RE: [PATCH v1] scsi: core: pair EH runtime PM get and put

[Fang Hongjie(方洪杰)]

> > @@ -2383,12 +2384,16 @@ int scsi_error_handler(void *data)
> >   		 * what we need to do to get it up and online again (if we
> can).
> >   		 * If we fail, we end up taking the thing offline.
> >   		 */
> > -		if (!shost->eh_noresume &&
> scsi_autopm_get_host(shost) != 0) {
> > -			SCSI_LOG_ERROR_RECOVERY(1,
> > -				shost_printk(KERN_ERR, shost,
> > -					     "scsi_eh_%d: unable to
> autoresume\n",
> > -					     shost->host_no));
> > -			continue;
> > +		autopm_put = false;
> > +		if (!shost->eh_noresume) {
> > +			if (scsi_autopm_get_host(shost) != 0) {
> > +				SCSI_LOG_ERROR_RECOVERY(1,
> > +					shost_printk(KERN_ERR, shost,
> > +						     "scsi_eh_%d: unable to
> autoresume\n",
> > +						     shost->host_no));
> > +				continue;
> > +			}
> > +			autopm_put = true;
> >   		}
> >
> >   		if (shost->transportt->eh_strategy_handler)
> > @@ -2407,7 +2412,7 @@ int scsi_error_handler(void *data)
> >   		 * which are still online.
> >   		 */
> >   		scsi_restart_operations(shost);
> > -		if (!shost->eh_noresume)
> > +		if (autopm_put)
> >   			scsi_autopm_put_host(shost);
> >   	}
> >   	__set_current_state(TASK_RUNNING);
> 
> Please capture the value of shost->eh_noresume in a local variable such
> that concurrent changes of shost->eh_noresume can't break the above
> code.


Thanks Bart. I will update the patch to snapshot shost->eh_noresume once
per EH iteration and use that value for both the get and put decisions.


>
> Thanks,
> 
> Bart.