linux/drivers/ata/pata_sl82c105.c

366 lines
9.3 KiB
C
Raw Normal View History

/*
* pata_sl82c105.c - SL82C105 PATA for new ATA layer
* (C) 2005 Red Hat Inc
* Alan Cox <alan@redhat.com>
*
* Based in part on linux/drivers/ide/pci/sl82c105.c
* SL82C105/Winbond 553 IDE driver
*
* and in part on the documentation and errata sheet
*
*
* Note: The controller like many controllers has shared timings for
* PIO and DMA. We thus flip to the DMA timings in dma_start and flip back
* in the dma_stop function. Thus we actually don't need a set_dmamode
* method as the PIO method is always called and will set the right PIO
* timing parameters.
*/
#include <linux/kernel.h>
#include <linux/module.h>
#include <linux/pci.h>
#include <linux/init.h>
#include <linux/blkdev.h>
#include <linux/delay.h>
#include <scsi/scsi_host.h>
#include <linux/libata.h>
#define DRV_NAME "pata_sl82c105"
#define DRV_VERSION "0.3.2"
enum {
/*
* SL82C105 PCI config register 0x40 bits.
*/
CTRL_IDE_IRQB = (1 << 30),
CTRL_IDE_IRQA = (1 << 28),
CTRL_LEGIRQ = (1 << 11),
CTRL_P1F16 = (1 << 5),
CTRL_P1EN = (1 << 4),
CTRL_P0F16 = (1 << 1),
CTRL_P0EN = (1 << 0)
};
/**
* sl82c105_pre_reset - probe begin
* @link: ATA link
libata: add deadline support to prereset and reset methods Add @deadline to prereset and reset methods and make them honor it. ata_wait_ready() which directly takes @deadline is implemented to be used as the wait function. This patch is in preparation for EH timing improvements. * ata_wait_ready() never does busy sleep. It's only used from EH and no wait in EH is that urgent. This function also prints 'be patient' message automatically after 5 secs of waiting if more than 3 secs is remaining till deadline. * ata_bus_post_reset() now fails with error code if any of its wait fails. This is important because earlier reset tries will have shorter timeout than the spec requires. If a device fails to respond before the short timeout, reset should be retried with longer timeout rather than silently ignoring the device. There are three behavior differences. 1. Timeout is applied to both devices at once, not separately. This is more consistent with what the spec says. 2. When a device passes devchk but fails to become ready before deadline. Previouly, post_reset would just succeed and let device classification remove the device. New code fails the reset thus causing reset retry. After a few times, EH will give up disabling the port. 3. When slave device passes devchk but fails to become accessible (TF-wise) after reset. Original code disables dev1 after 30s timeout and continues as if the device doesn't exist, while the patched code fails reset. When this happens, new code fails reset on whole port rather than proceeding with only the primary device. If the failing device is suffering transient problems, new code retries reset which is a better behavior. If the failing device is actually broken, the net effect is identical to it, but not to the other device sharing the channel. In the previous code, reset would have succeeded after 30s thus detecting the working one. In the new code, reset fails and whole port gets disabled. IMO, it's a pathological case anyway (broken device sharing bus with working one) and doesn't really matter. * ata_bus_softreset() is changed to return error code from ata_bus_post_reset(). It used to return 0 unconditionally. * Spin up waiting is to be removed and not converted to honor deadline. * To be on the safe side, deadline is set to 40s for the time being. Signed-off-by: Tejun Heo <htejun@gmail.com> Signed-off-by: Jeff Garzik <jeff@garzik.org>
