All of lore.kernel.org
 help / color / mirror / Atom feed
* Encountering errors while using RNBD over rxe for v5.14
@ 2022-04-20 10:28 Haris Iqbal
  2022-04-27 11:49 ` Haris Iqbal
  2022-04-29  8:26 ` Guoqing Jiang
  0 siblings, 2 replies; 6+ messages in thread
From: Haris Iqbal @ 2022-04-20 10:28 UTC (permalink / raw)
  To: RDMA mailing list, yanjun.zhu; +Cc: Jinpu Wang

Hello,

We are facing some issues with the rxe driver in v5.14 (tested with 5.14.21)

After mapping a single RNBD device with 2 rxe interfaces, and with the
below fio config,

[global]
description=Emulation of Storage Server Access Pattern
bssplit=512/20:1k/16:2k/9:4k/12:8k/19:16k/10:32k/8:64k/4
fadvise_hint=0
rw=randrw:2
direct=1
random_distribution=zipf:1.2
time_based=1
runtime=60
ramp_time=1
ioengine=libaio
iodepth=128
iodepth_batch_submit=128
iodepth_batch_complete_min=1
iodepth_batch_complete_max=128
numjobs=1
group_reporting

[job1]
filename=/dev/rnbd0


We observe the following error,

[Fri Mar 25 19:08:03 2022] rtrs_client L353: <blya>: Failed
IB_WR_LOCAL_INV: WR flushed
[Fri Mar 25 19:08:03 2022] rtrs_client L333: <blya>: Failed
IB_WR_REG_MR: WR flushed
[Fri Mar 25 19:08:03 2022] rtrs_client L333: <blya>: Failed
IB_WR_REG_MR: WR flushed
[Fri Mar 25 19:08:34 2022] rtrs_client L353: <blya>: Failed
IB_WR_LOCAL_INV: WR flushed
[Fri Mar 25 19:08:34 2022] rtrs_client L353: <blya>: Failed
IB_WR_LOCAL_INV: WR flushed
[Fri Mar 25 19:08:34 2022] rtrs_client L448: <blya>: IO request
failed: error=-103 path=ip:192.168.123.77@ip:192.168.123.206
[mlx4_1:1] notify=0
[Fri Mar 25 19:08:34 2022] rtrs_client L353: <blya>: Failed
IB_WR_LOCAL_INV: WR flushed
[Fri Mar 25 19:08:34 2022] rtrs_client L448: <blya>: IO request
failed: error=-103 path=ip:192.168.123.77@ip:192.168.123.206
[mlx4_1:1] notify=0

This is only happening for writes. If I change the above fio to rand
read only, there is no error.

I digged into the reason why this error is happening.

During the processing of an rxe packet, while doing lookup_mr, it
discovers that the state of the mr in not RXE_MR_STATE_VALID; and is
RXE_MR_STATE_FREE.

The code path of the error is
rxe_requester -> finish_packet -> copy_data -> lookup_mr

The mr for which this lookup is happening is the wqe dma.


I tried bisecting the branch to try to find the rouge commit, but it
lands up in the below commit.

$ git show e04360a2ea01bf42aa639b65aad81f502e896c7f
commit e04360a2ea01bf42aa639b65aad81f502e896c7f (refs/bisect/bad)
Merge: 514798d36572 3d8287544223
Author: Linus Torvalds <torvalds@linux-foundation.org>
Date: Thu Jul 1 14:54:03 2021 -0700

Merge tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma

So I found out all the commits that went into this merge commit. The
starting and the end commits of that merge commit are as follows,
591f762b2750
3d8287544223

So I checked out the last commit, (below), but that somehow doesn't
have the error.

commit 3d8287544223a3d2f37981c1f9ffd94d0b5e9ffc (HEAD)
Author: Leon Romanovsky <leonro@nvidia.com>
Date:   Tue Jun 29 09:49:33 2021 +0300

    RDMA/core: Always release restrack object

(At this point rxe is actually broken, so I have to backport 3 commits
for even rnbd mapping to work).

