != p && ether_addr_equal(op->dev->dev_addr, addr) && (!vid || nbp_vlan_find(op, vid))) { f->dst = op; + f->added_by_user = 0; return; } } @@ -111,6 +112,7 @@ static void fdb_delete_local(struct net_bridge *br, if (p && ether_addr_equal(br->dev->dev_addr, addr) && (!vid || br_vlan_find(br, vid))) { f->dst = NULL; + f->added_by_user = 0; return; } @@ -261,26 +263,11 @@ void br_fdb_delete_by_port(struct net_bridge *br, if (f->is_static && !do_all) continue; - /* - * if multiple ports all have the same device address - * then when one port is deleted, assign - * the local entry to other port - */ - if (f->is_local) { - struct net_bridge_port *op; - list_for_each_entry(op, &br->port_list, list) { - if (op != p && - ether_addr_equal(op->dev->dev_addr, - f->addr.addr)) { - f->dst = op; - f->added_by_user = 0; - goto skip_delete; - } - } - } - fdb_delete(br, f); - skip_delete: ; + if (f->is_local) + fdb_delete_local(br, p, f); + else + fdb_delete(br, f); } } spin_unlock_bh(&br->hash_lock); From 424bb9c97cc1992018950485ca7410779b8ea21d Mon Sep 17 00:00:00 2001 From: Toshiaki Makita Date: Fri, 7 Feb 2014 16:48:25 +0900 Subject: [PATCH 465/596] bridge: Properly check if local fdb entry can be deleted when deleting vlan Vlan codes unconditionally delete local fdb entries. We should consider the possibility that other ports have the same address and vlan. Example of problematic case: ip link set eth0 address 12:34:56:78:90:ab ip link set eth1 address aa:bb:cc:dd:ee:ff brctl addif br0 eth0 brctl addif br0 eth1 # br0 will have mac address 12:34:56:78:90:ab bridge vlan add dev eth0 vid 10 bridge vlan add dev eth1 vid 10 bridge vlan add dev br0 vid 10 self We will have fdb entry such that f->dst == eth0, f->vlan_id == 10 and f->addr == 12:34:56:78:90:ab at this time. Next, delete eth0 vlan 10. bridge vlan del dev eth0 vid 10 In this case, we still need the entry for br0, but it will be deleted. Note that br0 needs the entry even though its mac address is not set manually. To delete the entry with proper condition checking, fdb_delete_local() is suitable to use. Signed-off-by: Toshiaki Makita Acked-by: Vlad Yasevich Signed-off-by: David S. Miller --- net/bridge/br_fdb.c | 20 ++++++++++++++++++-- net/bridge/br_private.h | 4 +++- net/bridge/br_vlan.c | 8 ++------ 3 files changed, 23 insertions(+), 9 deletions(-) diff --git a/net/bridge/br_fdb.c b/net/bridge/br_fdb.c index 43e54d1ae956..0426dff4b3fd 100644 --- a/net/bridge/br_fdb.c +++ b/net/bridge/br_fdb.c @@ -27,6 +27,9 @@ #include "br_private.h" static struct kmem_cache *br_fdb_cache __read_mostly; +static struct net_bridge_fdb_entry *fdb_find(struct hlist_head *head, + const unsigned char *addr, + __u16 vid); static int fdb_insert(struct net_bridge *br, struct net_bridge_port *source, const unsigned char *addr, u16 vid); static void fdb_notify(struct net_bridge *br, @@ -119,6 +122,20 @@ static void fdb_delete_local(struct net_bridge *br, fdb_delete(br, f); } +void br_fdb_find_delete_local(struct net_bridge *br, + const struct net_bridge_port *p, + const unsigned char *addr, u16 vid) +{ + struct hlist_head *head = &br->hash[br_mac_hash(addr, vid)]; + struct net_bridge_fdb_entry *f; + + spin_lock_bh(&br->hash_lock); + f = fdb_find(head, addr, vid); + if (f && f->is_local && !f->added_by_user && f->dst == p) + fdb_delete_local(br, p, f); + spin_unlock_bh(&br->hash_lock); +} + void br_fdb_changeaddr(struct net_bridge_port *p, const unsigned char *newaddr) { struct net_bridge *br = p->br; @@ -770,8 +787,7 @@ int br_fdb_add(struct ndmsg *ndm, struct nlattr *tb[], return err; } -int fdb_delete_by_addr(struct net_bridge *br, const u8 *addr, - u16 vlan) +static int fdb_delete_by_addr(struct net_bridge *br, const u8 *addr, u16 vlan) { struct hlist_head *head = &br->hash[br_mac_hash(addr, vlan)]; struct net_bridge_fdb_entry *fdb; diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h index f91e1d9c8e92..3ba11bc99b65 100644 --- a/net/bridge/br_private.h +++ b/net/bridge/br_private.h @@ -371,6 +371,9 @@ static inline void br_netpoll_disable(struct net_bridge_port *p) int br_fdb_init(void); void br_fdb_fini(void); void br_fdb_flush(struct net_bridge *br); +void br_fdb_find_delete_local(struct net_bridge *br, + const struct net_bridge_port *p, + const unsigned char *addr, u16 vid); void br_fdb_changeaddr(struct net_bridge_port *p, const unsigned char *newaddr); void br_fdb_change_mac_address(struct net_bridge *br, const u8 *newaddr); void br_fdb_cleanup(unsigned long arg); @@ -385,7 +388,6 @@ int br_fdb_insert(struct net_bridge *br, struct net_bridge_port *source, const unsigned char *addr, u16 vid); void br_fdb_update(struct net_bridge *br, struct net_bridge_port *source, const unsigned char *addr, u16 vid, bool added_by_user); -int fdb_delete_by_addr(struct net_bridge *br, const u8 *addr, u16 vid); int br_fdb_delete(struct ndmsg *ndm, struct nlattr *tb[], struct net_device *dev, const unsigned char *addr); diff --git a/net/bridge/br_vlan.c b/net/bridge/br_vlan.c index 233ec1c6e9db..8249ca764c79 100644 --- a/net/bridge/br_vlan.c +++ b/net/bridge/br_vlan.c @@ -275,9 +275,7 @@ int br_vlan_delete(struct net_bridge *br, u16 vid) if (!pv) return -EINVAL; - spin_lock_bh(&br->hash_lock); - fdb_delete_by_addr(br, br->dev->dev_addr, vid); - spin_unlock_bh(&br->hash_lock); + br_fdb_find_delete_local(br, NULL, br->dev->dev_addr, vid); __vlan_del(pv, vid); return 0; @@ -378,9 +376,7 @@ int nbp_vlan_delete(struct net_bridge_port *port, u16 vid) if (!pv) return -EINVAL; - spin_lock_bh(&port->br->hash_lock); - fdb_delete_by_addr(port->br, port->dev->dev_addr, vid); - spin_unlock_bh(&port->br->hash_lock); + br_fdb_find_delete_local(port->br, port, port->dev->dev_addr, vid); return __vlan_del(pv, vid); } From ac4c8868837a7c70ebb3eaf2298324a3b61b214e Mon Sep 17 00:00:00 2001 From: Toshiaki Makita Date: Fri, 7 Feb 2014 16:48:26 +0900 Subject: [PATCH 466/596] bridge: Prevent possible race condition in br_fdb_change_mac_address br_fdb_change_mac_address() calls fdb_insert()/fdb_delete() without br->hash_lock. These hash list updates are racy with br_fdb_update()/br_fdb_cleanup(). Signed-off-by: Toshiaki Makita Acked-by: Vlad Yasevich Signed-off-by: David S. Miller --- net/bridge/br_fdb.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/net/bridge/br_fdb.c b/net/bridge/br_fdb.c index 0426dff4b3fd..9203d5a1943f 100644 --- a/net/bridge/br_fdb.c +++ b/net/bridge/br_fdb.c @@ -191,6 +191,8 @@ void br_fdb_change_mac_address(struct net_bridge *br, const u8 *newaddr) struct net_port_vlans *pv; u16 vid = 0; + spin_lock_bh(&br->hash_lock); + /* If old entry was unassociated with any port, then delete it. */ f = __br_fdb_get(br, br->dev->dev_addr, 0); if (f && f->is_local && !f->dst) @@ -204,7 +206,7 @@ void br_fdb_change_mac_address(struct net_bridge *br, const u8 *newaddr) */ pv = br_get_vlan_info(br); if (!pv) - return; + goto out; for_each_set_bit_from(vid, pv->vlan_bitmap, VLAN_N_VID) { f = __br_fdb_get(br, br->dev->dev_addr, vid); @@ -212,6 +214,8 @@ void br_fdb_change_mac_address(struct net_bridge *br, const u8 *newaddr) fdb_delete_local(br, NULL, f); fdb_insert(br, NULL, newaddr, vid); } +out: + spin_unlock_bh(&br->hash_lock); } void br_fdb_cleanup(unsigned long _data) From bf06200e732de613a1277984bf34d1a21c2de03d Mon Sep 17 00:00:00 2001 From: John Ogness Date: Sun, 9 Feb 2014 18:40:11 -0800 Subject: [PATCH 467/596] tcp: tsq: fix nonagle handling Commit 46d3ceabd8d9 ("tcp: TCP Small Queues") introduced a possible regression for applications using TCP_NODELAY. If TCP session is throttled because of tsq, we should consult tp->nonagle when TX completion is done and allow us to send additional segment, especially if this segment is not a full MSS. Otherwise this segment is sent after an RTO. [edumazet] : Cooked the changelog, added another fix about testing sk_wmem_alloc twice because TX completion can happen right before setting TSQ_THROTTLED bit. This problem is particularly visible with recent auto corking, but might also be triggered with low tcp_limit_output_bytes values or NIC drivers delaying TX completion by hundred of usec, and very low rtt. Thomas Glanzmann for example reported an iscsi regression, caused by tcp auto corking making this bug quite visible. Fixes: 46d3ceabd8d9 ("tcp: TCP Small Queues") Signed-off-by: John Ogness Signed-off-by: Eric Dumazet Reported-by: Thomas Glanzmann Signed-off-by: David S. Miller --- net/ipv4/tcp_output.c | 13 +++++++++++-- 1 file changed, 11 insertions(+), 2 deletions(-) diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 10435b3b9d0f..3be16727f058 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -698,7 +698,8 @@ static void tcp_tsq_handler(struct sock *sk) if ((1 << sk->sk_state) & (TCPF_ESTABLISHED | TCPF_FIN_WAIT1 | TCPF_CLOSING | TCPF_CLOSE_WAIT | TCPF_LAST_ACK)) - tcp_write_xmit(sk, tcp_current_mss(sk), 0, 0, GFP_ATOMIC); + tcp_write_xmit(sk, tcp_current_mss(sk), tcp_sk(sk)->nonagle, + 0, GFP_ATOMIC); } /* * One tasklet per cpu tries to send more skbs. @@ -1904,7 +1905,15 @@ static bool tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle, if (atomic_read(&sk->sk_wmem_alloc) > limit) { set_bit(TSQ_THROTTLED, &tp->tsq_flags); - break; + /* It is possible TX completion already happened + * before we set TSQ_THROTTLED, so we must + * test again the condition. + * We abuse smp_mb__after_clear_bit() because + * there is no smp_mb__after_set_bit() yet + */ + smp_mb__after_clear_bit(); + if (atomic_read(&sk->sk_wmem_alloc) > limit) + break; } limit = mss_now; From 49d3d6c37a322117b1eeb410a49165bb3d0441f7 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Mon, 10 Feb 2014 14:25:30 -0800 Subject: [PATCH 468/596] drivers/misc/sgi-gru/grukdump.c: unlocking should be conditional in gru_dump_context() I was reviewing this and noticed that unlocking should be conditional on the error path. I've changed it to unlock and return directly since we only do it once and it seems unlikely to change in the near future. Signed-off-by: Dan Carpenter Acked-by: Dimitri Sivanich Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/misc/sgi-gru/grukdump.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) diff --git a/drivers/misc/sgi-gru/grukdump.c b/drivers/misc/sgi-gru/grukdump.c index 9b2062d17327..2bef3f76032a 100644 --- a/drivers/misc/sgi-gru/grukdump.c +++ b/drivers/misc/sgi-gru/grukdump.c @@ -139,8 +139,11 @@ static int gru_dump_context(struct gru_state *gru, int ctxnum, ubuf += sizeof(hdr); ubufcch = ubuf; - if (gru_user_copy_handle(&ubuf, cch)) - goto fail; + if (gru_user_copy_handle(&ubuf, cch)) { + if (cch_locked) + unlock_cch_handle(cch); + return -EFAULT; + } if (cch_locked) ubufcch->delresp = 0; bytes = sizeof(hdr) + GRU_CACHE_LINE_BYTES; @@ -179,10 +182,6 @@ static int gru_dump_context(struct gru_state *gru, int ctxnum, ret = -EFAULT; return ret ? ret : bytes; - -fail: - unlock_cch_handle(cch); - return -EFAULT; } int gru_dump_chiplet_request(unsigned long arg) From ec83b616a71d76cc062371339472ed66ba405824 Mon Sep 17 00:00:00 2001 From: Richard Genoud Date: Mon, 10 Feb 2014 14:25:31 -0800 Subject: [PATCH 469/596] get_maintainer: fix detection of git repository Since git v1.7.7, the .git directory can be a file when, for example, the kernel is a submodule of another git super project. So, the check "-d .git" is not working anymore in this case. Using a more generic check like "-e .git" corrects this behaviour. Signed-off-by: Richard Genoud Cc: Andy Whitcroft Cc: Joe Perches Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- scripts/get_maintainer.pl | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/scripts/get_maintainer.pl b/scripts/get_maintainer.pl index 9c3986f4140c..41987885bd31 100755 --- a/scripts/get_maintainer.pl +++ b/scripts/get_maintainer.pl @@ -95,7 +95,7 @@ my %VCS_cmds; my %VCS_cmds_git = ( "execute_cmd" => \&git_execute_cmd, - "available" => '(which("git") ne "") && (-d ".git")', + "available" => '(which("git") ne "") && (-e ".git")', "find_signers_cmd" => "git log --no-color --follow --since=\$email_git_since " . '--numstat --no-merges ' . From 3645e3283b90524ed9a6b33b68c93ee29d1bfeb9 Mon Sep 17 00:00:00 2001 From: Richard Genoud Date: Mon, 10 Feb 2014 14:25:32 -0800 Subject: [PATCH 470/596] checkpatch: fix detection of git repository Since git v1.7.7, the .git directory can be a file when, for example, the kernel is a submodule of another git super project. So, the check "-d .git" is not working anymore in this case. Using a more generic check like "-e .git" corrects this behaviour. Signed-off-by: Richard Genoud Cc: Andy Whitcroft Cc: Joe Perches Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- scripts/checkpatch.pl | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl index 0ea2a1e24ade..464dcef79b35 100755 --- a/scripts/checkpatch.pl +++ b/scripts/checkpatch.pl @@ -471,7 +471,7 @@ sub seed_camelcase_includes { $camelcase_seeded = 1; - if (-d ".git") { + if (-e ".git") { my $git_last_include_commit = `git log --no-merges --pretty=format:"%h%n" -1 -- include`; chomp $git_last_include_commit; $camelcase_cache = ".checkpatch-camelcase.git.$git_last_include_commit"; @@ -499,7 +499,7 @@ sub seed_camelcase_includes { return; } - if (-d ".git") { + if (-e ".git") { $files = `git ls-files "include/*.h"`; @include_files = split('\n', $files); } From 3cf8ca1c250d3c3d8bf61f20bd71acd1876bf777 Mon Sep 17 00:00:00 2001 From: Henrik Austad Date: Mon, 10 Feb 2014 14:25:33 -0800 Subject: [PATCH 471/596] Documentation/: update 00-INDEX files Some of the 00-INDEX files are somewhat outdated and some folders does not contain 00-INDEX at all. Only outdated (with the notably exception of spi) indexes are touched here, the 169 folders without 00-INDEX has not been touched. New 00-INDEX - spi/* was added in a series of commits dating back to 2006 Added files (missing in (*/)00-INDEX) - dmatest.txt was added by commit 851b7e16a07d ("dmatest: run test via debugfs") - this_cpu_ops.txt was added by commit a1b2a555d637 ("percpu: add documentation on this_cpu operations") - ww-mutex-design.txt was added by commit 040a0a371005 ("mutex: Add support for wound/wait style locks") - bcache.txt was added by commit cafe56359144 ("bcache: A block layer cache") - kernel-per-CPU-kthreads.txt was added by commit 49717cb40410 ("kthread: Document ways of reducing OS jitter due to per-CPU kthreads") - phy.txt was added by commit ff764963479a ("drivers: phy: add generic PHY framework") - block/null_blk was added by commit 12f8f4fc0314 ("null_blk: documentation") - module-signing.txt was added by commit 3cafea307642 ("Add Documentation/module-signing.txt file") - assoc_array.txt was added by commit 3cb989501c26 ("Add a generic associative array implementation.") - arm/IXP4xx was part of the initial repo - arm/cluster-pm-race-avoidance.txt was added by commit 7fe31d28e839 ("ARM: mcpm: introduce helpers for platform coherency exit/setup") - arm/firmware.txt was added by commit 7366b92a77fc ("ARM: Add interface for registering and calling firmware-specific operations") - arm/kernel_mode_neon.txt was added by commit 2afd0a05241d ("ARM: 7825/1: document the use of NEON in kernel mode") - arm/tcm.txt was added by commit bc581770cfdd ("ARM: 5580/2: ARM TCM (Tightly-Coupled Memory) support v3") - arm/vlocks.txt was added by commit 9762f12d3e05 ("ARM: mcpm: Add baremetal voting mutexes") - blackfin/gptimers-example.c, Makefile was added by commit 4b60779d5ea7 ("Blackfin: add an example showing how to use the gptimers API") - devicetree/usage-model.txt was added by commit 31134efc681a ("dt: Linux DT usage model documentation") - fb/api.txt was added by commit fb21c2f42879 ("fbdev: Add FOURCC-based format configuration API") - fb/sm501.txt was added by commit e6a049807105 ("video, sm501: add edid and commandline support") - fb/udlfb.txt was added by commit 96f8d864afd6 ("fbdev: move udlfb out of staging.") - filesystems/Makefile was added by commit 1e0051ae48a2 ("Documentation/fs/: split txt and source files") - filesystems/nfs/nfsd-admin-interfaces.txt was added by commit 8a4c6e19cfed ("nfsd: document kernel interfaces for nfsd configuration") - ide/warm-plug-howto.txt was added by commit f74c91413ec6 ("ide: add warm-plug support for IDE devices (take 2)") - laptops/Makefile was added by commit d49129accc21 ("Documentation/laptop/: split txt and source files") - leds/leds-blinkm.txt was added by commit b54cf35a7f65 ("LEDS: add BlinkM RGB LED driver, documentation and update MAINTAINERS") - leds/ledtrig-oneshot.txt was added by commit 5e417281cde2 ("leds: add oneshot trigger") - leds/ledtrig-transient.txt was added by commit 44e1e9f8e705 ("leds: add new transient trigger for one shot timer activation") - m68k/README.buddha was part of the initial repo - networking/LICENSE.(qla3xxx|qlcnic|qlge) was added by commits 40839129f779, c4e84bde1d59, 5a4faa873782 - networking/Makefile was added by commit 3794f3e812ef ("docsrc: build Documentation/ sources") - networking/i40evf.txt was added by commit 105bf2fe6b32 ("i40evf: add driver to kernel build system") - networking/ipsec.txt was added by commit b3c6efbc36e2 ("xfrm: Add file to document IPsec corner case") - networking/mac80211-auth-assoc-deauth.txt was added by commit 3cd7920a2be8 ("mac80211: add auth/assoc/deauth flow diagram") - networking/netlink_mmap.txt was added by commit 5683264c3981 ("netlink: add documentation for memory mapped I/O") - networking/nf_conntrack-sysctl.txt was added by commit c9f9e0e1597f ("netfilter: doc: add nf_conntrack sysctl api documentation") lan) - networking/team.txt was added by commit 3d249d4ca7d0 ("net: introduce ethernet teaming device") - networking/vxlan.txt was added by commit d342894c5d2f ("vxlan: virtual extensible lan") - power/runtime_pm.txt was added by commit 5e928f77a09a ("PM: Introduce core framework for run-time PM of I/O devices (rev. 17)") - power/charger-manager.txt was added by commit 3bb3dbbd56ea ("power_supply: Add initial Charger-Manager driver") - RCU/lockdep-splat.txt was added by commit d7bd2d68aa2e ("rcu: Document interpretation of RCU-lockdep splats") - s390/kvm.txt was added by 5ecee4b (KVM: s390: API documentation) - s390/qeth.txt was added by commit b4d72c08b358 ("qeth: bridgeport support - basic control") - scheduler/sched-bwc.txt was added by commit 88ebc08ea9f7 ("sched: Add documentation for bandwidth control") - scsi/advansys.txt was added by commit 4bd6d7f35661 ("[SCSI] advansys: Move documentation to Documentation/scsi") - scsi/bfa.txt was added by commit 1ec90174bdb4 ("[SCSI] bfa: add readme file") - scsi/bnx2fc.txt was added by commit 12b8fc10eaf4 ("[SCSI] bnx2fc: Add driver documentation") - scsi/cxgb3i.txt was added by commit c3673464ebc0 ("[SCSI] cxgb3i: Add cxgb3i iSCSI driver.") - scsi/hpsa.txt was added by commit 992ebcf14f3c ("[SCSI] hpsa: Add hpsa.txt to Documentation/scsi") - scsi/link_power_management_policy.txt was added by commit ca77329fb713 ("[libata] Link power management infrastructure") - scsi/osd.txt was added by commit 78e0c621deca ("[SCSI] osd: Documentation for OSD library") - scsi/scsi-parameter.txt was created/moved by commit 163475fb111c ("Documentation: move SCSI parameters to their own text file") - serial/driver was part of the initial repo - serial/n_gsm.txt was added by commit 323e84122ec6 ("n_gsm: add a documentation") - timers/Makefile was added by commit 3794f3e812ef ("docsrc: build Documentation/ sources") - virt/kvm/s390.txt was added by commit d9101fca3d57 ("KVM: s390: diagnose call documentation") - vm/split_page_table_lock was added by commit 49076ec2ccaf ("mm: dynamically allocate page->ptl if it cannot be embedded to struct page") - w1/slaves/w1_ds28e04 was added by commit fbf7f7b4e2ae ("w1: Add 1-wire slave device driver for DS28E04-100") - w1/masters/omap-hdq was added by commit e0a29382c6f5 ("hdq: documentation for OMAP HDQ") - x86/early-microcode.txt was added by commit 0d91ea86a895 ("x86, doc: Documentation for early microcode loading") - x86/earlyprintk.txt was added by commit a1aade478862 ("x86/doc: mini-howto for using earlyprintk=dbgp") - x86/entry_64.txt was added by commit 8b4777a4b50c ("x86-64: Document some of entry_64.S") - x86/pat.txt was added by commit d27554d874c7 ("x86: PAT documentation") Moved files - arm/kernel_user_helpers.txt was moved out of arch/arm/kernel by commit 37b8304642c7 ("ARM: kuser: move interface documentation out of the source code") - efi-stub.txt was moved out of x86/ and down into Documentation/ in commit 4172fe2f8a47 ("EFI stub documentation updates") - laptops/hpfall.c was moved out of hwmon/ and into laptops/ in commit efcfed9bad88 ("Move hp_accel to drivers/platform/x86") - commit 5616c23ad9cd ("x86: doc: move x86-generic documentation from Doc/x86/i386"): * x86/usb-legacy-support.txt * x86/boot.txt * x86/zero_page.txt - power/video_extension.txt was moved to acpi in commit 70e66e4df191 ("ACPI / video: move video_extension.txt to Documentation/acpi") Removed files (left in 00-INDEX) - memory.txt was removed by commit 00ea8990aadf ("memory.txt: remove stray information") - gpio.txt was moved to gpio/ in commit fd8e198cfcaa ("Documentation: gpiolib: document new interface") - networking/DLINK.txt was removed by commit 168e06ae26dd ("drivers/net: delete old parallel port de600/de620 drivers") - serial/hayes-esp.txt was removed by commit f53a2ade0bb9 ("tty: esp: remove broken driver") - s390/TAPE was removed by commit 9e280f669308 ("[S390] remove tape block docu") - vm/locking was removed by commit 57ea8171d2bc ("mm: documentation: remove hopelessly out-of-date locking doc") - laptops/acer-wmi.txt was remvoed by commit 020036678e81 ("acer-wmi: Delete out-of-date documentation") Typos/misc issues - rpc-server-gss.txt was added as knfsd-rpcgss.txt in commit 030d794bf498 ("SUNRPC: Use gssproxy upcall for server RPCGSS authentication.") - commit b88cf73d9278 ("net: add missing entries to Documentation/networking/00-INDEX") * generic-hdlc.txt was added as generic_hdlc.txt * spider_net.txt was added as spider-net.txt - w1/master/mxc-w1 was added as mxc_w1 by commit a5fd9139f74c ("w1: add 1-wire master driver for i.MX27 / i.MX31") - s390/zfcpdump.txt was added as zfcpdump by commit 6920c12a407e ("[S390] Add Documentation/s390/00-INDEX.") Signed-off-by: Henrik Austad Reviewed-by: Paul E. McKenney [rcu bits] Acked-by: Rob Landley Cc: Jiri Kosina Cc: Thomas Gleixner Cc: Rob Herring Cc: David S. Miller Cc: Mark Brown Cc: "H. Peter Anvin" Cc: Ingo Molnar Cc: Gleb Natapov Cc: Linus Torvalds Cc: Len Brown Cc: James Bottomley Cc: Jean-Christophe Plagniol-Villard Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/00-INDEX | 24 +++++++++++++++++---- Documentation/RCU/00-INDEX | 2 ++ Documentation/arm/00-INDEX | 14 ++++++++++++ Documentation/blackfin/00-INDEX | 6 ++++-- Documentation/block/00-INDEX | 2 ++ Documentation/devicetree/00-INDEX | 2 ++ Documentation/fb/00-INDEX | 6 ++++++ Documentation/filesystems/00-INDEX | 2 ++ Documentation/filesystems/nfs/00-INDEX | 4 +++- Documentation/ide/00-INDEX | 2 ++ Documentation/laptops/00-INDEX | 6 ++++-- Documentation/leds/00-INDEX | 8 +++++++ Documentation/m68k/00-INDEX | 2 ++ Documentation/networking/00-INDEX | 30 +++++++++++++++++++++----- Documentation/power/00-INDEX | 6 ++++-- Documentation/s390/00-INDEX | 8 ++++--- Documentation/scheduler/00-INDEX | 2 ++ Documentation/scsi/00-INDEX | 16 ++++++++++++++ Documentation/serial/00-INDEX | 6 ++++-- Documentation/spi/00-INDEX | 22 +++++++++++++++++++ Documentation/timers/00-INDEX | 2 ++ Documentation/virtual/kvm/00-INDEX | 2 ++ Documentation/vm/00-INDEX | 4 ++-- Documentation/w1/masters/00-INDEX | 4 +++- Documentation/w1/slaves/00-INDEX | 2 ++ Documentation/x86/00-INDEX | 18 ++++++++++++++-- 26 files changed, 176 insertions(+), 26 deletions(-) create mode 100644 Documentation/spi/00-INDEX diff --git a/Documentation/00-INDEX b/Documentation/00-INDEX index 38f8444bdd0e..07de7e19b4ce 100644 --- a/Documentation/00-INDEX +++ b/Documentation/00-INDEX @@ -29,6 +29,8 @@ DMA-ISA-LPC.txt - How to do DMA with ISA (and LPC) devices. DMA-attributes.txt - listing of the various possible attributes a DMA region can have +dmatest.txt + - how to compile, configure and use the dmatest system. DocBook/ - directory with DocBook templates etc. for kernel documentation. EDID/ @@ -77,6 +79,8 @@ arm/ - directory with info about Linux on the ARM architecture. arm64/ - directory with info about Linux on the 64 bit ARM architecture. +assoc_array.txt + - generic associative array intro. atomic_ops.txt - semantics and behavior of atomic and bitmask operations. auxdisplay/ @@ -87,6 +91,8 @@ bad_memory.txt - how to use kernel parameters to exclude bad RAM regions. basic_profiling.txt - basic instructions for those who wants to profile Linux kernel. +bcache.txt + - Block-layer cache on fast SSDs to improve slow (raid) I/O performance. binfmt_misc.txt - info on the kernel support for extra binary formats. blackfin/ @@ -171,6 +177,8 @@ early-userspace/ - info about initramfs, klibc, and userspace early during boot. edac.txt - information on EDAC - Error Detection And Correction +efi-stub.txt + - How to use the EFI boot stub to bypass GRUB or elilo on EFI systems. eisa.txt - info on EISA bus support. email-clients.txt @@ -195,8 +203,8 @@ futex-requeue-pi.txt - info on requeueing of tasks from a non-PI futex to a PI futex gcov.txt - use of GCC's coverage testing tool "gcov" with the Linux kernel -gpio.txt - - overview of GPIO (General Purpose Input/Output) access conventions. +gpio/ + - gpio related documentation hid/ - directory with information on human interface devices highuid.txt @@ -255,6 +263,8 @@ kernel-docs.txt - listing of various WWW + books that document kernel internals. kernel-parameters.txt - summary listing of command line / boot prompt args for the kernel. +kernel-per-CPU-kthreads.txt + - List of all per-CPU kthreads and how they introduce jitter. kmemcheck.txt - info on dynamic checker that detects uses of uninitialized memory. kmemleak.txt @@ -299,8 +309,6 @@ memory-devices/ - directory with info on parts like the Texas Instruments EMIF driver memory-hotplug.txt - Hotpluggable memory support, how to use and current status. -memory.txt - - info on typical Linux memory problems. metag/ - directory with info about Linux on Meta architecture. mips/ @@ -311,6 +319,8 @@ mmc/ - directory with info about the MMC subsystem mn10300/ - directory with info about the mn10300 architecture port +module-signing.txt + - Kernel module signing for increased security when loading modules. mtd/ - directory with info about memory technology devices (flash) mono.txt @@ -343,6 +353,8 @@ pcmcia/ - info on the Linux PCMCIA driver. percpu-rw-semaphore.txt - RCU based read-write semaphore optimized for locking for reading +phy.txt + - Description of the generic PHY framework. pi-futex.txt - documentation on lightweight priority inheritance futexes. pinctrl.txt @@ -431,6 +443,8 @@ sysrq.txt - info on the magic SysRq key. target/ - directory with info on generating TCM v4 fabric .ko modules +this_cpu_ops.txt + - List rationale behind and the way to use this_cpu operations. thermal/ - directory with information on managing thermal issues (CPU/temp) trace/ @@ -469,6 +483,8 @@ wimax/ - directory with info about Intel Wireless Wimax Connections workqueue.txt - information on the Concurrency Managed Workqueue implementation +ww-mutex-design.txt + - Intro to Mutex wait/would deadlock handling.s x86/x86_64/ - directory with info on Linux support for AMD x86-64 (Hammer) machines. xtensa/ diff --git a/Documentation/RCU/00-INDEX b/Documentation/RCU/00-INDEX index 1d7a885761f5..fa57139f50bf 100644 --- a/Documentation/RCU/00-INDEX +++ b/Documentation/RCU/00-INDEX @@ -8,6 +8,8 @@ listRCU.txt - Using RCU to Protect Read-Mostly Linked Lists lockdep.txt - RCU and lockdep checking +lockdep-splat.txt + - RCU Lockdep splats explained. NMI-RCU.txt - Using RCU to Protect Dynamic NMI Handlers rcubarrier.txt diff --git a/Documentation/arm/00-INDEX b/Documentation/arm/00-INDEX index 36420e116c90..a94090cc785d 100644 --- a/Documentation/arm/00-INDEX +++ b/Documentation/arm/00-INDEX @@ -4,6 +4,8 @@ Booting - requirements for booting Interrupts - ARM Interrupt subsystem documentation +IXP4xx + - Intel IXP4xx Network processor. msm - MSM specific documentation Netwinder @@ -24,8 +26,16 @@ SPEAr - ST SPEAr platform Linux Overview VFP/ - Release notes for Linux Kernel Vector Floating Point support code +cluster-pm-race-avoidance.txt + - Algorithm for CPU and Cluster setup/teardown empeg/ - Ltd's Empeg MP3 Car Audio Player +firmware.txt + - Secure firmware registration and calling. +kernel_mode_neon.txt + - How to use NEON instructions in kernel mode +kernel_user_helpers.txt + - Helper functions in kernel space made available for userspace. mem_alignment - alignment abort handler documentation memory.txt @@ -34,3 +44,7 @@ nwfpe/ - NWFPE floating point emulator documentation swp_emulation - SWP/SWPB emulation handler/logging description +tcm.txt + - ARM Tightly Coupled Memory +vlocks.txt + - Voting locks, low-level mechanism relying on memory system atomic writes. diff --git a/Documentation/blackfin/00-INDEX b/Documentation/blackfin/00-INDEX index 2df0365f2dff..c54fcdd4ae9f 100644 --- a/Documentation/blackfin/00-INDEX +++ b/Documentation/blackfin/00-INDEX @@ -1,8 +1,10 @@ 00-INDEX - This file - +Makefile + - Makefile for gptimers example file. bfin-gpio-notes.txt - Notes in developing/using bfin-gpio driver. - bfin-spi-notes.txt - Notes for using bfin spi bus driver. +gptimers-example.c + - gptimers example diff --git a/Documentation/block/00-INDEX b/Documentation/block/00-INDEX index 929d9904f74b..e840b47613f7 100644 --- a/Documentation/block/00-INDEX +++ b/Documentation/block/00-INDEX @@ -14,6 +14,8 @@ deadline-iosched.txt - Deadline IO scheduler tunables ioprio.txt - Block io priorities (in CFQ scheduler) +null_blk.txt + - Null block for block-layer benchmarking. queue-sysfs.txt - Queue's sysfs entries request.txt diff --git a/Documentation/devicetree/00-INDEX b/Documentation/devicetree/00-INDEX index b78f691fd847..8c4102c6a5e7 100644 --- a/Documentation/devicetree/00-INDEX +++ b/Documentation/devicetree/00-INDEX @@ -8,3 +8,5 @@ https://lists.ozlabs.org/listinfo/devicetree-discuss - this file booting-without-of.txt - Booting Linux without Open Firmware, describes history and format of device trees. +usage-model.txt + - How Linux uses DT and what DT aims to solve. \ No newline at end of file diff --git a/Documentation/fb/00-INDEX b/Documentation/fb/00-INDEX index 30a70542e823..fe85e7c5907a 100644 --- a/Documentation/fb/00-INDEX +++ b/Documentation/fb/00-INDEX @@ -5,6 +5,8 @@ please mail me. 00-INDEX - this file. +api.txt + - The frame buffer API between applications and buffer devices. arkfb.txt - info on the fbdev driver for ARK Logic chips. aty128fb.txt @@ -51,12 +53,16 @@ sh7760fb.txt - info on the SH7760/SH7763 integrated LCDC Framebuffer driver. sisfb.txt - info on the framebuffer device driver for various SiS chips. +sm501.txt + - info on the framebuffer device driver for sm501 videoframebuffer. sstfb.txt - info on the frame buffer driver for 3dfx' Voodoo Graphics boards. tgafb.txt - info on the TGA (DECChip 21030) frame buffer driver. tridentfb.txt info on the framebuffer driver for some Trident chip based cards. +udlfb.txt + - Driver for DisplayLink USB 2.0 chips. uvesafb.txt - info on the userspace VESA (VBE2+ compliant) frame buffer device. vesafb.txt diff --git a/Documentation/filesystems/00-INDEX b/Documentation/filesystems/00-INDEX index 632211cbdd56..ac28149aede4 100644 --- a/Documentation/filesystems/00-INDEX +++ b/Documentation/filesystems/00-INDEX @@ -2,6 +2,8 @@ - this file (info on some of the filesystems supported by linux). Locking - info on locking rules as they pertain to Linux VFS. +Makefile + - Makefile for building the filsystems-part of DocBook. 9p.txt - 9p (v9fs) is an implementation of the Plan 9 remote fs protocol. adfs.txt diff --git a/Documentation/filesystems/nfs/00-INDEX b/Documentation/filesystems/nfs/00-INDEX index 66eb6c8c5334..53f3b596ac0d 100644 --- a/Documentation/filesystems/nfs/00-INDEX +++ b/Documentation/filesystems/nfs/00-INDEX @@ -12,6 +12,8 @@ nfs41-server.txt - info on the Linux server implementation of NFSv4 minor version 1. nfs-rdma.txt - how to install and setup the Linux NFS/RDMA client and server software +nfsd-admin-interfaces.txt + - Administrative interfaces for nfsd. nfsroot.txt - short guide on setting up a diskless box with NFS root filesystem. pnfs.txt @@ -20,5 +22,5 @@ rpc-cache.txt - introduction to the caching mechanisms in the sunrpc layer. idmapper.txt - information for configuring request-keys to be used by idmapper -knfsd-rpcgss.txt +rpc-server-gss.txt - Information on GSS authentication support in the NFS Server diff --git a/Documentation/ide/00-INDEX b/Documentation/ide/00-INDEX index d6b778842b75..22f98ca79539 100644 --- a/Documentation/ide/00-INDEX +++ b/Documentation/ide/00-INDEX @@ -10,3 +10,5 @@ ide-tape.txt - info on the IDE ATAPI streaming tape driver ide.txt - important info for users of ATA devices (IDE/EIDE disks and CD-ROMS). +warm-plug-howto.txt + - using sysfs to remove and add IDE devices. \ No newline at end of file diff --git a/Documentation/laptops/00-INDEX b/Documentation/laptops/00-INDEX index fa688538e757..d13b9a9a9e00 100644 --- a/Documentation/laptops/00-INDEX +++ b/Documentation/laptops/00-INDEX @@ -1,13 +1,15 @@ 00-INDEX - This file -acer-wmi.txt - - information on the Acer Laptop WMI Extras driver. +Makefile + - Makefile for building dslm example program. asus-laptop.txt - information on the Asus Laptop Extras driver. disk-shock-protection.txt - information on hard disk shock protection. dslm.c - Simple Disk Sleep Monitor program +hpfall.c + - (HP) laptop accelerometer program for disk protection. laptop-mode.txt - how to conserve battery power using laptop-mode. sony-laptop.txt diff --git a/Documentation/leds/00-INDEX b/Documentation/leds/00-INDEX index 1ecd1596633e..b4ef1f34e25f 100644 --- a/Documentation/leds/00-INDEX +++ b/Documentation/leds/00-INDEX @@ -1,3 +1,7 @@ +00-INDEX + - This file +leds-blinkm.txt + - Driver for BlinkM LED-devices. leds-class.txt - documents LED handling under Linux. leds-lp3944.txt @@ -12,3 +16,7 @@ leds-lp55xx.txt - description about lp55xx common driver. leds-lm3556.txt - notes on how to use the leds-lm3556 driver. +ledtrig-oneshot.txt + - One-shot LED trigger for both sporadic and dense events. +ledtrig-transient.txt + - LED Transient Trigger, one shot timer activation. diff --git a/Documentation/m68k/00-INDEX b/Documentation/m68k/00-INDEX index a014e9f00765..2be8c6b00e74 100644 --- a/Documentation/m68k/00-INDEX +++ b/Documentation/m68k/00-INDEX @@ -1,5 +1,7 @@ 00-INDEX - this file +README.buddha + - Amiga Buddha and Catweasel IDE Driver kernel-options.txt - command line options for Linux/m68k diff --git a/Documentation/networking/00-INDEX b/Documentation/networking/00-INDEX index f11580f8719a..557b6ef70c26 100644 --- a/Documentation/networking/00-INDEX +++ b/Documentation/networking/00-INDEX @@ -6,8 +6,14 @@ - information on the 3Com Etherlink III Series Ethernet cards. 6pack.txt - info on the 6pack protocol, an alternative to KISS for AX.25 -DLINK.txt - - info on the D-Link DE-600/DE-620 parallel port pocket adapters +LICENSE.qla3xxx + - GPLv2 for QLogic Linux Networking HBA Driver +LICENSE.qlge + - GPLv2 for QLogic Linux qlge NIC Driver +LICENSE.qlcnic + - GPLv2 for QLogic Linux qlcnic NIC Driver +Makefile + - Makefile for docsrc. PLIP.txt - PLIP: The Parallel Line Internet Protocol device driver README.ipw2100 @@ -17,7 +23,7 @@ README.ipw2200 README.sb1000 - info on General Instrument/NextLevel SURFboard1000 cable modem. alias.txt - - info on using alias network devices + - info on using alias network devices. arcnet-hardware.txt - tons of info on ARCnet, hubs, jumper settings for ARCnet cards, etc. arcnet.txt @@ -80,7 +86,7 @@ framerelay.txt - info on using Frame Relay/Data Link Connection Identifier (DLCI). gen_stats.txt - Generic networking statistics for netlink users. -generic_hdlc.txt +generic-hdlc.txt - The generic High Level Data Link Control (HDLC) layer. generic_netlink.txt - info on Generic Netlink @@ -88,6 +94,8 @@ gianfar.txt - Gianfar Ethernet Driver. i40e.txt - README for the Intel Ethernet Controller XL710 Driver (i40e). +i40evf.txt + - Short note on the Driver for the Intel(R) XL710 X710 Virtual Function ieee802154.txt - Linux IEEE 802.15.4 implementation, API and drivers igb.txt @@ -102,6 +110,8 @@ ipddp.txt - AppleTalk-IP Decapsulation and AppleTalk-IP Encapsulation iphase.txt - Interphase PCI ATM (i)Chip IA Linux driver info. +ipsec.txt + - Note on not compressing IPSec payload and resulting failed policy check. ipv6.txt - Options to the ipv6 kernel module. ipvs-sysctl.txt @@ -120,6 +130,8 @@ lapb-module.txt - programming information of the LAPB module. ltpc.txt - the Apple or Farallon LocalTalk PC card driver +mac80211-auth-assoc-deauth.txt + - authentication and association / deauth-disassoc with max80211 mac80211-injection.txt - HOWTO use packet injection with mac80211 multiqueue.txt @@ -134,6 +146,10 @@ netdevices.txt - info on network device driver functions exported to the kernel. netif-msg.txt - Design of the network interface message level setting (NETIF_MSG_*). +netlink_mmap.txt + - memory mapped I/O with netlink +nf_conntrack-sysctl.txt + - list of netfilter-sysctl knobs. nfc.txt - The Linux Near Field Communication (NFS) subsystem. openvswitch.txt @@ -176,7 +192,7 @@ skfp.txt - SysKonnect FDDI (SK-5xxx, Compaq Netelligent) driver info. smc9.txt - the driver for SMC's 9000 series of Ethernet cards -spider-net.txt +spider_net.txt - README for the Spidernet Driver (as found in PS3 / Cell BE). stmmac.txt - README for the STMicro Synopsys Ethernet driver. @@ -188,6 +204,8 @@ tcp.txt - short blurb on how TCP output takes place. tcp-thin.txt - kernel tuning options for low rate 'thin' TCP streams. +team.txt + - pointer to information for ethernet teaming devices. tlan.txt - ThunderLAN (Compaq Netelligent 10/100, Olicom OC-2xxx) driver info. tproxy.txt @@ -200,6 +218,8 @@ vortex.txt - info on using 3Com Vortex (3c590, 3c592, 3c595, 3c597) Ethernet cards. vxge.txt - README for the Neterion X3100 PCIe Server Adapter. +vxlan.txt + - Virtual extensible LAN overview x25.txt - general info on X.25 development. x25-iface.txt diff --git a/Documentation/power/00-INDEX b/Documentation/power/00-INDEX index a4d682f54231..ad04cc8097ed 100644 --- a/Documentation/power/00-INDEX +++ b/Documentation/power/00-INDEX @@ -4,6 +4,8 @@ apm-acpi.txt - basic info about the APM and ACPI support. basic-pm-debugging.txt - Debugging suspend and resume +charger-manager.txt + - Battery charger management. devices.txt - How drivers interact with system-wide power management drivers-testing.txt @@ -22,6 +24,8 @@ pm_qos_interface.txt - info on Linux PM Quality of Service interface power_supply_class.txt - Tells userspace about battery, UPS, AC or DC power supply properties +runtime_pm.txt + - Power management framework for I/O devices. s2ram.txt - How to get suspend to ram working (and debug it when it isn't) states.txt @@ -38,7 +42,5 @@ tricks.txt - How to trick software suspend (to disk) into working when it isn't userland-swsusp.txt - Experimental implementation of software suspend in userspace -video_extension.txt - - ACPI video extensions video.txt - Video issues during resume from suspend diff --git a/Documentation/s390/00-INDEX b/Documentation/s390/00-INDEX index 3a2b96302ecc..10c874ebdfe5 100644 --- a/Documentation/s390/00-INDEX +++ b/Documentation/s390/00-INDEX @@ -16,11 +16,13 @@ Debugging390.txt - hints for debugging on s390 systems. driver-model.txt - information on s390 devices and the driver model. +kvm.txt + - ioctl calls to /dev/kvm on s390. monreader.txt - information on accessing the z/VM monitor stream from Linux. +qeth.txt + - HiperSockets Bridge Port Support. s390dbf.txt - information on using the s390 debug feature. -TAPE - - information on the driver for channel-attached tapes. -zfcpdump +zfcpdump.txt - information on the s390 SCSI dump tool. diff --git a/Documentation/scheduler/00-INDEX b/Documentation/scheduler/00-INDEX index 46702e4f89c9..eccf7ad2e7f9 100644 --- a/Documentation/scheduler/00-INDEX +++ b/Documentation/scheduler/00-INDEX @@ -2,6 +2,8 @@ - this file. sched-arch.txt - CPU Scheduler implementation hints for architecture specific code. +sched-bwc.txt + - CFS bandwidth control overview. sched-design-CFS.txt - goals, design and implementation of the Completely Fair Scheduler. sched-domains.txt diff --git a/Documentation/scsi/00-INDEX b/Documentation/scsi/00-INDEX index 2044be565d93..c4b978a72f78 100644 --- a/Documentation/scsi/00-INDEX +++ b/Documentation/scsi/00-INDEX @@ -36,6 +36,8 @@ NinjaSCSI.txt - info on WorkBiT NinjaSCSI-32/32Bi driver aacraid.txt - Driver supporting Adaptec RAID controllers +advansys.txt + - List of Advansys Host Adapters aha152x.txt - info on driver for Adaptec AHA152x based adapters aic79xx.txt @@ -44,6 +46,12 @@ aic7xxx.txt - info on driver for Adaptec controllers arcmsr_spec.txt - ARECA FIRMWARE SPEC (for IOP331 adapter) +bfa.txt + - Brocade FC/FCOE adapter driver. +bnx2fc.txt + - FCoE hardware offload for Broadcom network interfaces. +cxgb3i.txt + - Chelsio iSCSI Linux Driver dc395x.txt - README file for the dc395x SCSI driver dpti.txt @@ -52,18 +60,24 @@ dtc3x80.txt - info on driver for DTC 2x80 based adapters g_NCR5380.txt - info on driver for NCR5380 and NCR53c400 based adapters +hpsa.txt + - HP Smart Array Controller SCSI driver. hptiop.txt - HIGHPOINT ROCKETRAID 3xxx RAID DRIVER in2000.txt - info on in2000 driver libsas.txt - Serial Attached SCSI management layer. +link_power_management_policy.txt + - Link power management options. lpfc.txt - LPFC driver release notes megaraid.txt - Common Management Module, shared code handling ioctls for LSI drivers ncr53c8xx.txt - info on driver for NCR53c8xx based adapters +osd.txt + Object-Based Storage Device, command set introduction. osst.txt - info on driver for OnStream SC-x0 SCSI tape ppa.txt @@ -74,6 +88,8 @@ scsi-changer.txt - README for the SCSI media changer driver scsi-generic.txt - info on the sg driver for generic (non-disk/CD/tape) SCSI devices. +scsi-parameters.txt + - List of SCSI-parameters to pass to the kernel at module load-time. scsi.txt - short blurb on using SCSI support as a module. scsi_mid_low_api.txt diff --git a/Documentation/serial/00-INDEX b/Documentation/serial/00-INDEX index 1f1b22fbd739..f9c6b5ed03e7 100644 --- a/Documentation/serial/00-INDEX +++ b/Documentation/serial/00-INDEX @@ -4,10 +4,12 @@ README.cycladesZ - info on Cyclades-Z firmware loading. digiepca.txt - info on Digi Intl. {PC,PCI,EISA}Xx and Xem series cards. -hayes-esp.txt - - info on using the Hayes ESP serial driver. +driver + - intro to the low level serial driver. moxa-smartio - file with info on installing/using Moxa multiport serial driver. +n_gsm.txt + - GSM 0710 tty multiplexer howto. riscom8.txt - notes on using the RISCom/8 multi-port serial driver. rocket.txt diff --git a/Documentation/spi/00-INDEX b/Documentation/spi/00-INDEX new file mode 100644 index 000000000000..a128fa835512 --- /dev/null +++ b/Documentation/spi/00-INDEX @@ -0,0 +1,22 @@ +00-INDEX + - this file. +Makefile + - Makefile for the example sourcefiles. +butterfly + - AVR Butterfly SPI driver overview and pin configuration. +ep93xx_spi + - Basic EP93xx SPI driver configuration. +pxa2xx + - PXA2xx SPI master controller build by spi_message fifo wq +spidev + - Intro to the userspace API for spi devices +spidev_fdx.c + - spidev example file +spi-lm70llp + - Connecting an LM70-LLP sensor to the kernel via the SPI subsys. +spi-sc18is602 + - NXP SC18IS602/603 I2C-bus to SPI bridge +spi-summary + - (Linux) SPI overview. If unsure about SPI or SPI in Linux, start here. +spidev_test.c + - SPI testing utility. diff --git a/Documentation/timers/00-INDEX b/Documentation/timers/00-INDEX index ef2ccbf77fa2..6d042dc1cce0 100644 --- a/Documentation/timers/00-INDEX +++ b/Documentation/timers/00-INDEX @@ -8,6 +8,8 @@ hpet_example.c - sample hpet timer test program hrtimers.txt - subsystem for high-resolution kernel timers +Makefile + - Build and link hpet_example NO_HZ.txt - Summary of the different methods for the scheduler clock-interrupts management. timers-howto.txt diff --git a/Documentation/virtual/kvm/00-INDEX b/Documentation/virtual/kvm/00-INDEX index 641ec9220179..fee9f2bf9c64 100644 --- a/Documentation/virtual/kvm/00-INDEX +++ b/Documentation/virtual/kvm/00-INDEX @@ -20,5 +20,7 @@ ppc-pv.txt - the paravirtualization interface on PowerPC. review-checklist.txt - review checklist for KVM patches. +s390-diag.txt + - Diagnose hypercall description (for IBM S/390) timekeeping.txt - timekeeping virtualization for x86-based architectures. diff --git a/Documentation/vm/00-INDEX b/Documentation/vm/00-INDEX index a39d06680e1c..081c49777abb 100644 --- a/Documentation/vm/00-INDEX +++ b/Documentation/vm/00-INDEX @@ -16,8 +16,6 @@ hwpoison.txt - explains what hwpoison is ksm.txt - how to use the Kernel Samepage Merging feature. -locking - - info on how locking and synchronization is done in the Linux vm code. numa - information about NUMA specific code in the Linux vm. numa_memory_policy.txt @@ -32,6 +30,8 @@ slub.txt - a short users guide for SLUB. soft-dirty.txt - short explanation for soft-dirty PTEs +split_page_table_lock + - Separate per-table lock to improve scalability of the old page_table_lock. transhuge.txt - Transparent Hugepage Support, alternative way of using hugepages. unevictable-lru.txt diff --git a/Documentation/w1/masters/00-INDEX b/Documentation/w1/masters/00-INDEX index d63fa024ac05..8330cf9325f0 100644 --- a/Documentation/w1/masters/00-INDEX +++ b/Documentation/w1/masters/00-INDEX @@ -4,7 +4,9 @@ ds2482 - The Maxim/Dallas Semiconductor DS2482 provides 1-wire busses. ds2490 - The Maxim/Dallas Semiconductor DS2490 builds USB <-> W1 bridges. -mxc_w1 +mxc-w1 - W1 master controller driver found on Freescale MX2/MX3 SoCs +omap-hdq + - HDQ/1-wire module of TI OMAP 2430/3430. w1-gpio - GPIO 1-wire bus master driver. diff --git a/Documentation/w1/slaves/00-INDEX b/Documentation/w1/slaves/00-INDEX index 75613c9ac4db..6e18c70c3474 100644 --- a/Documentation/w1/slaves/00-INDEX +++ b/Documentation/w1/slaves/00-INDEX @@ -4,3 +4,5 @@ w1_therm - The Maxim/Dallas Semiconductor ds18*20 temperature sensor. w1_ds2423 - The Maxim/Dallas Semiconductor ds2423 counter device. +w1_ds28e04 + - The Maxim/Dallas Semiconductor ds28e04 eeprom. diff --git a/Documentation/x86/00-INDEX b/Documentation/x86/00-INDEX index f37b46d34861..692264456f0f 100644 --- a/Documentation/x86/00-INDEX +++ b/Documentation/x86/00-INDEX @@ -1,6 +1,20 @@ 00-INDEX - this file -mtrr.txt - - how to use x86 Memory Type Range Registers to increase performance +boot.txt + - List of boot protocol versions +early-microcode.txt + - How to load microcode from an initrd-CPIO archive early to fix CPU issues. +earlyprintk.txt + - Using earlyprintk with a USB2 debug port key. +entry_64.txt + - Describe (some of the) kernel entry points for x86. exception-tables.txt - why and how Linux kernel uses exception tables on x86 +mtrr.txt + - how to use x86 Memory Type Range Registers to increase performance +pat.txt + - Page Attribute Table intro and API +usb-legacy-support.txt + - how to fix/avoid quirks when using emulated PS/2 mouse/keyboard. +zero-page.txt + - layout of the first page of memory. From a3eb7fbb41eb8d6e22d491741b8c5d5aa6cb069a Mon Sep 17 00:00:00 2001 From: Alexey Khoroshilov Date: Mon, 10 Feb 2014 14:25:35 -0800 Subject: [PATCH 472/596] drivers/message/i2o/i2o_config.c: fix deadlock in compat_ioctl(I2OGETIOPS) i2o_cfg_compat_ioctl(I2OGETIOPS) locks i2o_cfg_mutex and then calls i2o_cfg_ioctl(I2OGETIOPS) that locks i2o_cfg_mutex as well. A deadlock is guaranteed. Found by Linux Driver Verification project (linuxtesting.org). Signed-off-by: Alexey Khoroshilov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/message/i2o/i2o_config.