io_u requeue and resubmit

All of lore.kernel.org
 help / color / mirror / Atom feed

* io_u requeue and resubmit
@ 2013-09-09 18:32 Suresh P
  2013-09-09 21:18 ` Jens Axboe
  0 siblings, 1 reply; 3+ messages in thread
From: Suresh P @ 2013-09-09 18:32 UTC (permalink / raw)
  To: fio@vger.kernel.org

I have a question on requeue_io_u. Code source lines reference here are from version fio-2.1.2.

Does a requeued io_u work get submitted again through __get_io_u() ? If so, the io_u item is taken out of requeue list & then the io_u->file is set to NULL. We don't make a distinction about item being already setup Vs a new io_u in __get_io_u().ï¿½In get_io_u(), when an io_u may come out of requeue list, we depend on io_u->file to be non NULL to detect the work as old pending one.

Am I missing something here or is this a bug?

1274 ï¿½ ï¿½ ï¿½ ï¿½ io_u = __get_io_u(td);
1275 ï¿½ ï¿½ ï¿½ ï¿½ if (!io_u) {
1276 ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ dprint(FD_IO, "__get_io_u failed\n");
1277 ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ return NULL;
1278 ï¿½ ï¿½ ï¿½ ï¿½ }
1279ï¿½
1280 ï¿½ ï¿½ ï¿½ ï¿½ if (check_get_verify(td, io_u))
1281 ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ goto out;
1282 ï¿½ ï¿½ ï¿½ ï¿½ if (check_get_trim(td, io_u))
1283 ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ goto out;
1284ï¿½
1285 ï¿½ ï¿½ ï¿½ ï¿½ /*
1286 ï¿½ ï¿½ ï¿½ ï¿½ ï¿½* from a requeue, io_u already setup
1287 ï¿½ ï¿½ ï¿½ ï¿½ ï¿½*/
1288 ï¿½ ï¿½ ï¿½ ï¿½ if (io_u->file)
1289 ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ goto out;
1290ï¿½
"io_u.c" line 1286 of 1667 --77%-- col 3-10


1130 struct io_u *__get_io_u(struct thread_data *td)
1131 {
1132 ï¿½ ï¿½ ï¿½ ï¿½ struct io_u *io_u;
1133ï¿½
1134 ï¿½ ï¿½ ï¿½ ï¿½ td_io_u_lock(td);
1135ï¿½
1136 again:
1137 ï¿½ ï¿½ ï¿½ ï¿½ if (!io_u_rempty(&td->io_u_requeues))
1138 ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ io_u = io_u_rpop(&td->io_u_requeues);
1139 ï¿½ ï¿½ ï¿½ ï¿½ else if (!io_u_qempty(&td->io_u_freelist))
1140 ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ io_u = io_u_qpop(&td->io_u_freelist);
1141ï¿½
1142 ï¿½ ï¿½ ï¿½ ï¿½ if (io_u) {
1143 ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ io_u->buflen = 0;
1144 ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ io_u->resid = 0;
1145 ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ io_u->file = NULL;
1146 ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ io_u->end_io = NULL;
1147 ï¿½ ï¿½ ï¿½ ï¿½ }
"io_u.c" line 1130 of 1667 --67%-- col 1

Thanks,
Suresh 		 	   		  

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: io_u requeue and resubmit
  2013-09-09 18:32 io_u requeue and resubmit Suresh P
@ 2013-09-09 21:18 ` Jens Axboe
  2013-09-09 21:51   ` Suresh P
  0 siblings, 1 reply; 3+ messages in thread
From: Jens Axboe @ 2013-09-09 21:18 UTC (permalink / raw)
  To: Suresh P; +Cc: fio@vger.kernel.org