The 3 backported commits are as follows,

From db4657afd10e45855ac1d8437fcc9a86bd3d741d Mon Sep 17 00:00:00 2001
From: Mike Marciniszyn <mike.marciniszyn@cornelisnetworks.com>
Date: Thu, 29 Jul 2021 14:26:22 -0400
Subject: [PATCH 1/3] RDMA/cma: Revert INIT-INIT patch

From e2a05339fa1188b6b37540f4611893ac4c534fa2 Mon Sep 17 00:00:00 2001
From: Bob Pearson <rpearsonhpe@gmail.com>
Date: Thu, 29 Jul 2021 17:00:38 -0500
Subject: [PATCH 2/3] RDMA/rxe: Use the correct size of wqe when processing SRQ

From ef4b96a5773d7f6568363b3d0c3c3f371fb690bd Mon Sep 17 00:00:00 2001
From: Bob Pearson <rpearsonhpe@gmail.com>
Date: Thu, 29 Jul 2021 17:00:39 -0500
Subject: [PATCH 3/3] RDMA/rxe: Restore setting tot_len in the IPv4 header


Any thoughts as to why this error could be happening?

(PS, I tried the other newer branches, 5.15, 5.16 and 5.17, there are
some other issues with them).

Regards
-Haris

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: Encountering errors while using RNBD over rxe for v5.14
  2022-04-20 10:28 Encountering errors while using RNBD over rxe for v5.14 Haris Iqbal
@ 2022-04-27 11:49 ` Haris Iqbal
  2022-04-29  8:26 ` Guoqing Jiang
  1 sibling, 0 replies; 6+ messages in thread
From: Haris Iqbal @ 2022-04-27 11:49 UTC (permalink / raw)
  To: RDMA mailing list, yanjun.zhu; +Cc: Jinpu Wang

On Wed, Apr 20, 2022 at 12:28 PM Haris Iqbal <haris.iqbal@ionos.com> wrote:
>
> Hello,
>
> We are facing some issues with the rxe driver in v5.14 (tested with 5.14.21)
>
> After mapping a single RNBD device with 2 rxe interfaces, and with the
> below fio config,
>
> [global]
> description=Emulation of Storage Server Access Pattern
> bssplit=512/20:1k/16:2k/9:4k/12:8k/19:16k/10:32k/8:64k/4
> fadvise_hint=0
> rw=randrw:2
> direct=1
> random_distribution=zipf:1.2
> time_based=1
> runtime=60
> ramp_time=1
> ioengine=libaio
> iodepth=128
> iodepth_batch_submit=128
> iodepth_batch_complete_min=1
> iodepth_batch_complete_max=128
> numjobs=1
> group_reporting
>
> [job1]
> filename=/dev/rnbd0
>
>
> We observe the following error,
>
> [Fri Mar 25 19:08:03 2022] rtrs_client L353: <blya>: Failed
> IB_WR_LOCAL_INV: WR flushed
> [Fri Mar 25 19:08:03 2022] rtrs_client L333: <blya>: Failed
> IB_WR_REG_MR: WR flushed
> [Fri Mar 25 19:08:03 2022] rtrs_client L333: <blya>: Failed
> IB_WR_REG_MR: WR flushed
> [Fri Mar 25 19:08:34 2022] rtrs_client L353: <blya>: Failed
> IB_WR_LOCAL_INV: WR flushed
> [Fri Mar 25 19:08:34 2022] rtrs_client L353: <blya>: Failed
> IB_WR_LOCAL_INV: WR flushed
> [Fri Mar 25 19:08:34 2022] rtrs_client L448: <blya>: IO request
> failed: error=-103 path=ip:192.168.123.77@ip:192.168.123.206
> [mlx4_1:1] notify=0
> [Fri Mar 25 19:08:34 2022] rtrs_client L353: <blya>: Failed
> IB_WR_LOCAL_INV: WR flushed
> [Fri Mar 25 19:08:34 2022] rtrs_client L448: <blya>: IO request
> failed: error=-103 path=ip:192.168.123.77@ip:192.168.123.206
> [mlx4_1:1] notify=0
>
> This is only happening for writes. If I change the above fio to rand
> read only, there is no error.
>
> I digged into the reason why this error is happening.
>
> During the processing of an rxe packet, while doing lookup_mr, it
> discovers that the state of the mr in not RXE_MR_STATE_VALID; and is
> RXE_MR_STATE_FREE.
>
> The code path of the error is
> rxe_requester -> finish_packet -> copy_data -> lookup_mr
>
> The mr for which this lookup is happening is the wqe dma.
>
>
> I tried bisecting the branch to try to find the rouge commit, but it
> lands up in the below commit.
>
> $ git show e04360a2ea01bf42aa639b65aad81f502e896c7f
> commit e04360a2ea01bf42aa639b65aad81f502e896c7f (refs/bisect/bad)
> Merge: 514798d36572 3d8287544223
> Author: Linus Torvalds <torvalds@linux-foundation.org>
> Date: Thu Jul 1 14:54:03 2021 -0700
>
> Merge tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma
>
> So I found out all the commits that went into this merge commit. The
> starting and the end commits of that merge commit are as follows,
> 591f762b2750
> 3d8287544223
>
> So I checked out the last commit, (below), but that somehow doesn't
> have the error.
>
> commit 3d8287544223a3d2f37981c1f9ffd94d0b5e9ffc (HEAD)
> Author: Leon Romanovsky <leonro@nvidia.com>
> Date:   Tue Jun 29 09:49:33 2021 +0300
>
>     RDMA/core: Always release restrack object
>
> (At this point rxe is actually broken, so I have to backport 3 commits
> for even rnbd mapping to work).
>
> The 3 backported commits are as follows,
>
> From db4657afd10e45855ac1d8437fcc9a86bd3d741d Mon Sep 17 00:00:00 2001
> From: Mike Marciniszyn <mike.marciniszyn@cornelisnetworks.com>
> Date: Thu, 29 Jul 2021 14:26:22 -0400
> Subject: [PATCH 1/3] RDMA/cma: Revert INIT-INIT patch
>
> From e2a05339fa1188b6b37540f4611893ac4c534fa2 Mon Sep 17 00:00:00 2001
> From: Bob Pearson <rpearsonhpe@gmail.com>
> Date: Thu, 29 Jul 2021 17:00:38 -0500
> Subject: [PATCH 2/3] RDMA/rxe: Use the correct size of wqe when processing SRQ
>
> From ef4b96a5773d7f6568363b3d0c3c3f371fb690bd Mon Sep 17 00:00:00 2001
> From: Bob Pearson <rpearsonhpe@gmail.com>
> Date: Thu, 29 Jul 2021 17:00:39 -0500
> Subject: [PATCH 3/3] RDMA/rxe: Restore setting tot_len in the IPv4 header
>
>
> Any thoughts as to why this error could be happening?