2007-02-02 15:50:52 +08:00
* @deadline: deadline jiffies for the operation
*
* Set up cable type and use generic probe init
*/
static int sl82c105_pre_reset(struct ata_link *link, unsigned long deadline)
{
static const struct pci_bits sl82c105_enable_bits[] = {
{ 0x40, 1, 0x01, 0x01 },
{ 0x40, 1, 0x10, 0x10 }
};
struct ata_port *ap = link->ap;
struct pci_dev *pdev = to_pci_dev(ap->host->dev);
if (ap->port_no && !pci_test_config_bits(pdev, &sl82c105_enable_bits[ap->port_no]))
return -ENOENT;
return ata_std_prereset(link, deadline);
}
static void sl82c105_error_handler(struct ata_port *ap)
{
ata_bmdma_drive_eh(ap, sl82c105_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
}
/**
* sl82c105_configure_piomode - set chip PIO timing
* @ap: ATA interface
* @adev: ATA device
* @pio: PIO mode
*
* Called to do the PIO mode setup. Our timing registers are shared
* so a configure_dmamode call will undo any work we do here and vice
* versa
*/
static void sl82c105_configure_piomode(struct ata_port *ap, struct ata_device *adev, int pio)
{
struct pci_dev *pdev = to_pci_dev(ap->host->dev);
static u16 pio_timing[5] = {
0x50D, 0x407, 0x304, 0x242, 0x240
};
u16 dummy;
int timing = 0x44 + (8 * ap->port_no) + (4 * adev->devno);
pci_write_config_word(pdev, timing, pio_timing[pio]);
/* Can we lose this oddity of the old driver */
pci_read_config_word(pdev, timing, &dummy);
}
/**
* sl82c105_set_piomode - set initial PIO mode data
* @ap: ATA interface
* @adev: ATA device
*
* Called to do the PIO mode setup. Our timing registers are shared
* but we want to set the PIO timing by default.
*/
static void sl82c105_set_piomode(struct ata_port *ap, struct ata_device *adev)
{
sl82c105_configure_piomode(ap, adev, adev->pio_mode - XFER_PIO_0);
}
/**
* sl82c105_configure_dmamode - set DMA mode in chip
* @ap: ATA interface
* @adev: ATA device
*
* Load DMA cycle times into the chip ready for a DMA transfer
* to occur.
*/
static void sl82c105_configure_dmamode(struct ata_port *ap, struct ata_device *adev)
{
struct pci_dev *pdev = to_pci_dev(ap->host->dev);
static u16 dma_timing[3] = {
0x707, 0x201, 0x200
};
u16 dummy;
int timing = 0x44 + (8 * ap->port_no) + (4 * adev->devno);
int dma = adev->dma_mode - XFER_MW_DMA_0;
pci_write_config_word(pdev, timing, dma_timing[dma]);
/* Can we lose this oddity of the old driver */
pci_read_config_word(pdev, timing, &dummy);
}
/**
* sl82c105_reset_engine - Reset the DMA engine
* @ap: ATA interface
*
* The sl82c105 has some serious problems with the DMA engine
* when transfers don't run as expected or ATAPI is used. The
* recommended fix is to reset the engine each use using a chip
* test register.
*/
static void sl82c105_reset_engine(struct ata_port *ap)
{
struct pci_dev *pdev = to_pci_dev(ap->host->dev);
u16 val;
pci_read_config_word(pdev, 0x7E, &val);
pci_write_config_word(pdev, 0x7E, val | 4);
pci_write_config_word(pdev, 0x7E, val & ~4);
}
/**
* sl82c105_bmdma_start - DMA engine begin
* @qc: ATA command
*
* Reset the DMA engine each use as recommended by the errata
* document.
*
* FIXME: if we switch clock at BMDMA start/end we might get better
* PIO performance on DMA capable devices.
*/
static void sl82c105_bmdma_start(struct ata_queued_cmd *qc)
{
struct ata_port *ap = qc->ap;
udelay(100);
sl82c105_reset_engine(ap);
udelay(100);
/* Set the clocks for DMA */
sl82c105_configure_dmamode(ap, qc->dev);
/* Activate DMA */
ata_bmdma_start(qc);
}
/**
* sl82c105_bmdma_end - DMA engine stop
* @qc: ATA command
*
* Reset the DMA engine each use as recommended by the errata
* document.
