Hi all,

the attached patch tries to fix SATA ATAPI error handling.
The original code invokes scsi_finish_command() regardless whether this
command is processed normally or by scsi_eh.
This looks quite dangerous to me as this might trigger a recovery for a
command which already is in recovery (as it's processed by scsi_eh).
Plus the error handling _never_ clears eh_cmd_q, leaving failed command
forever on that queue.

The attached patch models the error handling closely to the existing
error handling in scsi_unjam_host(), with the execption that we rely on
a proper sense code being returned by the strategy handler.

Comments are welcome.

Cheers,

Hannes
-- 
Dr. Hannes Reinecke			hare@suse.de
SuSE Linux AG				S390 & zSeries
Maxfeldstraße 5				+49 911 74053 688
90409 Nürnberg				http://www.suse.de