Ping.

>
> (PS, I tried the other newer branches, 5.15, 5.16 and 5.17, there are
> some other issues with them).
>
> Regards
> -Haris

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: Encountering errors while using RNBD over rxe for v5.14
  2022-04-20 10:28 Encountering errors while using RNBD over rxe for v5.14 Haris Iqbal
  2022-04-27 11:49 ` Haris Iqbal
@ 2022-04-29  8:26 ` Guoqing Jiang
  2022-05-04 14:22   ` Haris Iqbal
  1 sibling, 1 reply; 6+ messages in thread
From: Guoqing Jiang @ 2022-04-29  8:26 UTC (permalink / raw)
  To: Haris Iqbal, RDMA mailing list, yanjun.zhu; +Cc: Jinpu Wang

Hi Haris,

On 4/20/22 6:28 PM, Haris Iqbal wrote:
> Hello,
>
> We are facing some issues with the rxe driver in v5.14 (tested with 5.14.21)
>
> After mapping a single RNBD device with 2 rxe interfaces, and with the
> below fio config,
>
> [global]
> description=Emulation of Storage Server Access Pattern
> bssplit=512/20:1k/16:2k/9:4k/12:8k/19:16k/10:32k/8:64k/4
> fadvise_hint=0
> rw=randrw:2
> direct=1
> random_distribution=zipf:1.2
> time_based=1
> runtime=60
> ramp_time=1
> ioengine=libaio
> iodepth=128
> iodepth_batch_submit=128
> iodepth_batch_complete_min=1
> iodepth_batch_complete_max=128
> numjobs=1
> group_reporting
>
> [job1]
> filename=/dev/rnbd0
>
>
> We observe the following error,
>
> [Fri Mar 25 19:08:03 2022] rtrs_client L353: <blya>: Failed
> IB_WR_LOCAL_INV: WR flushed
> [Fri Mar 25 19:08:03 2022] rtrs_client L333: <blya>: Failed
> IB_WR_REG_MR: WR flushed
> [Fri Mar 25 19:08:03 2022] rtrs_client L333: <blya>: Failed
> IB_WR_REG_MR: WR flushed
> [Fri Mar 25 19:08:34 2022] rtrs_client L353: <blya>: Failed
> IB_WR_LOCAL_INV: WR flushed
> [Fri Mar 25 19:08:34 2022] rtrs_client L353: <blya>:*Failed IB_WR_LOCAL_INV: WR flushed*