*
* This function is also called to turn off DMA when a timeout occurs
* during DMA operation. In both cases we need to reset the engine,
* so no actual eng_timeout handler is required.
*
* We assume bmdma_stop is always called if bmdma_start as called. If
* not then we may need to wrap qc_issue.
*/
static void sl82c105_bmdma_stop(struct ata_queued_cmd *qc)
{
struct ata_port *ap = qc->ap;
ata_bmdma_stop(qc);
sl82c105_reset_engine(ap);
udelay(100);
/* This will redo the initial setup of the DMA device to matching
PIO timings */
sl82c105_set_piomode(ap, qc->dev);
}
static struct scsi_host_template sl82c105_sht = {
.module = THIS_MODULE,
.name = DRV_NAME,
.ioctl = ata_scsi_ioctl,
.queuecommand = ata_scsi_queuecmd,
.can_queue = ATA_DEF_QUEUE,
.this_id = ATA_SHT_THIS_ID,
.sg_tablesize = LIBATA_MAX_PRD,
.cmd_per_lun = ATA_SHT_CMD_PER_LUN,
.emulated = ATA_SHT_EMULATED,
.use_clustering = ATA_SHT_USE_CLUSTERING,
.proc_name = DRV_NAME,
.dma_boundary = ATA_DMA_BOUNDARY,
.slave_configure = ata_scsi_slave_config,
.slave_destroy = ata_scsi_slave_destroy,
.bios_param = ata_std_bios_param,
};
static struct ata_port_operations sl82c105_port_ops = {
.set_piomode = sl82c105_set_piomode,
.mode_filter = ata_pci_default_filter,
.tf_load = ata_tf_load,
.tf_read = ata_tf_read,
.check_status = ata_check_status,
.exec_command = ata_exec_command,
.dev_select = ata_std_dev_select,
.freeze = ata_bmdma_freeze,
.thaw = ata_bmdma_thaw,
.error_handler = sl82c105_error_handler,
.post_internal_cmd = ata_bmdma_post_internal_cmd,
.cable_detect = ata_cable_40wire,
.bmdma_setup = ata_bmdma_setup,
.bmdma_start = sl82c105_bmdma_start,
.bmdma_stop = sl82c105_bmdma_stop,
.bmdma_status = ata_bmdma_status,
.qc_prep = ata_qc_prep,
.qc_issue = ata_qc_issue_prot,
.data_xfer = ata_data_xfer,
.irq_handler = ata_interrupt,
.irq_clear = ata_bmdma_irq_clear,
.irq_on = ata_irq_on,
.port_start = ata_port_start,
};
/**
* sl82c105_bridge_revision - find bridge version
* @pdev: PCI device for the ATA function
*
* Locates the PCI bridge associated with the ATA function and
* providing it is a Winbond 553 reports the revision. If it cannot
* find a revision or the right device it returns -1
*/
static int sl82c105_bridge_revision(struct pci_dev *pdev)
{
struct pci_dev *bridge;
/*
* The bridge should be part of the same device, but function 0.
*/
bridge = pci_get_slot(pdev->bus,
PCI_DEVFN(PCI_SLOT(pdev->devfn), 0));
if (!bridge)
return -1;
/*
* Make sure it is a Winbond 553 and is an ISA bridge.
