mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Chuck Lever III <chuck.lever@oracle.com>
To: "Marciniszyn, Mike" <mike.marciniszyn@cornelisnetworks.com>
Cc: "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	"linux-rdma@vger.kernel.org" <linux-rdma@vger.kernel.org>,
	Jason Gunthorpe <jgg@nvidia.com>,
	"Hillman, Richie" <Richie.Hillman@cornelisnetworks.com>,
	"Dalessandro, Dennis" <dennis.dalessandro@cornelisnetworks.com>,
	Linux NFS Mailing List <linux-nfs@vger.kernel.org>
Subject: Re: NFS trace new to 5.13.0 (GA)
Date: Tue, 29 Jun 2021 18:38:22 +0000	[thread overview]
Message-ID: <A9DA4F5A-9BA8-4395-82CF-24C071AF1F8C@oracle.com> (raw)
In-Reply-To: <CH0PR01MB71539295AEF1947518073D0FF2029@CH0PR01MB7153.prod.exchangelabs.com>

Hi Mike-

> On Jun 29, 2021, at 2:28 PM, Marciniszyn, Mike <mike.marciniszyn@cornelisnetworks.com> wrote:
> 
> During our continuous integration testing on 5.13.0 kernel our testing trips on NFS testing with the following trace on the client:
> 
> [32936.156848] INFO: task kworker/9:1:519 blocked for more than 122 seconds.
> [32936.165201]       Tainted: G S                5.13.0 #1
> [32936.171562] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
> [32936.180773] task:kworker/9:1     state:D stack:    0 pid:  519 ppid:     2 flags:0x00004000
> [32936.190565] Workqueue: events xprt_destroy_cb [sunrpc]
> [32936.196854] Call Trace:
> [32936.200107]  __schedule+0x38e/0x8b0
> [32936.204482]  schedule+0x3c/0xa0
> [32936.208464]  schedule_timeout+0x215/0x2b0
> [32936.213401]  ? check_preempt_curr+0x3f/0x70
> [32936.218518]  ? ttwu_do_wakeup+0x17/0x140
> [32936.223336]  wait_for_completion+0x98/0xf0
> [32936.228396]  __flush_work+0x128/0x1e0
> [32936.232942]  ? worker_attach_to_pool+0xb0/0xb0
> [32936.238351]  ? work_busy+0x80/0x80
> [32936.242555]  __cancel_work_timer+0x110/0x1a0
> [32936.247726]  ? xprt_rdma_bc_destroy+0xc6/0xe0 [rpcrdma]
> [32936.254034]  xprt_rdma_destroy+0x15/0x50 [rpcrdma]
> [32936.259873]  process_one_work+0x1cb/0x360
> [32936.264788]  ? process_one_work+0x360/0x360
> [32936.269915]  worker_thread+0x30/0x370
> [32936.274436]  ? process_one_work+0x360/0x360
> [32936.279526]  kthread+0x116/0x130
> [32936.283534]  ? __kthread_cancel_work+0x40/0x40
> [32936.288924]  ret_from_fork+0x22/0x30
> 
> The same tests and same servers see no such issue from rc4 to rc7, so the failure seems new.
> 
> Any thoughts?
> 
> I'm currently rerunning rc7 just to be sure.

The NFS server in v5.13 is afflicted by a late-breaking bug-fix
to the alloc_pages_bulk_array() API. It's been fixed in Linus'
tree, but that tree is otherwise unstable for me.

Have a look at commit 66d9282523b for a one-liner fix, it should
apply cleanly to v5.13.

--
Chuck Lever




  reply	other threads:[~2021-06-29 18:38 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-06-29 18:28 Marciniszyn, Mike
2021-06-29 18:38 ` Chuck Lever III [this message]
2021-06-29 19:34   ` Marciniszyn, Mike
2021-06-30 13:23     ` Marciniszyn, Mike

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=A9DA4F5A-9BA8-4395-82CF-24C071AF1F8C@oracle.com \
    --to=chuck.lever@oracle.com \
    --cc=Richie.Hillman@cornelisnetworks.com \
    --cc=dennis.dalessandro@cornelisnetworks.com \
    --cc=jgg@nvidia.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-nfs@vger.kernel.org \
    --cc=linux-rdma@vger.kernel.org \
    --cc=mike.marciniszyn@cornelisnetworks.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®