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/message/i2o/i2o_config.c b/drivers/message/i2o/i2o_config.c index a60c188c2bd9..04bd3b6de401 100644 --- a/drivers/message/i2o/i2o_config.c +++ b/drivers/message/i2o/i2o_config.c @@ -754,19 +754,19 @@ static long i2o_cfg_compat_ioctl(struct file *file, unsigned cmd, unsigned long arg) { int ret; - mutex_lock(&i2o_cfg_mutex); switch (cmd) { case I2OGETIOPS: ret = i2o_cfg_ioctl(file, cmd, arg); break; case I2OPASSTHRU32: + mutex_lock(&i2o_cfg_mutex); ret = i2o_cfg_passthru32(file, cmd, arg); + mutex_unlock(&i2o_cfg_mutex); break; default: ret = -ENOIOCTLCMD; break; } - mutex_unlock(&i2o_cfg_mutex); return ret; } From 38dfac843cb6d7be1874888839817404a15a6b3c Mon Sep 17 00:00:00 2001 From: Greg Pearson Date: Mon, 10 Feb 2014 14:25:36 -0800 Subject: [PATCH 473/596] vmcore: prevent PT_NOTE p_memsz overflow during header update Currently, update_note_header_size_elf64() and update_note_header_size_elf32() will add the size of a PT_NOTE entry to real_sz even if that causes real_sz to exceeds max_sz. This patch corrects the while loop logic in those routines to ensure that does not happen and prints a warning if a PT_NOTE entry is dropped. If zero PT_NOTE entries are found or this condition is encountered because the only entry was dropped, a warning is printed and an error is returned. One possible negative side effect of exceeding the max_sz limit is an allocation failure in merge_note_headers_elf64() or merge_note_headers_elf32() which would produce console output such as the following while booting the crash kernel. vmalloc: allocation failure: 14076997632 bytes swapper/0: page allocation failure: order:0, mode:0x80d2 CPU: 0 PID: 1 Comm: swapper/0 Not tainted 3.10.0-gbp1 #7 Call Trace: dump_stack+0x19/0x1b warn_alloc_failed+0xf0/0x160 __vmalloc_node_range+0x19e/0x250 vmalloc_user+0x4c/0x70 merge_note_headers_elf64.constprop.9+0x116/0x24a vmcore_init+0x2d4/0x76c do_one_initcall+0xe2/0x190 kernel_init_freeable+0x17c/0x207 kernel_init+0xe/0x180 ret_from_fork+0x7c/0xb0 Kdump: vmcore not initialized kdump: dump target is /dev/sda4 kdump: saving to /sysroot//var/crash/127.0.0.1-2014.01.28-13:58:52/ kdump: saving vmcore-dmesg.txt Cannot open /proc/vmcore: No such file or directory kdump: saving vmcore-dmesg.txt failed kdump: saving vmcore kdump: saving vmcore failed This type of failure has been seen on a four socket prototype system with certain memory configurations. Most PT_NOTE sections have a single entry similar to: n_namesz = 0x5 n_descsz = 0x150 n_type = 0x1 Occasionally, a second entry is encountered with very large n_namesz and n_descsz sizes: n_namesz = 0x80000008 n_descsz = 0x510ae163 n_type = 0x80000008 Not yet sure of the source of these extra entries, they seem bogus, but they shouldn't cause crash dump to fail. Signed-off-by: Greg Pearson Acked-by: Vivek Goyal Cc: HATAYAMA Daisuke Cc: Michael Holzheu Cc: "Eric W. Biederman" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/proc/vmcore.c | 26 ++++++++++++++++++++------ 1 file changed, 20 insertions(+), 6 deletions(-) diff --git a/fs/proc/vmcore.c b/fs/proc/vmcore.c index 2ca7ba047f04..88d4585b30f1 100644 --- a/fs/proc/vmcore.c +++ b/fs/proc/vmcore.c @@ -468,17 +468,24 @@ static int __init update_note_header_size_elf64(const Elf64_Ehdr *ehdr_ptr) return rc; } nhdr_ptr = notes_section; - while (real_sz < max_sz) { - if (nhdr_ptr->n_namesz == 0) - break; + while (nhdr_ptr->n_namesz != 0) { sz = sizeof(Elf64_Nhdr) + ((nhdr_ptr->n_namesz + 3) & ~3) + ((nhdr_ptr->n_descsz + 3) & ~3); + if ((real_sz + sz) > max_sz) { + pr_warn("Warning: Exceeded p_memsz, dropping PT_NOTE entry n_namesz=0x%x, n_descsz=0x%x\n", + nhdr_ptr->n_namesz, nhdr_ptr->n_descsz); + break; + } real_sz += sz; nhdr_ptr = (Elf64_Nhdr*)((char*)nhdr_ptr + sz); } kfree(notes_section); phdr_ptr->p_memsz = real_sz; + if (real_sz == 0) { + pr_warn("Warning: Zero PT_NOTE entries found\n"); + return -EINVAL; + } } return 0; @@ -648,17 +655,24 @@ static int __init update_note_header_size_elf32(const Elf32_Ehdr *ehdr_ptr) return rc; } nhdr_ptr = notes_section; - while (real_sz < max_sz) { - if (nhdr_ptr->n_namesz == 0) - break; + while (nhdr_ptr->n_namesz != 0) { sz = sizeof(Elf32_Nhdr) + ((nhdr_ptr->n_namesz + 3) & ~3) + ((nhdr_ptr->n_descsz + 3) & ~3); + if ((real_sz + sz) > max_sz) { + pr_warn("Warning: Exceeded p_memsz, dropping PT_NOTE entry n_namesz=0x%x, n_descsz=0x%x\n", + nhdr_ptr->n_namesz, nhdr_ptr->n_descsz); + break; + } real_sz += sz; nhdr_ptr = (Elf32_Nhdr*)((char*)nhdr_ptr + sz); } kfree(notes_section); phdr_ptr->p_memsz = real_sz; + if (real_sz == 0) { + pr_warn("Warning: Zero PT_NOTE entries found\n"); + return -EINVAL; + } } return 0; From bd180b4e2b0d5b8f14978e295741e69e6a01d398 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Mon, 10 Feb 2014 14:25:38 -0800 Subject: [PATCH 474/596] drivers/md/bcache/extents.c: use %zi to format size_t drivers/md/bcache/extents.c: In function `btree_ptr_bad_expensive': drivers/md/bcache/extents.c:196: warning: format `%li' expects type `long int', but argument 4 has type `size_t' Signed-off-by: Geert Uytterhoeven Cc: Kent Overstreet Cc: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/md/bcache/extents.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/md/bcache/extents.c b/drivers/md/bcache/extents.c index c3ead586dc27..416d1a3e028e 100644 --- a/drivers/md/bcache/extents.c +++ b/drivers/md/bcache/extents.c @@ -194,7 +194,7 @@ static bool btree_ptr_bad_expensive(struct btree *b, const struct bkey *k) mutex_unlock(&b->c->bucket_lock); bch_extent_to_text(buf, sizeof(buf), k); btree_bug(b, -"inconsistent btree pointer %s: bucket %li pin %i prio %i gen %i last_gc %i mark %llu gc_gen %i", +"inconsistent btree pointer %s: bucket %zi pin %i prio %i gen %i last_gc %i mark %llu gc_gen %i", buf, PTR_BUCKET_NR(b->c, k, i), atomic_read(&g->pin), g->prio, g->gen, g->last_gc, GC_MARK(g), g->gc_gen); return true; From 255d0884f5635122adb23866b242b4ca112f4bc8 Mon Sep 17 00:00:00 2001 From: David Rientjes Date: Mon, 10 Feb 2014 14:25:39 -0800 Subject: [PATCH 475/596] mm/slub.c: list_lock may not be held in some circumstances Commit c65c1877bd68 ("slub: use lockdep_assert_held") incorrectly required that add_full() and remove_full() hold n->list_lock. The lock is only taken when kmem_cache_debug(s), since that's the only time it actually does anything. Require that the lock only be taken under such a condition. Reported-by: Larry Finger Tested-by: Larry Finger Tested-by: Paul E. McKenney Acked-by: Christoph Lameter Cc: Pekka Enberg Signed-off-by: David Rientjes Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/slub.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/mm/slub.c b/mm/slub.c index 7e3e0458bce4..3d3a8a7a0f8c 100644 --- a/mm/slub.c +++ b/mm/slub.c @@ -1004,21 +1004,19 @@ static inline void slab_free_hook(struct kmem_cache *s, void *x) static void add_full(struct kmem_cache *s, struct kmem_cache_node *n, struct page *page) { - lockdep_assert_held(&n->list_lock); - if (!(s->flags & SLAB_STORE_USER)) return; + lockdep_assert_held(&n->list_lock); list_add(&page->lru, &n->full); } static void remove_full(struct kmem_cache *s, struct kmem_cache_node *n, struct page *page) { - lockdep_assert_held(&n->list_lock); - if (!(s->flags & SLAB_STORE_USER)) return; + lockdep_assert_held(&n->list_lock); list_del(&page->lru); } From a9c8e4beeeb64c22b84c803747487857fe424b68 Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Mon, 10 Feb 2014 14:25:40 -0800 Subject: [PATCH 476/596] xen: properly account for _PAGE_NUMA during xen pte translations Steven Noonan forwarded a users report where they had a problem starting vsftpd on a Xen paravirtualized guest, with this in dmesg: BUG: Bad page map in process vsftpd pte:8000000493b88165 pmd:e9cc01067 page:ffffea00124ee200 count:0 mapcount:-1 mapping: (null) index:0x0 page flags: 0x2ffc0000000014(referenced|dirty) addr:00007f97eea74000 vm_flags:00100071 anon_vma:ffff880e98f80380 mapping: (null) index:7f97eea74 CPU: 4 PID: 587 Comm: vsftpd Not tainted 3.12.7-1-ec2 #1 Call Trace: dump_stack+0x45/0x56 print_bad_pte+0x22e/0x250 unmap_single_vma+0x583/0x890 unmap_vmas+0x65/0x90 exit_mmap+0xc5/0x170 mmput+0x65/0x100 do_exit+0x393/0x9e0 do_group_exit+0xcc/0x140 SyS_exit_group+0x14/0x20 system_call_fastpath+0x1a/0x1f Disabling lock debugging due to kernel taint BUG: Bad rss-counter state mm:ffff880e9ca60580 idx:0 val:-1 BUG: Bad rss-counter state mm:ffff880e9ca60580 idx:1 val:1 The issue could not be reproduced under an HVM instance with the same kernel, so it appears to be exclusive to paravirtual Xen guests. He bisected the problem to commit 1667918b6483 ("mm: numa: clear numa hinting information on mprotect") that was also included in 3.12-stable. The problem was related to how xen translates ptes because it was not accounting for the _PAGE_NUMA bit. This patch splits pte_present to add a pteval_present helper for use by xen so both bare metal and xen use the same code when checking if a PTE is present. [mgorman@suse.de: wrote changelog, proposed minor modifications] [akpm@linux-foundation.org: fix typo in comment] Reported-by: Steven Noonan Tested-by: Steven Noonan Signed-off-by: Elena Ufimtseva Signed-off-by: Mel Gorman Reviewed-by: David Vrabel Acked-by: Konrad Rzeszutek Wilk Cc: [3.12+] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86/include/asm/pgtable.h | 14 ++++++++++++-- arch/x86/xen/mmu.c | 4 ++-- 2 files changed, 14 insertions(+), 4 deletions(-) diff --git a/arch/x86/include/asm/pgtable.h b/arch/x86/include/asm/pgtable.h index bbc8b12fa443..5ad38ad07890 100644 --- a/arch/x86/include/asm/pgtable.h +++ b/arch/x86/include/asm/pgtable.h @@ -445,10 +445,20 @@ static inline int pte_same(pte_t a, pte_t b) return a.pte == b.pte; } +static inline int pteval_present(pteval_t pteval) +{ + /* + * Yes Linus, _PAGE_PROTNONE == _PAGE_NUMA. Expressing it this + * way clearly states that the intent is that protnone and numa + * hinting ptes are considered present for the purposes of + * pagetable operations like zapping, protection changes, gup etc. + */ + return pteval & (_PAGE_PRESENT | _PAGE_PROTNONE | _PAGE_NUMA); +} + static inline int pte_present(pte_t a) { - return pte_flags(a) & (_PAGE_PRESENT | _PAGE_PROTNONE | - _PAGE_NUMA); + return pteval_present(pte_flags(a)); } #define pte_accessible pte_accessible diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c index 2423ef04ffea..256282e7888b 100644 --- a/arch/x86/xen/mmu.c +++ b/arch/x86/xen/mmu.c @@ -365,7 +365,7 @@ void xen_ptep_modify_prot_commit(struct mm_struct *mm, unsigned long addr, /* Assume pteval_t is equivalent to all the other *val_t types. */ static pteval_t pte_mfn_to_pfn(pteval_t val) { - if (val & _PAGE_PRESENT) { + if (pteval_present(val)) { unsigned long mfn = (val & PTE_PFN_MASK) >> PAGE_SHIFT; unsigned long pfn = mfn_to_pfn(mfn); @@ -381,7 +381,7 @@ static pteval_t pte_mfn_to_pfn(pteval_t val) static pteval_t pte_pfn_to_mfn(pteval_t val) { - if (val & _PAGE_PRESENT) { + if (pteval_present(val)) { unsigned long pfn = (val & PTE_PFN_MASK) >> PAGE_SHIFT; pteval_t flags = val & PTE_FLAGS_MASK; unsigned long mfn; From 96c7a2ff21501691587e1ae969b83cbec8b78e08 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Mon, 10 Feb 2014 14:25:41 -0800 Subject: [PATCH 477/596] fs/file.c:fdtable: avoid triggering OOMs from alloc_fdmem Recently due to a spike in connections per second memcached on 3 separate boxes triggered the OOM killer from accept. At the time the OOM killer was triggered there was 4GB out of 36GB free in zone 1. The problem was that alloc_fdtable was allocating an order 3 page (32KiB) to hold a bitmap, and there was sufficient fragmentation that the largest page available was 8KiB. I find the logic that PAGE_ALLOC_COSTLY_ORDER can't fail pretty dubious but I do agree that order 3 allocations are very likely to succeed. There are always pathologies where order > 0 allocations can fail when there are copious amounts of free memory available. Using the pigeon hole principle it is easy to show that it requires 1 page more than 50% of the pages being free to guarantee an order 1 (8KiB) allocation will succeed, 1 page more than 75% of the pages being free to guarantee an order 2 (16KiB) allocation will succeed and 1 page more than 87.5% of the pages being free to guarantee an order 3 allocate will succeed. A server churning memory with a lot of small requests and replies like memcached is a common case that if anything can will skew the odds against large pages being available. Therefore let's not give external applications a practical way to kill linux server applications, and specify __GFP_NORETRY to the kmalloc in alloc_fdmem. Unless I am misreading the code and by the time the code reaches should_alloc_retry in __alloc_pages_slowpath (where __GFP_NORETRY becomes signification). We have already tried everything reasonable to allocate a page and the only thing left to do is wait. So not waiting and falling back to vmalloc immediately seems like the reasonable thing to do even if there wasn't a chance of triggering the OOM killer. Signed-off-by: "Eric W. Biederman" Cc: Eric Dumazet Acked-by: David Rientjes Cc: Cong Wang Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/file.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/file.c b/fs/file.c index 771578b33fb6..db25c2bdfe46 100644 --- a/fs/file.c +++ b/fs/file.c @@ -34,7 +34,7 @@ static void *alloc_fdmem(size_t size) * vmalloc() if the allocation size will be considered "large" by the VM. */ if (size <= (PAGE_SIZE << PAGE_ALLOC_COSTLY_ORDER)) { - void *data = kmalloc(size, GFP_KERNEL|__GFP_NOWARN); + void *data = kmalloc(size, GFP_KERNEL|__GFP_NOWARN|__GFP_NORETRY); if (data != NULL) return data; } From 79040cad3f8235937e229f1b9401ba36dd5ad69b Mon Sep 17 00:00:00 2001 From: Prarit Bhargava Date: Mon, 10 Feb 2014 14:25:43 -0800 Subject: [PATCH 478/596] drivers/edac/edac_mc_sysfs.c: poll timeout cannot be zero If you do echo 0 > /sys/module/edac_core/parameters/edac_mc_poll_msec the following stack trace is output because the edac module is not designed to poll with a timeout of zero. WARNING: CPU: 12 PID: 0 at lib/list_debug.c:33 __list_add+0xac/0xc0() list_add corruption. prev->next should be next (ffff8808291dd1b8), but was (null). (prev=ffff8808286fe3f8). Modules linked in: sg nfsv3 rpcsec_gss_krb5 nfsv4 dns_resolver nfs fscache cfg80211 rfkill x86_pkg_temp_thermal coretemp kvm_intel kvm ixgbe e1000e crct10dif_pclmul crc32_pclmul crc32c_intel ghash_clmulni_intel aesni_intel lrw gf128mul glue_helper ablk_helper cryptd iTCO_wdt ptp sb_edac iTCO_vendor_support pps_core mdio ipmi_devintf edac_core ioatdma microcode shpchp lpc_ich pcspkr i2c_i801 dca mfd_core ipmi_si wmi ipmi_msghandler nfsd auth_rpcgss nfs_acl lockd sunrpc xfs libcrc32c sd_mod sr_mod cdrom crc_t10dif crct10dif_common mgag200 syscopyarea sysfillrect sysimgblt isci i2c_algo_bit drm_kms_helper ttm drm libsas ahci libahci scsi_transport_sas libata i2c_core dm_mirror dm_region_hash dm_log dm_mod CPU: 12 PID: 0 Comm: swapper/12 Not tainted 3.13.0+ #1 Hardware name: Intel Corporation LH Pass ........../SVRBD-ROW_T, BIOS SE5C600.86B.01.08.0003.022620131521 02/26/2013 Call Trace: __list_add+0xac/0xc0 __internal_add_timer+0xab/0x130 internal_add_timer+0x17/0x40 mod_timer_pinned+0xca/0x170 intel_pstate_timer_func+0x28a/0x380 call_timer_fn+0x36/0x100 run_timer_softirq+0x1ff/0x2f0 __do_softirq+0xf5/0x2e0 irq_exit+0x10d/0x120 smp_apic_timer_interrupt+0x45/0x60 apic_timer_interrupt+0x6d/0x80 cpuidle_idle_call+0xb9/0x1f0 arch_cpu_idle+0xe/0x30 cpu_startup_entry+0x9e/0x240 start_secondary+0x1e4/0x290 kernel BUG at kernel/timer.c:1084! invalid opcode: 0000 [#1] SMP Modules linked in: sg nfsv3 rpcsec_gss_krb5 nfsv4 dns_resolver nfs fscache cfg80211 rfkill x86_pkg_temp_thermal coretemp kvm_intel kvm ixgbe e1000e crct10dif_pclmul crc32_pclmul crc32c_intel ghash_clmulni_intel aesni_intel lrw gf128mul glue_helper ablk_helper cryptd iTCO_wdt ptp sb_edac iTCO_vendor_support pps_core mdio ipmi_devintf edac_core ioatdma microcode shpchp lpc_ich pcspkr i2c_i801 dca mfd_core ipmi_si wmi ipmi_msghandler nfsd auth_rpcgss nfs_acl lockd sunrpc xfs libcrc32c sd_mod sr_mod cdrom crc_t10dif crct10dif_common mgag200 syscopyarea sysfillrect sysimgblt isci i2c_algo_bit drm_kms_helper ttm drm libsas ahci libahci scsi_transport_sas libata i2c_core dm_mirror dm_region_hash dm_log dm_mod CPU: 12 PID: 0 Comm: swapper/12 Tainted: G W 3.13.0+ #1 Hardware name: Intel Corporation LH Pass ........../SVRBD-ROW_T, BIOS SE5C600.86B.01.08.0003.022620131521 02/26/2013 Call Trace: run_timer_softirq+0x245/0x2f0 __do_softirq+0xf5/0x2e0 irq_exit+0x10d/0x120 smp_apic_timer_interrupt+0x45/0x60 apic_timer_interrupt+0x6d/0x80 cpuidle_idle_call+0xb9/0x1f0 arch_cpu_idle+0xe/0x30 cpu_startup_entry+0x9e/0x240 start_secondary+0x1e4/0x290 RIP cascade+0x93/0xa0 WARNING: CPU: 36 PID: 1154 at kernel/workqueue.c:1461 __queue_delayed_work+0xed/0x1a0() Modules linked in: sg nfsv3 rpcsec_gss_krb5 nfsv4 dns_resolver nfs fscache cfg80211 rfkill x86_pkg_temp_thermal coretemp kvm_intel kvm ixgbe e1000e crct10dif_pclmul crc32_pclmul crc32c_intel ghash_clmulni_intel aesni_intel lrw gf128mul glue_helper ablk_helper cryptd iTCO_wdt ptp sb_edac iTCO_vendor_support pps_core mdio ipmi_devintf edac_core ioatdma microcode shpchp lpc_ich pcspkr i2c_i801 dca mfd_core ipmi_si wmi ipmi_msghandler nfsd auth_rpcgss nfs_acl lockd sunrpc xfs libcrc32c sd_mod sr_mod cdrom crc_t10dif crct10dif_common mgag200 syscopyarea sysfillrect sysimgblt isci i2c_algo_bit drm_kms_helper ttm drm libsas ahci libahci scsi_transport_sas libata i2c_core dm_mirror dm_region_hash dm_log dm_mod CPU: 36 PID: 1154 Comm: kworker/u481:3 Tainted: G W 3.13.0+ #1 Hardware name: Intel Corporation LH Pass ........../SVRBD-ROW_T, BIOS SE5C600.86B.01.08.0003.022620131521 02/26/2013 Workqueue: edac-poller edac_mc_workq_function [edac_core] Call Trace: dump_stack+0x45/0x56 warn_slowpath_common+0x7d/0xa0 warn_slowpath_null+0x1a/0x20 __queue_delayed_work+0xed/0x1a0 queue_delayed_work_on+0x27/0x50 edac_mc_workq_function+0x72/0xa0 [edac_core] process_one_work+0x17b/0x460 worker_thread+0x11b/0x400 kthread+0xd2/0xf0 ret_from_fork+0x7c/0xb0 This patch adds a range check in the edac_mc_poll_msec code to check for 0. Signed-off-by: Prarit Bhargava Cc: Doug Thompson Cc: Mauro Carvalho Chehab Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/edac/edac_mc_sysfs.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/edac/edac_mc_sysfs.c b/drivers/edac/edac_mc_sysfs.c index 51c0362acf5c..8ec1747b1c39 100644 --- a/drivers/edac/edac_mc_sysfs.c +++ b/drivers/edac/edac_mc_sysfs.c @@ -61,7 +61,7 @@ static int edac_set_poll_msec(const char *val, struct kernel_param *kp) ret = kstrtol(val, 0, &l); if (ret) return ret; - if ((int)l != l) + if (!l || ((int)l != l)) return -EINVAL; *((int *)kp->arg) = l; From a987c7ca7fc9225a587b1dc59d7d4ad2d9e2e08e Mon Sep 17 00:00:00 2001 From: Younger Liu Date: Mon, 10 Feb 2014 14:25:44 -0800 Subject: [PATCH 479/596] ocfs2: fix ocfs2_sync_file() if filesystem is readonly If filesystem is readonly, there is no need to flush drive's caches or force any uncommitted transactions. [akpm@linux-foundation.org: return -EROFS, not 0] Signed-off-by: Younger Liu Cc: Joel Becker Cc: Mark Fasheh Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ocfs2/file.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index d77d71ead8d1..fcd970636bbf 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -185,6 +185,9 @@ static int ocfs2_sync_file(struct file *file, loff_t start, loff_t end, file->f_path.dentry->d_name.name, (unsigned long long)datasync); + if (ocfs2_is_hard_readonly(osb) || ocfs2_is_soft_readonly(osb)) + return -EROFS; + err = filemap_write_and_wait_range(inode->i_mapping, start, end); if (err) return err; From 25fba9bebeb79447bbe08dc0e1f441caef9f5644 Mon Sep 17 00:00:00 2001 From: Borislav Petkov Date: Mon, 10 Feb 2014 14:25:45 -0800 Subject: [PATCH 480/596] gitignore: add all.config This is used by kbuild to load preset Kconfig options. We need to ignore it, otherwise git clean kills it. Signed-off-by: Borislav Petkov Cc: Michal Marek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- .gitignore | 3 +++ 1 file changed, 3 insertions(+) diff --git a/.gitignore b/.gitignore index 7e9932e55475..42fa0d5626a9 100644 --- a/.gitignore +++ b/.gitignore @@ -92,3 +92,6 @@ extra_certificates signing_key.priv signing_key.x509 x509.genkey + +# Kconfig presets +all.config From 1e4dd9461fabfbc780cdfaf103cec790f3a53325 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Mon, 10 Feb 2014 14:25:46 -0800 Subject: [PATCH 481/596] slub: do not assert not having lock in removing freed partial Vladimir reported the following issue: Commit c65c1877bd68 ("slub: use lockdep_assert_held") requires remove_partial() to be called with n->list_lock held, but free_partial() called from kmem_cache_close() on cache destruction does not follow this rule, leading to a warning: WARNING: CPU: 0 PID: 2787 at mm/slub.c:1536 __kmem_cache_shutdown+0x1b2/0x1f0() Modules linked in: CPU: 0 PID: 2787 Comm: modprobe Tainted: G W 3.14.0-rc1-mm1+ #1 Hardware name: 0000000000000600 ffff88003ae1dde8 ffffffff816d9583 0000000000000600 0000000000000000 ffff88003ae1de28 ffffffff8107c107 0000000000000000 ffff880037ab2b00 ffff88007c240d30 ffffea0001ee5280 ffffea0001ee52a0 Call Trace: __kmem_cache_shutdown+0x1b2/0x1f0 kmem_cache_destroy+0x43/0xf0 xfs_destroy_zones+0x103/0x110 [xfs] exit_xfs_fs+0x38/0x4e4 [xfs] SyS_delete_module+0x19a/0x1f0 system_call_fastpath+0x16/0x1b His solution was to add a spinlock in order to quiet lockdep. Although there would be no contention to adding the lock, that lock also requires disabling of interrupts which will have a larger impact on the system. Instead of adding a spinlock to a location where it is not needed for lockdep, make a __remove_partial() function that does not test if the list_lock is held, as no one should have it due to it being freed. Also added a __add_partial() function that does not do the lock validation either, as it is not needed for the creation of the cache. Signed-off-by: Steven Rostedt Reported-by: Vladimir Davydov Suggested-by: David Rientjes Acked-by: David Rientjes Acked-by: Vladimir Davydov Acked-by: Christoph Lameter Cc: Pekka Enberg Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/slub.c | 34 +++++++++++++++++++++------------- 1 file changed, 21 insertions(+), 13 deletions(-) diff --git a/mm/slub.c b/mm/slub.c index 3d3a8a7a0f8c..25f14ad8f817 100644 --- a/mm/slub.c +++ b/mm/slub.c @@ -1518,11 +1518,9 @@ static void discard_slab(struct kmem_cache *s, struct page *page) /* * Management of partially allocated slabs. */ -static inline void add_partial(struct kmem_cache_node *n, - struct page *page, int tail) +static inline void +__add_partial(struct kmem_cache_node *n, struct page *page, int tail) { - lockdep_assert_held(&n->list_lock); - n->nr_partial++; if (tail == DEACTIVATE_TO_TAIL) list_add_tail(&page->lru, &n->partial); @@ -1530,13 +1528,25 @@ static inline void add_partial(struct kmem_cache_node *n, list_add(&page->lru, &n->partial); } +static inline void add_partial(struct kmem_cache_node *n, + struct page *page, int tail) +{ + lockdep_assert_held(&n->list_lock); + __add_partial(n, page, tail); +} + +static inline void +__remove_partial(struct kmem_cache_node *n, struct page *page) +{ + list_del(&page->lru); + n->nr_partial--; +} + static inline void remove_partial(struct kmem_cache_node *n, struct page *page) { lockdep_assert_held(&n->list_lock); - - list_del(&page->lru); - n->nr_partial--; + __remove_partial(n, page); } /* @@ -2904,12 +2914,10 @@ static void early_kmem_cache_node_alloc(int node) inc_slabs_node(kmem_cache_node, node, page->objects); /* - * the lock is for lockdep's sake, not for any actual - * race protection + * No locks need to be taken here as it has just been + * initialized and there is no concurrent access. */ - spin_lock(&n->list_lock); - add_partial(n, page, DEACTIVATE_TO_HEAD); - spin_unlock(&n->list_lock); + __add_partial(n, page, DEACTIVATE_TO_HEAD); } static void free_kmem_cache_nodes(struct kmem_cache *s) @@ -3195,7 +3203,7 @@ static void free_partial(struct kmem_cache *s, struct kmem_cache_node *n) list_for_each_entry_safe(page, h, &n->partial, lru) { if (!page->inuse) { - remove_partial(n, page); + __remove_partial(n, page); discard_slab(s, page); } else { list_slab_objects(s, page, From a0b54adda3fe4b4cc6d28f2a9217cd35d1aa888c Mon Sep 17 00:00:00 2001 From: Rafael Aquini Date: Mon, 10 Feb 2014 14:25:48 -0800 Subject: [PATCH 482/596] mm: fix page leak at nfs_symlink() Changes in commit a0b8cab3b9b2 ("mm: remove lru parameter from __pagevec_lru_add and remove parts of pagevec API") have introduced a call to add_to_page_cache_lru() which causes a leak in nfs_symlink() as now the page gets an extra refcount that is not dropped. Jan Stancek observed and reported the leak effect while running test8 from Connectathon Testsuite. After several iterations over the test case, which creates several symlinks on a NFS mountpoint, the test system was quickly getting into an out-of-memory scenario. This patch fixes the page leak by dropping that extra refcount add_to_page_cache_lru() is grabbing. Signed-off-by: Jan Stancek Signed-off-by: Rafael Aquini Acked-by: Mel Gorman Acked-by: Rik van Riel Cc: Jeff Layton Cc: Trond Myklebust Cc: [3.11.x+] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfs/dir.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index be38b573495a..4a48fe4b84b6 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -1846,6 +1846,11 @@ int nfs_symlink(struct inode *dir, struct dentry *dentry, const char *symname) GFP_KERNEL)) { SetPageUptodate(page); unlock_page(page); + /* + * add_to_page_cache_lru() grabs an extra page refcount. + * Drop it here to avoid leaking this page later. + */ + page_cache_release(page); } else __free_page(page); From fb37bb04d6c8d6c44e61d9da189dcfa6aa0f135e Mon Sep 17 00:00:00 2001 From: Paul Gortmaker Date: Mon, 10 Feb 2014 14:25:49 -0800 Subject: [PATCH 483/596] smp.h: fix x86+cpu.c sparse warnings about arch nonboot CPU calls Use what we already do for arch_disable_smp_support() to fix these: arch/x86/kernel/smpboot.c:1155:6: warning: symbol 'arch_enable_nonboot_cpus_begin' was not declared. Should it be static? arch/x86/kernel/smpboot.c:1160:6: warning: symbol 'arch_enable_nonboot_cpus_end' was not declared. Should it be static? kernel/cpu.c:512:13: warning: symbol 'arch_enable_nonboot_cpus_begin' was not declared. Should it be static? kernel/cpu.c:516:13: warning: symbol 'arch_enable_nonboot_cpus_end' was not declared. Should it be static? Signed-off-by: Paul Gortmaker Cc: Thomas Gleixner Cc: Ingo Molnar Cc: "H. Peter Anvin" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/smp.h | 3 +++ 1 file changed, 3 insertions(+) diff --git a/include/linux/smp.h b/include/linux/smp.h index 3834f43f9993..6ae004e437ea 100644 --- a/include/linux/smp.h +++ b/include/linux/smp.h @@ -188,6 +188,9 @@ static inline void kick_all_cpus_sync(void) { } */ extern void arch_disable_smp_support(void); +extern void arch_enable_nonboot_cpus_begin(void); +extern void arch_enable_nonboot_cpus_end(void); + void smp_setup_processor_id(void); #endif /* __LINUX_SMP_H */ From 8d547ff4ac5927245e0833ac18528f939da0ee0e Mon Sep 17 00:00:00 2001 From: Naoya Horiguchi Date: Mon, 10 Feb 2014 14:25:50 -0800 Subject: [PATCH 484/596] mm/memory-failure.c: move refcount only in !MF_COUNT_INCREASED mce-test detected a test failure when injecting error to a thp tail page. This is because we take page refcount of the tail page in madvise_hwpoison() while the fix in commit a3e0f9e47d5e ("mm/memory-failure.c: transfer page count from head page to tail page after split thp") assumes that we always take refcount on the head page. When a real memory error happens we take refcount on the head page where memory_failure() is called without MF_COUNT_INCREASED set, so it seems to me that testing memory error on thp tail page using madvise makes little sense. This patch cancels moving refcount in !MF_COUNT_INCREASED for valid testing. [akpm@linux-foundation.org: s/&&/&/] Signed-off-by: Naoya Horiguchi Cc: Andi Kleen Cc: Wanpeng Li Cc: Chen Gong Cc: [3.9+: a3e0f9e47d5e] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/memory-failure.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/mm/memory-failure.c b/mm/memory-failure.c index 4f08a2d61487..2f2f34a4e77d 100644 --- a/mm/memory-failure.c +++ b/mm/memory-failure.c @@ -945,8 +945,10 @@ static int hwpoison_user_mappings(struct page *p, unsigned long pfn, * to it. Similarly, page lock is shifted. */ if (hpage != p) { - put_page(hpage); - get_page(p); + if (!(flags & MF_COUNT_INCREASED)) { + put_page(hpage); + get_page(p); + } lock_page(p); unlock_page(hpage); *hpagep = p; From d62e74be1270c89fbaf7aada8218bfdf62d00a58 Mon Sep 17 00:00:00 2001 From: Younger Liu Date: Mon, 10 Feb 2014 14:25:51 -0800 Subject: [PATCH 485/596] ocfs2: fix issue that ocfs2_setattr() does not deal with new_i_size==i_size The issue scenario is as following: - Create a small file and fallocate a large disk space for a file with FALLOC_FL_KEEP_SIZE option. - ftruncate the file back to the original size again. but the disk free space is not changed back. This is a real bug that be fixed in this patch. In order to solve the issue above, we modified ocfs2_setattr(), if attr->ia_size != i_size_read(inode), It calls ocfs2_truncate_file(), and truncate disk space to attr->ia_size. Signed-off-by: Younger Liu Reviewed-by: Jie Liu Tested-by: Jie Liu Cc: Joel Becker Reviewed-by: Mark Fasheh Cc: Sunil Mushran Reviewed-by: Jensen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ocfs2/alloc.c | 2 +- fs/ocfs2/file.c | 9 ++------- 2 files changed, 3 insertions(+), 8 deletions(-) diff --git a/fs/ocfs2/alloc.c b/fs/ocfs2/alloc.c index aada5801567a..e2edff38be52 100644 --- a/fs/ocfs2/alloc.c +++ b/fs/ocfs2/alloc.c @@ -7158,7 +7158,7 @@ int ocfs2_truncate_inline(struct inode *inode, struct buffer_head *di_bh, if (end > i_size_read(inode)) end = i_size_read(inode); - BUG_ON(start >= end); + BUG_ON(start > end); if (!(OCFS2_I(inode)->ip_dyn_features & OCFS2_INLINE_DATA_FL) || !(le16_to_cpu(di->i_dyn_features) & OCFS2_INLINE_DATA_FL) || diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index fcd970636bbf..9148353c5cf8 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -477,11 +477,6 @@ static int ocfs2_truncate_file(struct inode *inode, goto bail; } - /* lets handle the simple truncate cases before doing any more - * cluster locking. */ - if (new_i_size == le64_to_cpu(fe->i_size)) - goto bail; - down_write(&OCFS2_I(inode)->ip_alloc_sem); ocfs2_resv_discard(&osb->osb_la_resmap, @@ -1148,14 +1143,14 @@ int ocfs2_setattr(struct dentry *dentry, struct iattr *attr) goto bail_unlock_rw; } - if (size_change && attr->ia_size != i_size_read(inode)) { + if (size_change) { status = inode_newsize_ok(inode, attr->ia_size); if (status) goto bail_unlock; inode_dio_wait(inode); - if (i_size_read(inode) > attr->ia_size) { + if (i_size_read(inode) >= attr->ia_size) { if (ocfs2_should_order_data(inode)) { status = ocfs2_begin_ordered_truncate(inode, attr->ia_size); From c7d2cbc364b2a237b0ed1bdd7cbf8a24c8a89dfd Mon Sep 17 00:00:00 2001 From: Junxiao Bi Date: Mon, 10 Feb 2014 14:25:53 -0800 Subject: [PATCH 486/596] ocfs2: update inode size after zeroing the hole fs-writeback will release the dirty pages without page lock whose offset are over inode size, the release happens at block_write_full_page_endio(). If not update, dirty pages in file holes may be released before flushed to the disk, then file holes will contain some non-zero data, this will cause sparse file md5sum error. To reproduce the bug, find a big sparse file with many holes, like vm image file, its actual size should be bigger than available mem size to make writeback work more frequently, tar it with -S option, then keep untar it and check its md5sum again and again until you get a wrong md5sum. Signed-off-by: Junxiao Bi Cc: Younger Liu Reviewed-by: Mark Fasheh Cc: Joel Becker Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ocfs2/file.c | 40 ++++++++++++++++++++++++++++++++-------- 1 file changed, 32 insertions(+), 8 deletions(-) diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index 9148353c5cf8..8450262bcf2a 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -716,7 +716,8 @@ static int __ocfs2_extend_allocation(struct inode *inode, u32 logical_start, * While a write will already be ordering the data, a truncate will not. * Thus, we need to explicitly order the zeroed pages. */ -static handle_t *ocfs2_zero_start_ordered_transaction(struct inode *inode) +static handle_t *ocfs2_zero_start_ordered_transaction(struct inode *inode, + struct buffer_head *di_bh) { struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); handle_t *handle = NULL; @@ -733,7 +734,14 @@ static handle_t *ocfs2_zero_start_ordered_transaction(struct inode *inode) } ret = ocfs2_jbd2_file_inode(handle, inode); - if (ret < 0) + if (ret < 0) { + mlog_errno(ret); + goto out; + } + + ret = ocfs2_journal_access_di(handle, INODE_CACHE(inode), di_bh, + OCFS2_JOURNAL_ACCESS_WRITE); + if (ret) mlog_errno(ret); out: @@ -749,7 +757,7 @@ static handle_t *ocfs2_zero_start_ordered_transaction(struct inode *inode) * to be too fragile to do exactly what we need without us having to * worry about recursive locking in ->write_begin() and ->write_end(). */ static int ocfs2_write_zero_page(struct inode *inode, u64 abs_from, - u64 abs_to) + u64 abs_to, struct buffer_head *di_bh) { struct address_space *mapping = inode->i_mapping; struct page *page; @@ -757,6 +765,7 @@ static int ocfs2_write_zero_page(struct inode *inode, u64 abs_from, handle_t *handle = NULL; int ret = 0; unsigned zero_from, zero_to, block_start, block_end; + struct ocfs2_dinode *di = (struct ocfs2_dinode *)di_bh->b_data; BUG_ON(abs_from >= abs_to); BUG_ON(abs_to > (((u64)index + 1) << PAGE_CACHE_SHIFT)); @@ -799,7 +808,8 @@ static int ocfs2_write_zero_page(struct inode *inode, u64 abs_from, } if (!handle) { - handle = ocfs2_zero_start_ordered_transaction(inode); + handle = ocfs2_zero_start_ordered_transaction(inode, + di_bh); if (IS_ERR(handle)) { ret = PTR_ERR(handle); handle = NULL; @@ -816,8 +826,22 @@ static int ocfs2_write_zero_page(struct inode *inode, u64 abs_from, ret = 0; } - if (handle) + if (handle) { + /* + * fs-writeback will release the dirty pages without page lock + * whose offset are over inode size, the release happens at + * block_write_full_page_endio(). + */ + i_size_write(inode, abs_to); + inode->i_blocks = ocfs2_inode_sector_count(inode); + di->i_size = cpu_to_le64((u64)i_size_read(inode)); + inode->i_mtime = inode->i_ctime = CURRENT_TIME; + di->i_mtime = di->i_ctime = cpu_to_le64(inode->i_mtime.tv_sec); + di->i_ctime_nsec = cpu_to_le32(inode->i_mtime.tv_nsec); + di->i_mtime_nsec = di->i_ctime_nsec; + ocfs2_journal_dirty(handle, di_bh); ocfs2_commit_trans(OCFS2_SB(inode->i_sb), handle); + } out_unlock: unlock_page(page); @@ -913,7 +937,7 @@ static int ocfs2_zero_extend_get_range(struct inode *inode, * has made sure that the entire range needs zeroing. */ static int ocfs2_zero_extend_range(struct inode *inode, u64 range_start, - u64 range_end) + u64 range_end, struct buffer_head *di_bh) { int rc = 0; u64 next_pos; @@ -929,7 +953,7 @@ static int ocfs2_zero_extend_range(struct inode *inode, u64 range_start, next_pos = (zero_pos & PAGE_CACHE_MASK) + PAGE_CACHE_SIZE; if (next_pos > range_end) next_pos = range_end; - rc = ocfs2_write_zero_page(inode, zero_pos, next_pos); + rc = ocfs2_write_zero_page(inode, zero_pos, next_pos, di_bh); if (rc < 0) { mlog_errno(rc); break; @@ -975,7 +999,7 @@ int ocfs2_zero_extend(struct inode *inode, struct buffer_head *di_bh, range_end = zero_to_size; ret = ocfs2_zero_extend_range(inode, range_start, - range_end); + range_end, di_bh); if (ret) { mlog_errno(ret); break; From 0e048316ff577e12c748e2d0a2e4f0f7b006654d Mon Sep 17 00:00:00 2001 From: Xue jiufei Date: Mon, 10 Feb 2014 14:25:54 -0800 Subject: [PATCH 487/596] ocfs2: check existence of old dentry in ocfs2_link() System call linkat first calls user_path_at(), check the existence of old dentry, and then calls vfs_link()->ocfs2_link() to do the actual work. There may exist a race when Node A create a hard link for file while node B rm it. Node A Node B user_path_at() ->ocfs2_lookup(), find old dentry exist rm file, add inode say inodeA to orphan_dir call ocfs2_link(),create a hard link for inodeA. rm the link, add inodeA to orphan_dir again When orphan_scan work start, it calls ocfs2_queue_orphans() to do the main work. It first tranverses entrys in orphan_dir, linking all inodes in this orphan_dir to a list look like this: inodeA->inodeB->...