1
0
Fork 0
alistair23-linux/drivers/ata/pata_jmicron.c

254 lines
6.7 KiB
C
Raw Normal View History

/*
* pata_jmicron.c - JMicron ATA driver for non AHCI mode. This drives the
* PATA port of the controller. The SATA ports are
* driven by AHCI in the usual configuration although
* this driver can handle other setups if we need it.
*
* (c) 2006 Red Hat <alan@redhat.com>
*/
#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 <linux/device.h>
#include <scsi/scsi_host.h>
#include <linux/libata.h>
#include <linux/ata.h>
#define DRV_NAME "pata_jmicron"
#define DRV_VERSION "0.1.4"
typedef enum {
PORT_PATA0 = 0,
PORT_PATA1 = 1,
PORT_SATA = 2,
} port_type;
/**
* jmicron_pre_reset - check for 40/80 pin
* @ap: Port
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 00:50:52 -07:00
* @deadline: deadline jiffies for the operation
*
* Perform the PATA port setup we need.
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 00:50:52 -07:00
*
* On the Jmicron 361/363 there is a single PATA port that can be mapped
* either as primary or secondary (or neither). We don't do any policy
* and setup here. We assume that has been done by init_one and the
* BIOS.
*/
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 00:50:52 -07:00
static int jmicron_pre_reset(struct ata_port *ap, unsigned long deadline)
{
struct pci_dev *pdev = to_pci_dev(ap->host->dev);
u32 control;
u32 control5;
int port_mask = 1<< (4 * ap->port_no);
int port = ap->port_no;
port_type port_map[2];
/* Check if our port is enabled */
pci_read_config_dword(pdev, 0x40, &control);
if ((control & port_mask) == 0)
return -ENOENT;
/* There are two basic mappings. One has the two SATA ports merged
as master/slave and the secondary as PATA, the other has only the
SATA port mapped */
if (control & (1 << 23)) {
port_map[0] = PORT_SATA;
port_map[1] = PORT_PATA0;
} else {
port_map[0] = PORT_SATA;
port_map[1] = PORT_SATA;
}
/* The 365/366 may have this bit set to map the second PATA port
as the internal primary channel */
pci_read_config_dword(pdev, 0x80, &control5);
if (control5 & (1<<24))
port_map[0] = PORT_PATA1;
/* The two ports may then be logically swapped by the firmware */
if (control & (1 << 22))
port = port ^ 1;
/*
* Now we know which physical port we are talking about we can
* actually do our cable checking etc. Thankfully we don't need
* to do the plumbing for other cases.
*/
switch (port_map[port])
{
case PORT_PATA0:
if (control & (1 << 5))
return 0;
if (control & (1 << 3)) /* 40/80 pin primary */
ap->cbl = ATA_CBL_PATA40;
else
ap->cbl = ATA_CBL_PATA80;
break;
case PORT_PATA1:
/* Bit 21 is set if the port is enabled */
if ((control5 & (1 << 21)) == 0)
return 0;
if (control5 & (1 << 19)) /* 40/80 pin secondary */
ap->cbl = ATA_CBL_PATA40;
else
ap->cbl = ATA_CBL_PATA80;
break;
case PORT_SATA:
ap->cbl = ATA_CBL_SATA;
break;
}
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 00:50:52 -07:00
return ata_std_prereset(ap, deadline);
}
/**
* jmicron_error_handler - Setup and error handler
* @ap: Port to handle
*
* LOCKING:
* None (inherited from caller).
*/
static void jmicron_error_handler(struct ata_port *ap)
{
return ata_bmdma_drive_eh(ap, jmicron_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
}
/* No PIO or DMA methods needed for this device */
static struct scsi_host_template jmicron_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,
/* Use standard CHS mapping rules */
.bios_param = ata_std_bios_param,
};
static const struct ata_port_operations jmicron_ops = {
.port_disable = ata_port_disable,
/* Task file is PCI ATA format, use helpers */
.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 = jmicron_error_handler,
.post_internal_cmd = ata_bmdma_post_internal_cmd,
/* BMDMA handling is PCI ATA format, use helpers */
.bmdma_setup = ata_bmdma_setup,
.bmdma_start = ata_bmdma_start,
.bmdma_stop = ata_bmdma_stop,
.bmdma_status = ata_bmdma_status,
.qc_prep = ata_qc_prep,
.qc_issue = ata_qc_issue_prot,
.data_xfer = ata_data_xfer,
/* IRQ-related hooks */
.irq_handler = ata_interrupt,
.irq_clear = ata_bmdma_irq_clear,
.irq_on = ata_irq_on,
.irq_ack = ata_irq_ack,
/* Generic PATA PCI ATA helpers */
.port_start = ata_port_start,
};
/**
* jmicron_init_one - Register Jmicron ATA PCI device with kernel services
* @pdev: PCI device to register
* @ent: Entry in jmicron_pci_tbl matching with @pdev
*
* Called from kernel PCI layer.
*
* LOCKING:
* Inherited from PCI layer (may sleep).
*
* RETURNS:
* Zero on success, or -ERRNO value.
*/
static int jmicron_init_one (struct pci_dev *pdev, const struct pci_device_id *id)
{
static struct ata_port_info info = {
.sht = &jmicron_sht,
.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
.pio_mask = 0x1f,
.mwdma_mask = 0x07,
.udma_mask = 0x3f,
.port_ops = &jmicron_ops,
};
struct ata_port_info *port_info[2] = { &info, &info };
return ata_pci_init_one(pdev, port_info, 2);
}
static const struct pci_device_id jmicron_pci_tbl[] = {
{ PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB361,
PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_STORAGE_IDE << 8, 0xffff00, 361 },
{ PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB363,
PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_STORAGE_IDE << 8, 0xffff00, 363 },
{ PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB365,
PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_STORAGE_IDE << 8, 0xffff00, 365 },
{ PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB366,
PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_STORAGE_IDE << 8, 0xffff00, 366 },
{ PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB368,
PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_STORAGE_IDE << 8, 0xffff00, 368 },
{ } /* terminate list */
};
static struct pci_driver jmicron_pci_driver = {
.name = DRV_NAME,
.id_table = jmicron_pci_tbl,
.probe = jmicron_init_one,
.remove = ata_pci_remove_one,
#ifdef CONFIG_PM
.suspend = ata_pci_device_suspend,
.resume = ata_pci_device_resume,
#endif
};
static int __init jmicron_init(void)
{
return pci_register_driver(&jmicron_pci_driver);
}
static void __exit jmicron_exit(void)
{
pci_unregister_driver(&jmicron_pci_driver);
}
module_init(jmicron_init);
module_exit(jmicron_exit);
MODULE_AUTHOR("Alan Cox");
MODULE_DESCRIPTION("SCSI low-level driver for Jmicron PATA ports");
MODULE_LICENSE("GPL");
MODULE_DEVICE_TABLE(pci, jmicron_pci_tbl);
MODULE_VERSION(DRV_VERSION);