[PATCH] libata: give devices one last chance even if recovery failed with -EINVAL

[PATCH] libata: give devices one last chance even if recovery failed with -EINVAL

[Tejun Heo]

After certain errors, some devices report complete garbage on
IDENTIFY.  This can cause ata_dev_read_id() to fail with -EINVAL
resulting in immediate disabling of the device.  Give the device one
last chance after -EINVAL to allow recovery from such situations.  As
-EINVAL is triggered very rarely, this shouldn't cause any noticeable
affect on more common error paths.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Cc: Harald Dunkel <harald.dunkel@t-online.de>

diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c
index 8256655..d7f62c7 100644
--- a/drivers/ata/libata-eh.c
+++ b/drivers/ata/libata-eh.c
@@ -2201,14 +2201,10 @@ static int ata_eh_recover(struct ata_por
 	ehc->tries[dev->devno]--;
 
 	switch (rc) {
-	case -EINVAL:
-		/* eeek, something went very wrong, give up */
-		ehc->tries[dev->devno] = 0;
-		break;
-
 	case -ENODEV:
 		/* device missing or wrong IDENTIFY data, schedule probing */
 		ehc->i.probe_mask |= (1 << dev->devno);
+	case -EINVAL:
 		/* give it just one more chance */
 		ehc->tries[dev->devno] = min(ehc->tries[dev->devno], 1);
 	case -EIO:

Re: [PATCH] libata: give devices one last chance even if recovery failed with -EINVAL

[Jeff Garzik]

Tejun Heo wrote:
> After certain errors, some devices report complete garbage on
> IDENTIFY.  This can cause ata_dev_read_id() to fail with -EINVAL
> resulting in immediate disabling of the device.  Give the device one
> last chance after -EINVAL to allow recovery from such situations.  As
> -EINVAL is triggered very rarely, this shouldn't cause any noticeable
> affect on more common error paths.
> 
> Signed-off-by: Tejun Heo <htejun@gmail.com>
> Cc: Harald Dunkel <harald.dunkel@t-online.de>

applied