On 09/09/2013 12:32 PM, Suresh P wrote:
> I have a question on requeue_io_u. Code source lines reference here are from version fio-2.1.2.
> 
> Does a requeued io_u work get submitted again through __get_io_u() ? If so, the io_u item is taken out of requeue list & then the io_u->file is set to NULL. We don't make a distinction about item being already setup Vs a new io_u in __get_io_u(). In get_io_u(), when an io_u may come out of requeue list, we depend on io_u->file to be non NULL to detect the work as old pending one.
> 
> Am I missing something here or is this a bug?
> 
> 1274         io_u = __get_io_u(td);
> 1275         if (!io_u) {
> 1276                 dprint(FD_IO, "__get_io_u failed\n");
> 1277                 return NULL;
> 1278         }
> 1279 
> 1280         if (check_get_verify(td, io_u))
> 1281                 goto out;
> 1282         if (check_get_trim(td, io_u))
> 1283                 goto out;
> 1284 
> 1285         /*
> 1286          * from a requeue, io_u already setup
> 1287          */
> 1288         if (io_u->file)
> 1289                 goto out;
> 1290 
> "io_u.c" line 1286 of 1667 --77%-- col 3-10
> 
> 
> 1130 struct io_u *__get_io_u(struct thread_data *td)
> 1131 {
> 1132         struct io_u *io_u;
> 1133 
> 1134         td_io_u_lock(td);
> 1135 
> 1136 again:
> 1137         if (!io_u_rempty(&td->io_u_requeues))
> 1138                 io_u = io_u_rpop(&td->io_u_requeues);
> 1139         else if (!io_u_qempty(&td->io_u_freelist))
> 1140                 io_u = io_u_qpop(&td->io_u_freelist);
> 1141 
> 1142         if (io_u) {
> 1143                 io_u->buflen = 0;
> 1144                 io_u->resid = 0;
> 1145                 io_u->file = NULL;
> 1146                 io_u->end_io = NULL;
> 1147         }
> "io_u.c" line 1130 of 1667 --67%-- col 1

Good catch! It was actually broken a few revisions back with the ring
lists. Fixed:

http://git.kernel.dk/?p=fio.git;a=commit;h=f713350f74dbeb333ff1ad2790cb60381ceaf463

-- 
Jens Axboe


^ permalink raw reply	[flat|nested] 3+ messages in thread

* RE: io_u requeue and resubmit
  2013-09-09 21:18 ` Jens Axboe
@ 2013-09-09 21:51   ` Suresh P
  0 siblings, 0 replies; 3+ messages in thread
From: Suresh P @ 2013-09-09 21:51 UTC (permalink / raw)
  To: Jens Axboe; +Cc: fio@vger.kernel.org

Thanks Jens..that was quick!

----------------------------------------
> Date: Mon, 9 Sep 2013 15:18:37 -0600
> From: axboe@kernel.dk
> To: suresh_p_@hotmail.com
> CC: fio@vger.kernel.org
> Subject: Re: io_u requeue and resubmit
>
> On 09/09/2013 12:32 PM, Suresh P wrote:
>> I have a question on requeue_io_u. Code source lines reference here are from version fio-2.1.2.
>>
>> Does a requeued io_u work get submitted again through __get_io_u() ? If so, the io_u item is taken out of requeue list & then the io_u->file is set to NULL. We don't make a distinction about item being already setup Vs a new io_u in __get_io_u(). In get_io_u(), when an io_u may come out of requeue list, we depend on io_u->file to be non NULL to detect the work as old pending one.
>>
>> Am I missing something here or is this a bug?
>>
>> 1274 io_u = __get_io_u(td);
>> 1275 if (!io_u) {
>> 1276 dprint(FD_IO, "__get_io_u failed\n");
>> 1277 return NULL;
>> 1278 }
>> 1279
>> 1280 if (check_get_verify(td, io_u))
>> 1281 goto out;
>> 1282 if (check_get_trim(td, io_u))
>> 1283 goto out;
>> 1284
>> 1285 /*
>> 1286 * from a requeue, io_u already setup
>> 1287 */
>> 1288 if (io_u->file)
>> 1289 goto out;
>> 1290
>> "io_u.c" line 1286 of 1667 --77%-- col 3-10
>>
>>
>> 1130 struct io_u *__get_io_u(struct thread_data *td)
>> 1131 {
>> 1132 struct io_u *io_u;
>> 1133
>> 1134 td_io_u_lock(td);
>> 1135
>> 1136 again:
>> 1137 if (!io_u_rempty(&td->io_u_requeues))
>> 1138 io_u = io_u_rpop(&td->io_u_requeues);
>> 1139 else if (!io_u_qempty(&td->io_u_freelist))
>> 1140 io_u = io_u_qpop(&td->io_u_freelist);
>> 1141
>> 1142 if (io_u) {
>> 1143 io_u->buflen = 0;
>> 1144 io_u->resid = 0;
>> 1145 io_u->file = NULL;
>> 1146 io_u->end_io = NULL;
>> 1147 }
>> "io_u.c" line 1130 of 1667 --67%-- col 1
>
> Good catch! It was actually broken a few revisions back with the ring
> lists. Fixed:
>
> http://git.kernel.dk/?p=fio.git;a=commit;h=f713350f74dbeb333ff1ad2790cb60381ceaf463
>
> --
> Jens Axboe
>
> --
> To unsubscribe from this list: send the line "unsubscribe fio" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html 		 	   		  

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2013-09-09 21:51 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2013-09-09 18:32 io_u requeue and resubmit Suresh P
2013-09-09 21:18 ` Jens Axboe
2013-09-09 21:51   ` Suresh P

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.