/* * Copyright(c) 2004 - 2009 Intel Corporation. All rights reserved. * * 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. * * The full GNU General Public License is included in this distribution in the * file called COPYING. */ #ifndef IOATDMA_H #define IOATDMA_H #include #include #include #include #include #include #include #include "registers.h" #include "hw.h" #define IOAT_DMA_VERSION "4.00" #define IOAT_DMA_DCA_ANY_CPU ~0 #define to_ioatdma_device(dev) container_of(dev, struct ioatdma_device, dma_dev) #define to_dev(ioat_chan) (&(ioat_chan)->ioat_dma->pdev->dev) #define to_pdev(ioat_chan) ((ioat_chan)->ioat_dma->pdev) #define chan_num(ch) ((int)((ch)->reg_base - (ch)->ioat_dma->reg_base) / 0x80) /* * workaround for IOAT ver.3.0 null descriptor issue * (channel returns error when size is 0) */ #define NULL_DESC_BUFFER_SIZE 1 enum ioat_irq_mode { IOAT_NOIRQ = 0, IOAT_MSIX, IOAT_MSI, IOAT_INTX }; /** * struct ioatdma_device - internal representation of a IOAT device * @pdev: PCI-Express device * @reg_base: MMIO register space base address * @dma_pool: for allocating DMA descriptors * @dma_dev: embedded struct dma_device * @version: version of ioatdma device * @msix_entries: irq handlers * @idx: per channel data * @dca: direct cache access context * @intr_quirk: interrupt setup quirk (for ioat_v1 devices) * @enumerate_channels: hw version specific channel enumeration * @reset_hw: hw version specific channel (re)initialization * @cleanup_fn: select between the v2 and v3 cleanup routines * @timer_fn: select between the v2 and v3 timer watchdog routines * @self_test: hardware version specific self test for each supported op type * * Note: the v3 cleanup routine supports raid operations */ struct ioatdma_device { struct pci_dev *pdev; void __iomem *reg_base; struct pci_pool *dma_pool; struct pci_pool *completion_pool; #define MAX_SED_POOLS 5 struct dma_pool *sed_hw_pool[MAX_SED_POOLS]; struct dma_device dma_dev; u8 version; struct msix_entry msix_entries[4]; struct ioatdma_chan *idx[4]; struct dca_provider *dca; enum ioat_irq_mode irq_mode; u32 cap; void (*intr_quirk)(struct ioatdma_device *ioat_dma); int (*enumerate_channels)(struct ioatdma_device *ioat_dma); int (*reset_hw)(struct ioatdma_chan *ioat_chan); void (*cleanup_fn)(unsigned long data); void (*timer_fn)(unsigned long data); int (*self_test)(struct ioatdma_device *ioat_dma); }; struct ioatdma_chan { struct dma_chan dma_chan; void __iomem *reg_base; dma_addr_t last_completion; spinlock_t cleanup_lock; unsigned long state; #define IOAT_COMPLETION_PENDING 0 #define IOAT_COMPLETION_ACK 1 #define IOAT_RESET_PENDING 2 #define IOAT_KOBJ_INIT_FAIL 3 #define IOAT_RESHAPE_PENDING 4 #define IOAT_RUN 5 #define IOAT_CHAN_ACTIVE 6 struct timer_list timer; #define COMPLETION_TIMEOUT msecs_to_jiffies(100) #define IDLE_TIMEOUT msecs_to_jiffies(2000) #define RESET_DELAY msecs_to_jiffies(100) struct ioatdma_device *ioat_dma; dma_addr_t completion_dma; u64 *completion; struct tasklet_struct cleanup_task; struct kobject kobj; /* ioat v2 / v3 channel attributes * @xfercap_log; log2 of channel max transfer length (for fast division) * @head: allocated index * @issued: hardware notification point * @tail: cleanup index * @dmacount: identical to 'head' except for occasionally resetting to zero * @alloc_order: log2 of the number of allocated descriptors * @produce: number of descriptors to produce at submit time * @ring: software ring buffer implementation of hardware ring * @prep_lock: serializes descriptor preparation (producers) */ size_t xfercap_log; u16 head; u16 issued; u16 tail; u16 dmacount; u16 alloc_order; u16 produce; struct ioat_ring_ent **ring; spinlock_t prep_lock; }; struct ioat_sysfs_entry { struct attribute attr; ssize_t (*show)(struct dma_chan *, char *); }; /** * struct ioat_sed_ent - wrapper around super extended hardware descriptor * @hw: hardware SED * @sed_dma: dma address for the SED * @list: list member * @parent: point to the dma descriptor that's the parent */ struct