->inodeA When tranvering this list, it will fall into loop, calling iput() again and again. And finally trigger BUG_ON(inode->i_state & I_CLEAR). Signed-off-by: joyce Reviewed-by: Mark Fasheh Cc: Joel Becker Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ocfs2/namei.c | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) diff --git a/fs/ocfs2/namei.c b/fs/ocfs2/namei.c index f4d609be9400..3683643f3f0e 100644 --- a/fs/ocfs2/namei.c +++ b/fs/ocfs2/namei.c @@ -664,6 +664,7 @@ static int ocfs2_link(struct dentry *old_dentry, struct ocfs2_super *osb = OCFS2_SB(dir->i_sb); struct ocfs2_dir_lookup_result lookup = { NULL, }; sigset_t oldset; + u64 old_de_ino; trace_ocfs2_link((unsigned long long)OCFS2_I(inode)->ip_blkno, old_dentry->d_name.len, old_dentry->d_name.name, @@ -686,6 +687,22 @@ static int ocfs2_link(struct dentry *old_dentry, goto out; } + err = ocfs2_lookup_ino_from_name(dir, old_dentry->d_name.name, + old_dentry->d_name.len, &old_de_ino); + if (err) { + err = -ENOENT; + goto out; + } + + /* + * Check whether another node removed the source inode while we + * were in the vfs. + */ + if (old_de_ino != OCFS2_I(inode)->ip_blkno) { + err = -ENOENT; + goto out; + } + err = ocfs2_check_dir_for_entry(dir, dentry->d_name.name, dentry->d_name.len); if (err) From 0215b4aa069b989c963d594cf0f1c705e21d8ca5 Mon Sep 17 00:00:00 2001 From: Paul Gortmaker Date: Fri, 7 Feb 2014 14:50:58 -0500 Subject: [PATCH 488/596] powerpc: Fix build failure in sysdev/mpic.c for MPIC_WEIRD=y Commit 446f6d06fab0b49c61887ecbe8286d6aaa796637 ("powerpc/mpic: Properly set default triggers") breaks the mpc7447_hpc_defconfig as follows: CC arch/powerpc/sysdev/mpic.o arch/powerpc/sysdev/mpic.c: In function 'mpic_set_irq_type': arch/powerpc/sysdev/mpic.c:886:9: error: case label does not reduce to an integer constant arch/powerpc/sysdev/mpic.c:890:9: error: case label does not reduce to an integer constant arch/powerpc/sysdev/mpic.c:894:9: error: case label does not reduce to an integer constant arch/powerpc/sysdev/mpic.c:898:9: error: case label does not reduce to an integer constant Looking at the cpp output (gcc 4.7.3), I see: case mpic->hw_set[MPIC_IDX_VECPRI_SENSE_EDGE] | mpic->hw_set[MPIC_IDX_VECPRI_POLARITY_POSITIVE]: The pointer into an array appears because CONFIG_MPIC_WEIRD=y is set for this platform, thus enabling the following: ------------------- #ifdef CONFIG_MPIC_WEIRD static u32 mpic_infos[][MPIC_IDX_END] = { [0] = { /* Original OpenPIC compatible MPIC */ [...] #define MPIC_INFO(name) mpic->hw_set[MPIC_IDX_##name] #else /* CONFIG_MPIC_WEIRD */ #define MPIC_INFO(name) MPIC_##name #endif /* CONFIG_MPIC_WEIRD */ ------------------- Here we convert the case section to if/else if, and also add the equivalent of a default case to warn about unknown types. Boot tested on sbc8548, build tested on all defconfigs. Signed-off-by: Paul Gortmaker Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/sysdev/mpic.c | 38 +++++++++++++++++++------------------- 1 file changed, 19 insertions(+), 19 deletions(-) diff --git a/arch/powerpc/sysdev/mpic.c b/arch/powerpc/sysdev/mpic.c index 0e166ed4cd16..8209744b2829 100644 --- a/arch/powerpc/sysdev/mpic.c +++ b/arch/powerpc/sysdev/mpic.c @@ -886,25 +886,25 @@ int mpic_set_irq_type(struct irq_data *d, unsigned int flow_type) /* Default: read HW settings */ if (flow_type == IRQ_TYPE_DEFAULT) { - switch(vold & (MPIC_INFO(VECPRI_POLARITY_MASK) | - MPIC_INFO(VECPRI_SENSE_MASK))) { - case MPIC_INFO(VECPRI_SENSE_EDGE) | - MPIC_INFO(VECPRI_POLARITY_POSITIVE): - flow_type = IRQ_TYPE_EDGE_RISING; - break; - case MPIC_INFO(VECPRI_SENSE_EDGE) | - MPIC_INFO(VECPRI_POLARITY_NEGATIVE): - flow_type = IRQ_TYPE_EDGE_FALLING; - break; - case MPIC_INFO(VECPRI_SENSE_LEVEL) | - MPIC_INFO(VECPRI_POLARITY_POSITIVE): - flow_type = IRQ_TYPE_LEVEL_HIGH; - break; - case MPIC_INFO(VECPRI_SENSE_LEVEL) | - MPIC_INFO(VECPRI_POLARITY_NEGATIVE): - flow_type = IRQ_TYPE_LEVEL_LOW; - break; - } + int vold_ps; + + vold_ps = vold & (MPIC_INFO(VECPRI_POLARITY_MASK) | + MPIC_INFO(VECPRI_SENSE_MASK)); + + if (vold_ps == (MPIC_INFO(VECPRI_SENSE_EDGE) | + MPIC_INFO(VECPRI_POLARITY_POSITIVE))) + flow_type = IRQ_TYPE_EDGE_RISING; + else if (vold_ps == (MPIC_INFO(VECPRI_SENSE_EDGE) | + MPIC_INFO(VECPRI_POLARITY_NEGATIVE))) + flow_type = IRQ_TYPE_EDGE_FALLING; + else if (vold_ps == (MPIC_INFO(VECPRI_SENSE_LEVEL) | + MPIC_INFO(VECPRI_POLARITY_POSITIVE))) + flow_type = IRQ_TYPE_LEVEL_HIGH; + else if (vold_ps == (MPIC_INFO(VECPRI_SENSE_LEVEL) | + MPIC_INFO(VECPRI_POLARITY_NEGATIVE))) + flow_type = IRQ_TYPE_LEVEL_LOW; + else + WARN_ONCE(1, "mpic: unknown IRQ type %d\n", vold); } /* Apply to irq desc */ From 8cc6b6cd8713457be80202fc4264f05d20bc5e1b Mon Sep 17 00:00:00 2001 From: Thadeu Lima de Souza Cascardo Date: Wed, 5 Feb 2014 16:20:45 -0200 Subject: [PATCH 489/596] powerpc/eeh: Drop taken reference to driver on eeh_rmv_device Commit f5c57710dd62dd06f176934a8b4b8accbf00f9f8 ("powerpc/eeh: Use partial hotplug for EEH unaware drivers") introduces eeh_rmv_device, which may grab a reference to a driver, but not release it. That prevents a driver from being removed after it has gone through EEH recovery. This patch drops the reference if it was taken. Signed-off-by: Thadeu Lima de Souza Cascardo Acked-by: Gavin Shan Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/eeh_driver.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/arch/powerpc/kernel/eeh_driver.c b/arch/powerpc/kernel/eeh_driver.c index 7bb30dca4e19..fdc679d309ec 100644 --- a/arch/powerpc/kernel/eeh_driver.c +++ b/arch/powerpc/kernel/eeh_driver.c @@ -362,9 +362,13 @@ static void *eeh_rmv_device(void *data, void *userdata) */ if (!dev || (dev->hdr_type & PCI_HEADER_TYPE_BRIDGE)) return NULL; + driver = eeh_pcid_get(dev); - if (driver && driver->err_handler) - return NULL; + if (driver) { + eeh_pcid_put(dev); + if (driver->err_handler) + return NULL; + } /* Remove it from PCI subsystem */ pr_debug("EEH: Removing %s without EEH sensitive driver\n", From 3ec8b78fcc5aa7745026d8d85a4e9ab52c922765 Mon Sep 17 00:00:00 2001 From: Mahesh Salgaonkar Date: Fri, 31 Jan 2014 00:30:45 +0530 Subject: [PATCH 490/596] powerpc/pseries: Disable relocation on exception while going down during crash. Disable relocation on exception while going down even in kdump case. This is because we are about clear htab mappings while kexec-ing into kdump kernel and we may run into issues if we still have AIL ON. Signed-off-by: Mahesh Salgaonkar Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/platforms/pseries/setup.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/arch/powerpc/platforms/pseries/setup.c b/arch/powerpc/platforms/pseries/setup.c index 8e639d7cbda7..972df0ffd4dc 100644 --- a/arch/powerpc/platforms/pseries/setup.c +++ b/arch/powerpc/platforms/pseries/setup.c @@ -430,8 +430,7 @@ static void pSeries_machine_kexec(struct kimage *image) { long rc; - if (firmware_has_feature(FW_FEATURE_SET_MODE) && - (image->type != KEXEC_TYPE_CRASH)) { + if (firmware_has_feature(FW_FEATURE_SET_MODE)) { rc = pSeries_disable_reloc_on_exc(); if (rc != H_SUCCESS) pr_warning("Warning: Failed to disable relocation on " From 429d2e8342954d337abe370d957e78291032d867 Mon Sep 17 00:00:00 2001 From: Mahesh Salgaonkar Date: Fri, 31 Jan 2014 00:31:04 +0530 Subject: [PATCH 491/596] powerpc: Fix kdump hang issue on p8 with relocation on exception enabled. On p8 systems, with relocation on exception feature enabled we are seeing kdump kernel hang at interrupt vector 0xc*4400. The reason is, with this feature enabled, exception are raised with MMU (IR=DR=1) ON with the default offset of 0xc*4000. Since exception is raised in virtual mode it requires the vector region to be executable without which it fails to fetch and execute instruction at 0xc*4xxx. For default kernel since kernel is loaded at real 0, the htab mappings sets the entire kernel text region executable. But for relocatable kernel (e.g. kdump case) we only copy interrupt vectors down to real 0 and never marked that region as executable because in p7 and below we always get exception in real mode. This patch fixes this issue by marking htab mapping range as executable that overlaps with the interrupt vector region for relocatable kernel. Thanks to Ben who helped me to debug this issue and find the root cause. Signed-off-by: Mahesh Salgaonkar Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/include/asm/sections.h | 12 ++++++++++++ arch/powerpc/mm/hash_utils_64.c | 14 ++++++++++++++ 2 files changed, 26 insertions(+) diff --git a/arch/powerpc/include/asm/sections.h b/arch/powerpc/include/asm/sections.h index 4ee06fe15de4..d0e784e0ff48 100644 --- a/arch/powerpc/include/asm/sections.h +++ b/arch/powerpc/include/asm/sections.h @@ -8,6 +8,7 @@ #ifdef __powerpc64__ +extern char __start_interrupts[]; extern char __end_interrupts[]; extern char __prom_init_toc_start[]; @@ -21,6 +22,17 @@ static inline int in_kernel_text(unsigned long addr) return 0; } +static inline int overlaps_interrupt_vector_text(unsigned long start, + unsigned long end) +{ + unsigned long real_start, real_end; + real_start = __start_interrupts - _stext; + real_end = __end_interrupts - _stext; + + return start < (unsigned long)__va(real_end) && + (unsigned long)__va(real_start) < end; +} + static inline int overlaps_kernel_text(unsigned long start, unsigned long end) { return start < (unsigned long)__init_end && diff --git a/arch/powerpc/mm/hash_utils_64.c b/arch/powerpc/mm/hash_utils_64.c index de6881259aef..d766d6ee33fe 100644 --- a/arch/powerpc/mm/hash_utils_64.c +++ b/arch/powerpc/mm/hash_utils_64.c @@ -207,6 +207,20 @@ int htab_bolt_mapping(unsigned long vstart, unsigned long vend, if (overlaps_kernel_text(vaddr, vaddr + step)) tprot &= ~HPTE_R_N; + /* + * If relocatable, check if it overlaps interrupt vectors that + * are copied down to real 0. For relocatable kernel + * (e.g. kdump case) we copy interrupt vectors down to real + * address 0. Mark that region as executable. This is + * because on p8 system with relocation on exception feature + * enabled, exceptions are raised with MMU (IR=DR=1) ON. Hence + * in order to execute the interrupt handlers in virtual + * mode the vector region need to be marked as executable. + */ + if ((PHYSICAL_START > MEMORY_START) && + overlaps_interrupt_vector_text(vaddr, vaddr + step)) + tprot &= ~HPTE_R_N; + hash = hpt_hash(vpn, shift, ssize); hpteg = ((hash & htab_hash_mask) * HPTES_PER_GROUP); From 3b830c824a8b15bf032eafe09c48a09d2441f16d Mon Sep 17 00:00:00 2001 From: Laurent Dufour Date: Thu, 30 Jan 2014 16:58:42 +0100 Subject: [PATCH 492/596] powerpc/relocate fix relocate processing in LE mode Relocation's code is not working in little endian mode because the r_info field, which is a 64 bits value, should be read from the right offset. The current code is optimized to read the r_info field as a 32 bits value starting at the middle of the double word (offset 12). When running in LE mode, the read value is not correct since only the MSB is read. This patch removes this optimization which consist to deal with a 32 bits value instead of a 64 bits one. This way it works in big and little endian mode. Signed-off-by: Laurent Dufour Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/reloc_64.S | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/powerpc/kernel/reloc_64.S b/arch/powerpc/kernel/reloc_64.S index b47a0e1ab001..1482327cfeba 100644 --- a/arch/powerpc/kernel/reloc_64.S +++ b/arch/powerpc/kernel/reloc_64.S @@ -69,8 +69,8 @@ _GLOBAL(relocate) * R_PPC64_RELATIVE ones. */ mtctr r8 -5: lwz r0,12(9) /* ELF64_R_TYPE(reloc->r_info) */ - cmpwi r0,R_PPC64_RELATIVE +5: ld r0,8(9) /* ELF64_R_TYPE(reloc->r_info) */ + cmpdi r0,R_PPC64_RELATIVE bne 6f ld r6,0(r9) /* reloc->r_offset */ ld r0,16(r9) /* reloc->r_addend */ From 2fdd313f54eca8221125b7898e6a8393604b6fb6 Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Fri, 24 Jan 2014 15:50:51 +1100 Subject: [PATCH 493/596] powerpc/perf: Add Power8 cache & TLB events Signed-off-by: Michael Ellerman Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/perf/power8-pmu.c | 144 +++++++++++++++++++++++++++++++++ 1 file changed, 144 insertions(+) diff --git a/arch/powerpc/perf/power8-pmu.c b/arch/powerpc/perf/power8-pmu.c index a3f7abd2f13f..96cee20dcd34 100644 --- a/arch/powerpc/perf/power8-pmu.c +++ b/arch/powerpc/perf/power8-pmu.c @@ -25,6 +25,37 @@ #define PM_BRU_FIN 0x10068 #define PM_BR_MPRED_CMPL 0x400f6 +/* All L1 D cache load references counted at finish, gated by reject */ +#define PM_LD_REF_L1 0x100ee +/* Load Missed L1 */ +#define PM_LD_MISS_L1 0x3e054 +/* Store Missed L1 */ +#define PM_ST_MISS_L1 0x300f0 +/* L1 cache data prefetches */ +#define PM_L1_PREF 0x0d8b8 +/* Instruction fetches from L1 */ +#define PM_INST_FROM_L1 0x04080 +/* Demand iCache Miss */ +#define PM_L1_ICACHE_MISS 0x200fd +/* Instruction Demand sectors wriittent into IL1 */ +#define PM_L1_DEMAND_WRITE 0x0408c +/* Instruction prefetch written into IL1 */ +#define PM_IC_PREF_WRITE 0x0408e +/* The data cache was reloaded from local core's L3 due to a demand load */ +#define PM_DATA_FROM_L3 0x4c042 +/* Demand LD - L3 Miss (not L2 hit and not L3 hit) */ +#define PM_DATA_FROM_L3MISS 0x300fe +/* All successful D-side store dispatches for this thread */ +#define PM_L2_ST 0x17080 +/* All successful D-side store dispatches for this thread that were L2 Miss */ +#define PM_L2_ST_MISS 0x17082 +/* Total HW L3 prefetches(Load+store) */ +#define PM_L3_PREF_ALL 0x4e052 +/* Data PTEG reload */ +#define PM_DTLB_MISS 0x300fc +/* ITLB Reloaded */ +#define PM_ITLB_MISS 0x400fc + /* * Raw event encoding for POWER8: @@ -557,6 +588,8 @@ static int power8_generic_events[] = { [PERF_COUNT_HW_INSTRUCTIONS] = PM_INST_CMPL, [PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = PM_BRU_FIN, [PERF_COUNT_HW_BRANCH_MISSES] = PM_BR_MPRED_CMPL, + [PERF_COUNT_HW_CACHE_REFERENCES] = PM_LD_REF_L1, + [PERF_COUNT_HW_CACHE_MISSES] = PM_LD_MISS_L1, }; static u64 power8_bhrb_filter_map(u64 branch_sample_type) @@ -596,6 +629,116 @@ static void power8_config_bhrb(u64 pmu_bhrb_filter) mtspr(SPRN_MMCRA, (mfspr(SPRN_MMCRA) | pmu_bhrb_filter)); } +#define C(x) PERF_COUNT_HW_CACHE_##x + +/* + * Table of generalized cache-related events. + * 0 means not supported, -1 means nonsensical, other values + * are event codes. + */ +static int power8_cache_events[C(MAX)][C(OP_MAX)][C(RESULT_MAX)] = { + [ C(L1D) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = PM_LD_REF_L1, + [ C(RESULT_MISS) ] = PM_LD_MISS_L1, + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = 0, + [ C(RESULT_MISS) ] = PM_ST_MISS_L1, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = PM_L1_PREF, + [ C(RESULT_MISS) ] = 0, + }, + }, + [ C(L1I) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = PM_INST_FROM_L1, + [ C(RESULT_MISS) ] = PM_L1_ICACHE_MISS, + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = PM_L1_DEMAND_WRITE, + [ C(RESULT_MISS) ] = -1, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = PM_IC_PREF_WRITE, + [ C(RESULT_MISS) ] = 0, + }, + }, + [ C(LL) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = PM_DATA_FROM_L3, + [ C(RESULT_MISS) ] = PM_DATA_FROM_L3MISS, + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = PM_L2_ST, + [ C(RESULT_MISS) ] = PM_L2_ST_MISS, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = PM_L3_PREF_ALL, + [ C(RESULT_MISS) ] = 0, + }, + }, + [ C(DTLB) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0, + [ C(RESULT_MISS) ] = PM_DTLB_MISS, + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + }, + [ C(ITLB) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0, + [ C(RESULT_MISS) ] = PM_ITLB_MISS, + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + }, + [ C(BPU) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = PM_BRU_FIN, + [ C(RESULT_MISS) ] = PM_BR_MPRED_CMPL, + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + }, + [ C(NODE) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + }, +}; + +#undef C + static struct power_pmu power8_pmu = { .name = "POWER8", .n_counter = 6, @@ -611,6 +754,7 @@ static struct power_pmu power8_pmu = { .flags = PPMU_HAS_SSLOT | PPMU_HAS_SIER | PPMU_BHRB | PPMU_EBB, .n_generic = ARRAY_SIZE(power8_generic_events), .generic_events = power8_generic_events, + .cache_events = &power8_cache_events, .attr_groups = power8_pmu_attr_groups, .bhrb_nr = 32, }; From 8d4887ee30695d6b092e96029ae974505c43bfb6 Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Tue, 10 Dec 2013 14:02:49 +1100 Subject: [PATCH 494/596] powerpc/pseries: Select ARCH_RANDOM on pseries We have a driver for the ARCH_RANDOM hook in rng.c, so we should select ARCH_RANDOM on pseries. Without this the build breaks if you turn ARCH_RANDOM off. This hasn't broken the build because pseries_defconfig doesn't specify a value for PPC_POWERNV, which is default y, and selects ARCH_RANDOM. Signed-off-by: Michael Ellerman Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/platforms/pseries/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/powerpc/platforms/pseries/Kconfig b/arch/powerpc/platforms/pseries/Kconfig index 37300f6ee244..80b1d57c306a 100644 --- a/arch/powerpc/platforms/pseries/Kconfig +++ b/arch/powerpc/platforms/pseries/Kconfig @@ -20,6 +20,7 @@ config PPC_PSERIES select PPC_DOORBELL select HAVE_CONTEXT_TRACKING select HOTPLUG_CPU if SMP + select ARCH_RANDOM default y config PPC_SPLPAR From 0ba3e1011660042dd3ea08ae8dbdc173352f3d47 Mon Sep 17 00:00:00 2001 From: Nathan Fontenot Date: Wed, 29 Jan 2014 10:34:56 -0600 Subject: [PATCH 495/596] crypto/nx/nx-842: Fix handling of vmalloc addresses The powerpc specific nx-842 compression driver does not currently handle translating a vmalloc address to a physical address. The current driver uses __pa() for all addresses which does not properly handle vmalloc addresses and thus causes a failure since we do not pass a proper physical address to the hypervisor. This patch adds a routine to convert an address to a physical address by checking for vmalloc addresses and handling them properly. Signed-off-by: Nathan Fontenot --- drivers/crypto/nx/nx-842.c | 29 +++++++++++++++++++---------- 1 file changed, 19 insertions(+), 10 deletions(-) Signed-off-by: Benjamin Herrenschmidt --- drivers/crypto/nx/nx-842.c | 29 +++++++++++++++++++---------- 1 file changed, 19 insertions(+), 10 deletions(-) diff --git a/drivers/crypto/nx/nx-842.c b/drivers/crypto/nx/nx-842.c index 6c4c000671c5..1e5481d88a26 100644 --- a/drivers/crypto/nx/nx-842.c +++ b/drivers/crypto/nx/nx-842.c @@ -158,6 +158,15 @@ static inline unsigned long nx842_get_scatterlist_size( return sl->entry_nr * sizeof(struct nx842_slentry); } +static inline unsigned long nx842_get_pa(void *addr) +{ + if (is_vmalloc_addr(addr)) + return page_to_phys(vmalloc_to_page(addr)) + + offset_in_page(addr); + else + return __pa(addr); +} + static int nx842_build_scatterlist(unsigned long buf, int len, struct nx842_scatterlist *sl) { @@ -168,7 +177,7 @@ static int nx842_build_scatterlist(unsigned long buf, int len, entry = sl->entries; while (len) { - entry->ptr = __pa(buf); + entry->ptr = nx842_get_pa((void *)buf); nextpage = ALIGN(buf + 1, NX842_HW_PAGE_SIZE); if (nextpage < buf + len) { /* we aren't at the end yet */ @@ -370,8 +379,8 @@ int nx842_compress(const unsigned char *in, unsigned int inlen, op.flags = NX842_OP_COMPRESS; csbcpb = &workmem->csbcpb; memset(csbcpb, 0, sizeof(*csbcpb)); - op.csbcpb = __pa(csbcpb); - op.out = __pa(slout.entries); + op.csbcpb = nx842_get_pa(csbcpb); + op.out = nx842_get_pa(slout.entries); for (i = 0; i < hdr->blocks_nr; i++) { /* @@ -401,13 +410,13 @@ int nx842_compress(const unsigned char *in, unsigned int inlen, */ if (likely(max_sync_size == NX842_HW_PAGE_SIZE)) { /* Create direct DDE */ - op.in = __pa(inbuf); + op.in = nx842_get_pa((void *)inbuf); op.inlen = max_sync_size; } else { /* Create indirect DDE (scatterlist) */ nx842_build_scatterlist(inbuf, max_sync_size, &slin); - op.in = __pa(slin.entries); + op.in = nx842_get_pa(slin.entries); op.inlen = -nx842_get_scatterlist_size(&slin); } @@ -565,7 +574,7 @@ int nx842_decompress(const unsigned char *in, unsigned int inlen, op.flags = NX842_OP_DECOMPRESS; csbcpb = &workmem->csbcpb; memset(csbcpb, 0, sizeof(*csbcpb)); - op.csbcpb = __pa(csbcpb); + op.csbcpb = nx842_get_pa(csbcpb); /* * max_sync_size may have changed since compression, @@ -597,12 +606,12 @@ int nx842_decompress(const unsigned char *in, unsigned int inlen, if (likely((inbuf & NX842_HW_PAGE_MASK) == ((inbuf + hdr->sizes[i] - 1) & NX842_HW_PAGE_MASK))) { /* Create direct DDE */ - op.in = __pa(inbuf); + op.in = nx842_get_pa((void *)inbuf); op.inlen = hdr->sizes[i]; } else { /* Create indirect DDE (scatterlist) */ nx842_build_scatterlist(inbuf, hdr->sizes[i] , &slin); - op.in = __pa(slin.entries); + op.in = nx842_get_pa(slin.entries); op.inlen = -nx842_get_scatterlist_size(&slin); } @@ -613,12 +622,12 @@ int nx842_decompress(const unsigned char *in, unsigned int inlen, */ if (likely(max_sync_size == NX842_HW_PAGE_SIZE)) { /* Create direct DDE */ - op.out = __pa(outbuf); + op.out = nx842_get_pa((void *)outbuf); op.outlen = max_sync_size; } else { /* Create indirect DDE (scatterlist) */ nx842_build_scatterlist(outbuf, max_sync_size, &slout); - op.out = __pa(slout.entries); + op.out = nx842_get_pa(slout.entries); op.outlen = -nx842_get_scatterlist_size(&slout); } From b4d6c06c8de81517320a9964b0c26e096aeadd7b Mon Sep 17 00:00:00 2001 From: Anshuman Khandual Date: Wed, 18 Dec 2013 13:14:53 +1100 Subject: [PATCH 496/596] powerpc/perf: Configure BHRB filter before enabling PMU interrupts Right now the config_bhrb() PMU specific call happens after write_mmcr0(), which actually enables the PMU for event counting and interrupts. So there is a small window of time where the PMU and BHRB runs without the required HW branch filter (if any) enabled in BHRB. This can cause some of the branch samples to be collected through BHRB without any filter applied and hence affects the correctness of the results. This patch moves the BHRB config function call before enabling interrupts. Here are some data points captured via trace prints which depicts how we could get PMU interrupts with BHRB filter NOT enabled with a standard perf record command line (asking for branch record information as well). $ perf record -j any_call ls Before the patch:- ls-1962 [003] d... 2065.299590: .perf_event_interrupt: MMCRA: 40000000000 ls-1962 [003] d... 2065.299603: .perf_event_interrupt: MMCRA: 40000000000 ... All the PMU interrupts before this point did not have the requested HW branch filter enabled in the MMCRA. ls-1962 [003] d... 2065.299647: .perf_event_interrupt: MMCRA: 40040000000 ls-1962 [003] d... 2065.299662: .perf_event_interrupt: MMCRA: 40040000000 After the patch:- ls-1850 [008] d... 190.311828: .perf_event_interrupt: MMCRA: 40040000000 ls-1850 [008] d... 190.311848: .perf_event_interrupt: MMCRA: 40040000000 All the PMU interrupts have the requested HW BHRB branch filter enabled in MMCRA. Signed-off-by: Anshuman Khandual [mpe: Fixed up whitespace and cleaned up changelog] Signed-off-by: Michael Ellerman Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/perf/core-book3s.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/arch/powerpc/perf/core-book3s.c b/arch/powerpc/perf/core-book3s.c index 29b89e863d7c..67cf22083f4c 100644 --- a/arch/powerpc/perf/core-book3s.c +++ b/arch/powerpc/perf/core-book3s.c @@ -1147,6 +1147,9 @@ static void power_pmu_enable(struct pmu *pmu) mmcr0 = ebb_switch_in(ebb, cpuhw->mmcr[0]); mb(); + if (cpuhw->bhrb_users) + ppmu->config_bhrb(cpuhw->bhrb_filter); + write_mmcr0(cpuhw, mmcr0); /* @@ -1158,8 +1161,6 @@ static void power_pmu_enable(struct pmu *pmu) } out: - if (cpuhw->bhrb_users) - ppmu->config_bhrb(cpuhw->bhrb_filter); local_irq_restore(flags); } From 730efb6193f8568354fd80849612291afa9fa81e Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Mon, 23 Dec 2013 23:46:04 +1100 Subject: [PATCH 497/596] powerpc/xmon: Don't loop forever in get_output_lock() If we enter with xmon_speaker != 0 we skip the first cmpxchg(), we also skip the while loop because xmon_speaker != last_speaker (0) - meaning we skip the second cmpxchg() also. Following that code path the compiler sees no memory barriers and so is within its rights to never reload xmon_speaker. The end result is we loop forever. This manifests as all cpus being in xmon ('c' command), but they refuse to take control when you switch to them ('c x' for cpu # x). I have seen this deadlock in practice and also checked the generated code to confirm this is what's happening. The simplest fix is just to always try the cmpxchg(). Signed-off-by: Michael Ellerman Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/xmon/xmon.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/arch/powerpc/xmon/xmon.c b/arch/powerpc/xmon/xmon.c index a90731b3d44a..598cdc7c7adc 100644 --- a/arch/powerpc/xmon/xmon.c +++ b/arch/powerpc/xmon/xmon.c @@ -309,12 +309,12 @@ static void get_output_lock(void) if (xmon_speaker == me) return; + for (;;) { - if (xmon_speaker == 0) { - last_speaker = cmpxchg(&xmon_speaker, 0, me); - if (last_speaker == 0) - return; - } + last_speaker = cmpxchg(&xmon_speaker, 0, me); + if (last_speaker == 0) + return; + timeout = 10000000; while (xmon_speaker == last_speaker) { if (--timeout > 0) From 1507589787529b0d8e2a9e66e0c6f113ecab5181 Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Mon, 23 Dec 2013 23:46:05 +1100 Subject: [PATCH 498/596] powerpc/xmon: Fix timeout loop in get_output_lock() As far as I can tell, our 70s era timeout loop in get_output_lock() is generating no code. This leads to the hostile takeover happening more or less simultaneously on all cpus. The result is "interesting", some example output that is more readable than most: cpu 0x1: Vector: 100 (Scypsut e0mx bR:e setV)e catto xc0p:u[ c 00 c0:0 000t0o0V0erc0td:o5 rfc28050000]0c00 0 0 0 6t(pSrycsV1ppuot uxe 1m 2 0Rx21e3:0s0ce000c00000t00)00 60602oV2SerucSayt0y 0p 1sxs Fix it by using udelay() in the timeout loop. The wait time and check frequency are arbitrary, but seem to work OK. We already rely on udelay() working so this is not a new dependency. Signed-off-by: Michael Ellerman Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/xmon/xmon.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/arch/powerpc/xmon/xmon.c b/arch/powerpc/xmon/xmon.c index 598cdc7c7adc..d712b23974ec 100644 --- a/arch/powerpc/xmon/xmon.c +++ b/arch/powerpc/xmon/xmon.c @@ -315,10 +315,17 @@ static void get_output_lock(void) if (last_speaker == 0) return; - timeout = 10000000; + /* + * Wait a full second for the lock, we might be on a slow + * console, but check every 100us. + */ + timeout = 10000; while (xmon_speaker == last_speaker) { - if (--timeout > 0) + if (--timeout > 0) { + udelay(100); continue; + } + /* hostile takeover */ prev = cmpxchg(&xmon_speaker, last_speaker, me); if (prev == last_speaker) From d2b496e5e1fa7a6796534e435440eb9d3ed184dd Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Mon, 23 Dec 2013 23:46:06 +1100 Subject: [PATCH 499/596] powerpc/xmon: Don't signal we've entered until we're finished printing Currently we set our cpu's bit in cpus_in_xmon, and then we take the output lock and print the exception information. This can race with the master cpu entering the command loop and printing the backtrace. The result is that the backtrace gets garbled with another cpu's exception print out. Fix it by delaying the set of cpus_in_xmon until we are finished printing. Signed-off-by: Michael Ellerman Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/xmon/xmon.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/arch/powerpc/xmon/xmon.c b/arch/powerpc/xmon/xmon.c index d712b23974ec..b07909850f77 100644 --- a/arch/powerpc/xmon/xmon.c +++ b/arch/powerpc/xmon/xmon.c @@ -404,7 +404,6 @@ static int xmon_core(struct pt_regs *regs, int fromipi) } xmon_fault_jmp[cpu] = recurse_jmp; - cpumask_set_cpu(cpu, &cpus_in_xmon); bp = NULL; if ((regs->msr & (MSR_IR|MSR_PR|MSR_64BIT)) == (MSR_IR|MSR_64BIT)) @@ -426,6 +425,8 @@ static int xmon_core(struct pt_regs *regs, int fromipi) release_output_lock(); } + cpumask_set_cpu(cpu, &cpus_in_xmon); + waiting: secondary = 1; while (secondary && !xmon_gate) { From 04a341138da55daf47551308660a5fcc6b5b9077 Mon Sep 17 00:00:00 2001 From: Kevin Hao Date: Wed, 29 Jan 2014 18:24:54 +0800 Subject: [PATCH 500/596] powerpc/ppc32: Fix the bug in the init of non-base exception stack for UP We would allocate one specific exception stack for each kind of non-base exceptions for every CPU. For ppc32 the CPU hard ID is used as the subscript to get the specific exception stack for one CPU. But for an UP kernel, there is only one element in the each kind of exception stack array. We would get stuck if the CPU hard ID is not equal to '0'. So in this case we should use the subscript '0' no matter what the CPU hard ID is. Signed-off-by: Kevin Hao Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/irq.c | 5 +++++ arch/powerpc/kernel/setup_32.c | 5 +++++ 2 files changed, 10 insertions(+) diff --git a/arch/powerpc/kernel/irq.c b/arch/powerpc/kernel/irq.c index 9729b23bfb0a..1d0848bba049 100644 --- a/arch/powerpc/kernel/irq.c +++ b/arch/powerpc/kernel/irq.c @@ -559,8 +559,13 @@ void exc_lvl_ctx_init(void) #ifdef CONFIG_PPC64 cpu_nr = i; #else +#ifdef CONFIG_SMP cpu_nr = get_hard_smp_processor_id(i); +#else + cpu_nr = 0; #endif +#endif + memset((void *)critirq_ctx[cpu_nr], 0, THREAD_SIZE); tp = critirq_ctx[cpu_nr]; tp->cpu = cpu_nr; diff --git a/arch/powerpc/kernel/setup_32.c b/arch/powerpc/kernel/setup_32.c index 2b0da27eaee4..04cc4fcca78b 100644 --- a/arch/powerpc/kernel/setup_32.c +++ b/arch/powerpc/kernel/setup_32.c @@ -247,7 +247,12 @@ static void __init exc_lvl_early_init(void) /* interrupt stacks must be in lowmem, we get that for free on ppc32 * as the memblock is limited to lowmem by MEMBLOCK_REAL_LIMIT */ for_each_possible_cpu(i) { +#ifdef CONFIG_SMP hw_cpu = get_hard_smp_processor_id(i); +#else + hw_cpu = 0; +#endif + critirq_ctx[hw_cpu] = (struct thread_info *) __va(memblock_alloc(THREAD_SIZE, THREAD_SIZE)); #ifdef CONFIG_BOOKE From ea961a828fe7250e954f086d74d9323c3d44c3e4 Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Wed, 22 Jan 2014 08:40:28 +1100 Subject: [PATCH 501/596] powerpc: Fix endian issues in kexec and crash dump code We expose a number of OF properties in the kexec and crash dump code and these need to be big endian. Cc: stable@vger.kernel.org # v3.13 Signed-off-by: Anton Blanchard Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/machine_kexec.c | 14 ++++++++++---- arch/powerpc/kernel/machine_kexec_64.c | 6 ++++-- 2 files changed, 14 insertions(+), 6 deletions(-) diff --git a/arch/powerpc/kernel/machine_kexec.c b/arch/powerpc/kernel/machine_kexec.c index 75d4f7340da8..015ae55c1868 100644 --- a/arch/powerpc/kernel/machine_kexec.c +++ b/arch/powerpc/kernel/machine_kexec.c @@ -196,7 +196,9 @@ int overlaps_crashkernel(unsigned long start, unsigned long size) /* Values we need to export to the second kernel via the device tree. */ static phys_addr_t kernel_end; +static phys_addr_t crashk_base; static phys_addr_t crashk_size; +static unsigned long long mem_limit; static struct property kernel_end_prop = { .name = "linux,kernel-end", @@ -207,7 +209,7 @@ static struct property kernel_end_prop = { static struct property crashk_base_prop = { .name = "linux,crashkernel-base", .length = sizeof(phys_addr_t), - .value = &crashk_res.start, + .value = &crashk_base }; static struct property crashk_size_prop = { @@ -219,9 +221,11 @@ static struct property crashk_size_prop = { static struct property memory_limit_prop = { .name = "linux,memory-limit", .length = sizeof(unsigned long long), - .value = &memory_limit, + .value = &mem_limit, }; +#define cpu_to_be_ulong __PASTE(cpu_to_be, BITS_PER_LONG) + static void __init export_crashk_values(struct device_node *node) { struct property *prop; @@ -237,8 +241,9 @@ static void __init export_crashk_values(struct device_node *node) of_remove_property(node, prop); if (crashk_res.start != 0) { + crashk_base = cpu_to_be_ulong(crashk_res.start), of_add_property(node, &crashk_base_prop); - crashk_size = resource_size(&crashk_res); + crashk_size = cpu_to_be_ulong(resource_size(&crashk_res)); of_add_property(node, &crashk_size_prop); } @@ -246,6 +251,7 @@ static void __init export_crashk_values(struct device_node *node) * memory_limit is required by the kexec-tools to limit the * crash regions to the actual memory used. */ + mem_limit = cpu_to_be_ulong(memory_limit); of_update_property(node, &memory_limit_prop); } @@ -264,7 +270,7 @@ static int __init kexec_setup(void) of_remove_property(node, prop); /* information needed by userspace when using default_machine_kexec */ - kernel_end = __pa(_end); + kernel_end = cpu_to_be_ulong(__pa(_end)); of_add_property(node, &kernel_end_prop); export_crashk_values(node); diff --git a/arch/powerpc/kernel/machine_kexec_64.c b/arch/powerpc/kernel/machine_kexec_64.c index be4e6d648f60..59d229a2a3e0 100644 --- a/arch/powerpc/kernel/machine_kexec_64.c +++ b/arch/powerpc/kernel/machine_kexec_64.c @@ -369,6 +369,7 @@ void default_machine_kexec(struct kimage *image) /* Values we need to export to the second kernel via the device tree. */ static unsigned long htab_base; +static unsigned long htab_size; static struct property htab_base_prop = { .name = "linux,htab-base", @@ -379,7 +380,7 @@ static struct property htab_base_prop = { static struct property htab_size_prop = { .name = "linux,htab-size", .length = sizeof(unsigned long), - .value = &htab_size_bytes, + .value = &htab_size, }; static int __init export_htab_values(void) @@ -403,8 +404,9 @@ static int __init export_htab_values(void) if (prop) of_remove_property(node, prop); - htab_base = __pa(htab_address); + htab_base = cpu_to_be64(__pa(htab_address)); of_add_property(node, &htab_base_prop); + htab_size = cpu_to_be64(htab_size_bytes); of_add_property(node, &htab_size_prop); of_node_put(node); From 19a38d8e0aa33b4f4d11d3b4baa902ad169daa80 Mon Sep 17 00:00:00 2001 From: Liu Junliang Date: Mon, 10 Feb 2014 14:31:42 +0800 Subject: [PATCH 502/596] USB2NET : SR9800 : One chip USB2.0 USB2NET SR9800 Device Driver Support Signed-off-by: Liu Junliang Signed-off-by: David S. Miller --- drivers/net/usb/Kconfig | 16 + drivers/net/usb/Makefile | 1 + drivers/net/usb/sr9800.c | 870 +++++++++++++++++++++++++++++++++++++++ drivers/net/usb/sr9800.h | 202 +++++++++ 4 files changed, 1089 insertions(+) create mode 100644 drivers/net/usb/sr9800.c create mode 100644 drivers/net/usb/sr9800.h diff --git a/drivers/net/usb/Kconfig b/drivers/net/usb/Kconfig index 6b638a066c1d..409499fdb157 100644 --- a/drivers/net/usb/Kconfig +++ b/drivers/net/usb/Kconfig @@ -292,6 +292,22 @@ config USB_NET_SR9700 This option adds support for CoreChip-sz SR9700 based USB 1.1 10/100 Ethernet adapters. +config USB_NET_SR9800 + tristate "CoreChip-sz SR9800 based USB 2.0 10/100 ethernet devices" + depends on USB_USBNET + select CRC32 + default y + ---help--- + Say Y if you want to use one of the following 100Mbps USB Ethernet + device based on the CoreChip-sz SR9800 chip. + + This driver makes the adapter appear as a normal Ethernet interface, + typically on eth0, if it is the only ethernet device, or perhaps on + eth1, if you have a PCI or ISA ethernet card installed. + + To compile this driver as a module, choose M here: the + module will be called sr9800. + config USB_NET_SMSC75XX tristate "SMSC LAN75XX based USB 2.0 gigabit ethernet devices" depends on USB_USBNET diff --git a/drivers/net/usb/Makefile b/drivers/net/usb/Makefile index b17b5e88bbaf..433f0a00c683 100644 --- a/drivers/net/usb/Makefile +++ b/drivers/net/usb/Makefile @@ -15,6 +15,7 @@ obj-$(CONFIG_USB_NET_CDCETHER) += cdc_ether.o r815x.o obj-$(CONFIG_USB_NET_CDC_EEM) += cdc_eem.o obj-$(CONFIG_USB_NET_DM9601) += dm9601.o obj-$(CONFIG_USB_NET_SR9700) += sr9700.o +obj-$(CONFIG_USB_NET_SR9800) += sr9800.o obj-$(CONFIG_USB_NET_SMSC75XX) += smsc75xx.o obj-$(CONFIG_USB_NET_SMSC95XX) += smsc95xx.o obj-$(CONFIG_USB_NET_GL620A) += gl620a.o diff --git a/drivers/net/usb/sr9800.c b/drivers/net/usb/sr9800.c new file mode 100644 index 000000000000..4175eb9fdeca --- /dev/null +++ b/drivers/net/usb/sr9800.c @@ -0,0 +1,870 @@ +/* CoreChip-sz SR9800 one chip USB 2.0 Ethernet Devices + * + * Author : Liu Junliang + * + * Based on asix_common.c, asix_devices.c + * + * This file is licensed under the terms of the GNU General Public License + * version 2. This program is licensed "as is" without any warranty of any + * kind, whether express or implied.* + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "sr9800.h" + +static int sr_read_cmd(struct usbnet *dev, u8 cmd, u16 value, u16 index, + u16 size, void *data) +{ + int err; + + err = usbnet_read_cmd(dev, cmd, SR_REQ_RD_REG, value, index, + data, size); + if ((err != size) && (err >= 0)) + err = -EINVAL; + + return err; +} + +static int sr_write_cmd(struct usbnet *dev, u8 cmd, u16 value, u16 index, + u16 size, void *data) +{ + int err; + + err = usbnet_write_cmd(dev, cmd, SR_REQ_WR_REG, value, index, + data, size); + if ((err != size) && (err >= 0)) + err = -EINVAL; + + return err; +} + +static void +sr_write_cmd_async(struct usbnet *dev, u8 cmd, u16 value, u16 index, + u16 size, void *data) +{ + usbnet_write_cmd_async(dev, cmd, SR_REQ_WR_REG, value, index, data, + size); +} + +static int sr_rx_fixup(struct usbnet *dev, struct sk_buff *skb) +{ + int offset = 0; + + while (offset + sizeof(u32) < skb->len) { + struct sk_buff *sr_skb; + u16 size; + u32 header = get_unaligned_le32(skb->data + offset); + + offset += sizeof(u32); + /* get the packet length */ + size = (u16) (header & 0x7ff); + if (size != ((~header >> 16) & 0x07ff)) { + netdev_err(dev->net, "%s : Bad Header Length\n", + __func__); + return 0; + } + + if ((size > dev->net->mtu + ETH_HLEN + VLAN_HLEN) || + (size + offset > skb->len)) { + netdev_err(dev->net, "%s : Bad RX Length %d\n", + __func__, size); + return 0; + } + sr_skb = netdev_alloc_skb_ip_align(dev->net, size); + if (!sr_skb) + return 0; + + skb_put(sr_skb, size); + memcpy(sr_skb->data, skb->data + offset, size); + usbnet_skb_return(dev, sr_skb); + + offset += (size + 1) & 0xfffe; + } + + if (skb->len != offset) { + netdev_err(dev->net, "%s : Bad SKB Length %d\n", __func__, + skb->len); + return 0; + } + + return 1; +} + +static struct sk_buff *sr_tx_fixup(struct usbnet *dev, struct sk_buff *skb, + gfp_t flags) +{ + int headroom = skb_headroom(skb); + int tailroom = skb_tailroom(skb); + u32 padbytes = 0xffff0000; + u32 packet_len; + int padlen; + + padlen = ((skb->len + 4) % (dev->maxpacket - 1)) ? 0 : 4; + + if ((!skb_cloned(skb)) && ((headroom + tailroom) >= (4 + padlen))) { + if ((headroom < 4) || (tailroom < padlen)) { + skb->data = memmove(skb->head + 4, skb->data, + skb->len); + skb_set_tail_pointer(skb, skb->len); + } + } else { + struct sk_buff *skb2; + skb2 = skb_copy_expand(skb, 4, padlen, flags); + dev_kfree_skb_any(skb); + skb = skb2; + if (!skb) + return NULL; + } + + skb_push(skb, 4); + packet_len = (((skb->len - 4) ^ 0x0000ffff) << 16) + (skb->len - 4); + cpu_to_le32s(&packet_len); + skb_copy_to_linear_data(skb, &packet_len, sizeof(packet_len)); + + if (padlen) { + cpu_to_le32s(&padbytes); + memcpy(skb_tail_pointer(skb), &padbytes, sizeof(padbytes)); + skb_put(skb, sizeof(padbytes)); + } + + return skb; +} + +static void sr_status(struct usbnet *dev, struct urb *urb) +{ + struct sr9800_int_data *event; + int link; + + if (urb->actual_length < 8) + return; + + event = urb->transfer_buffer; + link = event->link & 0x01; + if (netif_carrier_ok(dev->net) != link) { + usbnet_link_change(dev, link, 1); + netdev_dbg(dev->net, "Link Status is: %d\n", link); + } + + return; +} + +static inline int sr_set_sw_mii(struct usbnet *dev) +{ + int ret; + + ret = sr_write_cmd(dev, SR_CMD_SET_SW_MII, 0x0000, 0, 0, NULL); + if (ret < 0) + netdev_err(dev->net, "Failed to enable software MII access\n"); + return ret; +} + +static inline int sr_set_hw_mii(struct usbnet *dev) +{ + int ret; + + ret = sr_write_cmd(dev, SR_CMD_SET_HW_MII, 0x0000, 0, 0, NULL); + if (ret < 0) + netdev_err(dev->net, "Failed to enable hardware MII access\n"); + return ret; +} + +static inline int sr_get_phy_addr(struct usbnet *dev) +{ + u8 buf[2]; + int ret; + + ret = sr_read_cmd(dev, SR_CMD_READ_PHY_ID, 0, 0, 2, buf); + if (ret < 0) { + netdev_err(dev->net, "%s : Error reading PHYID register:%02x\n", + __func__, ret); + goto out; + } + netdev_dbg(dev->net, "%s : returning 0x%04x\n", __func__, + *((__le16 *)buf)); + + ret = buf[1]; + +out: + return ret; +} + +static int sr_sw_reset(struct usbnet *dev, u8 flags) +{ + int ret; + + ret = sr_write_cmd(dev, SR_CMD_SW_RESET, flags, 0, 0, NULL); + if (ret < 0) + netdev_err(dev->net, "Failed to send software reset:%02x\n", + ret); + + return ret; +} + +static u16 sr_read_rx_ctl(struct usbnet *dev) +{ + __le16 v; + int ret; + + ret = sr_read_cmd(dev, SR_CMD_READ_RX_CTL, 0, 0, 2, &v); + if (ret < 0) { + netdev_err(dev->net, "Error reading RX_CTL register:%02x\n", + ret); + goto out; + } + + ret = le16_to_cpu(v); +out: + return ret; +} + +static int sr_write_rx_ctl(struct usbnet *dev, u16 mode) +{ + int ret; + + netdev_dbg(dev->net, "%s : mode = 0x%04x\n", __func__, mode); + ret = sr_write_cmd(dev, SR_CMD_WRITE_RX_CTL, mode, 0, 0, NULL); + if (ret < 0) + netdev_err(dev->net, + "Failed to write RX_CTL mode to 0x%04x:%02x\n", + mode, ret); + + return ret; +} + +static u16 sr_read_medium_status(struct usbnet *dev) +{ + __le16 v; + int ret; + + ret = sr_read_cmd(dev, SR_CMD_READ_MEDIUM_STATUS, 0, 0, 2, &v); + if (ret < 0) { + netdev_err(dev->net, + "Error reading Medium Status register:%02x\n", ret); + return ret; /* TODO: callers not checking for error ret */ + } + + return le16_to_cpu(v); +} + +static int sr_write_medium_mode(struct usbnet *dev, u16 mode) +{ + int ret; + + netdev_dbg(dev->net, "%s : mode = 0x%04x\n", __func__, mode); + ret = sr_write_cmd(dev, SR_CMD_WRITE_MEDIUM_MODE, mode, 0, 0, NULL); + if (ret < 0) + netdev_err(dev->net, + "Failed to write Medium Mode mode to 0x%04x:%02x\n", + mode, ret); + return ret; +} + +static int sr_write_gpio(struct usbnet *dev, u16 value, int sleep) +{ + int ret; + + netdev_dbg(dev->net, "%s : value = 0x%04x\n", __func__, value); + ret = sr_write_cmd(dev, SR_CMD_WRITE_GPIOS, value, 0, 0, NULL); + if (ret < 0) + netdev_err(dev->net, "Failed to write GPIO value 0x%04x:%02x\n", + value, ret); + if (sleep) + msleep(sleep); + + return ret; +} + +/* SR9800 have a 16-bit RX_CTL value */ +static void sr_set_multicast(struct net_device *net) +{ + struct usbnet *dev = netdev_priv(net); + struct sr_data *data = (struct sr_data *)&dev->data; + u16 rx_ctl = SR_DEFAULT_RX_CTL; + + if (net->flags & IFF_PROMISC) { + rx_ctl |= SR_RX_CTL_PRO; + } else if (net->flags & IFF_ALLMULTI || + netdev_mc_count(net) > SR_MAX_MCAST) { + rx_ctl |= SR_RX_CTL_AMALL; + } else if (netdev_mc_empty(net)) { + /* just broadcast and directed */ + } else { + /* We use the 20 byte dev->data + * for our 8 byte filter buffer + * to avoid allocating memory that + * is tricky to free later + */ + struct netdev_hw_addr *ha; + u32 crc_bits; + + memset(data->multi_filter, 0, SR_MCAST_FILTER_SIZE); + + /* Build the multicast hash filter. */ + netdev_for_each_mc_addr(ha, net) { + crc_bits = ether_crc(ETH_ALEN, ha->addr) >> 26; + data->multi_filter[crc_bits >> 3] |= + 1 << (crc_bits & 7); + } + + sr_write_cmd_async(dev, SR_CMD_WRITE_MULTI_FILTER, 0, 0, + SR_MCAST_FILTER_SIZE, data->multi_filter); + + rx_ctl |= SR_RX_CTL_AM; + } + + sr_write_cmd_async(dev, SR_CMD_WRITE_RX_CTL, rx_ctl, 0, 0, NULL); +} + +static int sr_mdio_read(struct net_device *net, int phy_id, int loc) +{ + struct usbnet *dev = netdev_priv(net); + __le16 res; + + mutex_lock(&dev->phy_mutex); + sr_set_sw_mii(dev); + sr_read_cmd(dev, SR_CMD_READ_MII_REG, phy_id, (__u16)loc, 2, &res); + sr_set_hw_mii(dev); + mutex_unlock(&dev->phy_mutex); + + netdev_dbg(dev->net, + "%s : phy_id=0x%02x, loc=0x%02x, returns=0x%04x\n", __func__, + phy_id, loc, le16_to_cpu(res)); + + return le16_to_cpu(res); +} + +static void +sr_mdio_write(struct net_device *net, int phy_id, int loc, int val) +{ + struct usbnet *dev = netdev_priv(net); + __le16 res = cpu_to_le16(val); + + netdev_dbg(dev->net, + "%s : phy_id=0x%02x, loc=0x%02x, val=0x%04x\n", __func__, + phy_id, loc, val); + mutex_lock(&dev->phy_mutex); + sr_set_sw_mii(dev); + sr_write_cmd(dev, SR_CMD_WRITE_MII_REG, phy_id, (__u16)loc, 2, &res); + sr_set_hw_mii(dev); + mutex_unlock(&dev->phy_mutex); +} + +/* Get the PHY Identifier from the PHYSID1 & PHYSID2 MII registers */ +static u32 sr_get_phyid(struct usbnet *dev) +{ + int phy_reg; + u32 phy_id; + int i; + + /* Poll for the rare case the FW or phy isn't ready yet. */ + for (i = 0; i < 100; i++) { + phy_reg = sr_mdio_read(dev->net, dev->mii.phy_id, MII_PHYSID1); + if (phy_reg != 0 && phy_reg != 0xFFFF) + break; + mdelay(1); + } + + if (phy_reg <= 0 || phy_reg == 0xFFFF) + return 0; + + phy_id = (phy_reg & 0xffff) << 16; + + phy_reg = sr_mdio_read(dev->net, dev->mii.phy_id, MII_PHYSID2); + if (phy_reg < 0) + return 0; + + phy_id |= (phy_reg & 0xffff); + + return phy_id; +} + +static void +sr_get_wol(struct net_device *net, struct ethtool_wolinfo *wolinfo) +{ + struct usbnet *dev = netdev_priv(net); + u8 opt; + + if (sr_read_cmd(dev, SR_CMD_READ_MONITOR_MODE, 0, 0, 1, &opt) < 0) { + wolinfo->supported = 0; + wolinfo->wolopts = 0; + return; + } + wolinfo->supported = WAKE_PHY | WAKE_MAGIC; + wolinfo->wolopts = 0; + if (opt & SR_MONITOR_LINK) + wolinfo->wolopts |= WAKE_PHY; + if (opt & SR_MONITOR_MAGIC) + wolinfo->wolopts |= WAKE_MAGIC; +} + +static int +sr_set_wol(struct net_device *net, struct ethtool_wolinfo *wolinfo) +{ + struct usbnet *dev = netdev_priv(net); + u8 opt = 0; + + if (wolinfo->wolopts & WAKE_PHY) + opt |= SR_MONITOR_LINK; + if (wolinfo->wolopts & WAKE_MAGIC) + opt |= SR_MONITOR_MAGIC; + + if (sr_write_cmd(dev, SR_CMD_WRITE_MONITOR_MODE, + opt, 0, 0, NULL) < 0) + return -EINVAL; + + return 0; +} + +static int sr_get_eeprom_len(struct net_device *net) +{ + struct usbnet *dev = netdev_priv(net); + struct sr_data *data = (struct sr_data *)&dev->data; + + return data->eeprom_len; +} + +static int sr_get_eeprom(struct net_device *net, + struct ethtool_eeprom *eeprom, u8 *data) +{ + struct usbnet *dev = netdev_priv(net); + __le16 *ebuf = (__le16 *)data; + int ret; + int i; + + /* Crude hack to ensure that we don't overwrite memory + * if an odd length is supplied + */ + if (eeprom->len % 2) + return -EINVAL; + + eeprom->magic = SR_EEPROM_MAGIC; + + /* sr9800 returns 2 bytes from eeprom on read */ + for (i = 0; i < eeprom->len / 2; i++) { + ret = sr_read_cmd(dev, SR_CMD_READ_EEPROM, eeprom->offset + i, + 0, 2, &ebuf[i]); + if (ret < 0) + return -EINVAL; + } + return 0; +} + +static void sr_get_drvinfo(struct net_device *net, + struct ethtool_drvinfo *info) +{ + struct usbnet *dev = netdev_priv(net); + struct sr_data *data = (struct sr_data *)&dev->data; + + /* Inherit standard device info */ + usbnet_get_drvinfo(net, info); + strncpy(info->driver, DRIVER_NAME, sizeof(info->driver)); + strncpy(info->version, DRIVER_VERSION, sizeof(info->version)); + info->eedump_len = data->eeprom_len; +} + +static u32 sr_get_link(struct net_device *net) +{ + struct usbnet *dev = netdev_priv(net); + + return mii_link_ok(&dev->mii); +} + +static int sr_ioctl(struct net_device *net, struct ifreq *rq, int cmd) +{ + struct usbnet *dev = netdev_priv(net); + + return generic_mii_ioctl(&dev->mii, if_mii(rq), cmd, NULL); +} + +static int sr_set_mac_address(struct net_device *net, void *p) +{ + struct usbnet *dev = netdev_priv(net); + struct sr_data *data = (struct sr_data *)&dev->data; + struct sockaddr *addr = p; + + if (netif_running(net)) + return -EBUSY; + if (!is_valid_ether_addr(addr->sa_data)) + return -EADDRNOTAVAIL; + + memcpy(net->dev_addr, addr->sa_data, ETH_ALEN); + + /* We use the 20 byte dev->data + * for our 6 byte mac buffer + * to avoid allocating memory that + * is tricky to free later + */ + memcpy(data->mac_addr, addr->sa_data, ETH_ALEN); + sr_write_cmd_async(dev, SR_CMD_WRITE_NODE_ID, 0, 0, ETH_ALEN, + data->mac_addr); + + return 0; +} + +static const struct ethtool_ops sr9800_ethtool_ops = { + .get_drvinfo = sr_get_drvinfo, + .get_link = sr_get_link, + .get_msglevel = usbnet_get_msglevel, + .set_msglevel = usbnet_set_msglevel, + .get_wol = sr_get_wol, + .set_wol = sr_set_wol, + .get_eeprom_len = sr_get_eeprom_len, + .get_eeprom = sr_get_eeprom, + .get_settings = usbnet_get_settings, + .set_settings = usbnet_set_settings, + .nway_reset = usbnet_nway_reset, +}; + +static int sr9800_link_reset(struct usbnet *dev) +{ + struct ethtool_cmd ecmd = { .cmd = ETHTOOL_GSET }; + u16 mode; + + mii_check_media(&dev->mii, 1, 1); + mii_ethtool_gset(&dev->mii, &ecmd); + mode = SR9800_MEDIUM_DEFAULT; + + if (ethtool_cmd_speed(&ecmd) != SPEED_100) + mode &= ~SR_MEDIUM_PS; + + if (ecmd.duplex != DUPLEX_FULL) + mode &= ~SR_MEDIUM_FD; + + netdev_dbg(dev->net, "%s : speed: %u duplex: %d mode: 0x%04x\n", + __func__, ethtool_cmd_speed(&ecmd), ecmd.duplex, mode); + + sr_write_medium_mode(dev, mode); + + return 0; +} + + +static int sr9800_set_default_mode(struct usbnet *dev) +{ + u16 rx_ctl; + int ret; + + sr_mdio_write(dev->net, dev->mii.phy_id, MII_BMCR, BMCR_RESET); + sr_mdio_write(dev->net, dev->mii.phy_id, MII_ADVERTISE, + ADVERTISE_ALL | ADVERTISE_CSMA); + mii_nway_restart(&dev->mii); + + ret = sr_write_medium_mode(dev, SR9800_MEDIUM_DEFAULT); + if (ret < 0) + goto out; + + ret = sr_write_cmd(dev, SR_CMD_WRITE_IPG012, + SR9800_IPG0_DEFAULT | SR9800_IPG1_DEFAULT, + SR9800_IPG2_DEFAULT, 0, NULL); + if (ret < 0) { + netdev_dbg(dev->net, "Write IPG,IPG1,IPG2 failed: %d\n", ret); + goto out; + } + + /* Set RX_CTL to default values with 2k buffer, and enable cactus */ + ret = sr_write_rx_ctl(dev, SR_DEFAULT_RX_CTL); + if (ret < 0) + goto out; + + rx_ctl = sr_read_rx_ctl(dev); + netdev_dbg(dev->net, "RX_CTL is 0x%04x after all initializations\n", + rx_ctl); + + rx_ctl = sr_read_medium_status(dev); + netdev_dbg(dev->net, "Medium Status:0x%04x after all initializations\n", + rx_ctl); + + return 0; +out: + return ret; +} + +static int sr9800_reset(struct usbnet *dev) +{ + struct sr_data *data = (struct sr_data *)&dev->data; + int ret, embd_phy; + u16 rx_ctl; + + ret = sr_write_gpio(dev, + SR_GPIO_RSE | SR_GPIO_GPO_2 | SR_GPIO_GPO2EN, 5); + if (ret < 0) + goto out; + + embd_phy = ((sr_get_phy_addr(dev) & 0x1f) == 0x10 ? 