Bart Van Assche <Bart.VanAssche@xxxxxxxxxxx> writes:
> On Thu, 2017-04-06 at 15:36 +0200, Hannes Reinecke wrote:
>> The block layer always calls the timeout function from a workqueue
>> context, so there is no need to have yet another workqueue for
>> running command aborts.
>>
>> [ ... ]
>> @@ -271,10 +266,14 @@ enum blk_eh_timer_return scsi_times_out(struct request *req)
>> rtn = host->hostt->eh_timed_out(scmd);
>>
>> if (rtn == BLK_EH_NOT_HANDLED) {
>> - if (scsi_abort_command(scmd) != SUCCESS) {
>> + int ret;
>> +
>> + ret = scsi_abort_command(scmd);
>> + if (ret == FAILED) {
>> set_host_byte(scmd, DID_TIME_OUT);
>> scsi_eh_scmd_add(scmd);
>> - }
>> + } else if (ret == FAST_IO_FAIL)
>> + rtn = BLK_EH_RESET_TIMER;
>> }
>
> Has this patch been tested with the traditional block layer? For the
> traditional block layer scsi_times_out() is called with the queue lock
> held. Does this patch cause .eh_abort_handler(), a function that may
> sleep, to be called with the queue lock held?
Hannes: Ping!
--
Martin K. Petersen Oracle Linux Engineering
