[PATCH] ata: libata-eh: Do not retry reset if the device is gone

Igor Pylypiv posted 1 patch 2 months, 1 week ago
There is a newer version of this series
drivers/ata/libata-eh.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
[PATCH] ata: libata-eh: Do not retry reset if the device is gone
Posted by Igor Pylypiv 2 months, 1 week ago
If a device is hot-unplugged or otherwise disappears during error handling,
ata_eh_reset() may fail with -ENODEV. Currently, the error handler will
continue to retry the reset operation up to max_tries times.

Prevent unnecessary reset retries by exiting the loop early when
ata_eh_reset() returns -ENODEV.

Signed-off-by: Igor Pylypiv <ipylypiv@google.com>
---
 drivers/ata/libata-eh.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c
index 23be85418b3b..e97a842005e9 100644
--- a/drivers/ata/libata-eh.c
+++ b/drivers/ata/libata-eh.c
@@ -3171,7 +3171,7 @@ int ata_eh_reset(struct ata_link *link, int classify,
 	    sata_scr_read(link, SCR_STATUS, &sstatus))
 		rc = -ERESTART;
 
-	if (try >= max_tries) {
+	if (try >= max_tries || rc == -ENODEV) {
 		/*
 		 * Thaw host port even if reset failed, so that the port
 		 * can be retried on the next phy event.  This risks
-- 
2.53.0.1185.g05d4b7b318-goog
Re: [PATCH] ata: libata-eh: Do not retry reset if the device is gone
Posted by Niklas Cassel 2 months, 1 week ago
On Wed, Apr 01, 2026 at 06:11:01PM -0700, Igor Pylypiv wrote:
> If a device is hot-unplugged or otherwise disappears during error handling,
> ata_eh_reset() may fail with -ENODEV. Currently, the error handler will
> continue to retry the reset operation up to max_tries times.
> 
> Prevent unnecessary reset retries by exiting the loop early when
> ata_eh_reset() returns -ENODEV.

ata_do_reset() ?

The loop is inside ata_eh_reset(), so it sounds a bit weird to exit
the loop early if the function itself returns -ENODEV.


I guess you could also write it like:
Prevent unnecessary reset retries by exiting the loop early when
a callee in ata_eh_reset() returns -ENODEV.


Kind regards,
Niklas
Re: [PATCH] ata: libata-eh: Do not retry reset if the device is gone
Posted by Igor Pylypiv 2 months, 1 week ago
On Thu, Apr 02, 2026 at 10:55:41AM +0200, Niklas Cassel wrote:
> On Wed, Apr 01, 2026 at 06:11:01PM -0700, Igor Pylypiv wrote:
> > If a device is hot-unplugged or otherwise disappears during error handling,
> > ata_eh_reset() may fail with -ENODEV. Currently, the error handler will
> > continue to retry the reset operation up to max_tries times.
> > 
> > Prevent unnecessary reset retries by exiting the loop early when
> > ata_eh_reset() returns -ENODEV.
> 
> ata_do_reset() ?
> 
> The loop is inside ata_eh_reset(), so it sounds a bit weird to exit
> the loop early if the function itself returns -ENODEV.
>
 
Thank you for catching this, Niklas!

Yes, it should be ata_do_reset(). I'll send a V2 to fix the function name.

Thanks,
Igor
> 
> I guess you could also write it like:
> Prevent unnecessary reset retries by exiting the loop early when
> a callee in ata_eh_reset() returns -ENODEV.
> 
> 
> Kind regards,
> Niklas
Re: [PATCH] ata: libata-eh: Do not retry reset if the device is gone
Posted by Damien Le Moal 2 months, 1 week ago
On 2026/04/02 10:11, Igor Pylypiv wrote:
> If a device is hot-unplugged or otherwise disappears during error handling,
> ata_eh_reset() may fail with -ENODEV. Currently, the error handler will
> continue to retry the reset operation up to max_tries times.
> 
> Prevent unnecessary reset retries by exiting the loop early when
> ata_eh_reset() returns -ENODEV.
> 
> Signed-off-by: Igor Pylypiv <ipylypiv@google.com>

Looks good.

Reviewed-by: Damien Le Moal <dlemoal@kernel.org>

-- 
Damien Le Moal
Western Digital Research