*/
if (bridge->vendor != PCI_VENDOR_ID_WINBOND ||
bridge->device != PCI_DEVICE_ID_WINBOND_83C553 ||
bridge->class >> 8 != PCI_CLASS_BRIDGE_ISA) {
pci_dev_put(bridge);
return -1;
}
/*
* We need to find function 0's revision, not function 1
*/
pci_dev_put(bridge);
return bridge->revision;
}
static int sl82c105_init_one(struct pci_dev *dev, const struct pci_device_id *id)
{
libata: clean up SFF init mess The intention of using port_mask in SFF init helpers was to eventually support exoctic configurations such as combination of legacy and native port on the same controller. This never became actually necessary and the related code always has been subtly broken one way or the other. Now that new init model is in place, there is no reason to make common helpers capable of handling all corner cases. Exotic cases can simply dealt within LLDs as necessary. This patch removes port_mask handling in SFF init helpers. SFF init helpers don't take n_ports argument and interpret it into port_mask anymore. All information is carried via port_info. n_ports argument is dropped and always two ports are allocated. LLD can tell SFF to skip certain port by marking it dummy. Note that SFF code has been treating unuvailable ports this way for a long time until recent breakage fix from Linus and is consistent with how other drivers handle with unavailable ports. This fixes 1-port legacy host handling still broken after the recent native mode fix and simplifies SFF init logic. The following changes are made... * ata_pci_init_native_host() and ata_init_legacy_host() both now try to initialized whatever they can and mark failed ports dummy. They return 0 if any port is successfully initialized. * ata_pci_prepare_native_host() and ata_pci_init_one() now doesn't take n_ports argument. All info should be specified via port_info array. Always two ports are allocated. * ata_pci_init_bmdma() exported to be used by LLDs in exotic cases. * port_info handling in all LLDs are standardized - all port_info arrays are const stack variable named ppi. Unless the second port is different from the first, its port_info is specified as NULL (tells libata that it's identical to the last non-NULL port_info). * pata_hpt37x/hpt3x2n: don't modify static variable directly. Make an on-stack copy instead as ata_piix does. * pata_uli: It has 4 ports instead of 2. Don't use ata_pci_prepare_native_host(). Allocate the host explicitly and use init helpers. It's simple enough. Signed-off-by: Tejun Heo <htejun@gmail.com> Signed-off-by: Jeff Garzik <jeff@garzik.org>
2007-05-04 18:43:58 +08:00
static const struct ata_port_info info_dma = {
.sht = &sl82c105_sht,
.flags = ATA_FLAG_SLAVE_POSS,
.pio_mask = 0x1f,
.mwdma_mask = 0x07,
.port_ops = &sl82c105_port_ops
};
libata: clean up SFF init mess The intention of using port_mask in SFF init helpers was to eventually support exoctic configurations such as combination of legacy and native port on the same controller. This never became actually necessary and the related code always has been subtly broken one way or the other. Now that new init model is in place, there is no reason to make common helpers capable of handling all corner cases. Exotic cases can simply dealt within LLDs as necessary. This patch removes port_mask handling in SFF init helpers. SFF init helpers don't take n_ports argument and interpret it into port_mask anymore. All information is carried via port_info. n_ports argument is dropped and always two ports are allocated. LLD can tell SFF to skip certain port by marking it dummy. Note that SFF code has been treating unuvailable ports this way for a long time until recent breakage fix from Linus and is consistent with how other drivers handle with unavailable ports. This fixes 1-port legacy host handling still broken after the recent native mode fix and simplifies SFF init logic. The following changes are made... * ata_pci_init_native_host() and ata_init_legacy_host() both now try to initialized whatever they can and mark failed ports dummy. They return 0 if any port is successfully initialized. * ata_pci_prepare_native_host() and ata_pci_init_one() now doesn't take n_ports argument. All info should be specified via port_info array. Always two ports are allocated. * ata_pci_init_bmdma() exported to be used by LLDs in exotic cases. * port_info handling in all LLDs are standardized - all port_info arrays are const stack variable named ppi. Unless the second port is different from the first, its port_info is specified as NULL (tells libata that it's identical to the last non-NULL port_info). * pata_hpt37x/hpt3x2n: don't modify static variable directly. Make an on-stack copy instead as ata_piix does. * pata_uli: It has 4 ports instead of 2. Don't use ata_pci_prepare_native_host(). Allocate the host explicitly and use init helpers. It's simple enough. Signed-off-by: Tejun Heo <htejun@gmail.com> Signed-off-by: Jeff Garzik <jeff@garzik.org>
