[PATCH #upstream-fixes] libata: don't normalize UNKNOWN to NONE after reset

[PATCH #upstream-fixes] libata: don't normalize UNKNOWN to NONE after reset

[Tejun Heo]

After non-classifying reset, ehc->classes[] could contain
ATA_DEV_UNKNOWN which used to be normalized to ATA_DEV_NONE for
consistency.  However, this causes unfortunate side effect for drivers
which have non-classifying hardresets (e.g. sata_nv) by making
hardreset report ATA_DEV_NONE for non-classifying resets and thus
makes EH believe that the port is unoccupied and recovery can be
skipped.  The end result is that after a device is swapped with
another one, the new device isn't attached after the old one is
detached.

This patch makes ata_eh_reset() not normalize UNKNOWN to NONE after
non-classifying resets.  This fixes the above problem.  As UNKNOWN and
NONE are handled differently by only EH hotplug logic, this doesn't
cause other behavior changes.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Cc: Robert Hancock <hancockr@shaw.ca>
---
Robert, missing new device was a bug in core EH logic not sata_nv.
BTW, I did quite a few hotplug tests with adma=0 and it works just
fine.  No lock up.

 drivers/ata/libata-eh.c |    4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c
index f0124a8..e7757dd 100644
--- a/drivers/ata/libata-eh.c
+++ b/drivers/ata/libata-eh.c
@@ -2157,13 +2157,11 @@ int ata_eh_reset(struct ata_link *link, int classify,
 		if (ata_link_offline(link))
 			continue;
 
-		/* apply class override and convert UNKNOWN to NONE */
+		/* apply class override */
 		if (lflags & ATA_LFLAG_ASSUME_ATA)
 			classes[dev->devno] = ATA_DEV_ATA;
 		else if (lflags & ATA_LFLAG_ASSUME_SEMB)
 			classes[dev->devno] = ATA_DEV_SEMB_UNSUP; /* not yet */
-		else if (classes[dev->devno] == ATA_DEV_UNKNOWN)
-			classes[dev->devno] = ATA_DEV_NONE;
 	}
 
 	/* record current link speed */

Re: [PATCH #upstream-fixes] libata: don't normalize UNKNOWN to NONE after reset

[Jeff Garzik]

Tejun Heo wrote:
> After non-classifying reset, ehc->classes[] could contain
> ATA_DEV_UNKNOWN which used to be normalized to ATA_DEV_NONE for
> consistency.  However, this causes unfortunate side effect for drivers
> which have non-classifying hardresets (e.g. sata_nv) by making
> hardreset report ATA_DEV_NONE for non-classifying resets and thus
> makes EH believe that the port is unoccupied and recovery can be
> skipped.  The end result is that after a device is swapped with
> another one, the new device isn't attached after the old one is
> detached.
> 
> This patch makes ata_eh_reset() not normalize UNKNOWN to NONE after
> non-classifying resets.  This fixes the above problem.  As UNKNOWN and
> NONE are handled differently by only EH hotplug logic, this doesn't
> cause other behavior changes.
> 
> Signed-off-by: Tejun Heo <htejun@gmail.com>
> Cc: Robert Hancock <hancockr@shaw.ca>
> ---
> Robert, missing new device was a bug in core EH logic not sata_nv.
> BTW, I did quite a few hotplug tests with adma=0 and it works just
> fine.  No lock up.
> 
>  drivers/ata/libata-eh.c |    4 +---
>  1 file changed, 1 insertion(+), 3 deletions(-)

applied