Commit e500a34b authored by Dan Williams's avatar Dan Williams Committed by James Bottomley

[SCSI] libsas: kill invocation of scsi_eh_finish_cmd from sas_ata_task_done

Prior to the conversion to the new-style libata-eh sas_ata_task_done()
may have been the last opportunity to clean up the scmd, but now
libata-eh explicitly handles this case.  It also races against sas-eh.
If a lldd completes a task after SAS_TASK_STATE_ABORTED is set it could
trigger a spurious decrement of shost->host_failed.  Current lldds have
the band-aid of checking SAS_TASK_STATE_ABORTED before calling
->task_done(), but better to just let the scmds escalate to libata for
race free cleanup.
Signed-off-by: default avatarDan Williams <dan.j.williams@intel.com>
Signed-off-by: default avatarJames Bottomley <JBottomley@Parallels.com>
parent b91bb296
...@@ -145,20 +145,6 @@ static void sas_ata_task_done(struct sas_task *task) ...@@ -145,20 +145,6 @@ static void sas_ata_task_done(struct sas_task *task)
ata_qc_complete(qc); ata_qc_complete(qc);
spin_unlock_irqrestore(dev->sata_dev.ap->lock, flags); spin_unlock_irqrestore(dev->sata_dev.ap->lock, flags);
/*
* If the sas_task has an ata qc, a scsi_cmnd and the aborted
* flag is set, then we must have come in via the libsas EH
* functions. When we exit this function, we need to put the
* scsi_cmnd on the list of finished errors. The ata_qc_complete
* call cleans up the libata side of things but we're protected
* from the scsi_cmnd going away because the scsi_cmnd is owned
* by the EH, making libata's call to scsi_done a NOP.
*/
spin_lock_irqsave(&task->task_state_lock, flags);
if (qc->scsicmd && task->task_state_flags & SAS_TASK_STATE_ABORTED)
scsi_eh_finish_cmd(qc->scsicmd, &sas_ha->eh_done_q);
spin_unlock_irqrestore(&task->task_state_lock, flags);
qc_already_gone: qc_already_gone:
list_del_init(&task->list); list_del_init(&task->list);
sas_free_task(task); sas_free_task(task);
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment