Re: [PATCH] [libata] Fix HDIO_DRIVE_CMD ioctl sense data check

[Ken Moffat <zarniwhoop@ntlworld.com>]

On Thu, Mar 28, 2013 at 10:56:49PM -0700, Gwendal Grignou wrote:

 Hmm, not sure.  Smartd started and was happy to monitor the disk,
but I got two new messages between 'found in smartd database' and
'is SMART capable. Adding to "monitor" list' -

Mar 29 17:26:42 ac4tv smartd[2481]: Device: /dev/sda, not capable of
SMART Health Status check
Mar 29 17:26:42 ac4tv smartd[2481]: Device: /dev/sda, enable SMART
Automatic Offline Testing failed.

 I've seen the first (intermittently) when a drive was starting to
fail, and apparently there was a taskfile issue in the days of 2.6.22
which also caused it to appear.  I don't think I've seen the second
of these before.

 After going back and forth between the kernel where I reverted your
original patch, and regular rc4 plus this new patch the output from
running smartctl as root all seems to be consistent (including
'Passed' for the health check).

 I'm now running with the patch again, and I've started a manual
'long' test (which will take 85 minutes, the default 'offline' is
about 150 minutes).

> commit 84a9a8cd9d0aa93c17e5815ab8a9cc4c0a765c63 changed the sense key
> used for returning task registers, but HDIO_DRIVE_CMD ioctl was
> not changed accordingly.
> 
> Tested: check that SMART ENABLE sent using HDIO_DRIVE_CMD returns 0
> instead of EIO.
> 
> Signed-off-by: Gwendal Grignou <gwendal@google.com>
> ---
>  drivers/ata/libata-scsi.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c
> index 318b413..e05bf4c 100644
> --- a/drivers/ata/libata-scsi.c
> +++ b/drivers/ata/libata-scsi.c
> @@ -532,8 +532,8 @@ int ata_cmd_ioctl(struct scsi_device *scsidev, void __user *arg)
>  			struct scsi_sense_hdr sshdr;
>  			scsi_normalize_sense(sensebuf, SCSI_SENSE_BUFFERSIZE,
>  					     &sshdr);
> -			if (sshdr.sense_key == 0 &&
> -			    sshdr.asc == 0 && sshdr.ascq == 0)
> +			if (sshdr.sense_key == RECOVERED_ERROR &&
> +			    sshdr.asc == 0 && sshdr.ascq == 0x1D)
>  				cmd_result &= ~SAM_STAT_CHECK_CONDITION;
>  		}
>  
> -- 
> 1.8.1.3

-- 
das eine Mal als Tragödie, das andere Mal als Farce