[PATCH #upstream-fixes] libata: set device class to NONE if phys_offline

[PATCH #upstream-fixes] libata: set device class to NONE if phys_offline

[Tejun Heo]

Reset methods don't have access to phys link status for slave links
and may incorrectly indicate device presence causing unnecessary probe
failures for unoccupied links.  This patch clears device class to NONE
during post-reset processing if phys link is offline.

As on/offliness semantics is strictly defined and used in multiple
places by the core layer, this won't change behavior for drivers which
don't use slave links.

Signed-off-by: Tejun Heo <tj@kernel.org>
---
vt8251 incorrectly reports the same signature for unoccupied device if
the other device on the channel is present.  This patch fixes the
problem.

I integrated this patch into separate-sff-bmdma git tree.

 drivers/ata/libata-eh.c |   16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

Index: work/drivers/ata/libata-eh.c
===================================================================
--- work.orig/drivers/ata/libata-eh.c
+++ work/drivers/ata/libata-eh.c
@@ -2455,14 +2455,14 @@ int ata_eh_reset(struct ata_link *link,
 		dev->pio_mode = XFER_PIO_0;
 		dev->flags &= ~ATA_DFLAG_SLEEPING;
 
-		if (ata_phys_link_offline(ata_dev_phys_link(dev)))
-			continue;
-
-		/* 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 */
+		if (!ata_phys_link_offline(ata_dev_phys_link(dev))) {
+			/* 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;
+		} else
+			classes[dev->devno] = ATA_DEV_NONE;
 	}
 
 	/* record current link speed */

Re: [PATCH #upstream-fixes] libata: set device class to NONE if phys_offline

[Sergei Shtylyov]

Hello.

Tejun Heo wrote:

> Reset methods don't have access to phys link status for slave links
> and may incorrectly indicate device presence causing unnecessary probe
> failures for unoccupied links.  This patch clears device class to NONE
> during post-reset processing if phys link is offline.
>
> As on/offliness semantics is strictly defined and used in multiple
>   

   More grammar nitpicking: it's either on/offline or on/offlineness. :-)

MBR, Sergei

Re: [PATCH #upstream-fixes] libata: set device class to NONE if phys_offline

[Tejun Heo]

Sergei Shtylyov wrote:
>> As on/offliness semantics is strictly defined and used in multiple
>>   
> 
>   More grammar nitpicking: it's either on/offline or on/offlineness. :-)

Right.  I'll keep it in mind.  I'm so glad that you're reviewing
libata changes.  Thanks.  :-)

-- 
tejun

Re: [PATCH #upstream-fixes] libata: set device class to NONE if phys_offline

[Jeff Garzik]

Tejun Heo wrote:
> Reset methods don't have access to phys link status for slave links
> and may incorrectly indicate device presence causing unnecessary probe
> failures for unoccupied links.  This patch clears device class to NONE
> during post-reset processing if phys link is offline.
> 
> As on/offliness semantics is strictly defined and used in multiple
> places by the core layer, this won't change behavior for drivers which
> don't use slave links.
> 
> Signed-off-by: Tejun Heo <tj@kernel.org>
> ---
> vt8251 incorrectly reports the same signature for unoccupied device if
> the other device on the channel is present.  This patch fixes the
> problem.
> 
> I integrated this patch into separate-sff-bmdma git tree.
> 
>  drivers/ata/libata-eh.c |   16 ++++++++--------
>  1 file changed, 8 insertions(+), 8 deletions(-)

applied