Re: [PATCH] Add 'tpgs' sysfs attribute for SCSI devices

[James Bottomley <James.Bottomley@HansenPartnership.com>]

On Wed, 2009-03-18 at 15:02 +0100, Hannes Reinecke wrote:
> James Bottomley wrote:
> > On Wed, 2009-03-18 at 13:02 +0100, Hannes Reinecke wrote:
> >> We need to export the 'TGPS' setting of the
> >> inquiry data to properly support ALUA.
> >>
> >> Signed-off-by: Hannes Reinecke <hare@suse.de>
> >> ---
> >>  drivers/scsi/scsi_scan.c   |    1 +
> >>  drivers/scsi/scsi_sysfs.c  |    2 ++
> >>  include/scsi/scsi_device.h |    3 ++-
> >>  3 files changed, 5 insertions(+), 1 deletions(-)
> >>
> >> diff --git a/drivers/scsi/scsi_scan.c b/drivers/scsi/scsi_scan.c
> >> index a14d245..d1619ed 100644
> >> --- a/drivers/scsi/scsi_scan.c
> >> +++ b/drivers/scsi/scsi_scan.c
> >> @@ -836,6 +836,7 @@ static int scsi_add_lun(struct scsi_device *sdev, unsigned char *inq_result,
> >>  	sdev->inq_periph_qual = (inq_result[0] >> 5) & 7;
> >>  	sdev->lockable = sdev->removable;
> >>  	sdev->soft_reset = (inq_result[7] & 1) && ((inq_result[3] & 7) == 2);
> >> +	sdev->tpgs = (inq_result[5] >> 4) & 3;
> > 
> > As far as I can tell, this is a read only parameter, isn't it?
> > 
> > In that case, why not just do what we do for the SPI inquiry data and
> > have a wrapper pulling it out of sdev->inquiry instead of a new flag in
> > sdev?
> > 
> Because we're evaluating it later on from the device_handler.
> But honestly I didn't inquire in that direction.
> I'll be checking.

As long as you have access to the sdev, you have the inquiry data.

The idea would just be to have a static function

static int scsi_device_tpgs(struct scsi_device *sdev)
{
	return (sdev->inquiry[5] >> 4) &3;
}

We have a few others like this in scsi_device.h (in scsi-misc,
otherwise, I think they're in scsi.h)

James