1 : 0); + + ret = sr_write_cmd(dev, SR_CMD_SW_PHY_SELECT, embd_phy, 0, 0, NULL); + if (ret < 0) { + netdev_dbg(dev->net, "Select PHY #1 failed: %d\n", ret); + goto out; + } + + ret = sr_sw_reset(dev, SR_SWRESET_IPPD | SR_SWRESET_PRL); + if (ret < 0) + goto out; + + msleep(150); + + ret = sr_sw_reset(dev, SR_SWRESET_CLEAR); + if (ret < 0) + goto out; + + msleep(150); + + if (embd_phy) { + ret = sr_sw_reset(dev, SR_SWRESET_IPRL); + if (ret < 0) + goto out; + } else { + ret = sr_sw_reset(dev, SR_SWRESET_PRTE); + if (ret < 0) + goto out; + } + + msleep(150); + rx_ctl = sr_read_rx_ctl(dev); + netdev_dbg(dev->net, "RX_CTL is 0x%04x after software reset\n", rx_ctl); + ret = sr_write_rx_ctl(dev, 0x0000); + if (ret < 0) + goto out; + + rx_ctl = sr_read_rx_ctl(dev); + netdev_dbg(dev->net, "RX_CTL is 0x%04x setting to 0x0000\n", rx_ctl); + + ret = sr_sw_reset(dev, SR_SWRESET_PRL); + if (ret < 0) + goto out; + + msleep(150); + + ret = sr_sw_reset(dev, SR_SWRESET_IPRL | SR_SWRESET_PRL); + if (ret < 0) + goto out; + + msleep(150); + + ret = sr9800_set_default_mode(dev); + if (ret < 0) + goto out; + + /* Rewrite MAC address */ + memcpy(data->mac_addr, dev->net->dev_addr, ETH_ALEN); + ret = sr_write_cmd(dev, SR_CMD_WRITE_NODE_ID, 0, 0, ETH_ALEN, + data->mac_addr); + if (ret < 0) + goto out; + + return 0; + +out: + return ret; +} + +static const struct net_device_ops sr9800_netdev_ops = { + .ndo_open = usbnet_open, + .ndo_stop = usbnet_stop, + .ndo_start_xmit = usbnet_start_xmit, + .ndo_tx_timeout = usbnet_tx_timeout, + .ndo_change_mtu = usbnet_change_mtu, + .ndo_set_mac_address = sr_set_mac_address, + .ndo_validate_addr = eth_validate_addr, + .ndo_do_ioctl = sr_ioctl, + .ndo_set_rx_mode = sr_set_multicast, +}; + +static int sr9800_phy_powerup(struct usbnet *dev) +{ + int ret; + + /* set the embedded Ethernet PHY in power-down state */ + ret = sr_sw_reset(dev, SR_SWRESET_IPPD | SR_SWRESET_IPRL); + if (ret < 0) { + netdev_err(dev->net, "Failed to power down PHY : %d\n", ret); + return ret; + } + msleep(20); + + /* set the embedded Ethernet PHY in power-up state */ + ret = sr_sw_reset(dev, SR_SWRESET_IPRL); + if (ret < 0) { + netdev_err(dev->net, "Failed to reset PHY: %d\n", ret); + return ret; + } + msleep(600); + + /* set the embedded Ethernet PHY in reset state */ + ret = sr_sw_reset(dev, SR_SWRESET_CLEAR); + if (ret < 0) { + netdev_err(dev->net, "Failed to power up PHY: %d\n", ret); + return ret; + } + msleep(20); + + /* set the embedded Ethernet PHY in power-up state */ + ret = sr_sw_reset(dev, SR_SWRESET_IPRL); + if (ret < 0) { + netdev_err(dev->net, "Failed to reset PHY: %d\n", ret); + return ret; + } + + return 0; +} + +static int sr9800_bind(struct usbnet *dev, struct usb_interface *intf) +{ + struct sr_data *data = (struct sr_data *)&dev->data; + u16 led01_mux, led23_mux; + int ret, embd_phy; + u32 phyid; + u16 rx_ctl; + + data->eeprom_len = SR9800_EEPROM_LEN; + + usbnet_get_endpoints(dev, intf); + + /* LED Setting Rule : + * AABB:CCDD + * AA : MFA0(LED0) + * BB : MFA1(LED1) + * CC : MFA2(LED2), Reserved for SR9800 + * DD : MFA3(LED3), Reserved for SR9800 + */ + led01_mux = (SR_LED_MUX_LINK_ACTIVE << 8) | SR_LED_MUX_LINK; + led23_mux = (SR_LED_MUX_LINK_ACTIVE << 8) | SR_LED_MUX_TX_ACTIVE; + ret = sr_write_cmd(dev, SR_CMD_LED_MUX, led01_mux, led23_mux, 0, NULL); + if (ret < 0) { + netdev_err(dev->net, "set LINK LED failed : %d\n", ret); + goto out; + } + + /* Get the MAC address */ + ret = sr_read_cmd(dev, SR_CMD_READ_NODE_ID, 0, 0, ETH_ALEN, + dev->net->dev_addr); + if (ret < 0) { + netdev_dbg(dev->net, "Failed to read MAC address: %d\n", ret); + return ret; + } + netdev_dbg(dev->net, "mac addr : %pM\n", dev->net->dev_addr); + + /* Initialize MII structure */ + dev->mii.dev = dev->net; + dev->mii.mdio_read = sr_mdio_read; + dev->mii.mdio_write = sr_mdio_write; + dev->mii.phy_id_mask = 0x1f; + dev->mii.reg_num_mask = 0x1f; + dev->mii.phy_id = sr_get_phy_addr(dev); + + dev->net->netdev_ops = &sr9800_netdev_ops; + dev->net->ethtool_ops = &sr9800_ethtool_ops; + + embd_phy = ((dev->mii.phy_id & 0x1f) == 0x10 ? 1 : 0); + /* Reset the PHY to normal operation mode */ + ret = sr_write_cmd(dev, SR_CMD_SW_PHY_SELECT, embd_phy, 0, 0, NULL); + if (ret < 0) { + netdev_dbg(dev->net, "Select PHY #1 failed: %d\n", ret); + return ret; + } + + /* Init PHY routine */ + ret = sr9800_phy_powerup(dev); + if (ret < 0) + goto out; + + rx_ctl = sr_read_rx_ctl(dev); + netdev_dbg(dev->net, "RX_CTL is 0x%04x after software reset\n", rx_ctl); + ret = sr_write_rx_ctl(dev, 0x0000); + if (ret < 0) + goto out; + + rx_ctl = sr_read_rx_ctl(dev); + netdev_dbg(dev->net, "RX_CTL is 0x%04x setting to 0x0000\n", rx_ctl); + + /* Read PHYID register *AFTER* the PHY was reset properly */ + phyid = sr_get_phyid(dev); + netdev_dbg(dev->net, "PHYID=0x%08x\n", phyid); + + /* medium mode setting */ + ret = sr9800_set_default_mode(dev); + if (ret < 0) + goto out; + + if (dev->udev->speed == USB_SPEED_HIGH) { + ret = sr_write_cmd(dev, SR_CMD_BULKIN_SIZE, + SR9800_BULKIN_SIZE[SR9800_MAX_BULKIN_4K].byte_cnt, + SR9800_BULKIN_SIZE[SR9800_MAX_BULKIN_4K].threshold, + 0, NULL); + if (ret < 0) { + netdev_err(dev->net, "Reset RX_CTL failed: %d\n", ret); + goto out; + } + dev->rx_urb_size = + SR9800_BULKIN_SIZE[SR9800_MAX_BULKIN_4K].size; + } else { + ret = sr_write_cmd(dev, SR_CMD_BULKIN_SIZE, + SR9800_BULKIN_SIZE[SR9800_MAX_BULKIN_2K].byte_cnt, + SR9800_BULKIN_SIZE[SR9800_MAX_BULKIN_2K].threshold, + 0, NULL); + if (ret < 0) { + netdev_err(dev->net, "Reset RX_CTL failed: %d\n", ret); + goto out; + } + dev->rx_urb_size = + SR9800_BULKIN_SIZE[SR9800_MAX_BULKIN_2K].size; + } + netdev_dbg(dev->net, "%s : setting rx_urb_size with : %ld\n", __func__, + dev->rx_urb_size); + return 0; + +out: + return ret; +} + +static const struct driver_info sr9800_driver_info = { + .description = "CoreChip SR9800 USB 2.0 Ethernet", + .bind = sr9800_bind, + .status = sr_status, + .link_reset = sr9800_link_reset, + .reset = sr9800_reset, + .flags = DRIVER_FLAG, + .rx_fixup = sr_rx_fixup, + .tx_fixup = sr_tx_fixup, +}; + +static const struct usb_device_id products[] = { + { + USB_DEVICE(0x0fe6, 0x9800), /* SR9800 Device */ + .driver_info = (unsigned long) &sr9800_driver_info, + }, + {}, /* END */ +}; + +MODULE_DEVICE_TABLE(usb, products); + +static struct usb_driver sr_driver = { + .name = DRIVER_NAME, + .id_table = products, + .probe = usbnet_probe, + .suspend = usbnet_suspend, + .resume = usbnet_resume, + .disconnect = usbnet_disconnect, + .supports_autosuspend = 1, +}; + +module_usb_driver(sr_driver); + +MODULE_AUTHOR("Liu Junliang + * + * This file is licensed under the terms of the GNU General Public License + * version 2. This program is licensed "as is" without any warranty of any + * kind, whether express or implied. + */ + +#ifndef _SR9800_H +#define _SR9800_H + +/* SR9800 spec. command table on Linux Platform */ + +/* command : Software Station Management Control Reg */ +#define SR_CMD_SET_SW_MII 0x06 +/* command : PHY Read Reg */ +#define SR_CMD_READ_MII_REG 0x07 +/* command : PHY Write Reg */ +#define SR_CMD_WRITE_MII_REG 0x08 +/* command : Hardware Station Management Control Reg */ +#define SR_CMD_SET_HW_MII 0x0a +/* command : SROM Read Reg */ +#define SR_CMD_READ_EEPROM 0x0b +/* command : SROM Write Reg */ +#define SR_CMD_WRITE_EEPROM 0x0c +/* command : SROM Write Enable Reg */ +#define SR_CMD_WRITE_ENABLE 0x0d +/* command : SROM Write Disable Reg */ +#define SR_CMD_WRITE_DISABLE 0x0e +/* command : RX Control Read Reg */ +#define SR_CMD_READ_RX_CTL 0x0f +#define SR_RX_CTL_PRO (1 << 0) +#define SR_RX_CTL_AMALL (1 << 1) +#define SR_RX_CTL_SEP (1 << 2) +#define SR_RX_CTL_AB (1 << 3) +#define SR_RX_CTL_AM (1 << 4) +#define SR_RX_CTL_AP (1 << 5) +#define SR_RX_CTL_ARP (1 << 6) +#define SR_RX_CTL_SO (1 << 7) +#define SR_RX_CTL_RH1M (1 << 8) +#define SR_RX_CTL_RH2M (1 << 9) +#define SR_RX_CTL_RH3M (1 << 10) +/* command : RX Control Write Reg */ +#define SR_CMD_WRITE_RX_CTL 0x10 +/* command : IPG0/IPG1/IPG2 Control Read Reg */ +#define SR_CMD_READ_IPG012 0x11 +/* command : IPG0/IPG1/IPG2 Control Write Reg */ +#define SR_CMD_WRITE_IPG012 0x12 +/* command : Node ID Read Reg */ +#define SR_CMD_READ_NODE_ID 0x13 +/* command : Node ID Write Reg */ +#define SR_CMD_WRITE_NODE_ID 0x14 +/* command : Multicast Filter Array Read Reg */ +#define SR_CMD_READ_MULTI_FILTER 0x15 +/* command : Multicast Filter Array Write Reg */ +#define SR_CMD_WRITE_MULTI_FILTER 0x16 +/* command : Eth/HomePNA PHY Address Reg */ +#define SR_CMD_READ_PHY_ID 0x19 +/* command : Medium Status Read Reg */ +#define SR_CMD_READ_MEDIUM_STATUS 0x1a +#define SR_MONITOR_LINK (1 << 1) +#define SR_MONITOR_MAGIC (1 << 2) +#define SR_MONITOR_HSFS (1 << 4) +/* command : Medium Status Write Reg */ +#define SR_CMD_WRITE_MEDIUM_MODE 0x1b +#define SR_MEDIUM_GM (1 << 0) +#define SR_MEDIUM_FD (1 << 1) +#define SR_MEDIUM_AC (1 << 2) +#define SR_MEDIUM_ENCK (1 << 3) +#define SR_MEDIUM_RFC (1 << 4) +#define SR_MEDIUM_TFC (1 << 5) +#define SR_MEDIUM_JFE (1 << 6) +#define SR_MEDIUM_PF (1 << 7) +#define SR_MEDIUM_RE (1 << 8) +#define SR_MEDIUM_PS (1 << 9) +#define SR_MEDIUM_RSV (1 << 10) +#define SR_MEDIUM_SBP (1 << 11) +#define SR_MEDIUM_SM (1 << 12) +/* command : Monitor Mode Status Read Reg */ +#define SR_CMD_READ_MONITOR_MODE 0x1c +/* command : Monitor Mode Status Write Reg */ +#define SR_CMD_WRITE_MONITOR_MODE 0x1d +/* command : GPIO Status Read Reg */ +#define SR_CMD_READ_GPIOS 0x1e +#define SR_GPIO_GPO0EN (1 << 0) /* GPIO0 Output enable */ +#define SR_GPIO_GPO_0 (1 << 1) /* GPIO0 Output value */ +#define SR_GPIO_GPO1EN (1 << 2) /* GPIO1 Output enable */ +#define SR_GPIO_GPO_1 (1 << 3) /* GPIO1 Output value */ +#define SR_GPIO_GPO2EN (1 << 4) /* GPIO2 Output enable */ +#define SR_GPIO_GPO_2 (1 << 5) /* GPIO2 Output value */ +#define SR_GPIO_RESERVED (1 << 6) /* Reserved */ +#define SR_GPIO_RSE (1 << 7) /* Reload serial EEPROM */ +/* command : GPIO Status Write Reg */ +#define SR_CMD_WRITE_GPIOS 0x1f +/* command : Eth PHY Power and Reset Control Reg */ +#define SR_CMD_SW_RESET 0x20 +#define SR_SWRESET_CLEAR 0x00 +#define SR_SWRESET_RR (1 << 0) +#define SR_SWRESET_RT (1 << 1) +#define SR_SWRESET_PRTE (1 << 2) +#define SR_SWRESET_PRL (1 << 3) +#define SR_SWRESET_BZ (1 << 4) +#define SR_SWRESET_IPRL (1 << 5) +#define SR_SWRESET_IPPD (1 << 6) +/* command : Software Interface Selection Status Read Reg */ +#define SR_CMD_SW_PHY_STATUS 0x21 +/* command : Software Interface Selection Status Write Reg */ +#define SR_CMD_SW_PHY_SELECT 0x22 +/* command : BULK in Buffer Size Reg */ +#define SR_CMD_BULKIN_SIZE 0x2A +/* command : LED_MUX Control Reg */ +#define SR_CMD_LED_MUX 0x70 +#define SR_LED_MUX_TX_ACTIVE (1 << 0) +#define SR_LED_MUX_RX_ACTIVE (1 << 1) +#define SR_LED_MUX_COLLISION (1 << 2) +#define SR_LED_MUX_DUP_COL (1 << 3) +#define SR_LED_MUX_DUP (1 << 4) +#define SR_LED_MUX_SPEED (1 << 5) +#define SR_LED_MUX_LINK_ACTIVE (1 << 6) +#define SR_LED_MUX_LINK (1 << 7) + +/* Register Access Flags */ +#define SR_REQ_RD_REG (USB_DIR_IN | USB_TYPE_VENDOR | USB_RECIP_DEVICE) +#define SR_REQ_WR_REG (USB_DIR_OUT | USB_TYPE_VENDOR | USB_RECIP_DEVICE) + +/* Multicast Filter Array size & Max Number */ +#define SR_MCAST_FILTER_SIZE 8 +#define SR_MAX_MCAST 64 + +/* IPG0/1/2 Default Value */ +#define SR9800_IPG0_DEFAULT 0x15 +#define SR9800_IPG1_DEFAULT 0x0c +#define SR9800_IPG2_DEFAULT 0x12 + +/* Medium Status Default Mode */ +#define SR9800_MEDIUM_DEFAULT \ + (SR_MEDIUM_FD | SR_MEDIUM_RFC | \ + SR_MEDIUM_TFC | SR_MEDIUM_PS | \ + SR_MEDIUM_AC | SR_MEDIUM_RE) + +/* RX Control Default Setting */ +#define SR_DEFAULT_RX_CTL \ + (SR_RX_CTL_SO | SR_RX_CTL_AB | SR_RX_CTL_RH1M) + +/* EEPROM Magic Number & EEPROM Size */ +#define SR_EEPROM_MAGIC 0xdeadbeef +#define SR9800_EEPROM_LEN 0xff + +/* SR9800 Driver Version and Driver Name */ +#define DRIVER_VERSION "11-Nov-2013" +#define DRIVER_NAME "CoreChips" +#define DRIVER_FLAG \ + (FLAG_ETHER | FLAG_FRAMING_AX | FLAG_LINK_INTR | FLAG_MULTI_PACKET) + +/* SR9800 BULKIN Buffer Size */ +#define SR9800_MAX_BULKIN_2K 0 +#define SR9800_MAX_BULKIN_4K 1 +#define SR9800_MAX_BULKIN_6K 2 +#define SR9800_MAX_BULKIN_8K 3 +#define SR9800_MAX_BULKIN_16K 4 +#define SR9800_MAX_BULKIN_20K 5 +#define SR9800_MAX_BULKIN_24K 6 +#define SR9800_MAX_BULKIN_32K 7 + +struct {unsigned short size, byte_cnt, threshold; } SR9800_BULKIN_SIZE[] = { + /* 2k */ + {2048, 0x8000, 0x8001}, + /* 4k */ + {4096, 0x8100, 0x8147}, + /* 6k */ + {6144, 0x8200, 0x81EB}, + /* 8k */ + {8192, 0x8300, 0x83D7}, + /* 16 */ + {16384, 0x8400, 0x851E}, + /* 20k */ + {20480, 0x8500, 0x8666}, + /* 24k */ + {24576, 0x8600, 0x87AE}, + /* 32k */ + {32768, 0x8700, 0x8A3D}, +}; + +/* This structure cannot exceed sizeof(unsigned long [5]) AKA 20 bytes */ +struct sr_data { + u8 multi_filter[SR_MCAST_FILTER_SIZE]; + u8 mac_addr[ETH_ALEN]; + u8 phymode; + u8 ledmode; + u8 eeprom_len; +}; + +struct sr9800_int_data { + __le16 res1; + u8 link; + __le16 res2; + u8 status; + __le16 res3; +} __packed; + +#endif /* _SR9800_H */ From 6b8790b5006b5ca3ee1c039c3c909833d7958716 Mon Sep 17 00:00:00 2001 From: dingtianhong Date: Mon, 10 Feb 2014 16:33:59 +0800 Subject: [PATCH 503/596] bonding: remove unwanted bond lock for enslave processing The bond enslave processing don't hold bond->lock anymore, so release an unlocked rw lock will cause warning message, remove the unwanted read_unlock(&bond->lock). Cc: Jay Vosburgh Cc: Veaceslav Falico Cc: Andy Gospodarek Signed-off-by: Ding Tianhong Acked-by: Veaceslav Falico Signed-off-by: David S. Miller --- drivers/net/bonding/bond_main.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 35f7e90aef72..71ba18efa15b 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -1510,7 +1510,6 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) slave_dev->npinfo = bond->dev->npinfo; if (slave_dev->npinfo) { if (slave_enable_netpoll(new_slave)) { - read_unlock(&bond->lock); pr_info("Error, %s: master_dev is using netpoll, " "but new slave device does not support netpoll.\n", bond_dev->name); From b6f52ae2f0d32387bde2b89883e3b64d88b9bfe8 Mon Sep 17 00:00:00 2001 From: Richard Yao Date: Sat, 8 Feb 2014 19:32:01 -0500 Subject: [PATCH 504/596] 9p/trans_virtio.c: Fix broken zero-copy on vmalloc() buffers The 9p-virtio transport does zero copy on things larger than 1024 bytes in size. It accomplishes this by returning the physical addresses of pages to the virtio-pci device. At present, the translation is usually a bit shift. That approach produces an invalid page address when we read/write to vmalloc buffers, such as those used for Linux kernel modules. Any attempt to load a Linux kernel module from 9p-virtio produces the following stack. [] p9_virtio_zc_request+0x45e/0x510 [] p9_client_zc_rpc.constprop.16+0xfd/0x4f0 [] p9_client_read+0x15d/0x240 [] v9fs_fid_readn+0x50/0xa0 [] v9fs_file_readn+0x10/0x20 [] v9fs_file_read+0x37/0x70 [] vfs_read+0x9b/0x160 [] kernel_read+0x41/0x60 [] copy_module_from_fd.isra.34+0xfb/0x180 Subsequently, QEMU will die printing: qemu-system-x86_64: virtio: trying to map MMIO memory This patch enables 9p-virtio to correctly handle this case. This not only enables us to load Linux kernel modules off virtfs, but also enables ZFS file-based vdevs on virtfs to be used without killing QEMU. Special thanks to both Avi Kivity and Alexander Graf for their interpretation of QEMU backtraces. Without their guidence, tracking down this bug would have taken much longer. Also, special thanks to Linus Torvalds for his insightful explanation of why this should use is_vmalloc_addr() instead of is_vmalloc_or_module_addr(): https://lkml.org/lkml/2014/2/8/272 Signed-off-by: Richard Yao Signed-off-by: David S. Miller --- net/9p/trans_virtio.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/net/9p/trans_virtio.c b/net/9p/trans_virtio.c index cd1e1ede73a4..ac2666c1d011 100644 --- a/net/9p/trans_virtio.c +++ b/net/9p/trans_virtio.c @@ -340,7 +340,10 @@ static int p9_get_mapped_pages(struct virtio_chan *chan, int count = nr_pages; while (nr_pages) { s = rest_of_page(data); - pages[index++] = kmap_to_page(data); + if (is_vmalloc_addr(data)) + pages[index++] = vmalloc_to_page(data); + else + pages[index++] = kmap_to_page(data); data += s; nr_pages--; } From 3e5ccc29f71b5dfdfb81dac8c19372af83923b7f Mon Sep 17 00:00:00 2001 From: John Greene Date: Mon, 10 Feb 2014 14:34:04 -0500 Subject: [PATCH 505/596] alx: add missing stats_lock spinlock init Trivial fix for init time stack trace occuring in alx_get_stats64 upon start up. Should have been part of commit adding the spinlock: f1b6b106 alx: add alx_get_stats64 operation Signed-off-by: John Greene Signed-off-by: David S. Miller --- drivers/net/ethernet/atheros/alx/main.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/atheros/alx/main.c b/drivers/net/ethernet/atheros/alx/main.c index e92ffd6e1c15..2e45f6ec1bf0 100644 --- a/drivers/net/ethernet/atheros/alx/main.c +++ b/drivers/net/ethernet/atheros/alx/main.c @@ -1292,6 +1292,7 @@ static int alx_probe(struct pci_dev *pdev, const struct pci_device_id *ent) alx = netdev_priv(netdev); spin_lock_init(&alx->hw.mdio_lock); spin_lock_init(&alx->irq_lock); + spin_lock_init(&alx->stats_lock); alx->dev = netdev; alx->hw.pdev = pdev; alx->msg_enable = NETIF_MSG_LINK | NETIF_MSG_HW | NETIF_MSG_IFUP | From 20e7c4e80dcd01dad5e6c8b32455228b8fe9c619 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 10 Feb 2014 11:42:35 -0800 Subject: [PATCH 506/596] 6lowpan: fix lockdep splats When a device ndo_start_xmit() calls again dev_queue_xmit(), lockdep can complain because dev_queue_xmit() is re-entered and the spinlocks protecting tx queues share a common lockdep class. Same issue was fixed for bonding/l2tp/ppp in commits 0daa2303028a6 ("[PATCH] bonding: lockdep annotation") 49ee49202b4ac ("bonding: set qdisc_tx_busylock to avoid LOCKDEP splat") 23d3b8bfb8eb2 ("net: qdisc busylock needs lockdep annotations ") 303c07db487be ("ppp: set qdisc_tx_busylock to avoid LOCKDEP splat ") Reported-by: Alexander Aring Signed-off-by: Eric Dumazet Tested-by: Alexander Aring Signed-off-by: David S. Miller --- net/ieee802154/6lowpan.c | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+) diff --git a/net/ieee802154/6lowpan.c b/net/ieee802154/6lowpan.c index 8bfb40153fe7..8edfea5da572 100644 --- a/net/ieee802154/6lowpan.c +++ b/net/ieee802154/6lowpan.c @@ -530,7 +530,27 @@ static struct header_ops lowpan_header_ops = { .create = lowpan_header_create, }; +static struct lock_class_key lowpan_tx_busylock; +static struct lock_class_key lowpan_netdev_xmit_lock_key; + +static void lowpan_set_lockdep_class_one(struct net_device *dev, + struct netdev_queue *txq, + void *_unused) +{ + lockdep_set_class(&txq->_xmit_lock, + &lowpan_netdev_xmit_lock_key); +} + + +static int lowpan_dev_init(struct net_device *dev) +{ + netdev_for_each_tx_queue(dev, lowpan_set_lockdep_class_one, NULL); + dev->qdisc_tx_busylock = &lowpan_tx_busylock; + return 0; +} + static const struct net_device_ops lowpan_netdev_ops = { + .ndo_init = lowpan_dev_init, .ndo_start_xmit = lowpan_xmit, .ndo_set_mac_address = lowpan_set_address, }; From cd15b048445d0a54f7147c35a86c5a16ef231554 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Tue, 11 Feb 2014 11:32:38 +1100 Subject: [PATCH 507/596] powerpc/powernv: Add iommu DMA bypass support for IODA2 This patch adds the support for to create a direct iommu "bypass" window on IODA2 bridges (such as Power8) allowing to bypass iommu page translation completely for 64-bit DMA capable devices, thus significantly improving DMA performances. Additionally, this adds a hook to the struct iommu_table so that the IOMMU API / VFIO can disable the bypass when external ownership is requested, since in that case, the device will be used by an environment such as userspace or a KVM guest which must not be allowed to bypass translations. Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/include/asm/dma-mapping.h | 1 + arch/powerpc/include/asm/iommu.h | 1 + arch/powerpc/kernel/dma.c | 10 ++- arch/powerpc/kernel/iommu.c | 12 ++++ arch/powerpc/platforms/powernv/pci-ioda.c | 84 +++++++++++++++++++++++ arch/powerpc/platforms/powernv/pci.c | 10 +++ arch/powerpc/platforms/powernv/pci.h | 6 +- arch/powerpc/platforms/powernv/powernv.h | 8 +++ arch/powerpc/platforms/powernv/setup.c | 9 +++ 9 files changed, 137 insertions(+), 4 deletions(-) diff --git a/arch/powerpc/include/asm/dma-mapping.h b/arch/powerpc/include/asm/dma-mapping.h index e27e9ad6818e..150866b2a3fe 100644 --- a/arch/powerpc/include/asm/dma-mapping.h +++ b/arch/powerpc/include/asm/dma-mapping.h @@ -134,6 +134,7 @@ static inline int dma_supported(struct device *dev, u64 mask) } extern int dma_set_mask(struct device *dev, u64 dma_mask); +extern int __dma_set_mask(struct device *dev, u64 dma_mask); #define dma_alloc_coherent(d,s,h,f) dma_alloc_attrs(d,s,h,f,NULL) diff --git a/arch/powerpc/include/asm/iommu.h b/arch/powerpc/include/asm/iommu.h index f7a8036579b5..42632c7a2a4e 100644 --- a/arch/powerpc/include/asm/iommu.h +++ b/arch/powerpc/include/asm/iommu.h @@ -77,6 +77,7 @@ struct iommu_table { #ifdef CONFIG_IOMMU_API struct iommu_group *it_group; #endif + void (*set_bypass)(struct iommu_table *tbl, bool enable); }; /* Pure 2^n version of get_order */ diff --git a/arch/powerpc/kernel/dma.c b/arch/powerpc/kernel/dma.c index 8032b97ccdcb..ee78f6e49d64 100644 --- a/arch/powerpc/kernel/dma.c +++ b/arch/powerpc/kernel/dma.c @@ -191,12 +191,10 @@ EXPORT_SYMBOL(dma_direct_ops); #define PREALLOC_DMA_DEBUG_ENTRIES (1 << 16) -int dma_set_mask(struct device *dev, u64 dma_mask) +int __dma_set_mask(struct device *dev, u64 dma_mask) { struct dma_map_ops *dma_ops = get_dma_ops(dev); - if (ppc_md.dma_set_mask) - return ppc_md.dma_set_mask(dev, dma_mask); if ((dma_ops != NULL) && (dma_ops->set_dma_mask != NULL)) return dma_ops->set_dma_mask(dev, dma_mask); if (!dev->dma_mask || !dma_supported(dev, dma_mask)) @@ -204,6 +202,12 @@ int dma_set_mask(struct device *dev, u64 dma_mask) *dev->dma_mask = dma_mask; return 0; } +int dma_set_mask(struct device *dev, u64 dma_mask) +{ + if (ppc_md.dma_set_mask) + return ppc_md.dma_set_mask(dev, dma_mask); + return __dma_set_mask(dev, dma_mask); +} EXPORT_SYMBOL(dma_set_mask); u64 dma_get_required_mask(struct device *dev) diff --git a/arch/powerpc/kernel/iommu.c b/arch/powerpc/kernel/iommu.c index d773dd440a45..88e3ec6e1d96 100644 --- a/arch/powerpc/kernel/iommu.c +++ b/arch/powerpc/kernel/iommu.c @@ -1088,6 +1088,14 @@ int iommu_take_ownership(struct iommu_table *tbl) memset(tbl->it_map, 0xff, sz); iommu_clear_tces_and_put_pages(tbl, tbl->it_offset, tbl->it_size); + /* + * Disable iommu bypass, otherwise the user can DMA to all of + * our physical memory via the bypass window instead of just + * the pages that has been explicitly mapped into the iommu + */ + if (tbl->set_bypass) + tbl->set_bypass(tbl, false); + return 0; } EXPORT_SYMBOL_GPL(iommu_take_ownership); @@ -1102,6 +1110,10 @@ void iommu_release_ownership(struct iommu_table *tbl) /* Restore bit#0 set by iommu_init_table() */ if (tbl->it_offset == 0) set_bit(0, tbl->it_map); + + /* The kernel owns the device now, we can restore the iommu bypass */ + if (tbl->set_bypass) + tbl->set_bypass(tbl, true); } EXPORT_SYMBOL_GPL(iommu_release_ownership); diff --git a/arch/powerpc/platforms/powernv/pci-ioda.c b/arch/powerpc/platforms/powernv/pci-ioda.c index 7d6dcc6d5fa9..3b2b4fb3585b 100644 --- a/arch/powerpc/platforms/powernv/pci-ioda.c +++ b/arch/powerpc/platforms/powernv/pci-ioda.c @@ -21,6 +21,7 @@ #include #include #include +#include #include #include @@ -460,9 +461,39 @@ static void pnv_pci_ioda_dma_dev_setup(struct pnv_phb *phb, struct pci_dev *pdev return; pe = &phb->ioda.pe_array[pdn->pe_number]; + WARN_ON(get_dma_ops(&pdev->dev) != &dma_iommu_ops); set_iommu_table_base_and_group(&pdev->dev, &pe->tce32_table); } +static int pnv_pci_ioda_dma_set_mask(struct pnv_phb *phb, + struct pci_dev *pdev, u64 dma_mask) +{ + struct pci_dn *pdn = pci_get_pdn(pdev); + struct pnv_ioda_pe *pe; + uint64_t top; + bool bypass = false; + + if (WARN_ON(!pdn || pdn->pe_number == IODA_INVALID_PE)) + return -ENODEV;; + + pe = &phb->ioda.pe_array[pdn->pe_number]; + if (pe->tce_bypass_enabled) { + top = pe->tce_bypass_base + memblock_end_of_DRAM() - 1; + bypass = (dma_mask >= top); + } + + if (bypass) { + dev_info(&pdev->dev, "Using 64-bit DMA iommu bypass\n"); + set_dma_ops(&pdev->dev, &dma_direct_ops); + set_dma_offset(&pdev->dev, pe->tce_bypass_base); + } else { + dev_info(&pdev->dev, "Using 32-bit DMA via iommu\n"); + set_dma_ops(&pdev->dev, &dma_iommu_ops); + set_iommu_table_base(&pdev->dev, &pe->tce32_table); + } + return 0; +} + static void pnv_ioda_setup_bus_dma(struct pnv_ioda_pe *pe, struct pci_bus *bus) { struct pci_dev *dev; @@ -657,6 +688,56 @@ static void pnv_pci_ioda_setup_dma_pe(struct pnv_phb *phb, __free_pages(tce_mem, get_order(TCE32_TABLE_SIZE * segs)); } +static void pnv_pci_ioda2_set_bypass(struct iommu_table *tbl, bool enable) +{ + struct pnv_ioda_pe *pe = container_of(tbl, struct pnv_ioda_pe, + tce32_table); + uint16_t window_id = (pe->pe_number << 1 ) + 1; + int64_t rc; + + pe_info(pe, "%sabling 64-bit DMA bypass\n", enable ? "En" : "Dis"); + if (enable) { + phys_addr_t top = memblock_end_of_DRAM(); + + top = roundup_pow_of_two(top); + rc = opal_pci_map_pe_dma_window_real(pe->phb->opal_id, + pe->pe_number, + window_id, + pe->tce_bypass_base, + top); + } else { + rc = opal_pci_map_pe_dma_window_real(pe->phb->opal_id, + pe->pe_number, + window_id, + pe->tce_bypass_base, + 0); + + /* + * We might want to reset the DMA ops of all devices on + * this PE. However in theory, that shouldn't be necessary + * as this is used for VFIO/KVM pass-through and the device + * hasn't yet been returned to its kernel driver + */ + } + if (rc) + pe_err(pe, "OPAL error %lld configuring bypass window\n", rc); + else + pe->tce_bypass_enabled = enable; +} + +static void pnv_pci_ioda2_setup_bypass_pe(struct pnv_phb *phb, + struct pnv_ioda_pe *pe) +{ + /* TVE #1 is selected by PCI address bit 59 */ + pe->tce_bypass_base = 1ull << 59; + + /* Install set_bypass callback for VFIO */ + pe->tce32_table.set_bypass = pnv_pci_ioda2_set_bypass; + + /* Enable bypass by default */ + pnv_pci_ioda2_set_bypass(&pe->tce32_table, true); +} + static void pnv_pci_ioda2_setup_dma_pe(struct pnv_phb *phb, struct pnv_ioda_pe *pe) { @@ -727,6 +808,8 @@ static void pnv_pci_ioda2_setup_dma_pe(struct pnv_phb *phb, else pnv_ioda_setup_bus_dma(pe, pe->pbus); + /* Also create a bypass window */ + pnv_pci_ioda2_setup_bypass_pe(phb, pe); return; fail: if (pe->tce32_seg >= 0) @@ -1286,6 +1369,7 @@ void __init pnv_pci_init_ioda_phb(struct device_node *np, /* Setup TCEs */ phb->dma_dev_setup = pnv_pci_ioda_dma_dev_setup; + phb->dma_set_mask = pnv_pci_ioda_dma_set_mask; /* Setup shutdown function for kexec */ phb->shutdown = pnv_pci_ioda_shutdown; diff --git a/arch/powerpc/platforms/powernv/pci.c b/arch/powerpc/platforms/powernv/pci.c index b555ebc57ef5..95633d79ef5d 100644 --- a/arch/powerpc/platforms/powernv/pci.c +++ b/arch/powerpc/platforms/powernv/pci.c @@ -634,6 +634,16 @@ static void pnv_pci_dma_dev_setup(struct pci_dev *pdev) pnv_pci_dma_fallback_setup(hose, pdev); } +int pnv_pci_dma_set_mask(struct pci_dev *pdev, u64 dma_mask) +{ + struct pci_controller *hose = pci_bus_to_host(pdev->bus); + struct pnv_phb *phb = hose->private_data; + + if (phb && phb->dma_set_mask) + return phb->dma_set_mask(phb, pdev, dma_mask); + return __dma_set_mask(&pdev->dev, dma_mask); +} + void pnv_pci_shutdown(void) { struct pci_controller *hose; diff --git a/arch/powerpc/platforms/powernv/pci.h b/arch/powerpc/platforms/powernv/pci.h index 13f1942a9a5f..cde169442775 100644 --- a/arch/powerpc/platforms/powernv/pci.h +++ b/arch/powerpc/platforms/powernv/pci.h @@ -54,7 +54,9 @@ struct pnv_ioda_pe { struct iommu_table tce32_table; phys_addr_t tce_inval_reg_phys; - /* XXX TODO: Add support for additional 64-bit iommus */ + /* 64-bit TCE bypass region */ + bool tce_bypass_enabled; + uint64_t tce_bypass_base; /* MSIs. MVE index is identical for for 32 and 64 bit MSI * and -1 if not supported. (It's actually identical to the @@ -113,6 +115,8 @@ struct pnv_phb { unsigned int hwirq, unsigned int virq, unsigned int is_64, struct msi_msg *msg); void (*dma_dev_setup)(struct pnv_phb *phb, struct pci_dev *pdev); + int (*dma_set_mask)(struct pnv_phb *phb, struct pci_dev *pdev, + u64 dma_mask); void (*fixup_phb)(struct pci_controller *hose); u32 (*bdfn_to_pe)(struct pnv_phb *phb, struct pci_bus *bus, u32 devfn); void (*shutdown)(struct pnv_phb *phb); diff --git a/arch/powerpc/platforms/powernv/powernv.h b/arch/powerpc/platforms/powernv/powernv.h index de6819be1f95..0051e108ef0f 100644 --- a/arch/powerpc/platforms/powernv/powernv.h +++ b/arch/powerpc/platforms/powernv/powernv.h @@ -7,12 +7,20 @@ extern void pnv_smp_init(void); static inline void pnv_smp_init(void) { } #endif +struct pci_dev; + #ifdef CONFIG_PCI extern void pnv_pci_init(void); extern void pnv_pci_shutdown(void); +extern int pnv_pci_dma_set_mask(struct pci_dev *pdev, u64 dma_mask); #else static inline void pnv_pci_init(void) { } static inline void pnv_pci_shutdown(void) { } + +static inline int pnv_pci_dma_set_mask(struct pci_dev *pdev, u64 dma_mask) +{ + return -ENODEV; +} #endif extern void pnv_lpc_init(void); diff --git a/arch/powerpc/platforms/powernv/setup.c b/arch/powerpc/platforms/powernv/setup.c index 21166f65c97c..110f4fbd319f 100644 --- a/arch/powerpc/platforms/powernv/setup.c +++ b/arch/powerpc/platforms/powernv/setup.c @@ -27,6 +27,7 @@ #include #include #include +#include #include #include @@ -141,6 +142,13 @@ static void pnv_progress(char *s, unsigned short hex) { } +static int pnv_dma_set_mask(struct device *dev, u64 dma_mask) +{ + if (dev_is_pci(dev)) + return pnv_pci_dma_set_mask(to_pci_dev(dev), dma_mask); + return __dma_set_mask(dev, dma_mask); +} + static void pnv_shutdown(void) { /* Let the PCI code clear up IODA tables */ @@ -238,6 +246,7 @@ define_machine(powernv) { .machine_shutdown = pnv_shutdown, .power_save = powernv_idle, .calibrate_decr = generic_calibrate_decr, + .dma_set_mask = pnv_dma_set_mask, #ifdef CONFIG_KEXEC .kexec_cpu_down = pnv_kexec_cpu_down, #endif From 2c45aada341121438affc4cb8d5b4cfaa2813d3d Mon Sep 17 00:00:00 2001 From: Paul Gortmaker Date: Mon, 10 Feb 2014 13:39:53 -0500 Subject: [PATCH 508/596] genirq: Add missing irq_to_desc export for CONFIG_SPARSE_IRQ=n In allmodconfig builds for sparc and any other arch which does not set CONFIG_SPARSE_IRQ, the following will be seen at modpost: CC [M] lib/cpu-notifier-error-inject.o CC [M] lib/pm-notifier-error-inject.o ERROR: "irq_to_desc" [drivers/gpio/gpio-mcp23s08.ko] undefined! make[2]: *** [__modpost] Error 1 This happens because commit 3911ff30f5 ("genirq: export handle_edge_irq() and irq_to_desc()") added one export for it, but there were actually two instances of it, in an if/else clause for CONFIG_SPARSE_IRQ. Add the second one. Signed-off-by: Paul Gortmaker Cc: Jiri Kosina Cc: stable@vger.kernel.org # 3.4+ Link: http://lkml.kernel.org/r/1392057610-11514-1-git-send-email-paul.gortmaker@windriver.com Signed-off-by: Thomas Gleixner --- kernel/irq/irqdesc.c | 1 + 1 file changed, 1 insertion(+) diff --git a/kernel/irq/irqdesc.c b/kernel/irq/irqdesc.c index 192a302d6cfd..8ab8e9390297 100644 --- a/kernel/irq/irqdesc.c +++ b/kernel/irq/irqdesc.c @@ -274,6 +274,7 @@ struct irq_desc *irq_to_desc(unsigned int irq) { return (irq < NR_IRQS) ? irq_desc + irq : NULL; } +EXPORT_SYMBOL(irq_to_desc); static void free_desc(unsigned int irq) { From a2d213dd77da4710bcb75f8efe85a32e3db8b39b Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Fri, 7 Feb 2014 16:34:05 +0100 Subject: [PATCH 509/596] drm/i915: Fix intel_pipe_to_cpu_transcoder for UMS MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit We don't have all the drm_crtc&co hanging around in that case. This regression has been introduced in commit 391f75e2bf13f105d9e4a120736ccdd8e3bc638b Author: Ville Syrjälä Date: Wed Sep 25 19:55:26 2013 +0300 drm/i915: Fix pre-CTG vblank counter Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=69521 Cc: stable@vger.kernel.org (for 3.13 only) Cc: Ville Syrjälä Reviewed-by: Paulo Zanoni Signed-off-by: Daniel Vetter --- drivers/gpu/drm/i915/i915_irq.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c index 17d8fcb1b6f7..9fec71175571 100644 --- a/drivers/gpu/drm/i915/i915_irq.c +++ b/drivers/gpu/drm/i915/i915_irq.c @@ -567,8 +567,7 @@ static u32 i915_get_vblank_counter(struct drm_device *dev, int pipe) vbl_start = mode->crtc_vblank_start * mode->crtc_htotal; } else { - enum transcoder cpu_transcoder = - intel_pipe_to_cpu_transcoder(dev_priv, pipe); + enum transcoder cpu_transcoder = (enum transcoder) pipe; u32 htotal; htotal = ((I915_READ(HTOTAL(cpu_transcoder)) >> 16) & 0x1fff) + 1; From 1d2cb9a54abc6e1d239f28f07661366d5662a94a Mon Sep 17 00:00:00 2001 From: Mika Kuoppala Date: Fri, 7 Feb 2014 17:40:50 +0200 Subject: [PATCH 510/596] drm/i915: Pair va_copy with va_end in i915_error_vprintf Each invocation of va_copy() must be matched by a corresponding invocation of va_end() in the same function. This regression has been introduced in commit e29bb4ebbf000ff9ac081d29784a3331618f012e Author: Chris Wilson Date: Fri Sep 20 10:20:59 2013 +0100 drm/i915: Use a temporary va_list for two-pass string handling Signed-off-by: Mika Kuoppala Reviewed-by: Chris Wilson Cc: stable@vger.kernel.org Signed-off-by: Daniel Vetter --- drivers/gpu/drm/i915/i915_gpu_error.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c index d7fd2fd2f0a5..990cf8f43efd 100644 --- a/drivers/gpu/drm/i915/i915_gpu_error.c +++ b/drivers/gpu/drm/i915/i915_gpu_error.c @@ -146,7 +146,10 @@ static void i915_error_vprintf(struct drm_i915_error_state_buf *e, va_list tmp; va_copy(tmp, args); - if (!__i915_error_seek(e, vsnprintf(NULL, 0, f, tmp))) + len = vsnprintf(NULL, 0, f, tmp); + va_end(tmp); + + if (!__i915_error_seek(e, len)) return; } From eec77da2744b1265653ab99af91c3b26c49501e9 Mon Sep 17 00:00:00 2001 From: Tomi Valkeinen Date: Mon, 27 Jan 2014 11:29:53 +0200 Subject: [PATCH 511/596] OMAPDSS: DISPC: decimation rounding fix The driver uses DIV_ROUND_UP when calculating decimated width & height. For example, when decimating with 3, the width is calculated as: width = DIV_ROUND_UP(width, decim_x); This yields bad results for some values. For example, 800/3=266.666..., which is rounded to 267. When the input width is set to 267, and pixel increment is set to 3, this causes the dispc to read a line of 801 pixels, i.e. it reads a wrong pixel at the end of the line. Even more pressing, the above rounding causes a BUG() in pixinc(), as the value of 801 is used to calculate row increment, leading to a bad value being passed to pixinc(). This patch fixes the decimation by removing the DIV_ROUND_UP()s when calculating width and height for decimation. Signed-off-by: Tomi Valkeinen --- drivers/video/omap2/dss/dispc.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/drivers/video/omap2/dss/dispc.c b/drivers/video/omap2/dss/dispc.c index bbeb8dd7f108..77d6221618f4 100644 --- a/drivers/video/omap2/dss/dispc.c +++ b/drivers/video/omap2/dss/dispc.c @@ -2160,8 +2160,8 @@ static int dispc_ovl_calc_scaling_24xx(unsigned long pclk, unsigned long lclk, *five_taps = false; do { - in_height = DIV_ROUND_UP(height, *decim_y); - in_width = DIV_ROUND_UP(width, *decim_x); + in_height = height / *decim_y; + in_width = width / *decim_x; *core_clk = dispc.feat->calc_core_clk(pclk, in_width, in_height, out_width, out_height, mem_to_mem); error = (in_width > maxsinglelinewidth || !*core_clk || @@ -2199,8 +2199,8 @@ static int dispc_ovl_calc_scaling_34xx(unsigned long pclk, unsigned long lclk, dss_feat_get_param_max(FEAT_PARAM_LINEWIDTH); do { - in_height = DIV_ROUND_UP(height, *decim_y); - in_width = DIV_ROUND_UP(width, *decim_x); + in_height = height / *decim_y; + in_width = width / *decim_x; *five_taps = in_height > out_height; if (in_width > maxsinglelinewidth) @@ -2268,7 +2268,7 @@ static int dispc_ovl_calc_scaling_44xx(unsigned long pclk, unsigned long lclk, { u16 in_width, in_width_max; int decim_x_min = *decim_x; - u16 in_height = DIV_ROUND_UP(height, *decim_y); + u16 in_height = height / *decim_y; const int maxsinglelinewidth = dss_feat_get_param_max(FEAT_PARAM_LINEWIDTH); const int maxdownscale = dss_feat_get_param_max(FEAT_PARAM_DOWNSCALE); @@ -2287,7 +2287,7 @@ static int dispc_ovl_calc_scaling_44xx(unsigned long pclk, unsigned long lclk, return -EINVAL; do { - in_width = DIV_ROUND_UP(width, *decim_x); + in_width = width / *decim_x; } while (*decim_x <= *x_predecim && in_width > maxsinglelinewidth && ++*decim_x); @@ -2466,8 +2466,8 @@ static int dispc_ovl_setup_common(enum omap_plane plane, if (r) return r; - in_width = DIV_ROUND_UP(in_width, x_predecim); - in_height = DIV_ROUND_UP(in_height, y_predecim); + in_width = in_width / x_predecim; + in_height = in_height / y_predecim; if (color_mode == OMAP_DSS_COLOR_YUV2 || color_mode == OMAP_DSS_COLOR_UYVY || From c56812fc855e29f44aeb5e166d2b512dab0a3e26 Mon Sep 17 00:00:00 2001 From: Tomi Valkeinen Date: Tue, 28 Jan 2014 08:50:47 +0200 Subject: [PATCH 512/596] OMAPDSS: fix fck field types 'fck' field in dpi and sdi clock calculation struct is 'unsigned long long', even though it should be 'unsigned long'. This hasn't caused any issues so far. Fix the field's type. Signed-off-by: Tomi Valkeinen --- drivers/video/omap2/dss/dpi.c | 2 +- drivers/video/omap2/dss/sdi.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/video/omap2/dss/dpi.c b/drivers/video/omap2/dss/dpi.c index 7411f2674e16..23ef21ffc2c4 100644 --- a/drivers/video/omap2/dss/dpi.c +++ b/drivers/video/omap2/dss/dpi.c @@ -117,7 +117,7 @@ struct dpi_clk_calc_ctx { /* outputs */ struct dsi_clock_info dsi_cinfo; - unsigned long long fck; + unsigned long fck; struct dispc_clock_info dispc_cinfo; }; diff --git a/drivers/video/omap2/dss/sdi.c b/drivers/video/omap2/dss/sdi.c index efb9ee9e3c96..ba806c9e7f54 100644 --- a/drivers/video/omap2/dss/sdi.c +++ b/drivers/video/omap2/dss/sdi.c @@ -46,7 +46,7 @@ static struct { struct sdi_clk_calc_ctx { unsigned long pck_min, pck_max; - unsigned long long fck; + unsigned long fck; struct dispc_clock_info dispc_cinfo; }; From 97253eeeb792d61ed22d397cfa236bd0311e4caf Mon Sep 17 00:00:00 2001 From: David Vrabel Date: Wed, 5 Feb 2014 14:13:10 +0000 Subject: [PATCH 513/596] xen/events: bind all new interdomain events to VCPU0 Commit fc087e10734a4d3e40693fc099461ec1270b3fff (xen/events: remove unnecessary init_evtchn_cpu_bindings()) causes a regression. The kernel-side VCPU binding was not being correctly set for newly allocated or bound interdomain events. In ARM guests where 2-level events were used, this would result in no interdomain events being handled because the kernel-side VCPU masks would all be clear. x86 guests would work because the irq affinity was set during irq setup and this would set the correct kernel-side VCPU binding. Fix this by properly initializing the kernel-side VCPU binding in bind_evtchn_to_irq(). Reported-and-tested-by: Julien Grall Signed-off-by: David Vrabel Signed-off-by: Konrad Rzeszutek Wilk --- drivers/xen/events/events_base.