ioat_sed_ent { struct ioat_sed_raw_descriptor *hw; dma_addr_t dma; struct ioat_ring_ent *parent; unsigned int hw_pool; }; /** * struct ioat_ring_ent - wrapper around hardware descriptor * @hw: hardware DMA descriptor (for memcpy) * @fill: hardware fill descriptor * @xor: hardware xor descriptor * @xor_ex: hardware xor extension descriptor * @pq: hardware pq descriptor * @pq_ex: hardware pq extension descriptor * @pqu: hardware pq update descriptor * @raw: hardware raw (un-typed) descriptor * @txd: the generic software descriptor for all engines * @len: total transaction length for unmap * @result: asynchronous result of validate operations * @id: identifier for debug */ struct ioat_ring_ent { union { struct ioat_dma_descriptor *hw; struct ioat_xor_descriptor *xor; struct ioat_xor_ext_descriptor *xor_ex; struct ioat_pq_descriptor *pq; struct ioat_pq_ext_descriptor *pq_ex; struct ioat_pq_update_descriptor *pqu; struct ioat_raw_descriptor *raw; }; size_t len; struct dma_async_tx_descriptor txd; enum sum_check_flags *result; #ifdef DEBUG int id; #endif struct ioat_sed_ent *sed; }; static inline struct ioatdma_chan *to_ioat_chan(struct dma_chan *c) { return container_of(c, struct ioatdma_chan, dma_chan); } /* wrapper around hardware descriptor format + additional software fields */ #ifdef DEBUG #define set_desc_id(desc, i) ((desc)->id = (i)) #define desc_id(desc) ((desc)->id) #else #define set_desc_id(desc, i) #define desc_id(desc) (0) #endif static inline void __dump_desc_dbg(struct ioatdma_chan *ioat_chan, struct ioat_dma_descriptor *hw, struct dma_async_tx_descriptor *tx, int id) { struct device *dev = to_dev(ioat_chan); dev_dbg(dev, "desc[%d]: (%#llx->%#llx) cookie: %d flags: %#x" " ctl: %#10.8x (op: %#x int_en: %d compl: %d)\n", id, (unsigned long long) tx->phys, (unsigned long long) hw->next, tx->cookie, tx->flags, hw->ctl, hw->ctl_f.op, hw->ctl_f.int_en, hw->ctl_f.compl_write); } #define dump_desc_dbg(c, d) \ ({ if (d) __dump_desc_dbg(c, d->hw, &d->txd, desc_id(d)); 0; }) static inline struct ioatdma_chan * ioat_chan_by_index(struct ioatdma_device *ioat_dma, int index) { return ioat_dma->idx[index]; } static inline u64 ioat_chansts_32(struct ioatdma_chan *ioat_chan) { u8 ver = ioat_chan->ioat_dma->version; u64 status; u32 status_lo; /* We need to read the low address first as this causes the * chipset to latch the upper bits for the subsequent read */ status_lo = readl(ioat_chan->reg_base + IOAT_CHANSTS_OFFSET_LOW(ver)); status = readl(ioat_chan->reg_base + IOAT_CHANSTS_OFFSET_HIGH(ver)); status <<= 32; status |= status_lo; return status; } #if BITS_PER_LONG == 64 static inline u64 ioat_chansts(struct ioatdma_chan *ioat_chan) { u8 ver = ioat_chan->ioat_dma->version; u64 status; /* With IOAT v3.3 the status register is 64bit. */ if (ver >= IOAT_VER_3_3) status = readq(ioat_chan->reg_base + IOAT_CHANSTS_OFFSET(ver)); else status = ioat_chansts_32(ioat_chan); return status; } #else #define ioat_chansts ioat_chansts_32 #endif static inline u64 ioat_chansts_to_addr(u64 status) { return status & IOAT_CHANSTS_COMPLETED_DESCRIPTOR_ADDR; } static inline u32 ioat_chanerr(struct ioatdma_chan *ioat_chan) { return readl(ioat_chan->reg_base + IOAT_CHANERR_OFFSET); } static inline void ioat_suspend(struct ioatdma_chan *ioat_chan) { u8 ver = ioat_chan->ioat_dma->version; writeb(IOAT_CHANCMD_SUSPEND, ioat_chan->reg_base + IOAT_CHANCMD_OFFSET(ver)); } static inline void ioat_reset(struct ioatdma_chan *ioat_chan) { u8 ver = ioat_chan->ioat_dma->version; writeb(IOAT_CHANCMD_RESET, ioat_chan->reg_base + IOAT_CHANCMD_OFFSET(ver)); } static inline bool ioat_reset_pending(struct ioatdma_chan *ioat_chan) { u8 ver = ioat_chan->ioat_dma->version; u8 cmd; cmd = readb(ioat_chan->reg_base + IOAT_CHANCMD_OFFSET(ver)); return (cmd & IOAT_CHANCMD_RESET) == IOAT_CHANCMD_RESET; } static inline bool is_ioat_active(unsigned long status) { return ((status & IOAT_CHANSTS_STATUS) == IOAT_CHANSTS_ACTIVE); } static inline bool is_ioat_idle(unsigned long status) { return ((status & IOAT_CHANSTS_STATUS) == IOAT_CHANSTS_DONE); } static inline bool is_ioat_halted(unsigned long status) { return ((status & IOAT_CHANSTS_STATUS) == IOAT_CHANSTS_HALTED); } static inline bool is_ioat_suspended(unsigned long status) { return ((status & IOAT_CHANSTS_STATUS) == IOAT_CHANSTS_SUSPENDED); } /* channel was fatally programmed */ static inline bool is_ioat_bug(unsigned long err) { return !!err; } #define IOAT_MAX_ORDER 16 #define ioat_get_alloc_order() \ (min(ioat_ring_alloc_order, IOAT_MAX_ORDER)) #define ioat_get_max_alloc_order() \ (min(ioat_ring_max_alloc_order, IOAT_MAX_ORDER)) static inline u32 ioat_ring_size(struct ioatdma_chan *ioat_chan) { return 1 << ioat_chan->alloc_order; } /* count of descriptors in flight with the engine */ static inline u16 ioat_ring_active(struct ioatdma_chan *ioat_chan) { return CIRC_CNT(ioat_chan->head, ioat_chan->tail, ioat_ring_size(ioat_chan)); } /* count of descriptors pending submission to hardware */ static inline u16 ioat_ring_pending(struct ioatdma_chan *ioat_chan) { return CIRC_CNT(ioat_chan->head, ioat_chan->issued, ioat_ring_size(ioat_chan)); } static inline u32 ioat_ring_space(struct ioatdma_chan *ioat_chan) { return ioat_ring_size(ioat_chan) - ioat_ring_active(ioat_chan); } static inline u16 ioat_xferlen_to_descs(struct ioatdma_chan *ioat_chan, size_t len) { u16 num_descs = len >> ioat_chan->xfercap_log; num_descs += !!(len & ((1 << ioat_chan->xfercap_log) - 1)); return num_descs; } static inline struct ioat_ring_ent * ioat_get_ring_ent(struct ioatdma_chan *ioat_chan, u16 idx) { return ioat_chan->ring[idx & (ioat_ring_size(ioat_chan) - 1)]; } static inline void ioat_set_chainaddr(struct ioatdma_chan *ioat_chan, u64 addr) { writel(addr & 0x00000000FFFFFFFF, ioat_chan->reg_base + IOAT2_CHAINADDR_OFFSET_LOW); writel(addr >> 32, ioat_chan->reg_base + IOAT2_CHAINADDR_OFFSET_HIGH); } int ioat_probe(struct ioatdma_device *ioat_dma); int ioat_register(struct ioatdma_device *ioat_dma); int ioat_dma_self_test(struct ioatdma_device *ioat_dma); void ioat_dma_remove(struct ioatdma_device *ioat_dma); struct dca_provider *ioat_dca_init(struct pci_dev *pdev, void __iomem *iobase); void ioat_init_channel(struct ioatdma_device *ioat_dma, struct ioatdma_chan *ioat_chan, int idx); enum dma_status ioat_dma_tx_status(struct dma_chan *c, dma_cookie_t cookie, struct dma_tx_state *txstate); bool ioat_cleanup_preamble(struct ioatdma_chan *ioat_chan, dma_addr_t *phys_complete); void ioat_kobject_add(struct ioatdma_device *ioat_dma, struct kobj_type *type); void ioat_kobject_del(struct ioatdma_device *ioat_dma); int ioat_dma_setup_interrupts(struct ioatdma_device *ioat_dma); void ioat_stop(struct ioatdma_chan *ioat_chan); int ioat_dma_probe(struct ioatdma_device *ioat_dma, int dca); int ioat3_dma_probe(struct ioatdma_device *ioat_dma, int dca); struct dca_provider *ioat3_dca_init(struct pci_dev *pdev, void __iomem *iobase); int ioat_check_space_lock(struct ioatdma_chan *ioat_chan, int num_descs); int ioat_enumerate_channels(struct ioatdma_device *ioat_dma); struct dma_async_tx_descriptor * ioat_dma_prep_memcpy_lock(struct dma_chan *c, dma_addr_t dma_dest, dma_addr_t dma_src, size_t len, unsigned long flags); void ioat_issue_pending(struct dma_chan *chan); int ioat_alloc_chan_resources(struct dma_chan *c); void ioat_free_chan_resources(struct dma_chan *c); void __ioat_restart_chan(struct ioatdma_chan *ioat_chan); bool reshape_ring(struct ioatdma_chan *ioat, int order); void __ioat_issue_pending(struct ioatdma_chan *ioat_chan); void ioat_timer_event(unsigned long data); int ioat_quiesce(struct ioatdma_chan *ioat_chan, unsigned long tmo); int ioat_reset_sync(struct ioatdma_chan *ioat_chan, unsigned long tmo); extern const struct sysfs_ops ioat_sysfs_ops; extern struct ioat_sysfs_entry ioat_version_attr; extern struct ioat_sysfs_entry ioat_cap_attr; extern int ioat_pending_level; extern int ioat_ring_alloc_order; extern struct kobj_type ioat_ktype; extern struct kmem_cache *ioat_cache; #endif /* IOATDMA_H */