* [PATCH 1/2] libata-dev: determine err_mask when error is found
@ 2005-12-26 8:40 Albert Lee
2006-01-18 0:55 ` Jeff Garzik
0 siblings, 1 reply; 2+ messages in thread
From: Albert Lee @ 2005-12-26 8:40 UTC (permalink / raw)
To: Jeff Garzik; +Cc: Linux IDE, Doug Maxey
Jeff,
Patch 1/2: Minor fix to determine err_mask directly when an error is found.
Changes:
- Determine err_mask directly when an error is found.
- Remove "qc->err_mask |= __ac_err_mask(status);" in ata_host_intr()
Patch against the libata-dev irq-pio branch
(3d0a59c02303df01848537b3bf938dc11e9a0ded).
For your review, thanks.
Albert
Signed-off-by: Albert Lee <albertcc@tw.ibm.com>
============
--- irq-pio/drivers/scsi/libata-core.c 2005-12-26 11:25:52.000000000 +0800
+++ err_mask/drivers/scsi/libata-core.c 2005-12-26 14:59:53.000000000 +0800
@@ -3209,6 +3209,7 @@ static int ata_pio_first_block(struct at
/* sleep-wait for BSY to clear */
DPRINTK("busy wait\n");
if (ata_busy_sleep(ap, ATA_TMOUT_DATAOUT_QUICK, ATA_TMOUT_DATAOUT)) {
+ qc->err_mask |= AC_ERR_ATA_BUS;
ap->hsm_task_state = HSM_ST_TMOUT;
goto err_out;
}
@@ -3217,6 +3218,7 @@ static int ata_pio_first_block(struct at
status = ata_chk_status(ap);
if ((status & (ATA_BUSY | ATA_DRQ)) != ATA_DRQ) {
/* device status error */
+ qc->err_mask |= AC_ERR_ATA_BUS;
ap->hsm_task_state = HSM_ST_ERR;
goto err_out;
}
@@ -4232,6 +4234,12 @@ inline unsigned int ata_host_intr (struc
/* before we do anything else, clear DMA-Start bit */
ap->ops->bmdma_stop(qc);
+
+ if (unlikely(host_stat & ATA_DMA_ERR)) {
+ /* error when transfering data to/from memory */
+ qc->err_mask |= AC_ERR_HOST_BUS;
+ ap->hsm_task_state = HSM_ST_ERR;
+ }
}
break;
case HSM_ST:
@@ -4257,8 +4265,10 @@ inline unsigned int ata_host_intr (struc
ap->ops->irq_clear(ap);
/* check error */
- if (unlikely((status & ATA_ERR) || (host_stat & ATA_DMA_ERR)))
+ if (unlikely(status & (ATA_ERR | ATA_DF))) {
+ qc->err_mask |= AC_ERR_DEV;
ap->hsm_task_state = HSM_ST_ERR;
+ }
fsm_start:
switch (ap->hsm_task_state) {
@@ -4270,6 +4280,7 @@ fsm_start:
/* check device status */
if (unlikely((status & (ATA_BUSY | ATA_DRQ)) != ATA_DRQ)) {
/* Wrong status. Let EH handle this */
+ qc->err_mask |= AC_ERR_ATA_BUS;
ap->hsm_task_state = HSM_ST_ERR;
goto fsm_start;
}
@@ -4298,6 +4309,7 @@ fsm_start:
/* ATA PIO protocol */
if (unlikely((status & ATA_DRQ) == 0)) {
/* handle BSY=0, DRQ=0 as error */
+ qc->err_mask |= AC_ERR_ATA_BUS;
ap->hsm_task_state = HSM_ST_ERR;
goto fsm_start;
}
@@ -4319,6 +4331,7 @@ fsm_start:
case HSM_ST_LAST:
if (unlikely(status & ATA_DRQ)) {
/* handle DRQ=1 as error */
+ qc->err_mask |= AC_ERR_ATA_BUS;
ap->hsm_task_state = HSM_ST_ERR;
goto fsm_start;
}
@@ -4338,8 +4351,12 @@ fsm_start:
printk(KERN_ERR "ata%u: command error, drv_stat 0x%x host_stat 0x%x\n",
ap->id, status, host_stat);
+ /* make sure qc->err_mask is available to
+ * know what's wrong and recover
+ */
+ assert(qc->err_mask);
+
ap->hsm_task_state = HSM_ST_IDLE;
- qc->err_mask |= __ac_err_mask(status);
ata_qc_complete(qc);
break;
default:
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: [PATCH 1/2] libata-dev: determine err_mask when error is found
2005-12-26 8:40 [PATCH 1/2] libata-dev: determine err_mask when error is found Albert Lee
@ 2006-01-18 0:55 ` Jeff Garzik
0 siblings, 0 replies; 2+ messages in thread
From: Jeff Garzik @ 2006-01-18 0:55 UTC (permalink / raw)
To: Albert Lee; +Cc: Linux IDE, Doug Maxey
Albert Lee wrote:
> Jeff,
>
> Patch 1/2: Minor fix to determine err_mask directly when an error is found.
>
> Changes:
> - Determine err_mask directly when an error is found.
> - Remove "qc->err_mask |= __ac_err_mask(status);" in ata_host_intr()
>
> Patch against the libata-dev irq-pio branch
> (3d0a59c02303df01848537b3bf938dc11e9a0ded).
applied 1-2 to irq-pio branch
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2006-01-18 0:55 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2005-12-26 8:40 [PATCH 1/2] libata-dev: determine err_mask when error is found Albert Lee
2006-01-18 0:55 ` Jeff Garzik
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).