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/xen/events/events_base.c b/drivers/xen/events/events_base.c index 4672e003c0ad..f4a9e3311297 100644 --- a/drivers/xen/events/events_base.c +++ b/drivers/xen/events/events_base.c @@ -862,6 +862,8 @@ int bind_evtchn_to_irq(unsigned int evtchn) irq = ret; goto out; } + /* New interdomain events are bound to VCPU 0. */ + bind_evtchn_to_cpu(evtchn, 0); } else { struct irq_info *info = info_for_irq(irq); WARN_ON(info == NULL || info->type != IRQT_EVTCHN); From 564eb714f5f09ac733c26860d5f0831f213fbdf1 Mon Sep 17 00:00:00 2001 From: David Vrabel Date: Mon, 10 Feb 2014 13:54:02 +0000 Subject: [PATCH 514/596] xen: install xen/gntdev.h and xen/gntalloc.h xen/gntdev.h and xen/gntalloc.h both provide userspace ABIs so they should be installed. CC: stable@vger.kernel.org Signed-off-by: David Vrabel Signed-off-by: Konrad Rzeszutek Wilk --- include/uapi/xen/Kbuild | 2 ++ include/{ => uapi}/xen/gntalloc.h | 0 include/{ => uapi}/xen/gntdev.h | 0 3 files changed, 2 insertions(+) rename include/{ => uapi}/xen/gntalloc.h (100%) rename include/{ => uapi}/xen/gntdev.h (100%) diff --git a/include/uapi/xen/Kbuild b/include/uapi/xen/Kbuild index 61257cb14653..5c459628e8c7 100644 --- a/include/uapi/xen/Kbuild +++ b/include/uapi/xen/Kbuild @@ -1,3 +1,5 @@ # UAPI Header export list header-y += evtchn.h +header-y += gntalloc.h +header-y += gntdev.h header-y += privcmd.h diff --git a/include/xen/gntalloc.h b/include/uapi/xen/gntalloc.h similarity index 100% rename from include/xen/gntalloc.h rename to include/uapi/xen/gntalloc.h diff --git a/include/xen/gntdev.h b/include/uapi/xen/gntdev.h similarity index 100% rename from include/xen/gntdev.h rename to include/uapi/xen/gntdev.h From d8320b2d2e7d1cda8216d496c7c685c5fdf74ff0 Mon Sep 17 00:00:00 2001 From: Paul Bolle Date: Sat, 8 Feb 2014 23:35:43 +0100 Subject: [PATCH 515/596] ia64/xen: Remove Xen support for ia64 even more Commit d52eefb47d4e ("ia64/xen: Remove Xen support for ia64") removed the Kconfig symbol XEN_XENCOMM. But it didn't remove the code depending on that symbol. Remove that code now. Signed-off-by: Paul Bolle Acked-by: David Vrabel Signed-off-by: Konrad Rzeszutek Wilk --- drivers/xen/Makefile | 1 - drivers/xen/xencomm.c | 219 -------------------------------- include/xen/interface/xencomm.h | 41 ------ include/xen/xencomm.h | 77 ----------- 4 files changed, 338 deletions(-) delete mode 100644 drivers/xen/xencomm.c delete mode 100644 include/xen/interface/xencomm.h delete mode 100644 include/xen/xencomm.h diff --git a/drivers/xen/Makefile b/drivers/xen/Makefile index d75c811bfa56..45e00afa7f2d 100644 --- a/drivers/xen/Makefile +++ b/drivers/xen/Makefile @@ -16,7 +16,6 @@ xen-pad-$(CONFIG_X86) += xen-acpi-pad.o dom0-$(CONFIG_X86) += pcpu.o obj-$(CONFIG_XEN_DOM0) += $(dom0-y) obj-$(CONFIG_BLOCK) += biomerge.o -obj-$(CONFIG_XEN_XENCOMM) += xencomm.o obj-$(CONFIG_XEN_BALLOON) += xen-balloon.o obj-$(CONFIG_XEN_SELFBALLOONING) += xen-selfballoon.o obj-$(CONFIG_XEN_DEV_EVTCHN) += xen-evtchn.o diff --git a/drivers/xen/xencomm.c b/drivers/xen/xencomm.c deleted file mode 100644 index 4793fc594549..000000000000 --- a/drivers/xen/xencomm.c +++ /dev/null @@ -1,219 +0,0 @@ -/* - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA - * - * Copyright (C) IBM Corp. 2006 - * - * Authors: Hollis Blanchard - */ - -#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt - -#include -#include -#include -#include -#include -#include /* for xencomm_is_phys_contiguous() */ - -static int xencomm_init(struct xencomm_desc *desc, - void *buffer, unsigned long bytes) -{ - unsigned long recorded = 0; - int i = 0; - - while ((recorded < bytes) && (i < desc->nr_addrs)) { - unsigned long vaddr = (unsigned long)buffer + recorded; - unsigned long paddr; - int offset; - int chunksz; - - offset = vaddr % PAGE_SIZE; /* handle partial pages */ - chunksz = min(PAGE_SIZE - offset, bytes - recorded); - - paddr = xencomm_vtop(vaddr); - if (paddr == ~0UL) { - printk(KERN_DEBUG "%s: couldn't translate vaddr %lx\n", - __func__, vaddr); - return -EINVAL; - } - - desc->address[i++] = paddr; - recorded += chunksz; - } - - if (recorded < bytes) { - printk(KERN_DEBUG - "%s: could only translate %ld of %ld bytes\n", - __func__, recorded, bytes); - return -ENOSPC; - } - - /* mark remaining addresses invalid (just for safety) */ - while (i < desc->nr_addrs) - desc->address[i++] = XENCOMM_INVALID; - - desc->magic = XENCOMM_MAGIC; - - return 0; -} - -static struct xencomm_desc *xencomm_alloc(gfp_t gfp_mask, - void *buffer, unsigned long bytes) -{ - struct xencomm_desc *desc; - unsigned long buffer_ulong = (unsigned long)buffer; - unsigned long start = buffer_ulong & PAGE_MASK; - unsigned long end = (buffer_ulong + bytes) | ~PAGE_MASK; - unsigned long nr_addrs = (end - start + 1) >> PAGE_SHIFT; - unsigned long size = sizeof(*desc) + - sizeof(desc->address[0]) * nr_addrs; - - /* - * slab allocator returns at least sizeof(void*) aligned pointer. - * When sizeof(*desc) > sizeof(void*), struct xencomm_desc might - * cross page boundary. - */ - if (sizeof(*desc) > sizeof(void *)) { - unsigned long order = get_order(size); - desc = (struct xencomm_desc *)__get_free_pages(gfp_mask, - order); - if (desc == NULL) - return NULL; - - desc->nr_addrs = - ((PAGE_SIZE << order) - sizeof(struct xencomm_desc)) / - sizeof(*desc->address); - } else { - desc = kmalloc(size, gfp_mask); - if (desc == NULL) - return NULL; - - desc->nr_addrs = nr_addrs; - } - return desc; -} - -void xencomm_free(struct xencomm_handle *desc) -{ - if (desc && !((ulong)desc & XENCOMM_INLINE_FLAG)) { - struct xencomm_desc *desc__ = (struct xencomm_desc *)desc; - if (sizeof(*desc__) > sizeof(void *)) { - unsigned long size = sizeof(*desc__) + - sizeof(desc__->address[0]) * desc__->nr_addrs; - unsigned long order = get_order(size); - free_pages((unsigned long)__va(desc), order); - } else - kfree(__va(desc)); - } -} - -static int xencomm_create(void *buffer, unsigned long bytes, - struct xencomm_desc **ret, gfp_t gfp_mask) -{ - struct xencomm_desc *desc; - int rc; - - pr_debug("%s: %p[%ld]\n", __func__, buffer, bytes); - - if (bytes == 0) { - /* don't create a descriptor; Xen recognizes NULL. */ - BUG_ON(buffer != NULL); - *ret = NULL; - return 0; - } - - BUG_ON(buffer == NULL); /* 'bytes' is non-zero */ - - desc = xencomm_alloc(gfp_mask, buffer, bytes); - if (!desc) { - printk(KERN_DEBUG "%s failure\n", "xencomm_alloc"); - return -ENOMEM; - } - - rc = xencomm_init(desc, buffer, bytes); - if (rc) { - printk(KERN_DEBUG "%s failure: %d\n", "xencomm_init", rc); - xencomm_free((struct xencomm_handle *)__pa(desc)); - return rc; - } - - *ret = desc; - return 0; -} - -static struct xencomm_handle *xencomm_create_inline(void *ptr) -{ - unsigned long paddr; - - BUG_ON(!xencomm_is_phys_contiguous((unsigned long)ptr)); - - paddr = (unsigned long)xencomm_pa(ptr); - BUG_ON(paddr & XENCOMM_INLINE_FLAG); - return (struct xencomm_handle *)(paddr | XENCOMM_INLINE_FLAG); -} - -/* "mini" routine, for stack-based communications: */ -static int xencomm_create_mini(void *buffer, - unsigned long bytes, struct xencomm_mini *xc_desc, - struct xencomm_desc **ret) -{ - int rc = 0; - struct xencomm_desc *desc; - BUG_ON(((unsigned long)xc_desc) % sizeof(*xc_desc) != 0); - - desc = (void *)xc_desc; - - desc->nr_addrs = XENCOMM_MINI_ADDRS; - - rc = xencomm_init(desc, buffer, bytes); - if (!rc) - *ret = desc; - - return rc; -} - -struct xencomm_handle *xencomm_map(void *ptr, unsigned long bytes) -{ - int rc; - struct xencomm_desc *desc; - - if (xencomm_is_phys_contiguous((unsigned long)ptr)) - return xencomm_create_inline(ptr); - - rc = xencomm_create(ptr, bytes, &desc, GFP_KERNEL); - - if (rc || desc == NULL) - return NULL; - - return xencomm_pa(desc); -} - -struct xencomm_handle *__xencomm_map_no_alloc(void *ptr, unsigned long bytes, - struct xencomm_mini *xc_desc) -{ - int rc; - struct xencomm_desc *desc = NULL; - - if (xencomm_is_phys_contiguous((unsigned long)ptr)) - return xencomm_create_inline(ptr); - - rc = xencomm_create_mini(ptr, bytes, xc_desc, - &desc); - - if (rc) - return NULL; - - return xencomm_pa(desc); -} diff --git a/include/xen/interface/xencomm.h b/include/xen/interface/xencomm.h deleted file mode 100644 index ac45e0712afa..000000000000 --- a/include/xen/interface/xencomm.h +++ /dev/null @@ -1,41 +0,0 @@ -/* - * Permission is hereby granted, free of charge, to any person obtaining a copy - * of this software and associated documentation files (the "Software"), to - * deal in the Software without restriction, including without limitation the - * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or - * sell copies of the Software, and to permit persons to whom the Software is - * furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in - * all copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE - * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER - * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING - * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER - * DEALINGS IN THE SOFTWARE. - * - * Copyright (C) IBM Corp. 2006 - */ - -#ifndef _XEN_XENCOMM_H_ -#define _XEN_XENCOMM_H_ - -/* A xencomm descriptor is a scatter/gather list containing physical - * addresses corresponding to a virtually contiguous memory area. The - * hypervisor translates these physical addresses to machine addresses to copy - * to and from the virtually contiguous area. - */ - -#define XENCOMM_MAGIC 0x58434F4D /* 'XCOM' */ -#define XENCOMM_INVALID (~0UL) - -struct xencomm_desc { - uint32_t magic; - uint32_t nr_addrs; /* the number of entries in address[] */ - uint64_t address[0]; -}; - -#endif /* _XEN_XENCOMM_H_ */ diff --git a/include/xen/xencomm.h b/include/xen/xencomm.h deleted file mode 100644 index e43b039be112..000000000000 --- a/include/xen/xencomm.h +++ /dev/null @@ -1,77 +0,0 @@ -/* - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA - * - * Copyright (C) IBM Corp. 2006 - * - * Authors: Hollis Blanchard - * Jerone Young - */ - -#ifndef _LINUX_XENCOMM_H_ -#define _LINUX_XENCOMM_H_ - -#include - -#define XENCOMM_MINI_ADDRS 3 -struct xencomm_mini { - struct xencomm_desc _desc; - uint64_t address[XENCOMM_MINI_ADDRS]; -}; - -/* To avoid additionnal virt to phys conversion, an opaque structure is - presented. */ -struct xencomm_handle; - -extern void xencomm_free(struct xencomm_handle *desc); -extern struct xencomm_handle *xencomm_map(void *ptr, unsigned long bytes); -extern struct xencomm_handle *__xencomm_map_no_alloc(void *ptr, - unsigned long bytes, struct xencomm_mini *xc_area); - -#if 0 -#define XENCOMM_MINI_ALIGNED(xc_desc, n) \ - struct xencomm_mini xc_desc ## _base[(n)] \ - __attribute__((__aligned__(sizeof(struct xencomm_mini)))); \ - struct xencomm_mini *xc_desc = &xc_desc ## _base[0]; -#else -/* - * gcc bug workaround: - * http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16660 - * gcc doesn't handle properly stack variable with - * __attribute__((__align__(sizeof(struct xencomm_mini)))) - */ -#define XENCOMM_MINI_ALIGNED(xc_desc, n) \ - unsigned char xc_desc ## _base[((n) + 1 ) * \ - sizeof(struct xencomm_mini)]; \ - struct xencomm_mini *xc_desc = (struct xencomm_mini *) \ - ((unsigned long)xc_desc ## _base + \ - (sizeof(struct xencomm_mini) - \ - ((unsigned long)xc_desc ## _base) % \ - sizeof(struct xencomm_mini))); -#endif -#define xencomm_map_no_alloc(ptr, bytes) \ - ({ XENCOMM_MINI_ALIGNED(xc_desc, 1); \ - __xencomm_map_no_alloc(ptr, bytes, xc_desc); }) - -/* provided by architecture code: */ -extern unsigned long xencomm_vtop(unsigned long vaddr); - -static inline void *xencomm_pa(void *ptr) -{ - return (void *)xencomm_vtop((unsigned long)ptr); -} - -#define xen_guest_handle(hnd) ((hnd).p) - -#endif /* _LINUX_XENCOMM_H_ */ From 8423ae3d7a3cfe084865262cfaeba1359d405182 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Mon, 10 Feb 2014 17:45:50 -0800 Subject: [PATCH 516/596] block: Fix cloning of discard/write same bios Immutable biovecs changed the way bio segments are treated in such a way that bio_for_each_segment() cannot now do what we want for discard/write same bios, since bi_size means something completely different for them. Fortunately discard and write same bios never have more than a single biovec, so bio_for_each_segment() is unnecessary and not terribly meaningful for them, but we still have to special case them in a few places. Signed-off-by: Kent Overstreet Tested-by: Richard W.M. Jones Signed-off-by: Jens Axboe --- fs/bio.c | 15 ++++++++++----- include/linux/bio.h | 11 +++++++++++ 2 files changed, 21 insertions(+), 5 deletions(-) diff --git a/fs/bio.c b/fs/bio.c index 75c49a382239..8754e7b6eb49 100644 --- a/fs/bio.c +++ b/fs/bio.c @@ -611,7 +611,6 @@ EXPORT_SYMBOL(bio_clone_fast); struct bio *bio_clone_bioset(struct bio *bio_src, gfp_t gfp_mask, struct bio_set *bs) { - unsigned nr_iovecs = 0; struct bvec_iter iter; struct bio_vec bv; struct bio *bio; @@ -638,10 +637,7 @@ struct bio *bio_clone_bioset(struct bio *bio_src, gfp_t gfp_mask, * __bio_clone_fast() anyways. */ - bio_for_each_segment(bv, bio_src, iter) - nr_iovecs++; - - bio = bio_alloc_bioset(gfp_mask, nr_iovecs, bs); + bio = bio_alloc_bioset(gfp_mask, bio_segments(bio_src), bs); if (!bio) return NULL; @@ -650,9 +646,18 @@ struct bio *bio_clone_bioset(struct bio *bio_src, gfp_t gfp_mask, bio->bi_iter.bi_sector = bio_src->bi_iter.bi_sector; bio->bi_iter.bi_size = bio_src->bi_iter.bi_size; + if (bio->bi_rw & REQ_DISCARD) + goto integrity_clone; + + if (bio->bi_rw & REQ_WRITE_SAME) { + bio->bi_io_vec[bio->bi_vcnt++] = bio_src->bi_io_vec[0]; + goto integrity_clone; + } + bio_for_each_segment(bv, bio_src, iter) bio->bi_io_vec[bio->bi_vcnt++] = bv; +integrity_clone: if (bio_integrity(bio_src)) { int ret; diff --git a/include/linux/bio.h b/include/linux/bio.h index d6791bba8264..5a4d39b4686b 100644 --- a/include/linux/bio.h +++ b/include/linux/bio.h @@ -250,6 +250,17 @@ static inline unsigned bio_segments(struct bio *bio) struct bio_vec bv; struct bvec_iter iter; + /* + * We special case discard/write same, because they interpret bi_size + * differently: + */ + + if (bio->bi_rw & REQ_DISCARD) + return 1; + + if (bio->bi_rw & REQ_WRITE_SAME) + return 1; + bio_for_each_segment(bv, bio, iter) segs++; From 1e93b8c274268038c93763dca65a73b42a081e10 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 11 Feb 2014 08:27:13 -0800 Subject: [PATCH 517/596] blk-mq: dont assume rq->errors is set when returning an error from ->queue_rq rq->errors never has been part of the communication protocol between drivers and the block stack and most drivers will not have initialized it. Return -EIO to upper layers when the driver returns BLK_MQ_RQ_QUEUE_ERROR unconditionally. If a driver want to return a different error it can easily do so by returning success after calling blk_mq_end_io itself. Signed-off-by: Christoph Hellwig Signed-off-by: Jens Axboe --- block/blk-mq.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/block/blk-mq.c b/block/blk-mq.c index a59b0565e940..0480710a8b45 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -605,8 +605,8 @@ static void __blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx) break; default: pr_err("blk-mq: bad return on queue: %d\n", ret); - rq->errors = -EIO; case BLK_MQ_RQ_QUEUE_ERROR: + rq->errors = -EIO; blk_mq_end_io(rq, rq->errors); break; } From 49f5baa5109897b8cee491e8a7c4d74052b6bc1e Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 11 Feb 2014 08:27:14 -0800 Subject: [PATCH 518/596] blk-mq: pair blk_mq_start_request / blk_mq_requeue_request Make sure we have a proper pairing between starting and requeueing requests. Move the dma drain and REQ_END setup into blk_mq_start_request, and make sure blk_mq_requeue_request properly undoes them, giving us a pair of function to prepare and unprepare a request without leaving side effects. Together this ensures we always clean up properly after BLK_MQ_RQ_QUEUE_BUSY returns from ->queue_rq. Signed-off-by: Christoph Hellwig Signed-off-by: Jens Axboe --- block/blk-mq.c | 49 ++++++++++++++++++++++++++----------------------- 1 file changed, 26 insertions(+), 23 deletions(-) diff --git a/block/blk-mq.c b/block/blk-mq.c index 0480710a8b45..1fa9dd153fde 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -377,7 +377,7 @@ void blk_mq_complete_request(struct request *rq) } EXPORT_SYMBOL(blk_mq_complete_request); -static void blk_mq_start_request(struct request *rq) +static void blk_mq_start_request(struct request *rq, bool last) { struct request_queue *q = rq->q; @@ -390,6 +390,25 @@ static void blk_mq_start_request(struct request *rq) */ rq->deadline = jiffies + q->rq_timeout; set_bit(REQ_ATOM_STARTED, &rq->atomic_flags); + + if (q->dma_drain_size && blk_rq_bytes(rq)) { + /* + * Make sure space for the drain appears. We know we can do + * this because max_hw_segments has been adjusted to be one + * fewer than the device can handle. + */ + rq->nr_phys_segments++; + } + + /* + * Flag the last request in the series so that drivers know when IO + * should be kicked off, if they don't do it on a per-request basis. + * + * Note: the flag isn't the only condition drivers should do kick off. + * If drive is busy, the last request might not have the bit set. + */ + if (last) + rq->cmd_flags |= REQ_END; } static void blk_mq_requeue_request(struct request *rq) @@ -398,6 +417,11 @@ static void blk_mq_requeue_request(struct request *rq) trace_block_rq_requeue(q, rq); clear_bit(REQ_ATOM_STARTED, &rq->atomic_flags); + + rq->cmd_flags &= ~REQ_END; + + if (q->dma_drain_size && blk_rq_bytes(rq)) + rq->nr_phys_segments--; } struct blk_mq_timeout_data { @@ -565,29 +589,8 @@ static void __blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx) rq = list_first_entry(&rq_list, struct request, queuelist); list_del_init(&rq->queuelist); - blk_mq_start_request(rq); - if (q->dma_drain_size && blk_rq_bytes(rq)) { - /* - * make sure space for the drain appears we - * know we can do this because max_hw_segments - * has been adjusted to be one fewer than the - * device can handle - */ - rq->nr_phys_segments++; - } - - /* - * Last request in the series. Flag it as such, this - * enables drivers to know when IO should be kicked off, - * if they don't do it on a per-request basis. - * - * Note: the flag isn't the only condition drivers - * should do kick off. If drive is busy, the last - * request might not have the bit set. - */ - if (list_empty(&rq_list)) - rq->cmd_flags |= REQ_END; + blk_mq_start_request(rq, list_empty(&rq_list)); ret = q->mq_ops->queue_rq(hctx, rq); switch (ret) { From 8858d88a25142544843869f0cd3e6654aa7b4aec Mon Sep 17 00:00:00 2001 From: Linus Walleij Date: Tue, 11 Feb 2014 14:02:37 +0100 Subject: [PATCH 519/596] ARM: ux500: disable msp2 device tree node Commit 70b41abc151f9 "ARM: ux500: move MSP pin control to the device tree" accidentally activated MSP2, giving rise to a boot scroll scream as the kernel attempts to probe a driver for it and fails to obtain DMA channel 14. Fix this up by marking the node disabled again. Acked-by: Lee Jones Signed-off-by: Linus Walleij Signed-off-by: Kevin Hilman --- arch/arm/boot/dts/ste-href.dtsi | 1 - 1 file changed, 1 deletion(-) diff --git a/arch/arm/boot/dts/ste-href.dtsi b/arch/arm/boot/dts/ste-href.dtsi index 0c1e8d871ed1..6cb9b68e2188 100644 --- a/arch/arm/boot/dts/ste-href.dtsi +++ b/arch/arm/boot/dts/ste-href.dtsi @@ -188,7 +188,6 @@ msp1: msp@80124000 { msp2: msp@80117000 { pinctrl-names = "default"; pinctrl-0 = <&msp2_default_mode>; - status = "okay"; }; msp3: msp@80125000 { From d651aa1d68a2f0a7ee65697b04c6a92f8c0a12f2 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Red Hat)" Date: Tue, 11 Feb 2014 13:38:54 -0500 Subject: [PATCH 520/596] ring-buffer: Fix first commit on sub-buffer having non-zero delta Each sub-buffer (buffer page) has a full 64 bit timestamp. The events on that page use a 27 bit delta against that timestamp in order to save on bits written to the ring buffer. If the time between events is larger than what the 27 bits can hold, a "time extend" event is added to hold the entire 64 bit timestamp again and the events after that hold a delta from that timestamp. As a "time extend" is always paired with an event, it is logical to just allocate the event with the time extend, to make things a bit more efficient. Unfortunately, when the pairing code was written, it removed the "delta = 0" from the first commit on a page, causing the events on the page to be slightly skewed. Fixes: 69d1b839f7ee "ring-buffer: Bind time extend and data events together" Cc: stable@vger.kernel.org # 2.6.37+ Signed-off-by: Steven Rostedt --- kernel/trace/ring_buffer.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index 294b8a271a04..fc4da2d97f9b 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -2397,6 +2397,13 @@ __rb_reserve_next(struct ring_buffer_per_cpu *cpu_buffer, write &= RB_WRITE_MASK; tail = write - length; + /* + * If this is the first commit on the page, then it has the same + * timestamp as the page itself. + */ + if (!tail) + delta = 0; + /* See if we shot pass the end of this buffer page */ if (unlikely(write > BUF_PAGE_SIZE)) return rb_move_tail(cpu_buffer, length, tail, From 09bdc2d70dedd0fc0358da93bca664c7b11ff907 Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Tue, 11 Feb 2014 11:29:05 -0500 Subject: [PATCH 521/596] nfsd4: fix acl buffer overrun 4ac7249ea5a0ceef9f8269f63f33cc873c3fac61 "nfsd: use get_acl and ->set_acl" forgets to set the size in the case get_acl() succeeds, so _posix_to_nfsv4_one() can then write past the end of its allocation. Symptoms were slab corruption warnings. Also, some minor cleanup while we're here. (Among other things, note that the first few lines guarantee that pacl is non-NULL.) Signed-off-by: J. Bruce Fields --- fs/nfsd/nfs4acl.c | 9 +++------ 1 file changed, 3 insertions(+), 6 deletions(-) diff --git a/fs/nfsd/nfs4acl.c b/fs/nfsd/nfs4acl.c index d3a587144222..d190e33d0ec2 100644 --- a/fs/nfsd/nfs4acl.c +++ b/fs/nfsd/nfs4acl.c @@ -151,17 +151,15 @@ nfsd4_get_nfs4_acl(struct svc_rqst *rqstp, struct dentry *dentry, pacl = posix_acl_from_mode(inode->i_mode, GFP_KERNEL); if (IS_ERR(pacl)) return PTR_ERR(pacl); - /* allocate for worst case: one (deny, allow) pair each: */ - size += 2 * pacl->a_count; } + /* allocate for worst case: one (deny, allow) pair each: */ + size += 2 * pacl->a_count; if (S_ISDIR(inode->i_mode)) { flags = NFS4_ACL_DIR; dpacl = get_acl(inode, ACL_TYPE_DEFAULT); if (dpacl) size += 2 * dpacl->a_count; - } else { - dpacl = NULL; } *acl = nfs4_acl_new(size); @@ -170,8 +168,7 @@ nfsd4_get_nfs4_acl(struct svc_rqst *rqstp, struct dentry *dentry, goto out; } - if (pacl) - _posix_to_nfsv4_one(pacl, *acl, flags & ~NFS4_ACL_TYPE_DEFAULT); + _posix_to_nfsv4_one(pacl, *acl, flags & ~NFS4_ACL_TYPE_DEFAULT); if (dpacl) _posix_to_nfsv4_one(dpacl, *acl, flags | NFS4_ACL_TYPE_DEFAULT); From 3635c7e2d59f7861afa6fa5e87e2a58860ff514d Mon Sep 17 00:00:00 2001 From: Raymond Wanyoike Date: Sun, 9 Feb 2014 11:59:46 +0300 Subject: [PATCH 522/596] usb: option: blacklist ZTE MF667 net interface Interface #5 of 19d2:1270 is a net interface which has been submitted to the qmi_wwan driver so consequently remove it from the option driver. Signed-off-by: Raymond Wanyoike Cc: stable Signed-off-by: Greg Kroah-Hartman --- drivers/usb/serial/option.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/usb/serial/option.c b/drivers/usb/serial/option.c index 5c86f57e4afa..216d20affba8 100644 --- a/drivers/usb/serial/option.c +++ b/drivers/usb/serial/option.c @@ -1362,7 +1362,8 @@ static const struct usb_device_id option_ids[] = { { USB_DEVICE_AND_INTERFACE_INFO(ZTE_VENDOR_ID, 0x1267, 0xff, 0xff, 0xff) }, { USB_DEVICE_AND_INTERFACE_INFO(ZTE_VENDOR_ID, 0x1268, 0xff, 0xff, 0xff) }, { USB_DEVICE_AND_INTERFACE_INFO(ZTE_VENDOR_ID, 0x1269, 0xff, 0xff, 0xff) }, - { USB_DEVICE_AND_INTERFACE_INFO(ZTE_VENDOR_ID, 0x1270, 0xff, 0xff, 0xff) }, + { USB_DEVICE_AND_INTERFACE_INFO(ZTE_VENDOR_ID, 0x1270, 0xff, 0xff, 0xff), + .driver_info = (kernel_ulong_t)&net_intf5_blacklist }, { USB_DEVICE_AND_INTERFACE_INFO(ZTE_VENDOR_ID, 0x1271, 0xff, 0xff, 0xff) }, { USB_DEVICE_AND_INTERFACE_INFO(ZTE_VENDOR_ID, 0x1272, 0xff, 0xff, 0xff) }, { USB_DEVICE_AND_INTERFACE_INFO(ZTE_VENDOR_ID, 0x1273, 0xff, 0xff, 0xff) }, From ddf5eb564d97c94e114b45e84c89ce0e7024a9ac Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Tue, 11 Feb 2014 15:25:24 -0800 Subject: [PATCH 523/596] staging/rtl8821ae: fix build, depends on MAC80211 rtl8821ae uses ieee80211 interfaces so it should depend on MAC80211. Fixes these build errors: ERROR: "ieee80211_rx_irqsafe" [drivers/staging/rtl8821ae/rtl8821ae.ko] undefined! ERROR: "ieee80211_beacon_get_tim" [drivers/staging/rtl8821ae/rtl8821ae.ko] undefined! ERROR: "ieee80211_unregister_hw" [drivers/staging/rtl8821ae/rtl8821ae.ko] undefined! ERROR: "rate_control_send_low" [drivers/staging/rtl8821ae/rtl8821ae.ko] undefined! ERROR: "ieee80211_stop_queue" [drivers/staging/rtl8821ae/rtl8821ae.ko] undefined! ERROR: "ieee80211_rate_control_register" [drivers/staging/rtl8821ae/rtl8821ae.ko] undefined! ERROR: "wiphy_to_ieee80211_hw" [drivers/staging/rtl8821ae/rtl8821ae.ko] undefined! ERROR: "ieee80211_stop_tx_ba_cb_irqsafe" [drivers/staging/rtl8821ae/rtl8821ae.ko] undefined! ERROR: "ieee80211_tx_status_irqsafe" [drivers/staging/rtl8821ae/rtl8821ae.ko] undefined! ERROR: "ieee80211_find_sta" [drivers/staging/rtl8821ae/rtl8821ae.ko] undefined! ERROR: "ieee80211_wake_queue" [drivers/staging/rtl8821ae/rtl8821ae.ko] undefined! ERROR: "ieee80211_rate_control_unregister" [drivers/staging/rtl8821ae/rtl8821ae.ko] undefined! ERROR: "ieee80211_register_hw" [drivers/staging/rtl8821ae/rtl8821ae.ko] undefined! ERROR: "ieee80211_start_tx_ba_session" [drivers/staging/rtl8821ae/rtl8821ae.ko] undefined! ERROR: "ieee80211_alloc_hw" [drivers/staging/rtl8821ae/rtl8821ae.ko] undefined! ERROR: "ieee80211_free_hw" [drivers/staging/rtl8821ae/rtl8821ae.ko] undefined! ERROR: "ieee80211_connection_loss" [drivers/staging/rtl8821ae/rtl8821ae.ko] undefined! ERROR: "ieee80211_start_tx_ba_cb_irqsafe" [drivers/staging/rtl8821ae/rtl8821ae.ko] undefined! Signed-off-by: Randy Dunlap Signed-off-by: Greg Kroah-Hartman --- drivers/staging/rtl8821ae/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/staging/rtl8821ae/Kconfig b/drivers/staging/rtl8821ae/Kconfig index 2aa5dac2f1df..abccc9dabd65 100644 --- a/drivers/staging/rtl8821ae/Kconfig +++ b/drivers/staging/rtl8821ae/Kconfig @@ -1,6 +1,6 @@ config R8821AE tristate "RealTek RTL8821AE Wireless LAN NIC driver" - depends on PCI && WLAN + depends on PCI && WLAN && MAC80211 depends on m select WIRELESS_EXT select WEXT_PRIV From 87fbb2ac6073a7039303517546a76074feb14c84 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Red Hat)" Date: Tue, 11 Feb 2014 20:19:44 -0500 Subject: [PATCH 524/596] ftrace/x86: Use breakpoints for converting function graph caller When the conversion was made to remove stop machine and use the breakpoint logic instead, the modification of the function graph caller is still done directly as though it was being done under stop machine. As it is not converted via stop machine anymore, there is a possibility that the code could be layed across cache lines and if another CPU is accessing that function graph call when it is being updated, it could cause a General Protection Fault. Convert the update of the function graph caller to use the breakpoint method as well. Cc: H. Peter Anvin Cc: stable@vger.kernel.org # 3.5+ Fixes: 08d636b6d4fb "ftrace/x86: Have arch x86_64 use breakpoints instead of stop machine" Signed-off-by: Steven Rostedt --- arch/x86/kernel/ftrace.c | 83 +++++++++++++++++++++++----------------- 1 file changed, 47 insertions(+), 36 deletions(-) diff --git a/arch/x86/kernel/ftrace.c b/arch/x86/kernel/ftrace.c index d4bdd253fea7..e6253195a301 100644 --- a/arch/x86/kernel/ftrace.c +++ b/arch/x86/kernel/ftrace.c @@ -77,8 +77,7 @@ within(unsigned long addr, unsigned long start, unsigned long end) return addr >= start && addr < end; } -static int -do_ftrace_mod_code(unsigned long ip, const void *new_code) +static unsigned long text_ip_addr(unsigned long ip) { /* * On x86_64, kernel text mappings are mapped read-only with @@ -91,7 +90,7 @@ do_ftrace_mod_code(unsigned long ip, const void *new_code) if (within(ip, (unsigned long)_text, (unsigned long)_etext)) ip = (unsigned long)__va(__pa_symbol(ip)); - return probe_kernel_write((void *)ip, new_code, MCOUNT_INSN_SIZE); + return ip; } static const unsigned char *ftrace_nop_replace(void) @@ -123,8 +122,10 @@ ftrace_modify_code_direct(unsigned long ip, unsigned const char *old_code, if (memcmp(replaced, old_code, MCOUNT_INSN_SIZE) != 0) return -EINVAL; + ip = text_ip_addr(ip); + /* replace the text with the new text */ - if (do_ftrace_mod_code(ip, new_code)) + if (probe_kernel_write((void *)ip, new_code, MCOUNT_INSN_SIZE)) return -EPERM; sync_core(); @@ -221,37 +222,51 @@ int ftrace_modify_call(struct dyn_ftrace *rec, unsigned long old_addr, return -EINVAL; } -int ftrace_update_ftrace_func(ftrace_func_t func) +static unsigned long ftrace_update_func; + +static int update_ftrace_func(unsigned long ip, void *new) { - unsigned long ip = (unsigned long)(&ftrace_call); - unsigned char old[MCOUNT_INSN_SIZE], *new; + unsigned char old[MCOUNT_INSN_SIZE]; int ret; - memcpy(old, &ftrace_call, MCOUNT_INSN_SIZE); - new = ftrace_call_replace(ip, (unsigned long)func); + memcpy(old, (void *)ip, MCOUNT_INSN_SIZE); + + ftrace_update_func = ip; + /* Make sure the breakpoints see the ftrace_update_func update */ + smp_wmb(); /* See comment above by declaration of modifying_ftrace_code */ atomic_inc(&modifying_ftrace_code); ret = ftrace_modify_code(ip, old, new); + atomic_dec(&modifying_ftrace_code); + + return ret; +} + +int ftrace_update_ftrace_func(ftrace_func_t func) +{ + unsigned long ip = (unsigned long)(&ftrace_call); + unsigned char *new; + int ret; + + new = ftrace_call_replace(ip, (unsigned long)func); + ret = update_ftrace_func(ip, new); + /* Also update the regs callback function */ if (!ret) { ip = (unsigned long)(&ftrace_regs_call); - memcpy(old, &ftrace_regs_call, MCOUNT_INSN_SIZE); new = ftrace_call_replace(ip, (unsigned long)func); - ret = ftrace_modify_code(ip, old, new); + ret = update_ftrace_func(ip, new); } - atomic_dec(&modifying_ftrace_code); - return ret; } static int is_ftrace_caller(unsigned long ip) { - if (ip == (unsigned long)(&ftrace_call) || - ip == (unsigned long)(&ftrace_regs_call)) + if (ip == ftrace_update_func) return 1; return 0; @@ -677,45 +692,41 @@ int __init ftrace_dyn_arch_init(void *data) #ifdef CONFIG_DYNAMIC_FTRACE extern void ftrace_graph_call(void); -static int ftrace_mod_jmp(unsigned long ip, - int old_offset, int new_offset) +static unsigned char *ftrace_jmp_replace(unsigned long ip, unsigned long addr) { - unsigned char code[MCOUNT_INSN_SIZE]; + static union ftrace_code_union calc; - if (probe_kernel_read(code, (void *)ip, MCOUNT_INSN_SIZE)) - return -EFAULT; + /* Jmp not a call (ignore the .e8) */ + calc.e8 = 0xe9; + calc.offset = ftrace_calc_offset(ip + MCOUNT_INSN_SIZE, addr); - if (code[0] != 0xe9 || old_offset != *(int *)(&code[1])) - return -EINVAL; + /* + * ftrace external locks synchronize the access to the static variable. + */ + return calc.code; +} - *(int *)(&code[1]) = new_offset; +static int ftrace_mod_jmp(unsigned long ip, void *func) +{ + unsigned char *new; - if (do_ftrace_mod_code(ip, &code)) - return -EPERM; + new = ftrace_jmp_replace(ip, (unsigned long)func); - return 0; + return update_ftrace_func(ip, new); } int ftrace_enable_ftrace_graph_caller(void) { unsigned long ip = (unsigned long)(&ftrace_graph_call); - int old_offset, new_offset; - old_offset = (unsigned long)(&ftrace_stub) - (ip + MCOUNT_INSN_SIZE); - new_offset = (unsigned long)(&ftrace_graph_caller) - (ip + MCOUNT_INSN_SIZE); - - return ftrace_mod_jmp(ip, old_offset, new_offset); + return ftrace_mod_jmp(ip, &ftrace_graph_caller); } int ftrace_disable_ftrace_graph_caller(void) { unsigned long ip = (unsigned long)(&ftrace_graph_call); - int old_offset, new_offset; - old_offset = (unsigned long)(&ftrace_graph_caller) - (ip + MCOUNT_INSN_SIZE); - new_offset = (unsigned long)(&ftrace_stub) - (ip + MCOUNT_INSN_SIZE); - - return ftrace_mod_jmp(ip, old_offset, new_offset); + return ftrace_mod_jmp(ip, &ftrace_stub); } #endif /* !CONFIG_DYNAMIC_FTRACE */ From abb97b8c502a270d59c0c2e1ecea78ad752372ee Mon Sep 17 00:00:00 2001 From: Roger Pau Monne Date: Tue, 11 Feb 2014 20:34:03 -0700 Subject: [PATCH 525/596] xen-blkback: init persistent_purge_work work_struct MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Initialize persistent_purge_work work_struct on xen_blkif_alloc (and remove the previous initialization done in purge_persistent_gnt). This prevents flush_work from complaining even if purge_persistent_gnt has not been used. Signed-off-by: Roger Pau Monné Reviewed-by: David Vrabel Tested-by: Sander Eikelenboom Signed-off-by: Jens Axboe --- drivers/block/xen-blkback/blkback.c | 3 +-- drivers/block/xen-blkback/common.h | 1 + drivers/block/xen-blkback/xenbus.c | 1 + 3 files changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/block/xen-blkback/blkback.c b/drivers/block/xen-blkback/blkback.c index 765fc7348b66..64c60edcdfbc 100644 --- a/drivers/block/xen-blkback/blkback.c +++ b/drivers/block/xen-blkback/blkback.c @@ -299,7 +299,7 @@ static void free_persistent_gnts(struct xen_blkif *blkif, struct rb_root *root, BUG_ON(num != 0); } -static void unmap_purged_grants(struct work_struct *work) +void xen_blkbk_unmap_purged_grants(struct work_struct *work) { struct gnttab_unmap_grant_ref unmap[BLKIF_MAX_SEGMENTS_PER_REQUEST]; struct page *pages[BLKIF_MAX_SEGMENTS_PER_REQUEST]; @@ -420,7 +420,6 @@ static void purge_persistent_gnt(struct xen_blkif *blkif) blkif->vbd.overflow_max_grants = 0; /* We can defer this work */ - INIT_WORK(&blkif->persistent_purge_work, unmap_purged_grants); schedule_work(&blkif->persistent_purge_work); pr_debug(DRV_PFX "Purged %u/%u\n", (total - num_clean), total); return; diff --git a/drivers/block/xen-blkback/common.h b/drivers/block/xen-blkback/common.h index 9eb34e24b4fe..be052773ad03 100644 --- a/drivers/block/xen-blkback/common.h +++ b/drivers/block/xen-blkback/common.h @@ -385,6 +385,7 @@ int xen_blkbk_flush_diskcache(struct xenbus_transaction xbt, int xen_blkbk_barrier(struct xenbus_transaction xbt, struct backend_info *be, int state); struct xenbus_device *xen_blkbk_xenbus(struct backend_info *be); +void xen_blkbk_unmap_purged_grants(struct work_struct *work); static inline void blkif_get_x86_32_req(struct blkif_request *dst, struct blkif_x86_32_request *src) diff --git a/drivers/block/xen-blkback/xenbus.c b/drivers/block/xen-blkback/xenbus.c index 84973c6a856a..9a547e6b6ebf 100644 --- a/drivers/block/xen-blkback/xenbus.c +++ b/drivers/block/xen-blkback/xenbus.c @@ -129,6 +129,7 @@ static struct xen_blkif *xen_blkif_alloc(domid_t domid) blkif->free_pages_num = 0; atomic_set(&blkif->persistent_gnt_in_use, 0); atomic_set(&blkif->inflight, 0); + INIT_WORK(&blkif->persistent_purge_work, xen_blkbk_unmap_purged_grants); INIT_LIST_HEAD(&blkif->pending_free); From 7282059489868e0ed1b0d79765730c6b233a8399 Mon Sep 17 00:00:00 2001 From: Mika Westerberg Date: Tue, 11 Feb 2014 12:42:37 +0200 Subject: [PATCH 526/596] ACPI / hotplug / PCI: Relax the checking of _STA return values The ACPI specification (ACPI 5.0A, Section 6.3.7) says: _STA may return bit 0 clear (not present) with bit 3 set (device is functional). This case is used to indicate a valid device for which no device driver should be loaded (for example, a bridge device.) Children of this device may be present and valid. OSPM should continue enumeration below a device whose _STA returns this bit combination. Evidently, some BIOSes follow that and return 0x0A from _STA, which causes problems to happen when they trigger bus check or device check notifications for those devices too. Namely, ACPIPHP thinks that they are gone and may drop them, for example, if such a notification is triggered during a resume from system suspend. To fix that, modify ACPICA to regard devies as present and functioning if _STA returns both the ACPI_STA_DEVICE_ENABLED and ACPI_STA_DEVICE_FUNCTIONING bits set for them. Reported-and-tested-by: Peter Wu Cc: 3.12+ # 3.12+ [rjw: Subject and changelog, minor code modifications] Signed-off-by: Rafael J. Wysocki --- drivers/pci/hotplug/acpiphp_glue.c | 15 +++++++++++++-- 1 file changed, 13 insertions(+), 2 deletions(-) diff --git a/drivers/pci/hotplug/acpiphp_glue.c b/drivers/pci/hotplug/acpiphp_glue.c index e2a783fdb98f..7c7a388c85ab 100644 --- a/drivers/pci/hotplug/acpiphp_glue.c +++ b/drivers/pci/hotplug/acpiphp_glue.c @@ -730,6 +730,17 @@ static unsigned int get_slot_status(struct acpiphp_slot *slot) return (unsigned int)sta; } +static inline bool device_status_valid(unsigned int sta) +{ + /* + * ACPI spec says that _STA may return bit 0 clear with bit 3 set + * if the device is valid but does not require a device driver to be + * loaded (Section 6.3.7 of ACPI 5.0A). + */ + unsigned int mask = ACPI_STA_DEVICE_ENABLED | ACPI_STA_DEVICE_FUNCTIONING; + return (sta & mask) == mask; +} + /** * trim_stale_devices - remove PCI devices that are not responding. * @dev: PCI device to start walking the hierarchy from. @@ -745,7 +756,7 @@ static void trim_stale_devices(struct pci_dev *dev) unsigned long long sta; status = acpi_evaluate_integer(handle, "_STA", NULL, &sta); - alive = (ACPI_SUCCESS(status) && sta == ACPI_STA_ALL) + alive = (ACPI_SUCCESS(status) && device_status_valid(sta)) || acpiphp_no_hotplug(handle); } if (!alive) { @@ -792,7 +803,7 @@ static void acpiphp_check_bridge(struct acpiphp_bridge *bridge) mutex_lock(&slot->crit_sect); if (slot_no_hotplug(slot)) { ; /* do nothing */ - } else if (get_slot_status(slot) == ACPI_STA_ALL) { + } else if (device_status_valid(get_slot_status(slot))) { /* remove stale devices if any */ list_for_each_entry_safe_reverse(dev, tmp, &bus->devices, bus_list) From 0f6aa09e40c38d734f5d70762056116ca4a76126 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Wed, 12 Feb 2014 00:52:49 +0100 Subject: [PATCH 527/596] ACPI / container: Fix error code path in container_device_attach() To avoid leaking memory on errors from device_register(), do a put_device() on the device object in question in the error code path of container_device_attach(). Fixes: caa73ea158de (ACPI / hotplug / driver core: Handle containers in a special way) Signed-off-by: Rafael J. Wysocki Reviewed-by: Yasuaki Ishimatsu --- drivers/acpi/container.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/acpi/container.c b/drivers/acpi/container.c index 0b6ae6eb5c4a..368f9ddb8480 100644 --- a/drivers/acpi/container.c +++ b/drivers/acpi/container.c @@ -79,9 +79,10 @@ static int container_device_attach(struct acpi_device *adev, ACPI_COMPANION_SET(dev, adev); dev->release = acpi_container_release; ret = device_register(dev); - if (ret) + if (ret) { + put_device(dev); return ret; - + } adev->driver_data = dev; return 1; } From c8123f8c9cb517403b51aa41c3c46ff5e10b2c17 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Wed, 12 Feb 2014 09:34:01 -0700 Subject: [PATCH 528/596] block: add cond_resched() to potentially long running ioctl discard loop When mkfs issues a full device discard and the device only supports discards of a smallish size, we can loop in blkdev_issue_discard() for a long time. If preempt isn't enabled, this can turn into a softlock situation and the kernel will start complaining. Add an explicit cond_resched() at the end of the loop to avoid that. Cc: stable@kernel.org Signed-off-by: Jens Axboe --- block/blk-lib.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/block/blk-lib.c b/block/blk-lib.c index 2da76c999ef3..97a733cf3d5f 100644 --- a/block/blk-lib.c +++ b/block/blk-lib.c @@ -119,6 +119,14 @@ int blkdev_issue_discard(struct block_device *bdev, sector_t sector, atomic_inc(&bb.done); submit_bio(type, bio); + + /* + * We can loop for a long time in here, if someone does + * full device discards (like mkfs). Be nice and allow + * us to schedule out to avoid softlocking if preempt + * is disabled. + */ + cond_resched(); } blk_finish_plug(&plug); From 2078600b1f8fc68b02f6ebdda1759fb0c9f51afb Mon Sep 17 00:00:00 2001 From: Martin Kepplinger Date: Wed, 12 Feb 2014 17:09:23 +0100 Subject: [PATCH 529/596] ALSA: Revert "ALSA: hda/realtek - Avoid invalid COEFs for ALC271X" This reverts commit d3c56568f43807135f2c2a09582a69f809f0d8b7. The reverted commit breaks audio through headphone line out on the Acer TravelMate B113 (Type1Sku0) Notebook, my main work machine. I don't know much about it but this fixes my problem. Bisected and tested. Fixes: d3c56568f438 ('ALSA: hda/realtek - Avoid invalid COEFs for ALC271X') Cc: Tested-by: Martin Kepplinger Signed-off-by: Martin Kepplinger Signed-off-by: Takashi Iwai --- sound/pci/hda/patch_realtek.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index 0ab0b9e05967..a9a83b85517a 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -4438,9 +4438,6 @@ static void alc269_fill_coef(struct hda_codec *codec) if (spec->codec_variant != ALC269_TYPE_ALC269VB) return; - /* ALC271X doesn't seem to support these COEFs (bko#52181) */ - if (!strcmp(codec->chip_name, "ALC271X")) - return; if ((alc_get_coef0(codec) & 0x00ff) < 0x015) { alc_write_coef_idx(codec, 0xf, 0x960b); From d07875bd0d1517185534b5f9eef469426ba42cb9 Mon Sep 17 00:00:00 2001 From: Julia Lawall Date: Sun, 29 Dec 2013 23:47:37 +0100 Subject: [PATCH 530/596] RDMA/nes: Fix error return code Set the return variable to an error code as done elsewhere in the function. A simplified version of the semantic match that finds this problem is as follows: (http://coccinelle.lip6.fr/) // ( if@p1 (\(ret < 0\|ret != 0\)) { ... return ret; } | ret@p1 = 0 ) ... when != ret = e1 when != &ret *if(...) { ... when != ret = e2 when forall return ret; } // Signed-off-by: Julia Lawall Signed-off-by: Roland Dreier --- drivers/infiniband/hw/nes/nes.