2007-05-04 18:43:58 +08:00
static const struct ata_port_info info_early = {
.sht = &sl82c105_sht,
.flags = ATA_FLAG_SLAVE_POSS,
.pio_mask = 0x1f,
.port_ops = &sl82c105_port_ops
};
libata: clean up SFF init mess The intention of using port_mask in SFF init helpers was to eventually support exoctic configurations such as combination of legacy and native port on the same controller. This never became actually necessary and the related code always has been subtly broken one way or the other. Now that new init model is in place, there is no reason to make common helpers capable of handling all corner cases. Exotic cases can simply dealt within LLDs as necessary. This patch removes port_mask handling in SFF init helpers. SFF init helpers don't take n_ports argument and interpret it into port_mask anymore. All information is carried via port_info. n_ports argument is dropped and always two ports are allocated. LLD can tell SFF to skip certain port by marking it dummy. Note that SFF code has been treating unuvailable ports this way for a long time until recent breakage fix from Linus and is consistent with how other drivers handle with unavailable ports. This fixes 1-port legacy host handling still broken after the recent native mode fix and simplifies SFF init logic. The following changes are made... * ata_pci_init_native_host() and ata_init_legacy_host() both now try to initialized whatever they can and mark failed ports dummy. They return 0 if any port is successfully initialized. * ata_pci_prepare_native_host() and ata_pci_init_one() now doesn't take n_ports argument. All info should be specified via port_info array. Always two ports are allocated. * ata_pci_init_bmdma() exported to be used by LLDs in exotic cases. * port_info handling in all LLDs are standardized - all port_info arrays are const stack variable named ppi. Unless the second port is different from the first, its port_info is specified as NULL (tells libata that it's identical to the last non-NULL port_info). * pata_hpt37x/hpt3x2n: don't modify static variable directly. Make an on-stack copy instead as ata_piix does. * pata_uli: It has 4 ports instead of 2. Don't use ata_pci_prepare_native_host(). Allocate the host explicitly and use init helpers. It's simple enough. Signed-off-by: Tejun Heo <htejun@gmail.com> Signed-off-by: Jeff Garzik <jeff@garzik.org>
2007-05-04 18:43:58 +08:00
/* for now use only the first port */
const struct ata_port_info *ppi[] = { &info_early,
&ata_dummy_port_info };
u32 val;
int rev;
rev = sl82c105_bridge_revision(dev);
if (rev == -1)
dev_printk(KERN_WARNING, &dev->dev, "pata_sl82c105: Unable to find bridge, disabling DMA.\n");
else if (rev <= 5)
dev_printk(KERN_WARNING, &dev->dev, "pata_sl82c105: Early bridge revision, no DMA available.\n");
libata: clean up SFF init mess The intention of using port_mask in SFF init helpers was to eventually support exoctic configurations such as combination of legacy and native port on the same controller. This never became actually necessary and the related code always has been subtly broken one way or the other. Now that new init model is in place, there is no reason to make common helpers capable of handling all corner cases. Exotic cases can simply dealt within LLDs as necessary. This patch removes port_mask handling in SFF init helpers. SFF init helpers don't take n_ports argument and interpret it into port_mask anymore. All information is carried via port_info. n_ports argument is dropped and always two ports are allocated. LLD can tell SFF to skip certain port by marking it dummy. Note that SFF code has been treating unuvailable ports this way for a long time until recent breakage fix from Linus and is consistent with