Re: [PATCH] libata: remove unlock+relock cycle in ata_scsi_queuecmd

[Tejun Heo <tj@kernel.org>]

Hello, Jeff, Linus.

On 11/17/2010 09:08 AM, Jeff Garzik wrote:
> Looking solely at the SCSI code (ie. ignoring LLD code), it seems
> like the magic number zero for serial_number is signaling a boolean
> condition "are we an EH command?"
> 
> EH tests this at the very beginning of the abort/reset/explode error
> handling sequence, presumably to avoid recursive EH invocations
> (scsi_try_to_abort_cmd).
>
> So maybe an EH expert (Tejun?) can correct me here, but I think we
> may be able to completely the lock/get-serial/unlock sequence from
> libata, as long as scsi_init_cmd_errh() reliably sets an "I am an EH
> command" flag.
> 
> Would be nice if true...

Yeah, it's actually nice (for once).  libata doesn't use or care about
scmd->serial_number at all.  The SCSI EH path you mentioned above is
not applicable as libata implements its eh_strategy_handler and SCSI
only calls scsi_try_to_abort_cmd() for the default EH handler,
scsi_unjam_host().

We'll need to test a bit to make sure everything is okay but I'm
fairly certain removing it won't break anything fundamental.  If
something breaks at all, it would be some silly easy-to-fix thing.

Thanks.

-- 
tejun