I got similar message but I am not certain it is the same one, pls see 
the previous report,

https://lore.kernel.org/linux-rdma/20220210073655.42281-1-guoqing.jiang@linux.dev/

Thanks,
Guoqing

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: Encountering errors while using RNBD over rxe for v5.14
  2022-04-29  8:26 ` Guoqing Jiang
@ 2022-05-04 14:22   ` Haris Iqbal
  2022-05-05  3:46     ` Guoqing Jiang
  0 siblings, 1 reply; 6+ messages in thread
From: Haris Iqbal @ 2022-05-04 14:22 UTC (permalink / raw)
  To: Guoqing Jiang; +Cc: RDMA mailing list, yanjun.zhu, Jinpu Wang

Hi Guoqing,

On Fri, Apr 29, 2022 at 10:26 AM Guoqing Jiang <guoqing.jiang@linux.dev> wrote:
>
> Hi Haris,
>
> On 4/20/22 6:28 PM, Haris Iqbal wrote:
> > Hello,
> >
> > We are facing some issues with the rxe driver in v5.14 (tested with 5.14.21)
> >
> > After mapping a single RNBD device with 2 rxe interfaces, and with the
> > below fio config,
> >
> > [global]
> > description=Emulation of Storage Server Access Pattern
> > bssplit=512/20:1k/16:2k/9:4k/12:8k/19:16k/10:32k/8:64k/4
> > fadvise_hint=0
> > rw=randrw:2
> > direct=1
> > random_distribution=zipf:1.2
> > time_based=1
> > runtime=60
> > ramp_time=1
> > ioengine=libaio
> > iodepth=128
> > iodepth_batch_submit=128
> > iodepth_batch_complete_min=1
> > iodepth_batch_complete_max=128
> > numjobs=1
> > group_reporting
> >
> > [job1]
> > filename=/dev/rnbd0
> >
> >
> > We observe the following error,
> >
> > [Fri Mar 25 19:08:03 2022] rtrs_client L353: <blya>: Failed
> > IB_WR_LOCAL_INV: WR flushed
> > [Fri Mar 25 19:08:03 2022] rtrs_client L333: <blya>: Failed
> > IB_WR_REG_MR: WR flushed
> > [Fri Mar 25 19:08:03 2022] rtrs_client L333: <blya>: Failed
> > IB_WR_REG_MR: WR flushed
> > [Fri Mar 25 19:08:34 2022] rtrs_client L353: <blya>: Failed
> > IB_WR_LOCAL_INV: WR flushed
> > [Fri Mar 25 19:08:34 2022] rtrs_client L353: <blya>:*Failed IB_WR_LOCAL_INV: WR flushed*
>
> I got similar message but I am not certain it is the same one, pls see
> the previous report,
>
> https://lore.kernel.org/linux-rdma/20220210073655.42281-1-guoqing.jiang@linux.dev/

I went through the emails, and the error looks similar but I am not
sure if its related. The change to write_lock_bh was done after v5.14
I think, and the code where I am seeing this error still has irq
versions of those lock (e.g. write_lock_irqsave).

I don't have much knowledge of rxe code, hence I cannot really say for sure.

Thanks for the response.

>
> Thanks,
> Guoqing

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: Encountering errors while using RNBD over rxe for v5.14
  2022-05-04 14:22   ` Haris Iqbal
@ 2022-05-05  3:46     ` Guoqing Jiang
  2022-05-05 11:20       ` Haris Iqbal
  0 siblings, 1 reply; 6+ messages in thread
From: Guoqing Jiang @ 2022-05-05  3:46 UTC (permalink / raw)
  To: Haris Iqbal; +Cc: RDMA mailing list, yanjun.zhu, Jinpu Wang



