All of lore.kernel.org
 help / color / mirror / Atom feed
* Re: [SPDK] Out-of-order completion of nvme command?
@ 2017-11-17  3:27 Chang, Cunyin
  0 siblings, 0 replies; 3+ messages in thread
From: Chang, Cunyin @ 2017-11-17  3:27 UTC (permalink / raw)
  To: spdk

[-- Attachment #1: Type: text/plain, Size: 1160 bytes --]

Hi Wang,

We use spdk_nvme_qpair_process_completions() to poll the completion commands, and we can "not"  assume that the commands finished in the
same order as they submitted. Which means if you submit commands A B C, they could be finished in any order.

-Cunyin

> -----Original Message-----
> From: SPDK [mailto:spdk-bounces(a)lists.01.org] On Behalf Of Li Wang
> Sent: Friday, November 17, 2017 11:04 AM
> To: Storage Performance Development Kit <spdk(a)lists.01.org>
> Subject: [SPDK] Out-of-order completion of nvme command?
> 
> Hi guys,
> 
> If I submit a set of read/write commands to a single queue pair using spdk
> nvme interface and call the process_completion() function afterwards, will
> the callback function be triggered in the same order as I submitted the
> commands? In other words, if I submit 3 IO commands, say C1, C2 and C3,
> with callback function F1, F2 and F3, respectively, will the sequence of
> callback function always be F1, F2 and F3?
> 
> Regards,
> Li Wang
> _______________________________________________
> SPDK mailing list
> SPDK(a)lists.01.org
> https://lists.01.org/mailman/listinfo/spdk

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

* Re: [SPDK] Out-of-order completion of nvme command?
@ 2017-11-17  4:50 Li Wang
  0 siblings, 0 replies; 3+ messages in thread
From: Li Wang @ 2017-11-17  4:50 UTC (permalink / raw)
  To: spdk

[-- Attachment #1: Type: text/plain, Size: 1508 bytes --]

Hi Cunyin,

Thanks for your reply. It is helpful.

Li

On 17 November 2017 at 11:27, Chang, Cunyin <cunyin.chang(a)intel.com> wrote:

> Hi Wang,
>
> We use spdk_nvme_qpair_process_completions() to poll the completion
> commands, and we can "not"  assume that the commands finished in the
> same order as they submitted. Which means if you submit commands A B C,
> they could be finished in any order.
>
> -Cunyin
>
> > -----Original Message-----
> > From: SPDK [mailto:spdk-bounces(a)lists.01.org] On Behalf Of Li Wang
> > Sent: Friday, November 17, 2017 11:04 AM
> > To: Storage Performance Development Kit <spdk(a)lists.01.org>
> > Subject: [SPDK] Out-of-order completion of nvme command?
> >
> > Hi guys,
> >
> > If I submit a set of read/write commands to a single queue pair using
> spdk
> > nvme interface and call the process_completion() function afterwards,
> will
> > the callback function be triggered in the same order as I submitted the
> > commands? In other words, if I submit 3 IO commands, say C1, C2 and C3,
> > with callback function F1, F2 and F3, respectively, will the sequence of
> > callback function always be F1, F2 and F3?
> >
> > Regards,
> > Li Wang
> > _______________________________________________
> > SPDK mailing list
> > SPDK(a)lists.01.org
> > https://lists.01.org/mailman/listinfo/spdk
> _______________________________________________
> SPDK mailing list
> SPDK(a)lists.01.org
> https://lists.01.org/mailman/listinfo/spdk
>

[-- Attachment #2: attachment.html --]
[-- Type: text/html, Size: 2341 bytes --]

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

* [SPDK] Out-of-order completion of nvme command?
@ 2017-11-17  3:04 Li Wang
  0 siblings, 0 replies; 3+ messages in thread
From: Li Wang @ 2017-11-17  3:04 UTC (permalink / raw)
  To: spdk

[-- Attachment #1: Type: text/plain, Size: 442 bytes --]

Hi guys,

If I submit a set of read/write commands to a single queue pair using spdk nvme interface and call the process_completion() function afterwards, will the callback function be triggered in the same order as I submitted the commands? In other words, if I submit 3 IO commands, say C1, C2 and C3, with callback function F1, F2 and F3, respectively, will the sequence of callback function always be F1, F2 and F3?

Regards,
Li Wang

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

end of thread, other threads:[~2017-11-17  4:50 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2017-11-17  3:27 [SPDK] Out-of-order completion of nvme command? Chang, Cunyin
  -- strict thread matches above, loose matches on Subject: below --
2017-11-17  4:50 Li Wang
2017-11-17  3:04 Li Wang

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.