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/infiniband/hw/nes/nes.c b/drivers/infiniband/hw/nes/nes.c index 429141078eec..353c7b05a90a 100644 --- a/drivers/infiniband/hw/nes/nes.c +++ b/drivers/infiniband/hw/nes/nes.c @@ -675,8 +675,11 @@ static int nes_probe(struct pci_dev *pcidev, const struct pci_device_id *ent) INIT_DELAYED_WORK(&nesdev->work, nes_recheck_link_status); /* Initialize network devices */ - if ((netdev = nes_netdev_init(nesdev, mmio_regs)) == NULL) + netdev = nes_netdev_init(nesdev, mmio_regs); + if (netdev == NULL) { + ret = -ENOMEM; goto bail7; + } /* Register network device */ ret = register_netdev(netdev); From ab576627c8f97c08297d81537be17df161171923 Mon Sep 17 00:00:00 2001 From: Julia Lawall Date: Sun, 29 Dec 2013 23:47:29 +0100 Subject: [PATCH 531/596] RDMA/amso1100: Fix error return code Set the return variable to an error code as done elsewhere in the function. A simplified version of the semantic match that finds this problem is as follows: (http://coccinelle.lip6.fr/) // ( if@p1 (\(ret < 0\|ret != 0\)) { ... return ret; } | ret@p1 = 0 ) ... when != ret = e1 when != &ret *if(...) { ... when != ret = e2 when forall return ret; } // Signed-off-by: Julia Lawall Signed-off-by: Roland Dreier --- drivers/infiniband/hw/amso1100/c2.c | 4 +++- drivers/infiniband/hw/amso1100/c2_rnic.c | 3 ++- 2 files changed, 5 insertions(+), 2 deletions(-) diff --git a/drivers/infiniband/hw/amso1100/c2.c b/drivers/infiniband/hw/amso1100/c2.c index d53cf519f42a..00400c352c1a 100644 --- a/drivers/infiniband/hw/amso1100/c2.c +++ b/drivers/infiniband/hw/amso1100/c2.c @@ -1082,6 +1082,7 @@ static int c2_probe(struct pci_dev *pcidev, const struct pci_device_id *ent) /* Initialize network device */ if ((netdev = c2_devinit(c2dev, mmio_regs)) == NULL) { + ret = -ENOMEM; iounmap(mmio_regs); goto bail4; } @@ -1151,7 +1152,8 @@ static int c2_probe(struct pci_dev *pcidev, const struct pci_device_id *ent) goto bail10; } - if (c2_register_device(c2dev)) + ret = c2_register_device(c2dev); + if (ret) goto bail10; return 0; diff --git a/drivers/infiniband/hw/amso1100/c2_rnic.c b/drivers/infiniband/hw/amso1100/c2_rnic.c index b7c986990053..d2a6d961344b 100644 --- a/drivers/infiniband/hw/amso1100/c2_rnic.c +++ b/drivers/infiniband/hw/amso1100/c2_rnic.c @@ -576,7 +576,8 @@ int c2_rnic_init(struct c2_dev *c2dev) goto bail4; /* Initialize cached the adapter limits */ - if (c2_rnic_query(c2dev, &c2dev->props)) + err = c2_rnic_query(c2dev, &c2dev->props); + if (err) goto bail5; /* Initialize the PD pool */ From fc09149df6e20cfbb0bb86f10899607c321a31eb Mon Sep 17 00:00:00 2001 From: Nicholas Bellinger Date: Thu, 30 Jan 2014 13:08:49 -0800 Subject: [PATCH 532/596] target: Fix free-after-use regression in PR unregister This patch addresses a >= v3.11 free-after-use regression in core_scsi3_emulate_pro_register() that was introduced in the following commit: commit bc118fe4c4a8cfa453491ba77c0a146a6d0e73e0 Author: Andy Grover Date: Thu May 16 10:41:04 2013 -0700 target: Further refactoring of core_scsi3_emulate_pro_register() To avoid the free-after-use, save an type value before hand, and only call core_scsi3_put_pr_reg() with a valid *pr_reg. Reported-by: Dan Carpenter Cc: Andy Grover Cc: #3.11+ Signed-off-by: Nicholas Bellinger --- drivers/target/target_core_pr.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/drivers/target/target_core_pr.c b/drivers/target/target_core_pr.c index 2f5d77932c80..3013287a2aaa 100644 --- a/drivers/target/target_core_pr.c +++ b/drivers/target/target_core_pr.c @@ -2009,7 +2009,7 @@ core_scsi3_emulate_pro_register(struct se_cmd *cmd, u64 res_key, u64 sa_res_key, struct t10_reservation *pr_tmpl = &dev->t10_pr; unsigned char isid_buf[PR_REG_ISID_LEN], *isid_ptr = NULL; sense_reason_t ret = TCM_NO_SENSE; - int pr_holder = 0; + int pr_holder = 0, type; if (!se_sess || !se_lun) { pr_err("SPC-3 PR: se_sess || struct se_lun is NULL!\n"); @@ -2131,6 +2131,7 @@ core_scsi3_emulate_pro_register(struct se_cmd *cmd, u64 res_key, u64 sa_res_key, ret = TCM_RESERVATION_CONFLICT; goto out; } + type = pr_reg->pr_res_type; spin_lock(&pr_tmpl->registration_lock); /* @@ -2161,6 +2162,7 @@ core_scsi3_emulate_pro_register(struct se_cmd *cmd, u64 res_key, u64 sa_res_key, * Release the calling I_T Nexus registration now.. */ __core_scsi3_free_registration(cmd->se_dev, pr_reg, NULL, 1); + pr_reg = NULL; /* * From spc4r17, section 5.7.11.3 Unregistering @@ -2174,8 +2176,8 @@ core_scsi3_emulate_pro_register(struct se_cmd *cmd, u64 res_key, u64 sa_res_key, * RESERVATIONS RELEASED. */ if (pr_holder && - (pr_reg->pr_res_type == PR_TYPE_WRITE_EXCLUSIVE_REGONLY || - pr_reg->pr_res_type == PR_TYPE_EXCLUSIVE_ACCESS_REGONLY)) { + (type == PR_TYPE_WRITE_EXCLUSIVE_REGONLY || + type == PR_TYPE_EXCLUSIVE_ACCESS_REGONLY)) { list_for_each_entry(pr_reg_p, &pr_tmpl->registration_list, pr_reg_list) { @@ -2194,7 +2196,8 @@ core_scsi3_emulate_pro_register(struct se_cmd *cmd, u64 res_key, u64 sa_res_key, ret = core_scsi3_update_and_write_aptpl(dev, aptpl); out: - core_scsi3_put_pr_reg(pr_reg); + if (pr_reg) + core_scsi3_put_pr_reg(pr_reg); return ret; } From cdf55949c10aed90d081342264d99b2b0ed0bbf3 Mon Sep 17 00:00:00 2001 From: Nicholas Bellinger Date: Thu, 30 Jan 2014 14:05:16 -0800 Subject: [PATCH 533/596] target: Fix 32-bit + CONFIG_LBDAF=n link error w/ sector_div This patch changes core_alua_state_lba_dependent() to use do_div() instead sector_div() to avoid the following link error on 32-bit with CONFIG_LBDAF=n as reported by Jim: buildlog-1391099072.txt-drivers/built-in.o: In function `target_alua_state_check': buildlog-1391099072.txt-(.text+0x928d93): undefined reference to `__umoddi3' buildlog-1391099072.txt:make: *** [vmlinux] Error 1 -- buildlog-1391101753.txt- CC init/version.o buildlog-1391101753.txt- LD init/built-in.o buildlog-1391101753.txt-drivers/built-in.o: In function `core_alua_state_lba_dependent': buildlog-1391101753.txt-/home/jim/linux/drivers/target/target_core_alua.c:503: undefined reference to `__umoddi3' buildlog-1391101753.txt:make: *** [vmlinux] Error 1 Reported-by: Jim Davis Cc: Hannes Reinecke Signed-off-by: Nicholas Bellinger --- drivers/target/target_core_alua.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/target/target_core_alua.c b/drivers/target/target_core_alua.c index 12da9b386169..c3d9df6aaf5f 100644 --- a/drivers/target/target_core_alua.c +++ b/drivers/target/target_core_alua.c @@ -500,7 +500,7 @@ static inline int core_alua_state_lba_dependent( if (segment_mult) { u64 tmp = lba; - start_lba = sector_div(tmp, segment_size * segment_mult); + start_lba = do_div(tmp, segment_size * segment_mult); last_lba = first_lba + segment_size - 1; if (start_lba >= first_lba && From 2d15025a568ba8ab2bc6120fb13c62a9ca322f62 Mon Sep 17 00:00:00 2001 From: Roland Dreier Date: Fri, 31 Jan 2014 13:11:36 -0800 Subject: [PATCH 534/596] qla2xxx: Remove last vestiges of qla_tgt_cmd.cmd_list The only place this struct member is touched is in one INIT_LIST_HEAD. Signed-off-by: Roland Dreier Signed-off-by: Nicholas Bellinger --- drivers/scsi/qla2xxx/qla_target.c | 2 -- drivers/scsi/qla2xxx/qla_target.h | 1 - 2 files changed, 3 deletions(-) diff --git a/drivers/scsi/qla2xxx/qla_target.c b/drivers/scsi/qla2xxx/qla_target.c index 9e80d61e5a3a..2eb97d7e8d12 100644 --- a/drivers/scsi/qla2xxx/qla_target.c +++ b/drivers/scsi/qla2xxx/qla_target.c @@ -2595,8 +2595,6 @@ static int qlt_handle_cmd_for_atio(struct scsi_qla_host *vha, return -ENOMEM; } - INIT_LIST_HEAD(&cmd->cmd_list); - memcpy(&cmd->atio, atio, sizeof(*atio)); cmd->state = QLA_TGT_STATE_NEW; cmd->tgt = vha->vha_tgt.qla_tgt; diff --git a/drivers/scsi/qla2xxx/qla_target.h b/drivers/scsi/qla2xxx/qla_target.h index 1d10eecad499..66e755cdde57 100644 --- a/drivers/scsi/qla2xxx/qla_target.h +++ b/drivers/scsi/qla2xxx/qla_target.h @@ -855,7 +855,6 @@ struct qla_tgt_cmd { uint16_t loop_id; /* to save extra sess dereferences */ struct qla_tgt *tgt; /* to save extra sess dereferences */ struct scsi_qla_host *vha; - struct list_head cmd_list; struct atio_from_isp atio; }; From 6a16d7be932a9df1024836ccbb448de73afd3dd0 Mon Sep 17 00:00:00 2001 From: Roland Dreier Date: Mon, 3 Feb 2014 00:35:03 -0800 Subject: [PATCH 535/596] target: Fix missing length check in spc_emulate_evpd_83() Commit fbfe858fea2a ("target_core_spc: Include target device descriptor in VPD page 83") added a new length variable, but (due to a cut and paste mistake?) just checks scsi_name_len against 256 twice. Fix this to check scsi_target_len for overflow too. Signed-off-by: Roland Dreier Signed-off-by: Nicholas Bellinger --- drivers/target/target_core_spc.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/target/target_core_spc.c b/drivers/target/target_core_spc.c index 43c5ca9878bc..3bebc71ea033 100644 --- a/drivers/target/target_core_spc.c +++ b/drivers/target/target_core_spc.c @@ -440,8 +440,8 @@ spc_emulate_evpd_83(struct se_cmd *cmd, unsigned char *buf) padding = ((-scsi_target_len) & 3); if (padding) scsi_target_len += padding; - if (scsi_name_len > 256) - scsi_name_len = 256; + if (scsi_target_len > 256) + scsi_target_len = 256; buf[off-1] = scsi_target_len; off += scsi_target_len; From 752d86805b18e58c56c6f3c7004c2dffd4430c17 Mon Sep 17 00:00:00 2001 From: Nicholas Bellinger Date: Mon, 3 Feb 2014 12:55:42 -0800 Subject: [PATCH 536/596] iscsi-target: Fix SNACK Type 1 + BegRun=0 handling This patch fixes Status SNACK handling of BegRun=0 to allow for all unacknowledged respones to be resent, instead of always assuming that BegRun would be an explicit value less than the current ExpStatSN. Reported-by: santosh kulkarni Signed-off-by: Nicholas Bellinger --- drivers/target/iscsi/iscsi_target_erl1.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/target/iscsi/iscsi_target_erl1.c b/drivers/target/iscsi/iscsi_target_erl1.c index e048d6439f4a..cda4d80cfaef 100644 --- a/drivers/target/iscsi/iscsi_target_erl1.c +++ b/drivers/target/iscsi/iscsi_target_erl1.c @@ -507,7 +507,9 @@ int iscsit_handle_status_snack( u32 last_statsn; int found_cmd; - if (conn->exp_statsn > begrun) { + if (!begrun) { + begrun = conn->exp_statsn; + } else if (conn->exp_statsn > begrun) { pr_err("Got Status SNACK Begrun: 0x%08x, RunLength:" " 0x%08x but already got ExpStatSN: 0x%08x on CID:" " %hu.\n", begrun, runlength, conn->exp_statsn, From a80e21b3b2d151d79bb9be42334ab10d40195324 Mon Sep 17 00:00:00 2001 From: Nicholas Bellinger Date: Mon, 3 Feb 2014 12:59:56 -0800 Subject: [PATCH 537/596] iser-target: Fix leak on failure in isert_conn_create_fastreg_pool This patch fixes a memory leak for fr_desc upon failure of isert_create_fr_desc() in isert_conn_create_fastreg_pool() code. As reported by Coverity 1166659: *** CID 1166659: Resource leak (RESOURCE_LEAK) /drivers/infiniband/ulp/isert/ib_isert.c: 470 in isert_conn_create_fastreg_pool() 464 isert_conn, isert_conn->conn_fr_pool_size); 465 466 return 0; 467 468 err: 469 isert_conn_free_fastreg_pool(isert_conn); >>> CID 1166659: Resource leak (RESOURCE_LEAK) >>> Variable "fr_desc" going out of scope leaks the storage it points to. 470 return ret; 471 } 472 473 static int 474 isert_connect_request(struct rdma_cm_id *cma_id, struct rdma_cm_event *event) 475 { Cc: Sagi Grimberg Signed-off-by: Nicholas Bellinger --- drivers/infiniband/ulp/isert/ib_isert.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/infiniband/ulp/isert/ib_isert.c b/drivers/infiniband/ulp/isert/ib_isert.c index 2b161be3c1a3..d18d08a076e8 100644 --- a/drivers/infiniband/ulp/isert/ib_isert.c +++ b/drivers/infiniband/ulp/isert/ib_isert.c @@ -453,6 +453,7 @@ isert_conn_create_fastreg_pool(struct isert_conn *isert_conn) if (ret) { pr_err("Failed to create fastreg descriptor err=%d\n", ret); + kfree(fr_desc); goto err; } From 3dca1471993f9b89f3184468f8bbab2b1e024451 Mon Sep 17 00:00:00 2001 From: Roland Dreier Date: Mon, 3 Feb 2014 14:08:26 -0800 Subject: [PATCH 538/596] target: Simplify command completion by removing CMD_T_FAILED flag The CMD_T_FAILED flag is set used in one place to record the result of a trivial test, and it is only tested once, few lines later. We might as well make the code simpler and easier to read by directly doing the test of "success" where we want to use it. Signed-off-by: Roland Dreier Signed-off-by: Nicholas Bellinger --- drivers/target/target_core_transport.c | 5 +---- include/target/target_core_base.h | 1 - 2 files changed, 1 insertion(+), 5 deletions(-) diff --git a/drivers/target/target_core_transport.c b/drivers/target/target_core_transport.c index c50fd9f11aab..24b4f65d8777 100644 --- a/drivers/target/target_core_transport.c +++ b/drivers/target/target_core_transport.c @@ -669,9 +669,6 @@ void target_complete_cmd(struct se_cmd *cmd, u8 scsi_status) return; } - if (!success) - cmd->transport_state |= CMD_T_FAILED; - /* * Check for case where an explicit ABORT_TASK has been received * and transport_wait_for_tasks() will be waiting for completion.. @@ -681,7 +678,7 @@ void target_complete_cmd(struct se_cmd *cmd, u8 scsi_status) spin_unlock_irqrestore(&cmd->t_state_lock, flags); complete(&cmd->t_transport_stop_comp); return; - } else if (cmd->transport_state & CMD_T_FAILED) { + } else if (!success) { INIT_WORK(&cmd->work, target_complete_failure_work); } else { INIT_WORK(&cmd->work, target_complete_ok_work); diff --git a/include/target/target_core_base.h b/include/target/target_core_base.h index c9c791209cd1..1772fadcff62 100644 --- a/include/target/target_core_base.h +++ b/include/target/target_core_base.h @@ -525,7 +525,6 @@ struct se_cmd { #define CMD_T_COMPLETE (1 << 2) #define CMD_T_SENT (1 << 4) #define CMD_T_STOP (1 << 5) -#define CMD_T_FAILED (1 << 6) #define CMD_T_DEV_ACTIVE (1 << 7) #define CMD_T_REQUEST_STOP (1 << 8) #define CMD_T_BUSY (1 << 9) From 9d8abf45944e4f1c18a04070fc3ed2f3ffcbbcb6 Mon Sep 17 00:00:00 2001 From: Jingoo Han Date: Wed, 5 Feb 2014 11:22:05 +0900 Subject: [PATCH 539/596] IB/srpt: replace strict_strtoul() with kstrtoul() The usage of strict_strtoul() is not preferred, because strict_strtoul() is obsolete. Thus, kstrtoul() should be used. Signed-off-by: Jingoo Han Signed-off-by: Nicholas Bellinger --- drivers/infiniband/ulp/srpt/ib_srpt.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/drivers/infiniband/ulp/srpt/ib_srpt.c b/drivers/infiniband/ulp/srpt/ib_srpt.c index 520a7e5a490b..0e537d8d0e47 100644 --- a/drivers/infiniband/ulp/srpt/ib_srpt.c +++ b/drivers/infiniband/ulp/srpt/ib_srpt.c @@ -3666,9 +3666,9 @@ static ssize_t srpt_tpg_attrib_store_srp_max_rdma_size( unsigned long val; int ret; - ret = strict_strtoul(page, 0, &val); + ret = kstrtoul(page, 0, &val); if (ret < 0) { - pr_err("strict_strtoul() failed with ret: %d\n", ret); + pr_err("kstrtoul() failed with ret: %d\n", ret); return -EINVAL; } if (val > MAX_SRPT_RDMA_SIZE) { @@ -3706,9 +3706,9 @@ static ssize_t srpt_tpg_attrib_store_srp_max_rsp_size( unsigned long val; int ret; - ret = strict_strtoul(page, 0, &val); + ret = kstrtoul(page, 0, &val); if (ret < 0) { - pr_err("strict_strtoul() failed with ret: %d\n", ret); + pr_err("kstrtoul() failed with ret: %d\n", ret); return -EINVAL; } if (val > MAX_SRPT_RSP_SIZE) { @@ -3746,9 +3746,9 @@ static ssize_t srpt_tpg_attrib_store_srp_sq_size( unsigned long val; int ret; - ret = strict_strtoul(page, 0, &val); + ret = kstrtoul(page, 0, &val); if (ret < 0) { - pr_err("strict_strtoul() failed with ret: %d\n", ret); + pr_err("kstrtoul() failed with ret: %d\n", ret); return -EINVAL; } if (val > MAX_SRPT_SRQ_SIZE) { @@ -3793,7 +3793,7 @@ static ssize_t srpt_tpg_store_enable( unsigned long tmp; int ret; - ret = strict_strtoul(page, 0, &tmp); + ret = kstrtoul(page, 0, &tmp); if (ret < 0) { printk(KERN_ERR "Unable to extract srpt_tpg_store_enable\n"); return -EINVAL; From d6a65fdc8903e632aa7bf86ee0f61a73969371f6 Mon Sep 17 00:00:00 2001 From: Sagi Grimberg Date: Wed, 12 Feb 2014 11:40:25 +0200 Subject: [PATCH 540/596] Target/sbc: Fix protection copy routine Need to take into account that protection sg_list (copy-buffer) may consist of multiple entries. Changes from v0: - Changed commit description Signed-off-by: Sagi Grimberg Signed-off-by: Nicholas Bellinger --- drivers/target/target_core_sbc.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/target/target_core_sbc.c b/drivers/target/target_core_sbc.c index fa3cae393e13..a4489444ffbc 100644 --- a/drivers/target/target_core_sbc.c +++ b/drivers/target/target_core_sbc.c @@ -1074,12 +1074,19 @@ sbc_dif_copy_prot(struct se_cmd *cmd, unsigned int sectors, bool read, struct scatterlist *psg; void *paddr, *addr; unsigned int i, len, left; + unsigned int offset = 0; left = sectors * dev->prot_length; for_each_sg(cmd->t_prot_sg, psg, cmd->t_prot_nents, i) { len = min(psg->length, left); + if (offset >= sg->length) { + sg = sg_next(sg); + offset = 0; + sg_off = sg->offset; + } + paddr = kmap_atomic(sg_page(psg)) + psg->offset; addr = kmap_atomic(sg_page(sg)) + sg_off; @@ -1089,6 +1096,7 @@ sbc_dif_copy_prot(struct se_cmd *cmd, unsigned int sectors, bool read, memcpy(addr, paddr, len); left -= len; + offset += len; kunmap_atomic(paddr); kunmap_atomic(addr); } From 31fd8f5b89f22a88940a9838dfd1564f3d440247 Mon Sep 17 00:00:00 2001 From: Olof Johansson Date: Mon, 3 Feb 2014 17:13:23 -0800 Subject: [PATCH 541/596] dma: mv_xor: Silence a bunch of LPAE-related warnings Enabling some of the mvebu platforms in the multiplatform config for ARM enabled these drivers, which also triggered a bunch of warnings when LPAE is enabled (thus making phys_addr_t 64-bit). Most changes are switching printk formats, but also a bit of changes to what used to be array-based pointer arithmetic that could just be done with the address types instead. The warnings were: drivers/dma/mv_xor.c: In function 'mv_xor_tx_submit': drivers/dma/mv_xor.c:500:3: warning: format '%x' expects argument of type 'unsigned int', but argument 4 has type 'dma_addr_t' [-Wformat] drivers/dma/mv_xor.c: In function 'mv_xor_alloc_chan_resources': drivers/dma/mv_xor.c:553:13: warning: cast to pointer from integer of different size [-Wint-to-pointer-cast] drivers/dma/mv_xor.c:555:4: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast] drivers/dma/mv_xor.c: In function 'mv_xor_prep_dma_memcpy': drivers/dma/mv_xor.c:584:2: warning: format '%x' expects argument of type 'unsigned int', but argument 5 has type 'dma_addr_t' [-Wformat] drivers/dma/mv_xor.c:584:2: warning: format '%x' expects argument of type 'unsigned int', but argument 6 has type 'dma_addr_t' [-Wformat] drivers/dma/mv_xor.c: In function 'mv_xor_prep_dma_xor': drivers/dma/mv_xor.c:628:2: warning: format '%u' expects argument of type 'unsigned int', but argument 7 has type 'dma_addr_t' [-Wformat] Acked-by: Vinod Koul Acked-by: Jason Cooper Signed-off-by: Olof Johansson --- drivers/dma/mv_xor.c | 24 ++++++++++++------------ 1 file changed, 12 insertions(+), 12 deletions(-) diff --git a/drivers/dma/mv_xor.c b/drivers/dma/mv_xor.c index 53fb0c8365b0..766b68ed505c 100644 --- a/drivers/dma/mv_xor.c +++ b/drivers/dma/mv_xor.c @@ -497,8 +497,8 @@ mv_xor_tx_submit(struct dma_async_tx_descriptor *tx) if (!mv_can_chain(grp_start)) goto submit_done; - dev_dbg(mv_chan_to_devp(mv_chan), "Append to last desc %x\n", - old_chain_tail->async_tx.phys); + dev_dbg(mv_chan_to_devp(mv_chan), "Append to last desc %pa\n", + &old_chain_tail->async_tx.phys); /* fix up the hardware chain */ mv_desc_set_next_desc(old_chain_tail, grp_start->async_tx.phys); @@ -527,7 +527,8 @@ mv_xor_tx_submit(struct dma_async_tx_descriptor *tx) /* returns the number of allocated descriptors */ static int mv_xor_alloc_chan_resources(struct dma_chan *chan) { - char *hw_desc; + void *virt_desc; + dma_addr_t dma_desc; int idx; struct mv_xor_chan *mv_chan = to_mv_xor_chan(chan); struct mv_xor_desc_slot *slot = NULL; @@ -542,17 +543,16 @@ static int mv_xor_alloc_chan_resources(struct dma_chan *chan) " %d descriptor slots", idx); break; } - hw_desc = (char *) mv_chan->dma_desc_pool_virt; - slot->hw_desc = (void *) &hw_desc[idx * MV_XOR_SLOT_SIZE]; + virt_desc = mv_chan->dma_desc_pool_virt; + slot->hw_desc = virt_desc + idx * MV_XOR_SLOT_SIZE; dma_async_tx_descriptor_init(&slot->async_tx, chan); slot->async_tx.tx_submit = mv_xor_tx_submit; INIT_LIST_HEAD(&slot->chain_node); INIT_LIST_HEAD(&slot->slot_node); INIT_LIST_HEAD(&slot->tx_list); - hw_desc = (char *) mv_chan->dma_desc_pool; - slot->async_tx.phys = - (dma_addr_t) &hw_desc[idx * MV_XOR_SLOT_SIZE]; + dma_desc = mv_chan->dma_desc_pool; + slot->async_tx.phys = dma_desc + idx * MV_XOR_SLOT_SIZE; slot->idx = idx++; spin_lock_bh(&mv_chan->lock); @@ -582,8 +582,8 @@ mv_xor_prep_dma_memcpy(struct dma_chan *chan, dma_addr_t dest, dma_addr_t src, int slot_cnt; dev_dbg(mv_chan_to_devp(mv_chan), - "%s dest: %x src %x len: %u flags: %ld\n", - __func__, dest, src, len, flags); + "%s dest: %pad src %pad len: %u flags: %ld\n", + __func__, &dest, &src, len, flags); if (unlikely(len < MV_XOR_MIN_BYTE_COUNT)) return NULL; @@ -626,8 +626,8 @@ mv_xor_prep_dma_xor(struct dma_chan *chan, dma_addr_t dest, dma_addr_t *src, BUG_ON(len > MV_XOR_MAX_BYTE_COUNT); dev_dbg(mv_chan_to_devp(mv_chan), - "%s src_cnt: %d len: dest %x %u flags: %ld\n", - __func__, src_cnt, len, dest, flags); + "%s src_cnt: %d len: %u dest %pad flags: %ld\n", + __func__, src_cnt, len, &dest, flags); spin_lock_bh(&mv_chan->lock); slot_cnt = mv_chan_xor_slot_count(len, src_cnt); From 709c078e176bd47227e89bb34de7c64b57aaaeab Mon Sep 17 00:00:00 2001 From: Dirk Brandewie Date: Wed, 12 Feb 2014 10:01:03 -0800 Subject: [PATCH 542/596] intel_pstate: Remove energy reporting from pstate_sample tracepoint Remove the reporting of energy since it does not provide any useful information about the state of the driver and will be a maintainance headache going forward since the RAPL energy units register is not architectural and subject to change between micro-architectures References: https://bugzilla.kernel.org/show_bug.cgi?id=69831 Fixes: b69880f9ccf7 (intel_pstate: Add trace point to report internal state.) Signed-off-by: Dirk Brandewie Signed-off-by: Rafael J. Wysocki --- drivers/cpufreq/intel_pstate.c | 9 --------- include/trace/events/power.h | 7 +------ 2 files changed, 1 insertion(+), 15 deletions(-) diff --git a/drivers/cpufreq/intel_pstate.c b/drivers/cpufreq/intel_pstate.c index 79606f473f48..c788abf1c457 100644 --- a/drivers/cpufreq/intel_pstate.c +++ b/drivers/cpufreq/intel_pstate.c @@ -51,8 +51,6 @@ static inline int32_t div_fp(int32_t x, int32_t y) return div_s64((int64_t)x << FRAC_BITS, (int64_t)y); } -static u64 energy_divisor; - struct sample { int32_t core_pct_busy; u64 aperf; @@ -630,12 +628,10 @@ static void intel_pstate_timer_func(unsigned long __data) { struct cpudata *cpu = (struct cpudata *) __data; struct sample *sample; - u64 energy; intel_pstate_sample(cpu); sample = &cpu->samples[cpu->sample_ptr]; - rdmsrl(MSR_PKG_ENERGY_STATUS, energy); intel_pstate_adjust_busy_pstate(cpu); @@ -644,7 +640,6 @@ static void intel_pstate_timer_func(unsigned long __data) cpu->pstate.current_pstate, sample->mperf, sample->aperf, - div64_u64(energy, energy_divisor), sample->freq); intel_pstate_set_sample_time(cpu); @@ -926,7 +921,6 @@ static int __init intel_pstate_init(void) int cpu, rc = 0; const struct x86_cpu_id *id; struct cpu_defaults *cpu_info; - u64 units; if (no_load) return -ENODEV; @@ -960,9 +954,6 @@ static int __init intel_pstate_init(void) if (rc) goto out; - rdmsrl(MSR_RAPL_POWER_UNIT, units); - energy_divisor = 1 << ((units >> 8) & 0x1f); /* bits{12:8} */ - intel_pstate_debug_expose_params(); intel_pstate_sysfs_expose_params(); diff --git a/include/trace/events/power.h b/include/trace/events/power.h index 9e9475c85de5..e5bf9a76f169 100644 --- a/include/trace/events/power.h +++ b/include/trace/events/power.h @@ -42,7 +42,6 @@ TRACE_EVENT(pstate_sample, u32 state, u64 mperf, u64 aperf, - u32 energy, u32 freq ), @@ -51,7 +50,6 @@ TRACE_EVENT(pstate_sample, state, mperf, aperf, - energy, freq ), @@ -61,7 +59,6 @@ TRACE_EVENT(pstate_sample, __field(u32, state) __field(u64, mperf) __field(u64, aperf) - __field(u32, energy) __field(u32, freq) ), @@ -72,17 +69,15 @@ TRACE_EVENT(pstate_sample, __entry->state = state; __entry->mperf = mperf; __entry->aperf = aperf; - __entry->energy = energy; __entry->freq = freq; ), - TP_printk("core_busy=%lu scaled=%lu state=%lu mperf=%llu aperf=%llu energy=%lu freq=%lu ", + TP_printk("core_busy=%lu scaled=%lu state=%lu mperf=%llu aperf=%llu freq=%lu ", (unsigned long)__entry->core_busy, (unsigned long)__entry->scaled_busy, (unsigned long)__entry->state, (unsigned long long)__entry->mperf, (unsigned long long)__entry->aperf, - (unsigned long)__entry->energy, (unsigned long)__entry->freq ) From 171d0ba869675dac9acbbb9f1966c54359e5aac8 Mon Sep 17 00:00:00 2001 From: Sudeep Holla Date: Tue, 11 Feb 2014 11:42:32 +0000 Subject: [PATCH 543/596] MAINTAINERS / cpufreq: update Sudeep's email address Though my old email address continues to work as alias, updating to the new address as it's shorter, easier and more appropriate. Signed-off-by: Sudeep Holla Signed-off-by: Rafael J. Wysocki --- MAINTAINERS | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index b2cf5cfb4d29..956f069a66e8 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2367,7 +2367,7 @@ F: include/linux/cpufreq.h CPU FREQUENCY DRIVERS - ARM BIG LITTLE M: Viresh Kumar -M: Sudeep KarkadaNagesha +M: Sudeep Holla L: cpufreq@vger.kernel.org L: linux-pm@vger.kernel.org W: http://www.arm.com/products/processors/technologies/biglittleprocessing.php From 789b5e0315284463617e106baad360cb9e8db3ac Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Thu, 6 Feb 2014 03:42:45 +0530 Subject: [PATCH 544/596] md/raid5: Fix CPU hotplug callback registration Subsystems that want to register CPU hotplug callbacks, as well as perform initialization for the CPUs that are already online, often do it as shown below: get_online_cpus(); for_each_online_cpu(cpu) init_cpu(cpu); register_cpu_notifier(&foobar_cpu_notifier); put_online_cpus(); This is wrong, since it is prone to ABBA deadlocks involving the cpu_add_remove_lock and the cpu_hotplug.lock (when running concurrently with CPU hotplug operations). Interestingly, the raid5 code can actually prevent double initialization and hence can use the following simplified form of callback registration: register_cpu_notifier(&foobar_cpu_notifier); get_online_cpus(); for_each_online_cpu(cpu) init_cpu(cpu); put_online_cpus(); A hotplug operation that occurs between registering the notifier and calling get_online_cpus(), won't disrupt anything, because the code takes care to perform the memory allocations only once. So reorganize the code in raid5 this way to fix the deadlock with callback registration. Cc: linux-raid@vger.kernel.org Cc: stable@vger.kernel.org (v2.6.32+) Fixes: 36d1c6476be51101778882897b315bd928c8c7b5 Signed-off-by: Oleg Nesterov [Srivatsa: Fixed the unregister_cpu_notifier() deadlock, added the free_scratch_buffer() helper to condense code further and wrote the changelog.] Signed-off-by: Srivatsa S. Bhat Signed-off-by: NeilBrown --- drivers/md/raid5.c | 92 +++++++++++++++++++++++----------------------- 1 file changed, 45 insertions(+), 47 deletions(-) diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index f1feadeb7bb2..16f5c21963db 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -5514,23 +5514,43 @@ raid5_size(struct mddev *mddev, sector_t sectors, int raid_disks) return sectors * (raid_disks - conf->max_degraded); } +static void free_scratch_buffer(struct r5conf *conf, struct raid5_percpu *percpu) +{ + safe_put_page(percpu->spare_page); + kfree(percpu->scribble); + percpu->spare_page = NULL; + percpu->scribble = NULL; +} + +static int alloc_scratch_buffer(struct r5conf *conf, struct raid5_percpu *percpu) +{ + if (conf->level == 6 && !percpu->spare_page) + percpu->spare_page = alloc_page(GFP_KERNEL); + if (!percpu->scribble) + percpu->scribble = kmalloc(conf->scribble_len, GFP_KERNEL); + + if (!percpu->scribble || (conf->level == 6 && !percpu->spare_page)) { + free_scratch_buffer(conf, percpu); + return -ENOMEM; + } + + return 0; +} + static void raid5_free_percpu(struct r5conf *conf) { - struct raid5_percpu *percpu; unsigned long cpu; if (!conf->percpu) return; - get_online_cpus(); - for_each_possible_cpu(cpu) { - percpu = per_cpu_ptr(conf->percpu, cpu); - safe_put_page(percpu->spare_page); - kfree(percpu->scribble); - } #ifdef CONFIG_HOTPLUG_CPU unregister_cpu_notifier(&conf->cpu_notify); #endif + + get_online_cpus(); + for_each_possible_cpu(cpu) + free_scratch_buffer(conf, per_cpu_ptr(conf->percpu, cpu)); put_online_cpus(); free_percpu(conf->percpu); @@ -5557,15 +5577,7 @@ static int raid456_cpu_notify(struct notifier_block *nfb, unsigned long action, switch (action) { case CPU_UP_PREPARE: case CPU_UP_PREPARE_FROZEN: - if (conf->level == 6 && !percpu->spare_page) - percpu->spare_page = alloc_page(GFP_KERNEL); - if (!percpu->scribble) - percpu->scribble = kmalloc(conf->scribble_len, GFP_KERNEL); - - if (!percpu->scribble || - (conf->level == 6 && !percpu->spare_page)) { - safe_put_page(percpu->spare_page); - kfree(percpu->scribble); + if (alloc_scratch_buffer(conf, percpu)) { pr_err("%s: failed memory allocation for cpu%ld\n", __func__, cpu); return notifier_from_errno(-ENOMEM); @@ -5573,10 +5585,7 @@ static int raid456_cpu_notify(struct notifier_block *nfb, unsigned long action, break; case CPU_DEAD: case CPU_DEAD_FROZEN: - safe_put_page(percpu->spare_page); - kfree(percpu->scribble); - percpu->spare_page = NULL; - percpu->scribble = NULL; + free_scratch_buffer(conf, per_cpu_ptr(conf->percpu, cpu)); break; default: break; @@ -5588,40 +5597,29 @@ static int raid456_cpu_notify(struct notifier_block *nfb, unsigned long action, static int raid5_alloc_percpu(struct r5conf *conf) { unsigned long cpu; - struct page *spare_page; - struct raid5_percpu __percpu *allcpus; - void *scribble; - int err; + int err = 0; - allcpus = alloc_percpu(struct raid5_percpu); - if (!allcpus) + conf->percpu = alloc_percpu(struct raid5_percpu); + if (!conf->percpu) return -ENOMEM; - conf->percpu = allcpus; - get_online_cpus(); - err = 0; - for_each_present_cpu(cpu) { - if (conf->level == 6) { - spare_page = alloc_page(GFP_KERNEL); - if (!spare_page) { - err = -ENOMEM; - break; - } - per_cpu_ptr(conf->percpu, cpu)->spare_page = spare_page; - } - scribble = kmalloc(conf->scribble_len, GFP_KERNEL); - if (!scribble) { - err = -ENOMEM; - break; - } - per_cpu_ptr(conf->percpu, cpu)->scribble = scribble; - } #ifdef CONFIG_HOTPLUG_CPU conf->cpu_notify.notifier_call = raid456_cpu_notify; conf->cpu_notify.priority = 0; - if (err == 0) - err = register_cpu_notifier(&conf->cpu_notify); + err = register_cpu_notifier(&conf->cpu_notify); + if (err) + return err; #endif + + get_online_cpus(); + for_each_present_cpu(cpu) { + err = alloc_scratch_buffer(conf, per_cpu_ptr(conf->percpu, cpu)); + if (err) { + pr_err("%s: failed memory allocation for cpu%ld\n", + __func__, cpu); + break; + } + } put_online_cpus(); return err; From a9f180345f5378ac87d80ed0bea55ba421d83859 Mon Sep 17 00:00:00 2001 From: Steven Noonan Date: Wed, 12 Feb 2014 23:01:07 -0800 Subject: [PATCH 545/596] compiler/gcc4: Make quirk for asm_volatile_goto() unconditional I started noticing problems with KVM guest destruction on Linux 3.12+, where guest memory wasn't being cleaned up. I bisected it down to the commit introducing the new 'asm goto'-based atomics, and found this quirk was later applied to those. Unfortunately, even with GCC 4.8.2 (which ostensibly fixed the known 'asm goto' bug) I am still getting some kind of miscompilation. If I enable the asm_volatile_goto quirk for my compiler, KVM guests are destroyed correctly and the memory is cleaned up. So make the quirk unconditional for now, until bug is found and fixed. Suggested-by: Linus Torvalds Signed-off-by: Steven Noonan Cc: Peter Zijlstra Cc: Steven Rostedt Cc: Jakub Jelinek Cc: Richard Henderson Cc: Andrew Morton Cc: Oleg Nesterov Cc: Link: http://lkml.kernel.org/r/1392274867-15236-1-git-send-email-steven@uplinklabs.net Link: http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58670 Signed-off-by: Ingo Molnar --- include/linux/compiler-gcc4.h | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) diff --git a/include/linux/compiler-gcc4.h b/include/linux/compiler-gcc4.h index ded429966c1f..2507fd2a1eb4 100644 --- a/include/linux/compiler-gcc4.h +++ b/include/linux/compiler-gcc4.h @@ -75,11 +75,7 @@ * * (asm goto is automatically volatile - the naming reflects this.) */ -#if GCC_VERSION <= 40801 -# define asm_volatile_goto(x...) do { asm goto(x); asm (""); } while (0) -#else -# define asm_volatile_goto(x...) do { asm goto(x); } while (0) -#endif +#define asm_volatile_goto(x...) do { asm goto(x); asm (""); } while (0) #ifdef CONFIG_ARCH_USE_BUILTIN_BSWAP #if GCC_VERSION >= 40400 From 03bbd596ac04fef47ce93a730b8f086d797c3021 Mon Sep 17 00:00:00 2001 From: "H. Peter Anvin" Date: Thu, 13 Feb 2014 07:34:30 -0800 Subject: [PATCH 546/596] x86, smap: Don't enable SMAP if CONFIG_X86_SMAP is disabled If SMAP support is not compiled into the kernel, don't enable SMAP in CR4 -- in fact, we should clear it, because the kernel doesn't contain the proper STAC/CLAC instructions for SMAP support. Found by Fengguang Wu's test system. Reported-by: Fengguang Wu Link: http://lkml.kernel.org/r/20140213124550.GA30497@localhost Signed-off-by: H. Peter Anvin Cc: # v3.7+ --- arch/x86/kernel/cpu/common.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index 24b6fd10625a..8e28bf2fc3ef 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -284,8 +284,13 @@ static __always_inline void setup_smap(struct cpuinfo_x86 *c) raw_local_save_flags(eflags); BUG_ON(eflags & X86_EFLAGS_AC); - if (cpu_has(c, X86_FEATURE_SMAP)) + if (cpu_has(c, X86_FEATURE_SMAP)) { +#ifdef CONFIG_X86_SMAP set_in_cr4(X86_CR4_SMAP); +#else + clear_in_cr4(X86_CR4_SMAP); +#endif + } } /* From 4640c7ee9b8953237d05a61ea3ea93981d1bc961 Mon Sep 17 00:00:00 2001 From: "H. Peter Anvin" Date: Thu, 13 Feb 2014 07:46:04 -0800 Subject: [PATCH 547/596] x86, smap: smap_violation() is bogus if CONFIG_X86_SMAP is off If CONFIG_X86_SMAP is disabled, smap_violation() tests for conditions which are incorrect (as the AC flag doesn't matter), causing spurious faults. The dynamic disabling of SMAP (nosmap on the command line) is fine because it disables X86_FEATURE_SMAP, therefore causing the static_cpu_has() to return false. Found by Fengguang Wu's test system. [ v3: move all predicates into smap_violation() ] [ v2: use IS_ENABLED() instead of #ifdef ] Reported-by: Fengguang Wu Link: http://lkml.kernel.org/r/20140213124550.GA30497@localhost Signed-off-by: H. Peter Anvin Cc: # v3.7+ --- arch/x86/mm/fault.c | 14 +++++++++----- 1 file changed, 9 insertions(+), 5 deletions(-) diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c index 9d591c895803..6dea040cc3a1 100644 --- a/arch/x86/mm/fault.c +++ b/arch/x86/mm/fault.c @@ -1001,6 +1001,12 @@ static int fault_in_kernel_space(unsigned long address) static inline bool smap_violation(int error_code, struct pt_regs *regs) { + if (!IS_ENABLED(CONFIG_X86_SMAP)) + return false; + + if (!static_cpu_has(X86_FEATURE_SMAP)) + return false; + if (error_code & PF_USER) return false; @@ -1087,11 +1093,9 @@ __do_page_fault(struct pt_regs *regs, unsigned long error_code) if (unlikely(error_code & PF_RSVD)) pgtable_bad(regs, error_code, address); - if (static_cpu_has(X86_FEATURE_SMAP)) { - if (unlikely(smap_violation(error_code, regs))) { - bad_area_nosemaphore(regs, error_code, address); - return; - } + if (unlikely(smap_violation(error_code, regs))) { + bad_area_nosemaphore(regs, error_code, address); + return; } /* From 4196670be786d529ab7f6c18f5077141ce1b787e Mon Sep 17 00:00:00 2001 From: Matan Barak Date: Sun, 2 Feb 2014 17:06:47 +0200 Subject: [PATCH 548/596] IB/mlx4: Don't allocate range of steerable UD QPs for Ethernet-only device When the device has only Ethernet ports, don't try to allocate range of steerable UD QPs since they aren't needed. This fixes an issue where mlx4 VFs tried to allocate a range of UD steerable QPs, but failed to do so. Fixes: c1c98501121e ("IB/mlx4: Add support for steerable IB UD QPs") Signed-off-by: Matan Barak Signed-off-by: Or Gerlitz Signed-off-by: Roland Dreier --- drivers/infiniband/hw/mlx4/main.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c index c2702f549f10..64ca4087fb52 100644 --- a/drivers/infiniband/hw/mlx4/main.c +++ b/drivers/infiniband/hw/mlx4/main.c @@ -1810,6 +1810,7 @@ static void *mlx4_ib_add(struct mlx4_dev *dev) int i, j; int err; struct mlx4_ib_iboe *iboe; + int ib_num_ports = 0; pr_info_once("%s", mlx4_ib_version); @@ -1985,10 +1986,14 @@ static void *mlx4_ib_add(struct mlx4_dev *dev) ibdev->counters[i] = -1; } + mlx4_foreach_port(i, dev, MLX4_PORT_TYPE_IB) + ib_num_ports++; + spin_lock_init(&ibdev->sm_lock); mutex_init(&ibdev->cap_mask_mutex); - if (ibdev->steering_support == MLX4_STEERING_MODE_DEVICE_MANAGED) { + if (ibdev->steering_support == MLX4_STEERING_MODE_DEVICE_MANAGED && + ib_num_ports) { ibdev->steer_qpn_count = MLX4_IB_UC_MAX_NUM_QPS; err = mlx4_qp_reserve_range(dev, ibdev->steer_qpn_count, MLX4_IB_UC_STEER_QPN_ALIGN, From ce6acca65ee42c06ac54ec9ba918865d06cee29d Mon Sep 17 00:00:00 2001 From: Markus Pargmann Date: Fri, 24 Jan 2014 18:09:41 +0100 Subject: [PATCH 549/596] serial: omap-serial: Move info message to probe function MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Currently the info message about a missing wakeirq for uart is printed every time the serial driver's startup function is called. This happens multiple times and not just once. This can cause lots of extra messages at boot time, slowing things down. It is caused by commit 2a0b965cfb6e (serial: omap: Add support for optional wake-up) which was applied for v3.13-rc1. This patch moves the infomessage to the probe function to display it only once. Reported-by: Uwe Kleine-König Acked-by: Tony Lindgren Signed-off-by: Markus Pargmann Acked-by: Uwe Kleine-König Cc: stable # 3.13 Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/omap-serial.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/tty/serial/omap-serial.c b/drivers/tty/serial/omap-serial.c index fa511ebab67c..205158173090 100644 --- a/drivers/tty/serial/omap-serial.c +++ b/drivers/tty/serial/omap-serial.c @@ -738,9 +738,6 @@ static int serial_omap_startup(struct uart_port *port) return retval; } disable_irq(up->wakeirq); - } else { - dev_info(up->port.dev, "no wakeirq for uart%d\n", - up->port.line); } dev_dbg(up->port.dev, "serial_omap_startup+%d\n", up->port.line); @@ -1687,6 +1684,9 @@ static int serial_omap_probe(struct platform_device *pdev) up->port.iotype = UPIO_MEM; up->port.irq = uartirq; up->wakeirq = wakeirq; + if (!up->wakeirq) + dev_info(up->port.dev, "no wakeirq for uart%d\n", + up->port.line); up->port.regshift = 2; up->port.fifosize = 64; From 13b949f03e642c5df36ebf5b4770d75f8c4a6b09 Mon Sep 17 00:00:00 2001 From: Mika Westerberg Date: Thu, 16 Jan 2014 14:55:57 +0200 Subject: [PATCH 550/596] serial: 8250_dw: fix compilation warning when !CONFIG_PM_SLEEP MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit CONFIG_PM will be set if either or both CONFIG_PM_SLEEP and CONFIG_PM_RUNTIME is set. Compiling the driver with !CONFIG_PM_SLEEP causes following compilation warnings: drivers/tty/serial/8250/8250_dw.c:404:12: warning: ‘dw8250_suspend’ defined but not used [-Wunused-function] drivers/tty/serial/8250/8250_dw.c:413:12: warning: ‘dw8250_resume’ defined but not used [-Wunused-function] Fix this by using CONFIG_PM_SLEEP instead. Signed-off-by: Mika Westerberg Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/8250/8250_dw.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/tty/serial/8250/8250_dw.c b/drivers/tty/serial/8250/8250_dw.c index faa64e646100..ed3113576740 100644 --- a/drivers/tty/serial/8250/8250_dw.c +++ b/drivers/tty/serial/8250/8250_dw.c @@ -391,7 +391,7 @@ static int dw8250_remove(struct platform_device *pdev) return 0; } -#ifdef CONFIG_PM +#ifdef CONFIG_PM_SLEEP static int dw8250_suspend(struct device *dev) { struct dw8250_data *data = dev_get_drvdata(dev); @@ -409,7 +409,7 @@ static int dw8250_resume(struct device *dev) return 0; } -#endif /* CONFIG_PM */ +#endif /* CONFIG_PM_SLEEP */ #ifdef CONFIG_PM_RUNTIME static int dw8250_runtime_suspend(struct device *dev) From a64c1a1c755674399da3689499b3f64c2a538120 Mon Sep 17 00:00:00 2001 From: Michael Grzeschik Date: Thu, 13 Feb 2014 10:52:03 +0100 Subject: [PATCH 551/596] serial: omap: fix rs485 probe on defered pinctrl MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit If the gpio is not yet available we better also defer the probing in the rs485 case. Signed-off-by: Michael Grzeschik Signed-off-by: Uwe Kleine-König Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/omap-serial.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/tty/serial/omap-serial.c b/drivers/tty/serial/omap-serial.c index 205158173090..77f035158d6c 100644 --- a/drivers/tty/serial/omap-serial.c +++ b/drivers/tty/serial/omap-serial.c @@ -1601,8 +1601,11 @@ static int serial_omap_probe_rs485(struct uart_omap_port *up, flags & SER_RS485_RTS_AFTER_SEND); if (ret < 0) return ret; - } else + } else if (up->rts_gpio == -EPROBE_DEFER) { + return -EPROBE_DEFER; + } else { up->rts_gpio = -EINVAL; + } if (of_property_read_u32_array(np, "rs485-rts-delay", rs485_delay, 2) == 0) { From a5934804a834f525c9e6289935ceef65b952b101 Mon Sep 17 00:00:00 2001 From: Peter Hurley Date: Tue, 11 Feb 2014 11:49:58 -0500 Subject: [PATCH 552/596] n_tty: Fix poll() when TIME_CHAR and MIN_CHAR == 0 Commit eafbe67f84761d787802e5113d895a316b6292fe, n_tty: Refactor input_available_p() by call site broke poll() when TIME_CHAR(tty) and MIN_CHAR(tty) are both 0. When TIME_CHAR and MIN_CHAR are both 0, input is available if the read_cnt is 1 (not 0). Reported-by: Eric Dumazet Tested-by: Eric Dumazet Reported-by: Stephane Eranian Tested-by: David Ahern Tested-by: Arnaldo Carvalho de Melo Signed-off-by: Peter Hurley Signed-off-by: Greg Kroah-Hartman --- drivers/tty/n_tty.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/tty/n_tty.c b/drivers/tty/n_tty.c index cb8017aa4434..2747a3b20a50 100644 --- a/drivers/tty/n_tty.c +++ b/drivers/tty/n_tty.c @@ -1896,7 +1896,7 @@ static int n_tty_open(struct tty_struct *tty) static inline int input_available_p(struct tty_struct *tty, int poll) { struct n_tty_data *ldata = tty->disc_data; - int amt = poll && !TIME_CHAR(tty) ? MIN_CHAR(tty) : 1; + int amt = poll && !TIME_CHAR(tty) && MIN_CHAR(tty) ? MIN_CHAR(tty) : 1; if (ldata->icanon && !L_EXTPROC(tty)) { if (ldata->canon_head != ldata->read_tail) From 333c085e4b0cd9df04b087d29f87969a4e26dfa9 Mon Sep 17 00:00:00 2001 From: Dmitry Eremin-Solenikov Date: Tue, 11 Feb 2014 14:18:13 +0400 Subject: [PATCH 553/596] serial: 8250_pci: unbreak last serial ports on NetMos 9865 cards Aparently 9865 uses standard BAR encoding scheme (unlike 99xx cards). Current pci_netmos_9900_setup() uses wrong BAR indices for the 9865 PCI device, function 2. Using standard BAR indices makes all 6 ports work for me. Thus disable the NetMos 9900 quirk for NetMos 9865 pci device. For the reference, here is the relevant part of lspci for my device: 02:07.0 Serial controller: MosChip Semiconductor Technology Ltd. PCI 9865 Multi-I/O Controller (prog-if 02 [16550]) Subsystem: Device a000:1000 Flags: bus master, medium devsel, latency 32, IRQ 17 I/O ports at ac00 [size=8] Memory at fcfff000 (32-bit, non-prefetchable) [size=4K] Memory at fcffe000 (32-bit, non-prefetchable) [size=4K] Capabilities: [48] Power Management version 2 Kernel driver in use: serial 02:07.1 Serial controller: MosChip Semiconductor Technology Ltd. PCI 9865 Multi-I/O Controller (prog-if 02 [16550]) Subsystem: Device a000:1000 Flags: bus master, medium devsel, latency 32, IRQ 18 I/O ports at a800 [size=8] Memory at fcffd000 (32-bit, non-prefetchable) [size=4K] Memory at fcffc000 (32-bit, non-prefetchable) [size=4K] Capabilities: [48] Power Management version 2 Kernel driver in use: serial 02:07.2 Communication controller: MosChip Semiconductor Technology Ltd. PCI 9865 Multi-I/O Controller Subsystem: Device a000:3004 Flags: bus master, medium devsel, latency 32, IRQ 19 I/O ports at a400 [size=8] I/O ports at a000 [size=8] I/O ports at 9c00 [size=8] I/O ports at 9800 [size=8] Memory at fcffb000 (32-bit, non-prefetchable) [size=4K] Capabilities: [48] Power Management version 2 Kernel driver in use: serial Signed-off-by: Dmitry Eremin-Solenikov Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/8250/8250_pci.