how other drivers handle with unavailable ports. This fixes 1-port legacy host handling still broken after the recent native mode fix and simplifies SFF init logic. The following changes are made... * ata_pci_init_native_host() and ata_init_legacy_host() both now try to initialized whatever they can and mark failed ports dummy. They return 0 if any port is successfully initialized. * ata_pci_prepare_native_host() and ata_pci_init_one() now doesn't take n_ports argument. All info should be specified via port_info array. Always two ports are allocated. * ata_pci_init_bmdma() exported to be used by LLDs in exotic cases. * port_info handling in all LLDs are standardized - all port_info arrays are const stack variable named ppi. Unless the second port is different from the first, its port_info is specified as NULL (tells libata that it's identical to the last non-NULL port_info). * pata_hpt37x/hpt3x2n: don't modify static variable directly. Make an on-stack copy instead as ata_piix does. * pata_uli: It has 4 ports instead of 2. Don't use ata_pci_prepare_native_host(). Allocate the host explicitly and use init helpers. It's simple enough. Signed-off-by: Tejun Heo <htejun@gmail.com> Signed-off-by: Jeff Garzik <jeff@garzik.org>
2007-05-04 18:43:58 +08:00
else
ppi[0] = &info_dma;
pci_read_config_dword(dev, 0x40, &val);
val |= CTRL_P0EN | CTRL_P0F16 | CTRL_P1F16;
pci_write_config_dword(dev, 0x40, val);
libata: clean up SFF init mess The intention of using port_mask in SFF init helpers was to eventually support exoctic configurations such as combination of legacy and native port on the same controller. This never became actually necessary and the related code always has been subtly broken one way or the other. Now that new init model is in place, there is no reason to make common helpers capable of handling all corner cases. Exotic cases can simply dealt within LLDs as necessary. This patch removes port_mask handling in SFF init helpers. SFF init helpers don't take n_ports argument and interpret it into port_mask anymore. All information is carried via port_info. n_ports argument is dropped and always two ports are allocated. LLD can tell SFF to skip certain port by marking it dummy. Note that SFF code has been treating unuvailable ports this way for a long time until recent breakage fix from Linus and is consistent with how other drivers handle with unavailable ports. This fixes 1-port legacy host handling still broken after the recent native mode fix and simplifies SFF init logic. The following changes are made... * ata_pci_init_native_host() and ata_init_legacy_host() both now try to initialized whatever they can and mark failed ports dummy. They return 0 if any port is successfully initialized. * ata_pci_prepare_native_host() and ata_pci_init_one() now doesn't take n_ports argument. All info should be specified via port_info array. Always two ports are allocated. * ata_pci_init_bmdma() exported to be used by LLDs in exotic cases. * port_info handling in all LLDs are standardized - all port_info arrays are const stack variable named ppi. Unless the second port is different from the first, its port_info is specified as NULL (tells libata that it's identical to the last non-NULL port_info). * pata_hpt37x/hpt3x2n: don't modify static variable directly. Make an on-stack copy instead as ata_piix does. * pata_uli: It has 4 ports instead of 2. Don't use ata_pci_prepare_native_host(). Allocate the host explicitly and use init helpers. It's simple enough. Signed-off-by: Tejun Heo <htejun@gmail.com> Signed-off-by: Jeff Garzik <jeff@garzik.org>
2007-05-04 18:43:58 +08:00
return ata_pci_init_one(dev, ppi);
}
static const struct pci_device_id sl82c105[] = {
{ PCI_VDEVICE(WINBOND, PCI_DEVICE_ID_WINBOND_82C105), },
{ },
};
static struct pci_driver sl82c105_pci_driver = {
.name = DRV_NAME,
.id_table = sl82c105,
.probe = sl82c105_init_one,
.remove = ata_pci_remove_one
};
static int __init sl82c105_init(void)
{
return pci_register_driver(&sl82c105_pci_driver);
}
static void __exit sl82c105_exit(void)
{
pci_unregister_driver(&sl82c105_pci_driver);
}
MODULE_AUTHOR("Alan Cox");
MODULE_DESCRIPTION("low-level driver for Sl82c105");
MODULE_LICENSE("GPL");
MODULE_DEVICE_TABLE(pci, sl82c105);
MODULE_VERSION(DRV_VERSION);
module_init(sl82c105_init);
module_exit(sl82c105_exit);