On 5/4/22 10:22 PM, Haris Iqbal wrote:
>
>>> We observe the following error,
>>>
>>> [Fri Mar 25 19:08:03 2022] rtrs_client L353: <blya>: Failed
>>> IB_WR_LOCAL_INV: WR flushed
>>> [Fri Mar 25 19:08:03 2022] rtrs_client L333: <blya>: Failed
>>> IB_WR_REG_MR: WR flushed
>>> [Fri Mar 25 19:08:03 2022] rtrs_client L333: <blya>: Failed
>>> IB_WR_REG_MR: WR flushed
>>> [Fri Mar 25 19:08:34 2022] rtrs_client L353: <blya>: Failed
>>> IB_WR_LOCAL_INV: WR flushed
>>> [Fri Mar 25 19:08:34 2022] rtrs_client L353: <blya>:*Failed IB_WR_LOCAL_INV: WR flushed*
>> I got similar message but I am not certain it is the same one, pls see
>> the previous report,
>>
>> https://lore.kernel.org/linux-rdma/20220210073655.42281-1-guoqing.jiang@linux.dev/
> I went through the emails, and the error looks similar but I am not
> sure if its related. The change to write_lock_bh was done after v5.14
> I think, and the code where I am seeing this error still has irq
> versions of those lock (e.g. write_lock_irqsave).

The lock is irrelevant, pls check

https://lore.kernel.org/linux-rdma/473a53b6-9ab2-0d48-a9cf-c84b8dc4c3f3@linux.dev/

and

https://lore.kernel.org/linux-rdma/3b6ddb23-6dfa-29e2-27fd-741c1e3e576d@linux.dev/

But I am not certain it is the same issue.

Thanks,
Guoqing

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: Encountering errors while using RNBD over rxe for v5.14
  2022-05-05  3:46     ` Guoqing Jiang
@ 2022-05-05 11:20       ` Haris Iqbal
  0 siblings, 0 replies; 6+ messages in thread
From: Haris Iqbal @ 2022-05-05 11:20 UTC (permalink / raw)
  To: Guoqing Jiang; +Cc: RDMA mailing list, yanjun.zhu, Jinpu Wang

On Thu, May 5, 2022 at 5:46 AM Guoqing Jiang <guoqing.jiang@linux.dev> wrote:
>
>
>
> On 5/4/22 10:22 PM, Haris Iqbal wrote:
> >
> >>> We observe the following error,
> >>>
> >>> [Fri Mar 25 19:08:03 2022] rtrs_client L353: <blya>: Failed
> >>> IB_WR_LOCAL_INV: WR flushed
> >>> [Fri Mar 25 19:08:03 2022] rtrs_client L333: <blya>: Failed
> >>> IB_WR_REG_MR: WR flushed
> >>> [Fri Mar 25 19:08:03 2022] rtrs_client L333: <blya>: Failed
> >>> IB_WR_REG_MR: WR flushed
> >>> [Fri Mar 25 19:08:34 2022] rtrs_client L353: <blya>: Failed
> >>> IB_WR_LOCAL_INV: WR flushed
> >>> [Fri Mar 25 19:08:34 2022] rtrs_client L353: <blya>:*Failed IB_WR_LOCAL_INV: WR flushed*
> >> I got similar message but I am not certain it is the same one, pls see
> >> the previous report,
> >>
> >> https://lore.kernel.org/linux-rdma/20220210073655.42281-1-guoqing.jiang@linux.dev/
> > I went through the emails, and the error looks similar but I am not
> > sure if its related. The change to write_lock_bh was done after v5.14
> > I think, and the code where I am seeing this error still has irq
> > versions of those lock (e.g. write_lock_irqsave).
>
> The lock is irrelevant, pls check
>
> https://lore.kernel.org/linux-rdma/473a53b6-9ab2-0d48-a9cf-c84b8dc4c3f3@linux.dev/
>
> and
>
> https://lore.kernel.org/linux-rdma/3b6ddb23-6dfa-29e2-27fd-741c1e3e576d@linux.dev/

I tried with always_invalidate off, and it still fails. Also, I
thought to try your small patch, but it turns out the commit
647bf13ce944 ("RDMA/rxe:
Create duplicate mapping tables for FMRs") is not present in the
version I am running.


>
> But I am not certain it is the same issue.
>
> Thanks,
> Guoqing

^ permalink raw reply	[flat|nested] 6+ messages in thread

end of thread, other threads:[~2022-05-05 11:20 UTC | newest]

Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-04-20 10:28 Encountering errors while using RNBD over rxe for v5.14 Haris Iqbal
2022-04-27 11:49 ` Haris Iqbal
2022-04-29  8:26 ` Guoqing Jiang
2022-05-04 14:22   ` Haris Iqbal
2022-05-05  3:46     ` Guoqing Jiang
2022-05-05 11:20       ` Haris Iqbal

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.