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/tty/serial/8250/8250_pci.c b/drivers/tty/serial/8250/8250_pci.c index 50228eed3b6f..0ff3e3624d4c 100644 --- a/drivers/tty/serial/8250/8250_pci.c +++ b/drivers/tty/serial/8250/8250_pci.c @@ -783,7 +783,8 @@ static int pci_netmos_9900_setup(struct serial_private *priv, { unsigned int bar; - if ((priv->dev->subsystem_device & 0xff00) == 0x3000) { + if ((priv->dev->device != PCI_DEVICE_ID_NETMOS_9865) && + (priv->dev->subsystem_device & 0xff00) == 0x3000) { /* netmos apparently orders BARs by datasheet layout, so serial * ports get BARs 0 and 3 (or 1 and 4 for memmapped) */ From fb78b811422cd2d8c8605949cc4cc13618347ad5 Mon Sep 17 00:00:00 2001 From: Qipan Li Date: Mon, 27 Jan 2014 14:23:39 +0800 Subject: [PATCH 554/596] serial: sirf: fix kernel panic caused by unpaired spinlock commit 8b9ade9f74f8a279 coming from Viresh Kumar "tty: serial: sirfsoc: drop uart_port->lock before calling tty_flip_buffer_push()" broke sirfsoc uart driver by knic: [ 5.129122] BUG: spinlock already unlocked on CPU#0, ip6tables/1331 [ 5.132554] lock: sirfsoc_uart_ports+0x4/0x8a0, .magic: dead4ead, .owner: /-1, .owner_cpu: -1 [ 5.141651] CPU: 0 PID: 1331 Comm: ip6tables Tainted: G W O 3.10.16 #3 [ 5.148866] [] (unwind_backtrace+0x0/0xe0) from [] (show_stack+0x10/0x14) [ 5.157362] [] (show_stack+0x10/0x14) from [] (do_raw_spin_unlock+0x40/0xc8) [ 5.166125] [] (do_raw_spin_unlock+0x40/0xc8) from [] (_raw_spin_unlock+0x8/0x40) [ 5.175322] [] (_raw_spin_unlock+0x8/0x40) from [] (sirfsoc_uart_pio_rx_chars+0xa4/0xc0) [ 5.185120] [] (sirfsoc_uart_pio_rx_chars+0xa4/0xc0) from [] (sirfsoc_rx_tmo_process_tl+0xdc/0x1e0) [ 5.195875] [] (sirfsoc_rx_tmo_process_tl+0xdc/0x1e0) from [] (tasklet_action+0x8c/0xec) [ 5.205673] [] (tasklet_action+0x8c/0xec) from [] (__do_softirq+0xec/0x1d4) [ 5.214347] [] (__do_softirq+0xec/0x1d4) from [] (do_softirq+0x48/0x54) [ 5.222674] [] (do_softirq+0x48/0x54) from [] (irq_exit+0x74/0xc0) [ 5.230573] [] (irq_exit+0x74/0xc0) from [] (handle_IRQ+0x6c/0x90) [ 5.238465] [] (handle_IRQ+0x6c/0x90) from [] (__irq_svc+0x40/0x70) [ 5.246446] [] (__irq_svc+0x40/0x70) from [] (mark_page_accessed+0xc/0x68) [ 5.255034] [] (mark_page_accessed+0xc/0x68) from [] (unmap_single_vma+0x3bc/0x550) [ 5.264402] [] (unmap_single_vma+0x3bc/0x550) from [] (unmap_vmas+0x44/0x54) [ 5.273164] [] (unmap_vmas+0x44/0x54) from [] (exit_mmap+0xc4/0x1e0) [ 5.281233] [] (exit_mmap+0xc4/0x1e0) from [] (mmput+0x3c/0xdc) [ 5.288868] [] (mmput+0x3c/0xdc) from [] (do_exit+0x30c/0x828) [ 5.296413] [] (do_exit+0x30c/0x828) from [] (do_group_exit+0x4c/0xb0) [ 5.304653] [] (do_group_exit+0x4c/0xb0) from [] (__wake_up_parent+0x0/0x18) Root cause: the commit dropped uart_port->lock before calling tty_flip_buffer_push(), but in sirfsoc-uart, sirfsoc_uart_pio_rx_chars() can be called by sirfsoc_rx_tmo_process_tl(). here uart_port->lock has not been taken yet. so that caused unpaired lock/unlock. Solution: This patch is doing a quick fix for that, it adds spin_lock/unlock(&port->lock) protect to sirfsoc_uart_pio_rx_chars() in sirfsoc_rx_tmo_process_tl() to keep spin_lock/unlock in pair. Signed-off-by: Qipan Li Signed-off-by: Barry Song Cc: stable # 3.12 Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/sirfsoc_uart.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/tty/serial/sirfsoc_uart.c b/drivers/tty/serial/sirfsoc_uart.c index 49a2ffd101a7..b7bfe24d4ebc 100644 --- a/drivers/tty/serial/sirfsoc_uart.c +++ b/drivers/tty/serial/sirfsoc_uart.c @@ -542,8 +542,10 @@ static void sirfsoc_rx_tmo_process_tl(unsigned long param) wr_regl(port, ureg->sirfsoc_rx_dma_io_ctrl, rd_regl(port, ureg->sirfsoc_rx_dma_io_ctrl) | SIRFUART_IO_MODE); - sirfsoc_uart_pio_rx_chars(port, 4 - sirfport->rx_io_count); spin_unlock_irqrestore(&sirfport->rx_lock, flags); + spin_lock(&port->lock); + sirfsoc_uart_pio_rx_chars(port, 4 - sirfport->rx_io_count); + spin_unlock(&port->lock); if (sirfport->rx_io_count == 4) { spin_lock_irqsave(&sirfport->rx_lock, flags); sirfport->rx_io_count = 0; From e2613be5093d04e6589924d36a1e363eef3c87c7 Mon Sep 17 00:00:00 2001 From: Peter Hurley Date: Tue, 11 Feb 2014 16:34:55 -0500 Subject: [PATCH 555/596] n_tty: Fix stale echo output When echoes cannot be flushed to output (usually because the tty has no more write room) and L_ECHO is subsequently turned off, then when L_ECHO is turned back on, stale echoes are output. Output completed echoes regardless of the L_ECHO setting: 1. before normal writes to that tty 2. if the tty was stopped by soft flow control and is being restarted Reported-by: Mikulas Patocka Cc: # 3.13.x Signed-off-by: Peter Hurley Signed-off-by: Greg Kroah-Hartman --- drivers/tty/n_tty.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/drivers/tty/n_tty.c b/drivers/tty/n_tty.c index 2747a3b20a50..d15624c1b751 100644 --- a/drivers/tty/n_tty.c +++ b/drivers/tty/n_tty.c @@ -817,8 +817,7 @@ static void process_echoes(struct tty_struct *tty) struct n_tty_data *ldata = tty->disc_data; size_t echoed; - if ((!L_ECHO(tty) && !L_ECHONL(tty)) || - ldata->echo_mark == ldata->echo_tail) + if (ldata->echo_mark == ldata->echo_tail) return; mutex_lock(&ldata->output_lock); @@ -1244,7 +1243,8 @@ n_tty_receive_signal_char(struct tty_struct *tty, int signal, unsigned char c) if (L_ECHO(tty)) { echo_char(c, tty); commit_echoes(tty); - } + } else + process_echoes(tty); isig(signal, tty); return; } @@ -1274,7 +1274,7 @@ n_tty_receive_char_special(struct tty_struct *tty, unsigned char c) if (I_IXON(tty)) { if (c == START_CHAR(tty)) { start_tty(tty); - commit_echoes(tty); + process_echoes(tty); return 0; } if (c == STOP_CHAR(tty)) { @@ -1820,8 +1820,10 @@ static void n_tty_set_termios(struct tty_struct *tty, struct ktermios *old) * Fix tty hang when I_IXON(tty) is cleared, but the tty * been stopped by STOP_CHAR(tty) before it. */ - if (!I_IXON(tty) && old && (old->c_iflag & IXON) && !tty->flow_stopped) + if (!I_IXON(tty) && old && (old->c_iflag & IXON) && !tty->flow_stopped) { start_tty(tty); + process_echoes(tty); + } /* The termios change make the tty ready for I/O */ if (waitqueue_active(&tty->write_wait)) From 45a7bd635930baa960b6817674c64bdfb8f39570 Mon Sep 17 00:00:00 2001 From: Joe Schultz Date: Tue, 11 Feb 2014 18:30:01 -0600 Subject: [PATCH 556/596] serial: 8250: Support XR17V35x fraction divisor The Exar XR17V35x family of UARTs have an additional fractional divisor register (DLD) which was not being used. Calculate and set this register for these devices to reduce their baud rate error. Signed-off-by: Joe Schultz Signed-off-by: Aaron Sierra Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/8250/8250_core.c | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+) diff --git a/drivers/tty/serial/8250/8250_core.c b/drivers/tty/serial/8250/8250_core.c index 61ecd709a722..69932b7556cf 100644 --- a/drivers/tty/serial/8250/8250_core.c +++ b/drivers/tty/serial/8250/8250_core.c @@ -2432,6 +2432,24 @@ serial8250_do_set_termios(struct uart_port *port, struct ktermios *termios, serial_dl_write(up, quot); + /* + * XR17V35x UARTs have an extra fractional divisor register (DLD) + * + * We need to recalculate all of the registers, because DLM and DLL + * are already rounded to a whole integer. + * + * When recalculating we use a 32x clock instead of a 16x clock to + * allow 1-bit for rounding in the fractional part. + */ + if (up->port.type == PORT_XR17V35X) { + unsigned int baud_x32 = (port->uartclk * 2) / baud; + u16 quot = baud_x32 / 32; + u8 quot_frac = DIV_ROUND_CLOSEST(baud_x32 % 32, 2); + + serial_dl_write(up, quot); + serial_port_out(port, 0x2, quot_frac & 0xf); + } + /* * LCR DLAB must be set to enable 64-byte FIFO mode. If the FCR * is written without DLAB set, this mode will be disabled. From 0930b0950a8996aa88b0d2ba4bb2bab27cc36bc7 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Petr=20P=C3=ADsa=C5=99?= Date: Thu, 6 Feb 2014 21:01:23 +0100 Subject: [PATCH 557/596] vt: Fix secure clear screen MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit \E[3J console code (secure clear screen) needs to update_screen(vc) in order to write-through blanks into off-screen video memory. This has been removed accidentally in 3.6 by: commit 81732c3b2fede049a692e58a7ceabb6d18ffb18c Author: Jean-François Moine Date: Thu Sep 6 19:24:13 2012 +0200 tty vt: Fix line garbage in virtual console on command line edition Signed-off-by: Petr Písař Cc: stable # 3.6 Signed-off-by: Greg Kroah-Hartman --- drivers/tty/vt/vt.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/tty/vt/vt.c b/drivers/tty/vt/vt.c index 61b1137d7e56..23b5d32954bf 100644 --- a/drivers/tty/vt/vt.c +++ b/drivers/tty/vt/vt.c @@ -1164,6 +1164,8 @@ static void csi_J(struct vc_data *vc, int vpar) scr_memsetw(vc->vc_screenbuf, vc->vc_video_erase_char, vc->vc_screenbuf_size >> 1); set_origin(vc); + if (CON_IS_VISIBLE(vc)) + update_screen(vc); /* fall through */ case 2: /* erase whole display */ count = vc->vc_cols * vc->vc_rows; From 2ec197db1a56c9269d75e965f14c344b58b2a4f6 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Fri, 7 Feb 2014 17:10:26 +1100 Subject: [PATCH 558/596] lockd: send correct lock when granting a delayed lock. If an NFS client attempts to get a lock (using NLM) and the lock is not available, the server will remember the request and when the lock becomes available it will send a GRANT request to the client to provide the lock. If the client already held an adjacent lock, the GRANT callback will report the union of the existing and new locks, which can confuse the client. This happens because __posix_lock_file (called by vfs_lock_file) updates the passed-in file_lock structure when adjacent or over-lapping locks are found. To avoid this problem we take a copy of the two fields that can be changed (fl_start and fl_end) before the call and restore them afterwards. An alternate would be to allocate a 'struct file_lock', initialise it, use locks_copy_lock() to take a copy, then locks_release_private() after the vfs_lock_file() call. But that is a lot more work. Reported-by: Olaf Kirch Signed-off-by: NeilBrown Cc: stable@vger.kernel.org Signed-off-by: J. Bruce Fields -- v1 had a couple of issues (large on-stack struct and didn't really work properly). This version is much better tested. Signed-off-by: J. Bruce Fields --- fs/lockd/svclock.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/fs/lockd/svclock.c b/fs/lockd/svclock.c index e066a3902973..ab798a88ec1d 100644 --- a/fs/lockd/svclock.c +++ b/fs/lockd/svclock.c @@ -779,6 +779,7 @@ nlmsvc_grant_blocked(struct nlm_block *block) struct nlm_file *file = block->b_file; struct nlm_lock *lock = &block->b_call->a_args.lock; int error; + loff_t fl_start, fl_end; dprintk("lockd: grant blocked lock %p\n", block); @@ -796,9 +797,16 @@ nlmsvc_grant_blocked(struct nlm_block *block) } /* Try the lock operation again */ + /* vfs_lock_file() can mangle fl_start and fl_end, but we need + * them unchanged for the GRANT_MSG + */ lock->fl.fl_flags |= FL_SLEEP; + fl_start = lock->fl.fl_start; + fl_end = lock->fl.fl_end; error = vfs_lock_file(file->f_file, F_SETLK, &lock->fl, NULL); lock->fl.fl_flags &= ~FL_SLEEP; + lock->fl.fl_start = fl_start; + lock->fl.fl_end = fl_end; switch (error) { case 0: From dd5fd9b91a77b4c9c28b7ef9c181b1a875820d0a Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 11 Feb 2014 14:35:40 +0100 Subject: [PATCH 559/596] tick: Clear broadcast pending bit when switching to oneshot AMD systems which use the C1E workaround in the amd_e400_idle routine trigger the WARN_ON_ONCE in the broadcast code when onlining a CPU. The reason is that the idle routine of those AMD systems switches the cpu into forced broadcast mode early on before the newly brought up CPU can switch over to high resolution / NOHZ mode. The timer related CPU1 bringup looks like this: clockevent_register_device(local_apic); tick_setup(local_apic); ... idle() tick_broadcast_on_off(FORCE); tick_broadcast_oneshot_control(ENTER) cpumask_set(cpu, broadcast_oneshot_mask); halt(); Now the broadcast interrupt on CPU0 sets CPU1 in the broadcast_pending_mask and wakes CPU1. So CPU1 continues: local_apic_timer_interrupt() tick_handle_periodic(); softirq() tick_init_highres(); cpumask_clr(cpu, broadcast_oneshot_mask); tick_broadcast_oneshot_control(ENTER) WARN_ON(cpumask_test(cpu, broadcast_pending_mask); So while we remove CPU1 from the broadcast_oneshot_mask when we switch over to highres mode, we do not clear the pending bit, which then triggers the warning when we go back to idle. The reason why this is only visible on C1E affected AMD systems is that the other machines enter the deep sleep states via acpi_idle/intel_idle and exit the broadcast mode before executing the remote triggered local_apic_timer_interrupt. So the pending bit is already cleared when the switch over to highres mode is clearing the oneshot mask. The solution is simple: Clear the pending bit together with the mask bit when we switch over to highres mode. Stanislaw came up independently with the same patch by enforcing the C1E workaround and debugging the fallout. I picked mine, because mine has a changelog :) Reported-by: poma Debugged-by: Stanislaw Gruszka Signed-off-by: Thomas Gleixner Cc: Olaf Hering Cc: Dave Jones Cc: Justin M. Forbes Cc: Josh Boyer Link: http://lkml.kernel.org/r/alpine.DEB.2.02.1402111434180.21991@ionos.tec.linutronix.de Cc: stable@vger.kernel.org # 3.10+ Signed-off-by: Thomas Gleixner --- kernel/time/tick-broadcast.c | 1 + 1 file changed, 1 insertion(+) diff --git a/kernel/time/tick-broadcast.c b/kernel/time/tick-broadcast.c index 43780ab5e279..98977a57ac72 100644 --- a/kernel/time/tick-broadcast.c +++ b/kernel/time/tick-broadcast.c @@ -756,6 +756,7 @@ void tick_broadcast_oneshot_control(unsigned long reason) static void tick_broadcast_clear_oneshot(int cpu) { cpumask_clear_cpu(cpu, tick_broadcast_oneshot_mask); + cpumask_clear_cpu(cpu, tick_broadcast_pending_mask); } static void tick_broadcast_init_next_event(struct cpumask *mask, From acc4fccf4eff5b29e545995b75de77e60ea44aae Mon Sep 17 00:00:00 2001 From: Moni Shoua Date: Wed, 5 Feb 2014 15:12:58 +0200 Subject: [PATCH 560/596] IB/mlx4: Make sure GID index 0 is always occupied Make sure that for Ethernet ports, the port GID table index 0 is always occupied with a default GID of the relevant IPv6 link-local adderss. This provides better user experience for legacy applications that don't use the RDMA CM and were working on index 0 prior to the IP addressing change. Also, as GIDs are generated from IP addresses of the network devices that are associated with the port, it's basically possible that the GID table will be empty if no IP address was assigned. This doesn't comply with the IB spec section 4.1.1 "GID usage and properties". Signed-off-by: Moni Shoua Signed-off-by: Or Gerlitz Signed-off-by: Roland Dreier --- drivers/infiniband/hw/mlx4/main.c | 92 +++++++++++++++++++++++-------- 1 file changed, 68 insertions(+), 24 deletions(-) diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c index 64ca4087fb52..b99d0ecc6a89 100644 --- a/drivers/infiniband/hw/mlx4/main.c +++ b/drivers/infiniband/hw/mlx4/main.c @@ -1357,6 +1357,21 @@ static struct device_attribute *mlx4_class_attributes[] = { &dev_attr_board_id }; +static void mlx4_addrconf_ifid_eui48(u8 *eui, u16 vlan_id, + struct net_device *dev) +{ + memcpy(eui, dev->dev_addr, 3); + memcpy(eui + 5, dev->dev_addr + 3, 3); + if (vlan_id < 0x1000) { + eui[3] = vlan_id >> 8; + eui[4] = vlan_id & 0xff; + } else { + eui[3] = 0xff; + eui[4] = 0xfe; + } + eui[0] ^= 2; +} + static void update_gids_task(struct work_struct *work) { struct update_gid_work *gw = container_of(work, struct update_gid_work, work); @@ -1425,7 +1440,8 @@ static void reset_gids_task(struct work_struct *work) } static int update_gid_table(struct mlx4_ib_dev *dev, int port, - union ib_gid *gid, int clear) + union ib_gid *gid, int clear, + int default_gid) { struct update_gid_work *work; int i; @@ -1434,26 +1450,31 @@ static int update_gid_table(struct mlx4_ib_dev *dev, int port, int found = -1; int max_gids; - max_gids = dev->dev->caps.gid_table_len[port]; - for (i = 0; i < max_gids; ++i) { - if (!memcmp(&dev->iboe.gid_table[port - 1][i], gid, - sizeof(*gid))) - found = i; - - if (clear) { - if (found >= 0) { - need_update = 1; - dev->iboe.gid_table[port - 1][found] = zgid; - break; - } - } else { - if (found >= 0) - break; - - if (free < 0 && - !memcmp(&dev->iboe.gid_table[port - 1][i], &zgid, + if (default_gid) { + free = 0; + } else { + max_gids = dev->dev->caps.gid_table_len[port]; + for (i = 1; i < max_gids; ++i) { + if (!memcmp(&dev->iboe.gid_table[port - 1][i], gid, sizeof(*gid))) - free = i; + found = i; + + if (clear) { + if (found >= 0) { + need_update = 1; + dev->iboe.gid_table[port - 1][found] = + zgid; + break; + } + } else { + if (found >= 0) + break; + + if (free < 0 && + !memcmp(&dev->iboe.gid_table[port - 1][i], + &zgid, sizeof(*gid))) + free = i; + } } } @@ -1478,6 +1499,13 @@ static int update_gid_table(struct mlx4_ib_dev *dev, int port, return 0; } +static void mlx4_make_default_gid(struct net_device *dev, union ib_gid *gid) +{ + gid->global.subnet_prefix = cpu_to_be64(0xfe80000000000000LL); + mlx4_addrconf_ifid_eui48(&gid->raw[8], 0xffff, dev); +} + + static int reset_gid_table(struct mlx4_ib_dev *dev) { struct update_gid_work *work; @@ -1502,6 +1530,12 @@ static int mlx4_ib_addr_event(int event, struct net_device *event_netdev, struct net_device *real_dev = rdma_vlan_dev_real_dev(event_netdev) ? rdma_vlan_dev_real_dev(event_netdev) : event_netdev; + union ib_gid default_gid; + + mlx4_make_default_gid(real_dev, &default_gid); + + if (!memcmp(gid, &default_gid, sizeof(*gid))) + return 0; if (event != NETDEV_DOWN && event != NETDEV_UP) return 0; @@ -1520,7 +1554,7 @@ static int mlx4_ib_addr_event(int event, struct net_device *event_netdev, (!netif_is_bond_master(real_dev) && (real_dev == iboe->netdevs[port - 1]))) update_gid_table(ibdev, port, gid, - event == NETDEV_DOWN); + event == NETDEV_DOWN, 0); spin_unlock(&iboe->lock); return 0; @@ -1607,7 +1641,7 @@ static void mlx4_ib_get_dev_addr(struct net_device *dev, /*ifa->ifa_address;*/ ipv6_addr_set_v4mapped(ifa->ifa_address, (struct in6_addr *)&gid); - update_gid_table(ibdev, port, &gid, 0); + update_gid_table(ibdev, port, &gid, 0, 0); } endfor_ifa(in_dev); in_dev_put(in_dev); @@ -1619,7 +1653,7 @@ static void mlx4_ib_get_dev_addr(struct net_device *dev, read_lock_bh(&in6_dev->lock); list_for_each_entry(ifp, &in6_dev->addr_list, if_list) { pgid = (union ib_gid *)&ifp->addr; - update_gid_table(ibdev, port, pgid, 0); + update_gid_table(ibdev, port, pgid, 0, 0); } read_unlock_bh(&in6_dev->lock); in6_dev_put(in6_dev); @@ -1627,6 +1661,14 @@ static void mlx4_ib_get_dev_addr(struct net_device *dev, #endif } +static void mlx4_ib_set_default_gid(struct mlx4_ib_dev *ibdev, + struct net_device *dev, u8 port) +{ + union ib_gid gid; + mlx4_make_default_gid(dev, &gid); + update_gid_table(ibdev, port, &gid, 0, 1); +} + static int mlx4_ib_init_gid_table(struct mlx4_ib_dev *ibdev) { struct net_device *dev; @@ -1660,7 +1702,9 @@ static void mlx4_ib_scan_netdevs(struct mlx4_ib_dev *ibdev) struct net_device *curr_master; iboe->netdevs[port - 1] = mlx4_get_protocol_dev(ibdev->dev, MLX4_PROT_ETH, port); - + if (iboe->netdevs[port - 1]) + mlx4_ib_set_default_gid(ibdev, + iboe->netdevs[port - 1], port); if (iboe->netdevs[port - 1] && netif_is_bond_slave(iboe->netdevs[port - 1])) { rtnl_lock(); From 4ce5a5744a2f5479e58c6788cbe3987b8071b62e Mon Sep 17 00:00:00 2001 From: Moni Shoua Date: Wed, 5 Feb 2014 15:12:59 +0200 Subject: [PATCH 561/596] IB/mlx4: Move rtnl locking to the right place On the one hand, the invocation of netdev_master_upper_dev_get() within mlx4_ib_scan_netdevs() must be done with rtnl lock held. On the other hand, it's wrong to call rtnl_lock() from within this function since it's also called by our netdev notifier callback. Therefore move the locking to mlx4_ib_add() so that both cases are covered. Signed-off-by: Moni Shoua Signed-off-by: Or Gerlitz Signed-off-by: Roland Dreier --- drivers/infiniband/hw/mlx4/main.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c index b99d0ecc6a89..93f492f62997 100644 --- a/drivers/infiniband/hw/mlx4/main.c +++ b/drivers/infiniband/hw/mlx4/main.c @@ -1707,10 +1707,8 @@ static void mlx4_ib_scan_netdevs(struct mlx4_ib_dev *ibdev) iboe->netdevs[port - 1], port); if (iboe->netdevs[port - 1] && netif_is_bond_slave(iboe->netdevs[port - 1])) { - rtnl_lock(); iboe->masters[port - 1] = netdev_master_upper_dev_get( iboe->netdevs[port - 1]); - rtnl_unlock(); } curr_master = iboe->masters[port - 1]; @@ -2100,7 +2098,9 @@ static void *mlx4_ib_add(struct mlx4_dev *dev) } } #endif + rtnl_lock(); mlx4_ib_scan_netdevs(ibdev); + rtnl_unlock(); mlx4_ib_init_gid_table(ibdev); } From ddf8bd349115c2bc85a62e3d94018c9976ac72f7 Mon Sep 17 00:00:00 2001 From: Moni Shoua Date: Wed, 5 Feb 2014 15:13:00 +0200 Subject: [PATCH 562/596] IB/mlx4: Do IBoE locking earlier when initializing the GID table Updating the GID table under IBoE requires read/write from/to shared data structures. These data structures are protected with the device iboe lock. The flows that modify the GID table start from 1. Initializing the GID table 2. NETDEV events 3. INET or INET6 events This patch makes sure that the flow of initializing the GID table is consistent with the other two flows w.r.t on what step the lock is taken. Signed-off-by: Moni Shoua Signed-off-by: Or Gerlitz Signed-off-by: Roland Dreier --- drivers/infiniband/hw/mlx4/main.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c index 93f492f62997..3bafd0bebd4c 100644 --- a/drivers/infiniband/hw/mlx4/main.c +++ b/drivers/infiniband/hw/mlx4/main.c @@ -1570,7 +1570,6 @@ static u8 mlx4_ib_get_dev_port(struct net_device *dev, rdma_vlan_dev_real_dev(dev) : dev; iboe = &ibdev->iboe; - spin_lock(&iboe->lock); for (port = 1; port <= MLX4_MAX_PORTS; ++port) if ((netif_is_bond_master(real_dev) && @@ -1579,8 +1578,6 @@ static u8 mlx4_ib_get_dev_port(struct net_device *dev, (real_dev == iboe->netdevs[port - 1]))) break; - spin_unlock(&iboe->lock); - if ((port == 0) || (port > MLX4_MAX_PORTS)) return 0; else @@ -1672,11 +1669,13 @@ static void mlx4_ib_set_default_gid(struct mlx4_ib_dev *ibdev, static int mlx4_ib_init_gid_table(struct mlx4_ib_dev *ibdev) { struct net_device *dev; + struct mlx4_ib_iboe *iboe = &ibdev->iboe; if (reset_gid_table(ibdev)) return -1; read_lock(&dev_base_lock); + spin_lock(&iboe->lock); for_each_netdev(&init_net, dev) { u8 port = mlx4_ib_get_dev_port(dev, ibdev); @@ -1684,6 +1683,7 @@ static int mlx4_ib_init_gid_table(struct mlx4_ib_dev *ibdev) mlx4_ib_get_dev_addr(dev, ibdev, port); } + spin_unlock(&iboe->lock); read_unlock(&dev_base_lock); return 0; From 5071456fe244e409adcda7226e5f0b5d1b879fd3 Mon Sep 17 00:00:00 2001 From: Moni Shoua Date: Wed, 5 Feb 2014 15:13:01 +0200 Subject: [PATCH 563/596] IB/mlx4: Do IBoE GID table resets per-port The IBoE code used to reset the GID table did it for all Ethernet ports of the device. Since the whole architecture of generating GIDs and responding to events is port-based, this is inefficient and can lead to wrong content in the GID table. Change the reset flow to be per-port. Signed-off-by: Moni Shoua Signed-off-by: Or Gerlitz Signed-off-by: Roland Dreier --- drivers/infiniband/hw/mlx4/main.c | 35 ++++++++++++++++--------------- 1 file changed, 18 insertions(+), 17 deletions(-) diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c index 3bafd0bebd4c..8e10630561c1 100644 --- a/drivers/infiniband/hw/mlx4/main.c +++ b/drivers/infiniband/hw/mlx4/main.c @@ -1408,7 +1408,6 @@ static void reset_gids_task(struct work_struct *work) struct mlx4_cmd_mailbox *mailbox; union ib_gid *gids; int err; - int i; struct mlx4_dev *dev = gw->dev->dev; mailbox = mlx4_alloc_cmd_mailbox(dev); @@ -1420,18 +1419,16 @@ static void reset_gids_task(struct work_struct *work) gids = mailbox->buf; memcpy(gids, gw->gids, sizeof(gw->gids)); - for (i = 1; i < gw->dev->num_ports + 1; i++) { - if (mlx4_ib_port_link_layer(&gw->dev->ib_dev, i) == - IB_LINK_LAYER_ETHERNET) { - err = mlx4_cmd(dev, mailbox->dma, - MLX4_SET_PORT_GID_TABLE << 8 | i, - 1, MLX4_CMD_SET_PORT, - MLX4_CMD_TIME_CLASS_B, - MLX4_CMD_WRAPPED); - if (err) - pr_warn(KERN_WARNING - "set port %d command failed\n", i); - } + if (mlx4_ib_port_link_layer(&gw->dev->ib_dev, gw->port) == + IB_LINK_LAYER_ETHERNET) { + err = mlx4_cmd(dev, mailbox->dma, + MLX4_SET_PORT_GID_TABLE << 8 | gw->port, + 1, MLX4_CMD_SET_PORT, + MLX4_CMD_TIME_CLASS_B, + MLX4_CMD_WRAPPED); + if (err) + pr_warn(KERN_WARNING + "set port %d command failed\n", gw->port); } mlx4_free_cmd_mailbox(dev, mailbox); @@ -1506,7 +1503,7 @@ static void mlx4_make_default_gid(struct net_device *dev, union ib_gid *gid) } -static int reset_gid_table(struct mlx4_ib_dev *dev) +static int reset_gid_table(struct mlx4_ib_dev *dev, u8 port) { struct update_gid_work *work; @@ -1514,10 +1511,12 @@ static int reset_gid_table(struct mlx4_ib_dev *dev) work = kzalloc(sizeof(*work), GFP_ATOMIC); if (!work) return -ENOMEM; - memset(dev->iboe.gid_table, 0, sizeof(dev->iboe.gid_table)); + + memset(dev->iboe.gid_table[port - 1], 0, sizeof(work->gids)); memset(work->gids, 0, sizeof(work->gids)); INIT_WORK(&work->work, reset_gids_task); work->dev = dev; + work->port = port; queue_work(wq, &work->work); return 0; } @@ -1670,9 +1669,11 @@ static int mlx4_ib_init_gid_table(struct mlx4_ib_dev *ibdev) { struct net_device *dev; struct mlx4_ib_iboe *iboe = &ibdev->iboe; + int i; - if (reset_gid_table(ibdev)) - return -1; + for (i = 1; i <= ibdev->num_ports; ++i) + if (reset_gid_table(ibdev, i)) + return -1; read_lock(&dev_base_lock); spin_lock(&iboe->lock); From ad4885d279b63c65347220236d07669a2f59634b Mon Sep 17 00:00:00 2001 From: Moni Shoua Date: Wed, 5 Feb 2014 15:13:02 +0200 Subject: [PATCH 564/596] IB/mlx4: Build the port IBoE GID table properly under bonding When scanning netdevices we need to check a few more conditions and cases to build the IBoE GID table properly. For example, under bonding we must make sure that when a port is down, the bond IP address isn't programmed as a GID, since doing so will cause failure with IB core flows that selects ports by GID. Signed-off-by: Moni Shoua Signed-off-by: Or Gerlitz Signed-off-by: Roland Dreier --- drivers/infiniband/hw/mlx4/main.c | 41 ++++++++++++++++++++++++++++--- 1 file changed, 37 insertions(+), 4 deletions(-) diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c index 8e10630561c1..fdb5c7c9e127 100644 --- a/drivers/infiniband/hw/mlx4/main.c +++ b/drivers/infiniband/hw/mlx4/main.c @@ -1507,7 +1507,6 @@ static int reset_gid_table(struct mlx4_ib_dev *dev, u8 port) { struct update_gid_work *work; - work = kzalloc(sizeof(*work), GFP_ATOMIC); if (!work) return -ENOMEM; @@ -1699,25 +1698,57 @@ static void mlx4_ib_scan_netdevs(struct mlx4_ib_dev *ibdev) spin_lock(&iboe->lock); mlx4_foreach_ib_transport_port(port, ibdev->dev) { + enum ib_port_state port_state = IB_PORT_NOP; struct net_device *old_master = iboe->masters[port - 1]; + struct net_device *curr_netdev; struct net_device *curr_master; + iboe->netdevs[port - 1] = mlx4_get_protocol_dev(ibdev->dev, MLX4_PROT_ETH, port); if (iboe->netdevs[port - 1]) mlx4_ib_set_default_gid(ibdev, iboe->netdevs[port - 1], port); + curr_netdev = iboe->netdevs[port - 1]; + if (iboe->netdevs[port - 1] && netif_is_bond_slave(iboe->netdevs[port - 1])) { iboe->masters[port - 1] = netdev_master_upper_dev_get( iboe->netdevs[port - 1]); + } else { + iboe->masters[port - 1] = NULL; } curr_master = iboe->masters[port - 1]; + if (curr_netdev) { + port_state = (netif_running(curr_netdev) && netif_carrier_ok(curr_netdev)) ? + IB_PORT_ACTIVE : IB_PORT_DOWN; + mlx4_ib_set_default_gid(ibdev, curr_netdev, port); + } else { + reset_gid_table(ibdev, port); + } + /* if using bonding/team and a slave port is down, we don't the bond IP + * based gids in the table since flows that select port by gid may get + * the down port. + */ + if (curr_master && (port_state == IB_PORT_DOWN)) { + reset_gid_table(ibdev, port); + mlx4_ib_set_default_gid(ibdev, curr_netdev, port); + } /* if bonding is used it is possible that we add it to masters - only after IP address is assigned to the net bonding - interface */ - if (curr_master && (old_master != curr_master)) + * only after IP address is assigned to the net bonding + * interface. + */ + if (curr_master && (old_master != curr_master)) { + reset_gid_table(ibdev, port); + mlx4_ib_set_default_gid(ibdev, curr_netdev, port); mlx4_ib_get_dev_addr(curr_master, ibdev, port); + } + + if (!curr_master && (old_master != curr_master)) { + reset_gid_table(ibdev, port); + mlx4_ib_set_default_gid(ibdev, curr_netdev, port); + mlx4_ib_get_dev_addr(curr_netdev, ibdev, port); + } } spin_unlock(&iboe->lock); @@ -2099,6 +2130,8 @@ static void *mlx4_ib_add(struct mlx4_dev *dev) } } #endif + for (i = 1 ; i <= ibdev->num_ports ; ++i) + reset_gid_table(ibdev, i); rtnl_lock(); mlx4_ib_scan_netdevs(ibdev); rtnl_unlock(); From b4a26a27287a7f81933ba016aeed6c69dd155323 Mon Sep 17 00:00:00 2001 From: Moni Shoua Date: Sun, 9 Feb 2014 11:54:34 +0200 Subject: [PATCH 565/596] IB: Report using RoCE IP based gids in port caps For userspace RoCE UD QPs we need to know the GID format that the kernel uses, e.g when working over older kernels. For that end, add a new port capability IB_PORT_IP_BASED_GIDS and report it when query port is issued. Signed-off-by: Moni Shoua Signed-off-by: Matan Barak Signed-off-by: Or Gerlitz Signed-off-by: Roland Dreier --- drivers/infiniband/hw/mlx4/main.c | 2 +- drivers/infiniband/hw/ocrdma/ocrdma_verbs.c | 2 +- include/rdma/ib_verbs.h | 3 ++- 3 files changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c index c2702f549f10..5d4088ad6bd5 100644 --- a/drivers/infiniband/hw/mlx4/main.c +++ b/drivers/infiniband/hw/mlx4/main.c @@ -347,7 +347,7 @@ static int eth_link_query_port(struct ib_device *ibdev, u8 port, props->active_width = (((u8 *)mailbox->buf)[5] == 0x40) ? IB_WIDTH_4X : IB_WIDTH_1X; props->active_speed = IB_SPEED_QDR; - props->port_cap_flags = IB_PORT_CM_SUP; + props->port_cap_flags = IB_PORT_CM_SUP | IB_PORT_IP_BASED_GIDS; props->gid_tbl_len = mdev->dev->caps.gid_table_len[port]; props->max_msg_sz = mdev->dev->caps.max_msg_sz; props->pkey_tbl_len = 1; diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c b/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c index aa92f40c9d50..7890e7b14d81 100644 --- a/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c +++ b/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c @@ -176,7 +176,7 @@ int ocrdma_query_port(struct ib_device *ibdev, props->port_cap_flags = IB_PORT_CM_SUP | IB_PORT_REINIT_SUP | - IB_PORT_DEVICE_MGMT_SUP | IB_PORT_VENDOR_CLASS_SUP; + IB_PORT_DEVICE_MGMT_SUP | IB_PORT_VENDOR_CLASS_SUP | IB_PORT_IP_BASED_GIDS; props->gid_tbl_len = OCRDMA_MAX_SGID; props->pkey_tbl_len = 1; props->bad_pkey_cntr = 0; diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h index 8d4a1c06f7e4..6793f32ccb58 100644 --- a/include/rdma/ib_verbs.h +++ b/include/rdma/ib_verbs.h @@ -226,7 +226,8 @@ enum ib_port_cap_flags { IB_PORT_CAP_MASK_NOTICE_SUP = 1 << 22, IB_PORT_BOOT_MGMT_SUP = 1 << 23, IB_PORT_LINK_LATENCY_SUP = 1 << 24, - IB_PORT_CLIENT_REG_SUP = 1 << 25 + IB_PORT_CLIENT_REG_SUP = 1 << 25, + IB_PORT_IP_BASED_GIDS = 1 << 26 }; enum ib_port_width { From 0f0132001fd239bb67c1f68436b95cc79de89736 Mon Sep 17 00:00:00 2001 From: Kumar Sanghvi Date: Thu, 6 Feb 2014 16:00:16 +0530 Subject: [PATCH 566/596] RDMA/cxgb4: Add missing neigh_release in LE-Workaround path Signed-off-by: Kumar Sanghvi Signed-off-by: Hariprasad Shenai Signed-off-by: Roland Dreier --- drivers/infiniband/hw/cxgb4/cm.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c index 45126879ad28..d286bdebe2ab 100644 --- a/drivers/infiniband/hw/cxgb4/cm.c +++ b/drivers/infiniband/hw/cxgb4/cm.c @@ -3352,6 +3352,7 @@ static int rx_pkt(struct c4iw_dev *dev, struct sk_buff *skb) goto free_dst; } + neigh_release(neigh); step = dev->rdev.lldi.nrxq / dev->rdev.lldi.nchan; rss_qid = dev->rdev.lldi.rxq_ids[pi->port_id * step]; window = (__force u16) htons((__force u16)tcph->window); From 2f75e12c4457a9b3d042c0a0d748fa198dc2ffaf Mon Sep 17 00:00:00 2001 From: Mike Marciniszyn Date: Wed, 12 Feb 2014 11:54:15 -0500 Subject: [PATCH 567/596] IB/qib: Add missing serdes init sequence Research has shown that commit a77fcf895046 ("IB/qib: Use a single txselect module parameter for serdes tuning") missed a key serdes init sequence. This patch add that sequence. Cc: Reviewed-by: Dennis Dalessandro Signed-off-by: Mike Marciniszyn Signed-off-by: Roland Dreier --- drivers/infiniband/hw/qib/qib_iba7322.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/infiniband/hw/qib/qib_iba7322.c b/drivers/infiniband/hw/qib/qib_iba7322.c index 5bfc02f450e6..d1bd21319d7d 100644 --- a/drivers/infiniband/hw/qib/qib_iba7322.c +++ b/drivers/infiniband/hw/qib/qib_iba7322.c @@ -2395,6 +2395,11 @@ static int qib_7322_bringup_serdes(struct qib_pportdata *ppd) qib_write_kreg_port(ppd, krp_ibcctrl_a, ppd->cpspec->ibcctrl_a); qib_write_kreg(dd, kr_scratch, 0ULL); + /* ensure previous Tx parameters are not still forced */ + qib_write_kreg_port(ppd, krp_tx_deemph_override, + SYM_MASK(IBSD_TX_DEEMPHASIS_OVERRIDE_0, + reset_tx_deemphasis_override)); + if (qib_compat_ddr_negotiate) { ppd->cpspec->ibdeltainprog = 1; ppd->cpspec->ibsymsnap = read_7322_creg32_port(ppd, From d3d89c468ceebbcf9423d1a3d66c5bf91f569570 Mon Sep 17 00:00:00 2001 From: Doug Anderson Date: Thu, 13 Feb 2014 14:39:34 -0800 Subject: [PATCH 568/596] hwmon: (ntc_thermistor) Avoid math overflow The ntc thermistor code was doing math whose temporary result might have overflowed 32-bits. We need some casts in there to make it safe. In one example I found: - pullup_uV: 1800000 - result of iio_read_channel_raw: 3226 - 1800000 * 3226 => 0x15a1cbc80 Signed-off-by: Doug Anderson Cc: stable@vger.kernel.org # 3.10+ Signed-off-by: Guenter Roeck --- drivers/hwmon/ntc_thermistor.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/hwmon/ntc_thermistor.c b/drivers/hwmon/ntc_thermistor.c index 8c23203915af..8a17f01e8672 100644 --- a/drivers/hwmon/ntc_thermistor.c +++ b/drivers/hwmon/ntc_thermistor.c @@ -145,7 +145,7 @@ struct ntc_data { static int ntc_adc_iio_read(struct ntc_thermistor_platform_data *pdata) { struct iio_channel *channel = pdata->chan; - unsigned int result; + s64 result; int val, ret; ret = iio_read_channel_raw(channel, &val); @@ -155,10 +155,10 @@ static int ntc_adc_iio_read(struct ntc_thermistor_platform_data *pdata) } /* unit: mV */ - result = pdata->pullup_uv * val; + result = pdata->pullup_uv * (s64) val; result >>= 12; - return result; + return (int)result; } static const struct of_device_id ntc_match[] = { From 6ecde51dd7894ffe2f959cca1fea3ea2b9ee2394 Mon Sep 17 00:00:00 2001 From: Roland Dreier Date: Thu, 13 Feb 2014 20:45:17 -0800 Subject: [PATCH 569/596] mlx5: Add include of because of kzalloc()/kfree() use On some architectures (for example, arm), we don't end up indirectly pulling in the declaration of kzalloc() and kfree(), and so building anything that includes breaks. Fix this by adding an explicit include to get the declaration. Reported-by: kbuild test robot Signed-off-by: Roland Dreier --- include/linux/mlx5/driver.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index 32cb18c399c2..130bc8d77fa5 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -38,8 +38,10 @@ #include #include #include +#include #include #include + #include #include From 0861565f501ce3fcea9394d4b98c02b1f6de6b9e Mon Sep 17 00:00:00 2001 From: Eli Cohen Date: Fri, 14 Feb 2014 10:29:49 +1100 Subject: [PATCH 570/596] IB/mlx5: Remove dependency on X86 Remove Kconfig dependency of mlx5_ib/mlx5_core on X86, since there is no such dependency in reality. Signed-off-by: Eli Cohen Signed-off-by: Michael Neuling Signed-off-by: Roland Dreier --- drivers/infiniband/hw/mlx5/Kconfig | 2 +- drivers/net/ethernet/mellanox/mlx5/core/Kconfig | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/infiniband/hw/mlx5/Kconfig b/drivers/infiniband/hw/mlx5/Kconfig index 8e6aebfaf8a4..10df386c6344 100644 --- a/drivers/infiniband/hw/mlx5/Kconfig +++ b/drivers/infiniband/hw/mlx5/Kconfig @@ -1,6 +1,6 @@ config MLX5_INFINIBAND tristate "Mellanox Connect-IB HCA support" - depends on NETDEVICES && ETHERNET && PCI && X86 + depends on NETDEVICES && ETHERNET && PCI select NET_VENDOR_MELLANOX select MLX5_CORE ---help--- diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Kconfig b/drivers/net/ethernet/mellanox/mlx5/core/Kconfig index 157fe8df2c3e..8ff57e8e3e91 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/Kconfig +++ b/drivers/net/ethernet/mellanox/mlx5/core/Kconfig @@ -4,5 +4,5 @@ config MLX5_CORE tristate - depends on PCI && X86 + depends on PCI default n From 31d6af29066bfaf30cf4f90e7cc0ab9409cee71f Mon Sep 17 00:00:00 2001 From: Sachin Kamat Date: Wed, 12 Feb 2014 15:19:43 +0530 Subject: [PATCH 571/596] video: exynos: Fix S6E8AX0 LCD driver build error Enable S6E8AX0 LCD driver only if LCD_CLASS_DEVICE is a built-in driver. Else we get the following errors due to missing symbols: drivers/built-in.o: In function `s6e8ax0_probe': :(.text+0x51aec): undefined reference to `lcd_device_register' :(.text+0x51c44): undefined reference to `lcd_device_unregister' Signed-off-by: Sachin Kamat Signed-off-by: Tomi Valkeinen --- drivers/video/exynos/Kconfig | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/video/exynos/Kconfig b/drivers/video/exynos/Kconfig index 1129d0e9e640..75c8a8e7efc0 100644 --- a/drivers/video/exynos/Kconfig +++ b/drivers/video/exynos/Kconfig @@ -22,7 +22,8 @@ config EXYNOS_MIPI_DSI config EXYNOS_LCD_S6E8AX0 bool "S6E8AX0 MIPI AMOLED LCD Driver" - depends on (EXYNOS_MIPI_DSI && BACKLIGHT_CLASS_DEVICE && LCD_CLASS_DEVICE) + depends on EXYNOS_MIPI_DSI && BACKLIGHT_CLASS_DEVICE + depends on (LCD_CLASS_DEVICE = y) default n help If you have an S6E8AX0 MIPI AMOLED LCD Panel, say Y to enable its From b359bb0d9c5132abb6fec1ef555c48daf267de59 Mon Sep 17 00:00:00 2001 From: Denis Carikli Date: Wed, 22 Jan 2014 18:09:36 +0100 Subject: [PATCH 572/596] video: Kconfig: Allow more broad selection of the imxfb framebuffer driver. MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Without that patch, a user can't select the imxfb driver when the i.MX25 and/or the i.MX27 device tree board are selected and that no boards that selects IMX_HAVE_PLATFORM_IMX_FB are compiled in. Cc: Eric Bénard Cc: Jean-Christophe Plagniol-Villard Cc: Sascha Hauer Cc: Tomi Valkeinen Cc: linux-arm-kernel@lists.infradead.org Cc: linux-fbdev@vger.kernel.org Signed-off-by: Denis Carikli Acked-by: Shawn Guo Acked-by: Jean-Christophe PLAGNIOL-VILLARD Acked-by: Sascha Hauer Signed-off-by: Tomi Valkeinen --- drivers/video/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/video/Kconfig b/drivers/video/Kconfig index 22262a3a0e2d..dade5b7699bc 100644 --- a/drivers/video/Kconfig +++ b/drivers/video/Kconfig @@ -364,7 +364,7 @@ config FB_SA1100 config FB_IMX tristate "Freescale i.MX1/21/25/27 LCD support" - depends on FB && IMX_HAVE_PLATFORM_IMX_FB + depends on FB && ARCH_MXC select FB_CFB_FILLRECT select FB_CFB_COPYAREA select FB_CFB_IMAGEBLIT From c55d016f7a930dd1c995336017123b469a8c8f5a Mon Sep 17 00:00:00 2001 From: Borislav Petkov Date: Fri, 14 Feb 2014 08:24:24 +0100 Subject: [PATCH 573/596] x86/efi: Fix 32-bit fallout We do not enable the new efi memmap on 32-bit and thus we need to run runtime_code_page_mkexec() unconditionally there. Fix that. Reported-and-tested-by: Lejun Zhu Signed-off-by: Borislav Petkov Signed-off-by: Matt Fleming --- arch/x86/include/asm/efi.h | 2 ++ arch/x86/platform/efi/efi.c | 5 ++--- arch/x86/platform/efi/efi_32.c | 6 ++++++ arch/x86/platform/efi/efi_64.c | 9 +++++++++ 4 files changed, 19 insertions(+), 3 deletions(-) diff --git a/arch/x86/include/asm/efi.h b/arch/x86/include/asm/efi.h index 3b978c472d08..3d6b9f81cc68 100644 --- a/arch/x86/include/asm/efi.h +++ b/arch/x86/include/asm/efi.h @@ -132,6 +132,8 @@ extern void __init efi_map_region_fixed(efi_memory_desc_t *md); extern void efi_sync_low_kernel_mappings(void); extern void efi_setup_page_tables(void); extern void __init old_map_region(efi_memory_desc_t *md); +extern void __init runtime_code_page_mkexec(void); +extern void __init efi_runtime_mkexec(void); struct efi_setup_data { u64 fw_vendor; diff --git a/arch/x86/platform/efi/efi.c b/arch/x86/platform/efi/efi.c index d62ec87a2b26..1a201ac7cef8 100644 --- a/arch/x86/platform/efi/efi.c +++ b/arch/x86/platform/efi/efi.c @@ -792,7 +792,7 @@ void __init efi_set_executable(efi_memory_desc_t *md, bool executable) set_memory_nx(addr, npages); } -static void __init runtime_code_page_mkexec(void) +void __init runtime_code_page_mkexec(void) { efi_memory_desc_t *md; void *p; @@ -1069,8 +1069,7 @@ void __init efi_enter_virtual_mode(void) efi.update_capsule = virt_efi_update_capsule; efi.query_capsule_caps = virt_efi_query_capsule_caps; - if (efi_enabled(EFI_OLD_MEMMAP) && (__supported_pte_mask & _PAGE_NX)) - runtime_code_page_mkexec(); + efi_runtime_mkexec(); kfree(new_memmap); diff --git a/arch/x86/platform/efi/efi_32.c b/arch/x86/platform/efi/efi_32.c index 249b183cf417..0b74cdf7f816 100644 --- a/arch/x86/platform/efi/efi_32.c +++ b/arch/x86/platform/efi/efi_32.c @@ -77,3 +77,9 @@ void efi_call_phys_epilog(void) local_irq_restore(efi_rt_eflags); } + +void __init efi_runtime_mkexec(void) +{ + if (__supported_pte_mask & _PAGE_NX) + runtime_code_page_mkexec(); +} diff --git a/arch/x86/platform/efi/efi_64.c b/arch/x86/platform/efi/efi_64.c index 6284f158a47d..0c2a234fef1e 100644 --- a/arch/x86/platform/efi/efi_64.c +++ b/arch/x86/platform/efi/efi_64.c @@ -233,3 +233,12 @@ void __init parse_efi_setup(u64 phys_addr, u32 data_len) { efi_setup = phys_addr + sizeof(struct setup_data); } + +void __init efi_runtime_mkexec(void) +{ + if (!efi_enabled(EFI_OLD_MEMMAP)) + return; + + if (__supported_pte_mask & _PAGE_NX) + runtime_code_page_mkexec(); +} From 9da21b1509d8aa7ab4846722817d16c72d656c91 Mon Sep 17 00:00:00 2001 From: Borislav Petkov Date: Mon, 3 Feb 2014 15:05:13 -0500 Subject: [PATCH 574/596] EDAC: Poll timeout cannot be zero, p2 Sanitize code even more to accept unsigned longs only and to not allow polling intervals below 1 second as this is unnecessary and doesn't make much sense anyway for polling errors. Signed-off-by: Borislav Petkov Link: http://lkml.kernel.org/r/1391457913-881-1-git-send-email-prarit@redhat.com Cc: Doug Thompson Cc: --- drivers/edac/edac_mc.c | 4 ++-- drivers/edac/edac_mc_sysfs.c | 10 ++++++---- drivers/edac/edac_module.h | 2 +- 3 files changed, 9 insertions(+), 7 deletions(-) diff --git a/drivers/edac/edac_mc.c b/drivers/edac/edac_mc.c index e8c9ef03495b..aef5ec24908e 100644 --- a/drivers/edac/edac_mc.c +++ b/drivers/edac/edac_mc.c @@ -601,7 +601,7 @@ static void edac_mc_workq_teardown(struct mem_ctl_info *mci) * user space has updated our poll period value, need to * reset our workq delays */ -void edac_mc_reset_delay_period(int value) +void edac_mc_reset_delay_period(unsigned long value) { struct mem_ctl_info *mci; struct list_head *item; @@ -611,7 +611,7 @@ void edac_mc_reset_delay_period(int value) list_for_each(item, &mc_devices) { mci = list_entry(item, struct mem_ctl_info, link); - edac_mc_workq_setup(mci, (unsigned long) value); + edac_mc_workq_setup(mci, value); } mutex_unlock(&mem_ctls_mutex); diff --git a/drivers/edac/edac_mc_sysfs.c b/drivers/edac/edac_mc_sysfs.c index 8ec1747b1c39..b335c6ab5efe 100644 --- a/drivers/edac/edac_mc_sysfs.c +++ b/drivers/edac/edac_mc_sysfs.c @@ -52,18 +52,20 @@ int edac_mc_get_poll_msec(void) static int edac_set_poll_msec(const char *val, struct kernel_param *kp) { - long l; + unsigned long l; int ret; if (!val) return -EINVAL; - ret = kstrtol(val, 0, &l); + ret = kstrtoul(val, 0, &l); if (ret) return ret; - if (!l || ((int)l != l)) + + if (l < 1000) return -EINVAL; - *((int *)kp->arg) = l; + + *((unsigned long *)kp->arg) = l; /* notify edac_mc engine to reset the poll period */ edac_mc_reset_delay_period(l); diff --git a/drivers/edac/edac_module.h b/drivers/edac/edac_module.h index 3d139c6e7fe3..f2118bfcf8df 100644 --- a/drivers/edac/edac_module.h +++ b/drivers/edac/edac_module.h @@ -52,7 +52,7 @@ extern void edac_device_workq_setup(struct edac_device_ctl_info *edac_dev, extern void edac_device_workq_teardown(struct edac_device_ctl_info *edac_dev); extern void edac_device_reset_delay_period(struct edac_device_ctl_info *edac_dev, unsigned long value); -extern void edac_mc_reset_delay_period(int value); +extern void edac_mc_reset_delay_period(unsigned long value); extern void *edac_align_ptr(void **p, unsigned size, int n_elems); From cb6ef42e516cb8948f15e4b70dc03af8020050a2 Mon Sep 17 00:00:00 2001 From: Borislav Petkov Date: Wed, 12 Feb 2014 18:15:00 +0100 Subject: [PATCH 575/596] EDAC: Correct workqueue setup path We're using edac_mc_workq_setup() both on the init path, when we load an edac driver and when we change the polling period (edac_mc_reset_delay_period) through /sys/.../edac_mc_poll_msec. On that second path we don't need to init the workqueue which has been initialized already. Thanks to Tejun for workqueue insights. Signed-off-by: Borislav Petkov Link: http://lkml.kernel.org/r/1391457913-881-1-git-send-email-prarit@redhat.com Cc: --- drivers/edac/edac_mc.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/drivers/edac/edac_mc.c b/drivers/edac/edac_mc.c index aef5ec24908e..33edd6766344 100644 --- a/drivers/edac/edac_mc.c +++ b/drivers/edac/edac_mc.c @@ -559,7 +559,8 @@ static void edac_mc_workq_function(struct work_struct *work_req) * * called with the mem_ctls_mutex held */ -static void edac_mc_workq_setup(struct mem_ctl_info *mci, unsigned msec) +static void edac_mc_workq_setup(struct mem_ctl_info *mci, unsigned msec, + bool init) { edac_dbg(0, "\n"); @@ -567,7 +568,9 @@ static void edac_mc_workq_setup(struct mem_ctl_info *mci, unsigned msec) if (mci->op_state != OP_RUNNING_POLL) return; - INIT_DELAYED_WORK(&mci->work, edac_mc_workq_function); + if (init) + INIT_DELAYED_WORK(&mci->work, edac_mc_workq_function); + mod_delayed_work(edac_workqueue, &mci->work, msecs_to_jiffies(msec)); } @@ -611,7 +614,7 @@ void edac_mc_reset_delay_period(unsigned long value) list_for_each(item, &mc_devices) { mci = list_entry(item, struct mem_ctl_info, link); - edac_mc_workq_setup(mci, value); + edac_mc_workq_setup(mci, value, false); } mutex_unlock(&mem_ctls_mutex); @@ -782,7 +785,7 @@ int edac_mc_add_mc(struct mem_ctl_info *mci) /* This instance is NOW RUNNING */ mci->op_state = OP_RUNNING_POLL; - edac_mc_workq_setup(mci, edac_mc_get_poll_msec()); + edac_mc_workq_setup(mci, edac_mc_get_poll_msec(), true); } else { mci->op_state = OP_RUNNING_INTERRUPT; } From 09503379dc99535b1bbfa51aa1aeef340f5d82ec Mon Sep 17 00:00:00 2001 From: Matt Fleming Date: Thu, 13 Feb 2014 17:17:54 +0000 Subject: [PATCH 576/596] x86/efi: Check status field to validate BGRT header Madper reported seeing the following crash, BUG: unable to handle kernel paging request at ffffffffff340003 IP: [] efi_bgrt_init+0x9d/0x133 Call Trace: [] efi_late_init+0x9/0xb [] start_kernel+0x436/0x450 [] ? repair_env_string+0x5c/0x5c [] ? early_idt_handlers+0x120/0x120 [] x86_64_start_reservations+0x2a/0x2c [] x86_64_start_kernel+0x13e/0x14d This is caused because the layout of the ACPI BGRT header on this system doesn't match the definition from the ACPI spec, and so we get a bogus physical address when dereferencing ->image_address in efi_bgrt_init(). Luckily the status field in the BGRT header clearly marks it as invalid, so we can check that field and skip BGRT initialisation. Reported-by: Madper Xie Suggested-by: Toshi Kani Cc: "Rafael J. Wysocki" Cc: Matthew Garrett Cc: Josh Triplett Signed-off-by: Matt Fleming --- arch/x86/platform/efi/efi-bgrt.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/platform/efi/efi-bgrt.c b/arch/x86/platform/efi/efi-bgrt.c index 4df9591eadad..f15103dff4b4 100644 --- a/arch/x86/platform/efi/efi-bgrt.c +++ b/arch/x86/platform/efi/efi-bgrt.c @@ -42,7 +42,7 @@ void __init efi_bgrt_init(void) if (bgrt_tab->header.length < sizeof(*bgrt_tab)) return; - if (bgrt_tab->version != 1) + if (bgrt_tab->version != 1 || bgrt_tab->status != 1) return; if (bgrt_tab->image_type != 0 || !bgrt_tab->image_address) return; From 2a2f3e269c75edf916de5967079069aeb6a601cb Mon Sep 17 00:00:00 2001 From: Christoffer Dall Date: Sun, 2 Feb 2014 13:41:02 -0800 Subject: [PATCH 577/596] arm64: KVM: Add VGIC device control for arm64 This fixes the build breakage introduced by c07a0191ef2de1f9510f12d1f88e3b0b5cd8d66f and adds support for the device control API and save/restore of the VGIC state for ARMv8. The defines were simply missing from the arm64 header files and uaccess.h must be implicitly imported from somewhere else on arm. Signed-off-by: Christoffer Dall Signed-off-by: Paolo Bonzini --- arch/arm64/include/uapi/asm/kvm.h | 9 +++++++++ virt/kvm/arm/vgic.c | 1 + 2 files changed, 10 insertions(+) diff --git a/arch/arm64/include/uapi/asm/kvm.h b/arch/arm64/include/uapi/asm/kvm.h index 495ab6f84a61..eaf54a30bedc 100644 --- a/arch/arm64/include/uapi/asm/kvm.h +++ b/arch/arm64/include/uapi/asm/kvm.h @@ -148,6 +148,15 @@ struct kvm_arch_memory_slot { #define KVM_REG_ARM_TIMER_CNT ARM64_SYS_REG(3, 3, 14, 3, 2) #define KVM_REG_ARM_TIMER_CVAL ARM64_SYS_REG(3, 3, 14, 0, 2) +/* Device Control API: ARM VGIC */ +#define KVM_DEV_ARM_VGIC_GRP_ADDR 0 +#define KVM_DEV_ARM_VGIC_GRP_DIST_REGS 1 +#define KVM_DEV_ARM_VGIC_GRP_CPU_REGS 2 +#define KVM_DEV_ARM_VGIC_CPUID_SHIFT 32 +#define KVM_DEV_ARM_VGIC_CPUID_MASK (0xffULL << KVM_DEV_ARM_VGIC_CPUID_SHIFT) +#define KVM_DEV_ARM_VGIC_OFFSET_SHIFT 0 +#define KVM_DEV_ARM_VGIC_OFFSET_MASK (0xffffffffULL << KVM_DEV_ARM_VGIC_OFFSET_SHIFT) + /* KVM_IRQ_LINE irq field index values */ #define KVM_ARM_IRQ_TYPE_SHIFT 24 #define KVM_ARM_IRQ_TYPE_MASK 0xff diff --git a/virt/kvm/arm/vgic.c b/virt/kvm/arm/vgic.c index be456ce264d0..8ca405cd7c1a 100644 --- a/virt/kvm/arm/vgic.c +++ b/virt/kvm/arm/vgic.c @@ -24,6 +24,7 @@ #include #include #include +#include #include From f809309a251a13bd97cc189c3fa428782aab9716 Mon Sep 17 00:00:00 2001 From: Upinder Malhi Date: Thu, 23 Jan 2014 22:38:04 +0000 Subject: [PATCH 578/596] IB/usnic: Fix smatch endianness error Error reported at http://marc.info/?l=linux-rdma&m=138995755801039&w=2 Fix short to int cast for big endian systems. Signed-off-by: Upinder Malhi Signed-off-by: Roland Dreier --- drivers/infiniband/hw/usnic/usnic_ib_qp_grp.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/drivers/infiniband/hw/usnic/usnic_ib_qp_grp.c b/drivers/infiniband/hw/usnic/usnic_ib_qp_grp.c index 7ecc6061f1f4..f8dfd76be89f 100644 --- a/drivers/infiniband/hw/usnic/usnic_ib_qp_grp.c +++ b/drivers/infiniband/hw/usnic/usnic_ib_qp_grp.c @@ -629,6 +629,7 @@ static int qp_grp_id_from_flow(struct usnic_ib_qp_grp_flow *qp_flow, { enum usnic_transport_type trans_type = qp_flow->trans_type; int err; + uint16_t port_num = 0; switch (trans_type) { case USNIC_TRANSPORT_ROCE_CUSTOM: @@ -637,9 +638,15 @@ static int qp_grp_id_from_flow(struct usnic_ib_qp_grp_flow *qp_flow, case USNIC_TRANSPORT_IPV4_UDP: err = usnic_transport_sock_get_addr(qp_flow->udp.sock, NULL, NULL, - (uint16_t *) id); + &port_num); if (err) return err; + /* + * Copy port_num to stack first and then to *id, + * so that the short to int cast works for little + * and big endian systems. + */ + *id = port_num; break; default: usnic_err("Unsupported transport %u\n", trans_type); From 7d9eacf9457efc6b614665e1095336c11ad83f0d Mon Sep 17 00:00:00 2001 From: Roi Dayan Date: Tue, 4 Feb 2014 16:54:54 +0200 Subject: [PATCH 579/596] IB/iser: Avoid dereferencing iscsi_iser conn object when not bound to iser connection Fix a possible NULL pointer dereference in disconnection flow. This can happen if the target disconnected/rejected the connection request, e.g before the binding stage between iscsi connection to the transport connection. Signed-off-by: Alex Tabachnik Signed-off-by: Roi Dayan Signed-off-by: Or Gerlitz Signed-off-by: Roland Dreier --- drivers/infiniband/ulp/iser/iser_verbs.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/drivers/infiniband/ulp/iser/iser_verbs.c b/drivers/infiniband/ulp/iser/iser_verbs.c index afe95674008b..ca37edef2791 100644 --- a/drivers/infiniband/ulp/iser/iser_verbs.c +++ b/drivers/infiniband/ulp/iser/iser_verbs.c @@ -652,9 +652,13 @@ static int iser_disconnected_handler(struct rdma_cm_id *cma_id) /* getting here when the state is UP means that the conn is being * * terminated asynchronously from the iSCSI layer's perspective. */ if (iser_conn_state_comp_exch(ib_conn, ISER_CONN_UP, - ISER_CONN_TERMINATING)) - iscsi_conn_failure(ib_conn->iser_conn->iscsi_conn, - ISCSI_ERR_CONN_FAILED); + ISER_CONN_TERMINATING)){ + if (ib_conn->iser_conn) + iscsi_conn_failure(ib_conn->iser_conn->iscsi_conn, + ISCSI_ERR_CONN_FAILED); + else + iser_err("iscsi_iser connection isn't bound\n"); + } /* Complete the termination process if no posts are pending */ if (ib_conn->post_recv_buf_count == 0 && From fd8b48b22a2b7cdf21f15b01cae379e6159a7eea Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Wed, 29 Jan 2014 16:18:51 +0300 Subject: [PATCH 580/596] IB/iser: Fix use after free in iser_snd_completion() We use "tx_desc" again after we free it. Signed-off-by: Dan Carpenter Acked-by: Or Gerlitz Signed-off-by: Roland Dreier --- drivers/infiniband/ulp/iser/iser_initiator.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/infiniband/ulp/iser/iser_initiator.c b/drivers/infiniband/ulp/iser/iser_initiator.c index 538822684d5b..334f34b1cd46 100644 --- a/drivers/infiniband/ulp/iser/iser_initiator.c +++ b/drivers/infiniband/ulp/iser/iser_initiator.c @@ -610,11 +610,12 @@ void iser_snd_completion(struct iser_tx_desc *tx_desc, ib_dma_unmap_single(device->ib_device, tx_desc->dma_addr, ISER_HEADERS_LEN, DMA_TO_DEVICE); kmem_cache_free(ig.desc_cache, tx_desc); + tx_desc = NULL; } atomic_dec(&ib_conn->post_send_buf_count); - if (tx_desc->type == ISCSI_TX_CONTROL) { + if (tx_desc && tx_desc->type == ISCSI_TX_CONTROL) { /* this arithmetic is legal by libiscsi dd_data allocation */ task = (void *) ((long)(void *)tx_desc - sizeof(struct iscsi_task)); From a61d93d92f5c9533898098abb5f187840900aeb5 Mon Sep 17 00:00:00 2001 From: Devesh Sharma Date: Mon, 10 Feb 2014 13:48:58 +0530 Subject: [PATCH 581/596] RDMA/ocrdma: Fix traffic class shift Use correct value for obtaining traffic class from device response for Query QP request. Signed-off-by: Devesh Sharma Reported-by: Dan Carpenter Signed-off-by: Roland Dreier --- drivers/infiniband/hw/ocrdma/ocrdma_verbs.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c b/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c index aa92f40c9d50..77e38ecd634e 100644 --- a/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c +++ b/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c @@ -1416,7 +1416,7 @@ int ocrdma_query_qp(struct ib_qp *ibqp, OCRDMA_QP_PARAMS_HOP_LMT_MASK) >> OCRDMA_QP_PARAMS_HOP_LMT_SHIFT; qp_attr->ah_attr.grh.traffic_class = (params.tclass_sq_psn & - OCRDMA_QP_PARAMS_SQ_PSN_MASK) >> + OCRDMA_QP_PARAMS_TCLASS_MASK) >> OCRDMA_QP_PARAMS_TCLASS_SHIFT; qp_attr->ah_attr.ah_flags = IB_AH_GRH; From 09de3f1313a30d8a22e488c9a5b96a9560cae96d Mon Sep 17 00:00:00 2001 From: Devesh Sharma Date: Tue, 4 Feb 2014 12:10:48 +0530 Subject: [PATCH 582/596] RDMA/ocrdma: Fix load time panic during GID table init We should use rdma_vlan_dev_real_dev() instead of using vlan_dev_real_dev() when building the GID table for a vlan interface. Signed-off-by: Devesh Sharma Signed-off-by: Roland Dreier --- drivers/infiniband/hw/ocrdma/ocrdma_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_main.c b/drivers/infiniband/hw/ocrdma/ocrdma_main.c index 2ca86ca818bd..1a8a945efa60 100644 --- a/drivers/infiniband/hw/ocrdma/ocrdma_main.c +++ b/drivers/infiniband/hw/ocrdma/ocrdma_main.c @@ -127,7 +127,7 @@ static int ocrdma_addr_event(unsigned long event, struct net_device *netdev, is_vlan = netdev->priv_flags & IFF_802_1Q_VLAN; if (is_vlan) - netdev = vlan_dev_real_dev(netdev); + netdev = rdma_vlan_dev_real_dev(netdev); rcu_read_lock(); list_for_each_entry_rcu(dev, &ocrdma_dev_list, entry) { From 8c442794b2e7e3e2c7c15cc7d3a1b550ce99ac88 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Fri, 14 Feb 2014 11:15:40 -0800 Subject: [PATCH 583/596] Revert "ARM: sunxi: dt: Convert to the new SID compatibles" This reverts commit 01ab1167cd2d861d20195eda08505652c536df97, it is incorrect, a future patch will fix this up properly. Cc: Maxime Ripard Signed-off-by: Greg Kroah-Hartman --- arch/arm/boot/dts/sun4i-a10.dtsi | 2 +- arch/arm/boot/dts/sun5i-a10s.dtsi | 2 +- arch/arm/boot/dts/sun5i-a13.dtsi | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) diff --git a/arch/arm/boot/dts/sun4i-a10.dtsi b/arch/arm/boot/dts/sun4i-a10.dtsi index 7b10f85ef988..040bb0eba152 100644 --- a/arch/arm/boot/dts/sun4i-a10.dtsi +++ b/arch/arm/boot/dts/sun4i-a10.dtsi @@ -421,7 +421,7 @@ rtc: rtc@01c20d00 { }; sid: eeprom@01c23800 { - compatible = "allwinner,sun4i-a10-sid"; + compatible = "allwinner,sun4i-sid"; reg = <0x01c23800 0x10>; }; diff --git a/arch/arm/boot/dts/sun5i-a10s.dtsi b/arch/arm/boot/dts/sun5i-a10s.dtsi index 6d287a0dac5d..ea16054857a4 100644 --- a/arch/arm/boot/dts/sun5i-a10s.dtsi +++ b/arch/arm/boot/dts/sun5i-a10s.dtsi @@ -378,7 +378,7 @@ wdt: watchdog@01c20c90 { }; sid: eeprom@01c23800 { - compatible = "allwinner,sun4i-a10-sid"; + compatible = "allwinner,sun4i-sid"; reg = <0x01c23800 0x10>; }; diff --git a/arch/arm/boot/dts/sun5i-a13.dtsi b/arch/arm/boot/dts/sun5i-a13.dtsi index f49eb13c33c3..320335abfccd 100644 --- a/arch/arm/boot/dts/sun5i-a13.dtsi +++ b/arch/arm/boot/dts/sun5i-a13.dtsi @@ -341,7 +341,7 @@ wdt: watchdog@01c20c90 { }; sid: eeprom@01c23800 { - compatible = "allwinner,sun4i-a10-sid"; + compatible = "allwinner,sun4i-sid"; reg = <0x01c23800 0x10>; }; From bf7471038840547c9328291b4d9d91c55581dcb8 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Fri, 14 Feb 2014 11:16:08 -0800 Subject: [PATCH 584/596] Revert "misc: eeprom: sunxi: Add new compatibles" This reverts commit f0de8e04a7201a2000f3c6d09732c11e7f35d42d, it is incorrect, a future patch will fix this up properly. Cc: Maxime Ripard Signed-off-by: Greg Kroah-Hartman --- .../devicetree/bindings/misc/allwinner,sunxi-sid.txt | 5 ++--- drivers/misc/eeprom/sunxi_sid.c | 5 +---- 2 files changed, 3 insertions(+), 7 deletions(-) diff --git a/Documentation/devicetree/bindings/misc/allwinner,sunxi-sid.txt b/Documentation/devicetree/bindings/misc/allwinner,sunxi-sid.txt index c10a61ad5f95..68ba37295565 100644 --- a/Documentation/devicetree/bindings/misc/allwinner,sunxi-sid.txt +++ b/Documentation/devicetree/bindings/misc/allwinner,sunxi-sid.txt @@ -1,13 +1,12 @@ Allwinner sunxi-sid Required properties: -- compatible: "allwinner,sun4i-a10-sid" or "allwinner,sun7i-a20-sid" - (Deprecated: "allwinner,sun4i-sid"). +- compatible: "allwinner,sun4i-sid" or "allwinner,sun7i-a20-sid". - reg: Should contain registers location and length Example for sun4i: sid@01c23800 { - compatible = "allwinner,sun4i-a10-sid"; + compatible = "allwinner,sun4i-sid"; reg = <0x01c23800 0x10> }; diff --git a/drivers/misc/eeprom/sunxi_sid.c b/drivers/misc/eeprom/sunxi_sid.c index e137e75a89e5..9c34e5704304 100644 --- a/drivers/misc/eeprom/sunxi_sid.c +++ b/drivers/misc/eeprom/sunxi_sid.c @@ -96,11 +96,8 @@ static int sunxi_sid_remove(struct platform_device *pdev) } static const struct of_device_id sunxi_sid_of_match[] = { - { .compatible = "allwinner,sun4i-a10-sid", .data = (void *)16}, - { .compatible = "allwinner,sun7i-a20-sid", .data = (void *)512}, - - /* Deprecated */ { .compatible = "allwinner,sun4i-sid", .data = (void *)16}, + { .compatible = "allwinner,sun7i-a20-sid", .data = (void *)512}, {/* sentinel */}, }; MODULE_DEVICE_TABLE(of, sunxi_sid_of_match); From 4e8ca6ee3a5d0e0f4c0cb32e43dc3d69e79e5a76 Mon Sep 17 00:00:00 2001 From: Kevin Hao Date: Fri, 14 Feb 2014 13:22:45 +0800 Subject: [PATCH 585/596] Revert "OF: base: match each node compatible against all given matches first" This reverts commit 105353145eafb3ea919f5cdeb652a9d8f270228e. Stephen Chivers reported this is broken as we will get a match entry '.type = "serial"' instead of the '.compatible = "ns16550"' in the following scenario: serial0: serial@4500 { compatible = "fsl,ns16550", "ns16550"; } struct of_device_id of_platform_serial_table[] = { { .compatible = "ns8250", .data = (void *)PORT_8250, }, { .compatible = "ns16450", .data = (void *)PORT_16450, }, { .compatible = "ns16550a", .data = (void *)PORT_16550A, }, { .compatible = "ns16550", .data = (void *)PORT_16550, }, { .compatible = "ns16750", .data = (void *)PORT_16750, }, { .compatible = "ns16850", .data = (void *)PORT_16850, }, ... { .type = "serial", .data = (void *)PORT_UNKNOWN, }, { /* end of list */ }, }; So just revert this patch, we will use another implementation to find the best compatible match in a follow-on patch. Reported-by: Stephen N Chivers Cc: Sebastian Hesselbarth Signed-off-by: Kevin Hao Signed-off-by: Rob Herring --- drivers/of/base.c | 53 ++++++++++++++--------------------------------- 1 file changed, 16 insertions(+), 37 deletions(-) diff --git a/drivers/of/base.c b/drivers/of/base.c index ff85450d5683..ba195fbce4c6 100644 --- a/drivers/of/base.c +++ b/drivers/of/base.c @@ -734,42 +734,24 @@ static const struct of_device_id *__of_match_node(const struct of_device_id *matches, const struct device_node *node) { - const char *cp; - int cplen, l; - if (!matches) return NULL; - cp = __of_get_property(node, "compatible", &cplen); - do { - const struct of_device_id *m = matches; - - /* Check against matches with current compatible string */ - while (m->name[0] || m->type[0] || m->compatible[0]) { - int match = 1; - if (m->name[0]) - match &= node->name - && !strcmp(m->name, node->name); - if (m->type[0]) - match &= node->type - && !strcmp(m->type, node->type); - if (m->compatible[0]) - match &= cp - && !of_compat_cmp(m->compatible, cp, - strlen(m->compatible)); - if (match) - return m; - m++; - } - - /* Get node's next compatible string */ - if (cp) { - l = strlen(cp) + 1; - cp += l; - cplen -= l; - } - } while (cp && (cplen > 0)); - + while (matches->name[0] || matches->type[0] || matches->compatible[0]) { + int match = 1; + if (matches->name[0]) + match &= node->name + && !strcmp(matches->name, node->name); + if (matches->type[0]) + match &= node->type + && !strcmp(matches->type, node->type); + if (matches->compatible[0]) + match &= __of_device_is_compatible(node, + matches->compatible); + if (match) + return matches; + matches++; + } return NULL; } @@ -778,10 +760,7 @@ const struct of_device_id *__of_match_node(const struct of_device_id *matches, * @matches: array of of device match structures to search in * @node: the of device structure to match against * - * Low level utility function used by device matching. Matching order - * is to compare each of the node's compatibles with all given matches - * first. This implies node's compatible is sorted from specific to - * generic while matches can be in any order. + * Low level utility function used by device matching. */ const struct of_device_id *of_match_node(const struct of_device_id *matches, const struct device_node *node) From 11bcac89c0d73dea42f1cb8646b532035796a5d6 Mon Sep 17 00:00:00 2001 From: Chris Mason Date: Fri, 14 Feb 2014 13:42:13 -0800 Subject: [PATCH 586/596] Revert "btrfs: add ioctl to export size of global metadata reservation" This reverts commit 01e219e8069516cdb98594d417b8bb8d906ed30d. David Sterba found a different way to provide these features without adding a new ioctl. We haven't released any progs with this ioctl yet, so I'm taking this out for now until we finalize things. Signed-off-by: Chris Mason Signed-off-by: David Sterba CC: Jeff Mahoney --- fs/btrfs/ioctl.c | 16 ---------------- include/uapi/linux/btrfs.h | 1 - 2 files changed, 17 deletions(-) diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index ebdd866d4cfd..9a9044585da7 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -3559,20 +3559,6 @@ static long btrfs_ioctl_space_info(struct btrfs_root *root, void __user *arg) return ret; } -static long btrfs_ioctl_global_rsv(struct btrfs_root *root, void __user *arg) -{ - struct btrfs_block_rsv *block_rsv = &root->fs_info->global_block_rsv; - u64 reserved; - - spin_lock(&block_rsv->lock); - reserved = block_rsv->reserved; - spin_unlock(&block_rsv->lock); - - if (arg && copy_to_user(arg, &reserved, sizeof(reserved))) - return -EFAULT; - return 0; -} - /* * there are many ways the trans_start and trans_end ioctls can lead * to deadlocks. They should only be used by applications that @@ -4779,8 +4765,6 @@ long btrfs_ioctl(struct file *file, unsigned int return btrfs_ioctl_logical_to_ino(root, argp); case BTRFS_IOC_SPACE_INFO: return btrfs_ioctl_space_info(root, argp); - case BTRFS_IOC_GLOBAL_RSV: - return btrfs_ioctl_global_rsv(root, argp); case BTRFS_IOC_SYNC: { int ret; diff --git a/include/uapi/linux/btrfs.h b/include/uapi/linux/btrfs.h index 1b8a0f4c9590..b4d69092fbdb 100644 --- a/include/uapi/linux/btrfs.h +++ b/include/uapi/linux/btrfs.h @@ -558,7 +558,6 @@ static inline char *btrfs_err_str(enum btrfs_err_code err_code) #define BTRFS_IOC_DEFAULT_SUBVOL _IOW(BTRFS_IOCTL_MAGIC, 19, __u64) #define BTRFS_IOC_SPACE_INFO _IOWR(BTRFS_IOCTL_MAGIC, 20, \ struct btrfs_ioctl_space_args) -#define BTRFS_IOC_GLOBAL_RSV _IOR(BTRFS_IOCTL_MAGIC, 20, __u64) #define BTRFS_IOC_START_SYNC _IOR(BTRFS_IOCTL_MAGIC, 24, __u64) #define BTRFS_IOC_WAIT_SYNC _IOW(BTRFS_IOCTL_MAGIC, 22, __u64) #define BTRFS_IOC_SNAP_CREATE_V2 _IOW(BTRFS_IOCTL_MAGIC, 23, \ From a9d2d4adb6a87e515912a7033b227acbd9c8835e Mon Sep 17 00:00:00 2001 From: Liu Bo Date: Sat, 8 Feb 2014 15:33:08 +0800 Subject: [PATCH 587/596] Btrfs: fix a lockdep warning when cleaning up aborted transaction Given now we have 2 spinlock for management of delayed refs, CONFIG_DEBUG_SPINLOCK=y helped me find this, [ 4723.413809] BUG: spinlock wrong CPU on CPU#1, btrfs-transacti/2258 [ 4723.414882] lock: 0xffff880048377670, .magic: dead4ead, .owner: btrfs-transacti/2258, .owner_cpu: 2 [ 4723.417146] CPU: 1 PID: 2258 Comm: btrfs-transacti Tainted: G W O 3.12.0+ #4 [ 4723.421321] Call Trace: [ 4723.421872] [] dump_stack+0x54/0x74 [ 4723.422753] [] spin_dump+0x8c/0x91 [ 4723.424979] [] spin_bug+0x21/0x26 [ 4723.425846] [] do_raw_spin_unlock+0x66/0x90 [ 4723.434424] [] _raw_spin_unlock+0x27/0x40 [ 4723.438747] [] btrfs_cleanup_one_transaction+0x35e/0x710 [btrfs] [ 4723.443321] [] btrfs_cleanup_transaction+0x104/0x570 [btrfs] [ 4723.444692] [] ? trace_hardirqs_on_caller+0xfd/0x1c0 [ 4723.450336] [] ? trace_hardirqs_on+0xd/0x10 [ 4723.451332] [] transaction_kthread+0x22e/0x270 [btrfs] [ 4723.452543] [] ? btrfs_cleanup_transaction+0x570/0x570 [btrfs] [ 4723.457833] [] kthread+0xea/0xf0 [ 4723.458990] [] ? kthread_create_on_node+0x140/0x140 [ 4723.460133] [] ret_from_fork+0x7c/0xb0 [ 4723.460865] [] ? kthread_create_on_node+0x140/0x140 [ 4723.496521] ------------[ cut here ]------------ ---------------------------------------------------------------------- The reason is that we get to call cond_resched_lock(&head_ref->lock) while still holding @delayed_refs->lock. So it's different with __btrfs_run_delayed_refs(), where we do drop-acquire dance before and after actually processing delayed refs. Here we don't drop the lock, others are not able to add new delayed refs to head_ref, so cond_resched_lock(&head_ref->lock) is not necessary here. Signed-off-by: Liu Bo Signed-off-by: Chris Mason --- fs/btrfs/disk-io.c | 1 - 1 file changed, 1 deletion(-) diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 3903bd3f8d2b..fcf367581073 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -3842,7 +3842,6 @@ static int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans, rb_erase(&ref->rb_node, &head->ref_root); atomic_dec(&delayed_refs->num_entries); btrfs_put_delayed_ref(ref); - cond_resched_lock(&head->lock); } if (head->must_insert_reserved) pin_bytes = true; From feb5f96589302b39a2b10fc210db7c47a73e4168 Mon Sep 17 00:00:00 2001 From: Mitch Harder Date: Thu, 13 Feb 2014 09:13:16 -0600 Subject: [PATCH 588/596] Btrfs: fix max_inline mount option Currently, the only mount option for max_inline that has any effect is max_inline=0. Any other value that is supplied to max_inline will be adjusted to a minimum of 4k. Since max_inline has an effective maximum of ~3900 bytes due to page size limitations, the current behaviour only has meaning for max_inline=0. This patch will allow the the max_inline mount option to accept non-zero values as indicated in the documentation. Signed-off-by: Mitch Harder Signed-off-by: Chris Mason --- fs/btrfs/super.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c index 97cc24198554..e73c80eec11b 100644 --- a/fs/btrfs/super.c +++ b/fs/btrfs/super.c @@ -566,7 +566,7 @@ int btrfs_parse_options(struct btrfs_root *root, char *options) kfree(num); if (info->max_inline) { - info->max_inline = max_t(u64, + info->max_inline = min_t(u64, info->max_inline, root->sectorsize); } From 3a0dfa6a12e4bb64a434426ecb17d4842092db5e Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Fri, 14 Feb 2014 13:43:48 -0500 Subject: [PATCH 589/596] Btrfs: unset DCACHE_DISCONNECTED when mounting default subvol A user was running into errors from an NFS export of a subvolume that had a default subvol set. When we mount a default subvol we will use d_obtain_alias() to find an existing dentry for the subvolume in the case that the root subvol has already been mounted, or a dummy one is allocated in the case that the root subvol has not already been mounted. This allows us to connect the dentry later on if we wander into the path. However if we don't ever wander into the path we will keep DCACHE_DISCONNECTED set for a long time, which angers NFS. It doesn't appear to cause any problems but it is annoying nonetheless, so simply unset DCACHE_DISCONNECTED in the get_default_root case and switch btrfs_lookup() to use d_materialise_unique() instead which will make everything play nicely together and reconnect stuff if we wander into the defaul subvol path from a different way. With this patch I'm no longer getting the NFS errors when exporting a volume that has been mounted with a default subvol set. Thanks, cc: bfields@fieldses.org cc: ebiederm@xmission.com Signed-off-by: Josef Bacik Acked-by: "Eric W. Biederman" Signed-off-by: Chris Mason --- fs/btrfs/inode.c | 2 +- fs/btrfs/super.c | 9 ++++++++- 2 files changed, 9 insertions(+), 2 deletions(-) diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 1af34d0c744b..4ffb6d79f9f0 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -5150,7 +5150,7 @@ static struct dentry *btrfs_lookup(struct inode *dir, struct dentry *dentry, return ERR_CAST(inode); } - return d_splice_alias(inode, dentry); + return d_materialise_unique(dentry, inode); } unsigned char btrfs_filetype_table[] = { diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c index e73c80eec11b..d04db817be5c 100644 --- a/fs/btrfs/super.c +++ b/fs/btrfs/super.c @@ -855,6 +855,7 @@ static struct dentry *get_default_root(struct super_block *sb, struct btrfs_path *path; struct btrfs_key location; struct inode *inode; + struct dentry *dentry; u64 dir_id; int new = 0; @@ -925,7 +926,13 @@ static struct dentry *get_default_root(struct super_block *sb, return dget(sb->s_root); } - return d_obtain_alias(inode); + dentry = d_obtain_alias(inode); + if (!IS_ERR(dentry)) { + spin_lock(&dentry->d_lock); + dentry->d_flags &= ~DCACHE_DISCONNECTED; + spin_unlock(&dentry->d_lock); + } + return dentry; } static int btrfs_fill_super(struct super_block *sb, From 79970db213344b4a4034645db5ebfc31571f3fa3 Mon Sep 17 00:00:00 2001 From: Wolfram Sang Date: Thu, 13 Feb 2014 21:36:29 +0100 Subject: [PATCH 590/596] i2c: mv64xxx: refactor message start to ensure proper initialization Because the offload mechanism can fall back to a standard transfer, having two seperate initialization states is unfortunate. Let's just have one state which does things consistently. This fixes a bug where some preparation was missing when the fallback happened. And it makes the code much easier to follow. To implement this, we put the check if offload is possible at the top of the offload setup function. Signed-off-by: Wolfram Sang Tested-by: Gregory CLEMENT Cc: stable@vger.kernel.org # v3.12+ Fixes: 930ab3d403ae (i2c: mv64xxx: Add I2C Transaction Generator support) --- drivers/i2c/busses/i2c-mv64xxx.c | 33 ++++++++++++++------------------ 1 file changed, 14 insertions(+), 19 deletions(-) diff --git a/drivers/i2c/busses/i2c-mv64xxx.c b/drivers/i2c/busses/i2c-mv64xxx.c index b8c5187b9ee0..d52d84937ad3 100644 --- a/drivers/i2c/busses/i2c-mv64xxx.c +++ b/drivers/i2c/busses/i2c-mv64xxx.c @@ -97,7 +97,6 @@ enum { enum { MV64XXX_I2C_ACTION_INVALID, MV64XXX_I2C_ACTION_CONTINUE, - MV64XXX_I2C_ACTION_OFFLOAD_SEND_START, MV64XXX_I2C_ACTION_SEND_START, MV64XXX_I2C_ACTION_SEND_RESTART, MV64XXX_I2C_ACTION_OFFLOAD_RESTART, @@ -204,6 +203,9 @@ static int mv64xxx_i2c_offload_msg(struct mv64xxx_i2c_data *drv_data) unsigned long ctrl_reg; struct i2c_msg *msg = drv_data->msgs; + if (!drv_data->offload_enabled) + return -EOPNOTSUPP; + drv_data->msg = msg; drv_data->byte_posn = 0; drv_data->bytes_left = msg->len; @@ -433,8 +435,7 @@ mv64xxx_i2c_do_action(struct mv64xxx_i2c_data *drv_data) drv_data->msgs++; drv_data->num_msgs--; - if (!(drv_data->offload_enabled && - mv64xxx_i2c_offload_msg(drv_data))) { + if (mv64xxx_i2c_offload_msg(drv_data) < 0) { drv_data->cntl_bits |= MV64XXX_I2C_REG_CONTROL_START; writel(drv_data->cntl_bits, drv_data->reg_base + drv_data->reg_offsets.control); @@ -458,15 +459,14 @@ mv64xxx_i2c_do_action(struct mv64xxx_i2c_data *drv_data) drv_data->reg_base + drv_data->reg_offsets.control); break; - case MV64XXX_I2C_ACTION_OFFLOAD_SEND_START: - if (!mv64xxx_i2c_offload_msg(drv_data)) - break; - else - drv_data->action = MV64XXX_I2C_ACTION_SEND_START; - /* FALLTHRU */ case MV64XXX_I2C_ACTION_SEND_START: - writel(drv_data->cntl_bits | MV64XXX_I2C_REG_CONTROL_START, - drv_data->reg_base + drv_data->reg_offsets.control); + /* Can we offload this msg ? */ + if (mv64xxx_i2c_offload_msg(drv_data) < 0) { + /* No, switch to standard path */ + mv64xxx_i2c_prepare_for_io(drv_data, drv_data->msgs); + writel(drv_data->cntl_bits | MV64XXX_I2C_REG_CONTROL_START, + drv_data->reg_base + drv_data->reg_offsets.control); + } break; case MV64XXX_I2C_ACTION_SEND_ADDR_1: @@ -625,15 +625,10 @@ mv64xxx_i2c_execute_msg(struct mv64xxx_i2c_data *drv_data, struct i2c_msg *msg, unsigned long flags; spin_lock_irqsave(&drv_data->lock, flags); - if (drv_data->offload_enabled) { - drv_data->action = MV64XXX_I2C_ACTION_OFFLOAD_SEND_START; - drv_data->state = MV64XXX_I2C_STATE_WAITING_FOR_START_COND; - } else { - mv64xxx_i2c_prepare_for_io(drv_data, msg); - drv_data->action = MV64XXX_I2C_ACTION_SEND_START; - drv_data->state = MV64XXX_I2C_STATE_WAITING_FOR_START_COND; - } + drv_data->action = MV64XXX_I2C_ACTION_SEND_START; + drv_data->state = MV64XXX_I2C_STATE_WAITING_FOR_START_COND; + drv_data->send_stop = is_last; drv_data->block = 1; mv64xxx_i2c_do_action(drv_data); From f085381e6d08f4c8d6882825f31accd455c54d70 Mon Sep 17 00:00:00 2001 From: Anand Jain Date: Wed, 15 Jan 2014 17:22:28 +0800 Subject: [PATCH 591/596] btrfs: fix null pointer deference at btrfs_sysfs_add_one+0x105 bdev is null when disk has disappeared and mounted with the degrade option stack trace --------- btrfs_sysfs_add_one+0x105/0x1c0 [btrfs] open_ctree+0x15f3/0x1fe0 [btrfs] btrfs_mount+0x5db/0x790 [btrfs] ? alloc_pages_current+0xa4/0x160 mount_fs+0x34/0x1b0 vfs_kern_mount+0x62/0xf0 do_mount+0x22e/0xa80 ? __get_free_pages+0x9/0x40 ? copy_mount_options+0x31/0x170 SyS_mount+0x7e/0xc0 system_call_fastpath+0x16/0x1b --------- reproducer: ------- mkfs.btrfs -draid1 -mraid1 /dev/sdc /dev/sdd (detach a disk) devmgt detach /dev/sdc [1] mount -o degrade /dev/sdd /btrfs ------- [1] github.com/anajain/devmgt.git Signed-off-by: Anand Jain Tested-by: Hidetoshi Seto Signed-off-by: Chris Mason --- fs/btrfs/sysfs.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/fs/btrfs/sysfs.c b/fs/btrfs/sysfs.c index 782374d8fd19..865f4cf9a769 100644 --- a/fs/btrfs/sysfs.c +++ b/fs/btrfs/sysfs.c @@ -578,8 +578,14 @@ static int add_device_membership(struct btrfs_fs_info *fs_info) return -ENOMEM; list_for_each_entry(dev, &fs_devices->devices, dev_list) { - struct hd_struct *disk = dev->bdev->bd_part; - struct kobject *disk_kobj = &part_to_dev(disk)->kobj; + struct hd_struct *disk; + struct kobject *disk_kobj; + + if (!dev->bdev) + continue; + + disk = dev->bdev->bd_part; + disk_kobj = &part_to_dev(disk)->kobj; error = sysfs_create_link(fs_info->device_dir_kobj, disk_kobj, disk_kobj->name); From 93de4ba86480a9e0d1062cb1d535fa97fb81af48 Mon Sep 17 00:00:00 2001 From: Filipe David Borba Manana Date: Sat, 15 Feb 2014 15:53:16 +0000 Subject: [PATCH 592/596] Btrfs: use right clone root offset for compressed extents For non compressed extents, iterate_extent_inodes() gives us offsets that take into account the data offset from the file extent items, while for compressed extents it doesn't. Therefore we have to adjust them before placing them in a send clone instruction. Not doing this adjustment leads to the receiving end requesting for a wrong a file range to the clone ioctl, which results in different file content from the one in the original send root. Issue reproducible with the following excerpt from the test I made for xfstests: _scratch_mkfs _scratch_mount "-o compress-force=lzo" $XFS_IO_PROG -f -c "truncate 118811" $SCRATCH_MNT/foo $XFS_IO_PROG -c "pwrite -S 0x0d -b 39987 92267 39987" $SCRATCH_MNT/foo $BTRFS_UTIL_PROG subvolume snapshot -r $SCRATCH_MNT $SCRATCH_MNT/mysnap1 $XFS_IO_PROG -c "pwrite -S 0x3e -b 80000 200000 80000" $SCRATCH_MNT/foo $BTRFS_UTIL_PROG filesystem sync $SCRATCH_MNT $XFS_IO_PROG -c "pwrite -S 0xdc -b 10000 250000 10000" $SCRATCH_MNT/foo $XFS_IO_PROG -c "pwrite -S 0xff -b 10000 300000 10000" $SCRATCH_MNT/foo # will be used for incremental send to be able to issue clone operations $BTRFS_UTIL_PROG subvolume snapshot -r $SCRATCH_MNT $SCRATCH_MNT/clones_snap $BTRFS_UTIL_PROG subvolume snapshot -r $SCRATCH_MNT $SCRATCH_MNT/mysnap2 $FSSUM_PROG -A -f -w $tmp/1.fssum $SCRATCH_MNT/mysnap1 $FSSUM_PROG -A -f -w $tmp/2.fssum -x $SCRATCH_MNT/mysnap2/mysnap1 \ -x $SCRATCH_MNT/mysnap2/clones_snap $SCRATCH_MNT/mysnap2 $FSSUM_PROG -A -f -w $tmp/clones.fssum $SCRATCH_MNT/clones_snap \ -x $SCRATCH_MNT/clones_snap/mysnap1 -x $SCRATCH_MNT/clones_snap/mysnap2 $BTRFS_UTIL_PROG send $SCRATCH_MNT/mysnap1 -f $tmp/1.snap $BTRFS_UTIL_PROG send $SCRATCH_MNT/clones_snap -f $tmp/clones.snap $BTRFS_UTIL_PROG send -p $SCRATCH_MNT/mysnap1 \ -c $SCRATCH_MNT/clones_snap $SCRATCH_MNT/mysnap2 -f $tmp/2.snap _scratch_unmount _scratch_mkfs _scratch_mount $BTRFS_UTIL_PROG receive $SCRATCH_MNT -f $tmp/1.snap $FSSUM_PROG -r $tmp/1.fssum $SCRATCH_MNT/mysnap1 2>> $seqres.full $BTRFS_UTIL_PROG receive $SCRATCH_MNT -f $tmp/clones.snap $FSSUM_PROG -r $tmp/clones.fssum $SCRATCH_MNT/clones_snap 2>> $seqres.full $BTRFS_UTIL_PROG receive $SCRATCH_MNT -f $tmp/2.snap $FSSUM_PROG -r $tmp/2.fssum $SCRATCH_MNT/mysnap2 2>> $seqres.full Signed-off-by: Filipe David Borba Manana Signed-off-by: Chris Mason --- fs/btrfs/send.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c index 9c8d1a3fdc3a..9dde9717c1b9 100644 --- a/fs/btrfs/send.c +++ b/fs/btrfs/send.c @@ -1332,6 +1332,16 @@ verbose_printk(KERN_DEBUG "btrfs: find_extent_clone: data_offset=%llu, " } if (cur_clone_root) { + if (compressed != BTRFS_COMPRESS_NONE) { + /* + * Offsets given by iterate_extent_inodes() are relative + * to the start of the extent, we need to add logical + * offset from the file extent item. + * (See why at backref.c:check_extent_in_eb()) + */ + cur_clone_root->offset += btrfs_file_extent_offset(eb, + fi); + } *found = cur_clone_root; ret = 0; } else { From aeca0fe62ac19217cb8158a0c5f6f74e3b8d304b Mon Sep 17 00:00:00 2001 From: Wolfram Sang Date: Mon, 10 Feb 2014 11:03:55 +0100 Subject: [PATCH 593/596] Documentation: i2c: describe devicetree method for instantiating devices Signed-off-by: Wolfram Sang --- Documentation/i2c/instantiating-devices | 34 +++++++++++++++++++++++-- 1 file changed, 32 insertions(+), 2 deletions(-) diff --git a/Documentation/i2c/instantiating-devices b/Documentation/i2c/instantiating-devices index c70e7a7638d1..6df095a6071e 100644 --- a/Documentation/i2c/instantiating-devices +++ b/Documentation/i2c/instantiating-devices @@ -8,8 +8,8 @@ reason, the kernel code must instantiate I2C devices explicitly. There are several ways to achieve this, depending on the context and requirements. -Method 1: Declare the I2C devices by bus number ------------------------------------------------ +Method 1a: Declare the I2C devices by bus number +------------------------------------------------ This method is appropriate when the I2C bus is a system bus as is the case for many embedded systems. On such systems, each I2C bus has a number @@ -51,6 +51,36 @@ The devices will be automatically unbound and destroyed when the I2C bus they sit on goes away (if ever.) +Method 1b: Declare the I2C devices via devicetree +------------------------------------------------- + +This method has the same implications as method 1a. The declaration of I2C +devices is here done via devicetree as subnodes of the master controller. + +Example: + + i2c1: i2c@400a0000 { + /* ... master properties skipped ... */ + clock-frequency = <100000>; + + flash@50 { + compatible = "atmel,24c256"; + reg = <0x50>; + }; + + pca9532: gpio@60 { + compatible = "nxp,pca9532"; + gpio-controller; + #gpio-cells = <2>; + reg = <0x60>; + }; + }; + +Here, two devices are attached to the bus using a speed of 100kHz. For +additional properties which might be needed to set up the device, please refer +to its devicetree documentation in Documentation/devicetree/bindings/. + + Method 2: Instantiate the devices explicitly -------------------------------------------- From fde1e418ab2c4269035f7f576e678b6ad3117d7d Mon Sep 17 00:00:00 2001 From: Wolfram Sang Date: Sat, 15 Feb 2014 15:58:35 +0100 Subject: [PATCH 594/596] Documentation: i2c: mention ACPI method for instantiating devices Signed-off-by: Wolfram Sang Acked-by: Guenter Roeck --- Documentation/i2c/instantiating-devices | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/Documentation/i2c/instantiating-devices b/Documentation/i2c/instantiating-devices index 6df095a6071e..0d85ac1935b7 100644 --- a/Documentation/i2c/instantiating-devices +++ b/Documentation/i2c/instantiating-devices @@ -81,6 +81,13 @@ additional properties which might be needed to set up the device, please refer to its devicetree documentation in Documentation/devicetree/bindings/. +Method 1c: Declare the I2C devices via ACPI +------------------------------------------- + +ACPI can also describe I2C devices. There is special documentation for this +which is currently located at Documentation/acpi/enumeration.txt. + + Method 2: Instantiate the devices explicitly -------------------------------------------- From 06b29e76a74b2373e6f8b5a7938b3630b9ae98b2 Mon Sep 17 00:00:00 2001 From: Kevin Hao Date: Fri, 14 Feb 2014 13:22:46 +0800 Subject: [PATCH 595/596] of: search the best compatible match first in __of_match_node() Currently, of_match_node compares each given match against all node's compatible strings with of_device_is_compatible. To achieve multiple compatible strings per node with ordering from specific to generic, this requires given matches to be ordered from specific to generic. For most of the drivers this is not true and also an alphabetical ordering is more sane there. Therefore, this patch introduces a function to match each of the node's compatible strings against all given compatible matches without type and name first, before checking the next compatible string. This implies that node's compatibles are ordered from specific to generic while given matches can be in any order. If we fail to find such a match entry, then fall-back to the old method in order to keep compatibility. Cc: Sebastian Hesselbarth Signed-off-by: Kevin Hao Tested-by: Stephen Chivers Signed-off-by: Rob Herring --- drivers/of/base.c | 43 ++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 42 insertions(+), 1 deletion(-) diff --git a/drivers/of/base.c b/drivers/of/base.c index ba195fbce4c6..10b51106c854 100644 --- a/drivers/of/base.c +++ b/drivers/of/base.c @@ -730,13 +730,49 @@ struct device_node *of_find_node_with_property(struct device_node *from, } EXPORT_SYMBOL(of_find_node_with_property); +static const struct of_device_id * +of_match_compatible(const struct of_device_id *matches, + const struct device_node *node) +{ + const char *cp; + int cplen, l; + const struct of_device_id *m; + + cp = __of_get_property(node, "compatible", &cplen); + while (cp && (cplen > 0)) { + m = matches; + while (m->name[0] || m->type[0] || m->compatible[0]) { + /* Only match for the entries without type and name */ + if (m->name[0] || m->type[0] || + of_compat_cmp(m->compatible, cp, + strlen(m->compatible))) + m++; + else + return m; + } + + /* Get node's next compatible string */ + l = strlen(cp) + 1; + cp += l; + cplen -= l; + } + + return NULL; +} + static const struct of_device_id *__of_match_node(const struct of_device_id *matches, const struct device_node *node) { + const struct of_device_id *m; + if (!matches) return NULL; + m = of_match_compatible(matches, node); + if (m) + return m; + while (matches->name[0] || matches->type[0] || matches->compatible[0]) { int match = 1; if (matches->name[0]) @@ -760,7 +796,12 @@ const struct of_device_id *__of_match_node(const struct of_device_id *matches, * @matches: array of of device match structures to search in * @node: the of device structure to match against * - * Low level utility function used by device matching. + * Low level utility function used by device matching. We have two ways + * of matching: + * - Try to find the best compatible match by comparing each compatible + * string of device node with all the given matches respectively. + * - If the above method failed, then try to match the compatible by using + * __of_device_is_compatible() besides the match in type and name. */ const struct of_device_id *of_match_node(const struct of_device_id *matches, const struct device_node *node) From 6d0abeca3242a88cab8232e4acd7e2bf088f3bc2 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 16 Feb 2014 13:30:25 -0800 Subject: [PATCH 596/596] Linux 3.14-rc3 --- Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Makefile b/Makefile index 933e1def6baf..893d6f0e875b 100644 --- a/Makefile +++ b/Makefile @@ -1,7 +1,7 @@ VERSION = 3 PATCHLEVEL = 14 SUBLEVEL = 0 -EXTRAVERSION = -rc2 +EXTRAVERSION = -rc3 NAME = Shuffling Zombie Juror # *DOCUMENTATION*