public inbox for linux-block@vger.kernel.org
 help / color / mirror / Atom feed
From: Jens Axboe <axboe@kernel.dk>
To: Christoph Hellwig <hch@lst.de>
Cc: Ming Lei <ming.lei@redhat.com>, linux-block@vger.kernel.org
Subject: Re: [PATCH 2/2] blk-mq: ensure a q_usage_counter reference is held when splitting bios
Date: Thu, 11 Jan 2024 09:17:41 -0700	[thread overview]
Message-ID: <d91cb6ee-50bd-4397-b82f-d34dfc135b4a@kernel.dk> (raw)
In-Reply-To: <20240111161440.GA16626@lst.de>

On 1/11/24 9:14 AM, Christoph Hellwig wrote:
> On Thu, Jan 11, 2024 at 09:12:23AM -0700, Jens Axboe wrote:
>> On 1/11/24 6:57 AM, Christoph Hellwig wrote:
>>> q_usage_counter is the only thing preventing us from the limits changing
>>> under us in __bio_split_to_limits, but blk_mq_submit_bio doesn't hold it.
>>>
>>> Change __submit_bio to always acquire the q_usage_counter counter before
>>> branching out into bio vs request based helper, and let blk_mq_submit_bio
>>> tell it if it consumed the reference by handing it off to the request.
>>
>> This causes hangs for me on shutdown/reset:
> 
>> which seems to indicate that a reference is being leaked. Haven't poked
>> any further at it, I'll drop these two for now.
> 
> Can you send me your .config?

Don't think it's .config related, hit it on my test box and then in my
vm as well. It's likely due to batched allocations, would be my guess.
I can start/halt the vm fine with just a boot, but if I do:

$ ~/git/fio/t/io_uring -p1 -d128 -b512 -s32 -c32 -F1 -B1 -R0 -X1 -P1 /dev/nvme0n1

for just a brief moment, nvme0 will hang on shutdown.

-- 
Jens Axboe



  reply	other threads:[~2024-01-11 16:17 UTC|newest]

Thread overview: 33+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-01-11 13:57 ensure q_usage_counter is held over bio splits Christoph Hellwig
2024-01-11 13:57 ` [PATCH 1/2] blk-mq: rename blk_mq_can_use_cached_rq Christoph Hellwig
2024-01-11 13:57 ` [PATCH 2/2] blk-mq: ensure a q_usage_counter reference is held when splitting bios Christoph Hellwig
2024-01-11 16:12   ` Jens Axboe
2024-01-11 16:14     ` Christoph Hellwig
2024-01-11 16:17       ` Jens Axboe [this message]
2024-01-11 16:18         ` Christoph Hellwig
2024-01-11 17:10         ` Christoph Hellwig
2024-01-11 17:18           ` Jens Axboe
2024-01-11 17:24             ` Christoph Hellwig
2024-01-11 20:06               ` Jens Axboe
2024-01-12  5:44                 ` Christoph Hellwig
2024-01-12 14:22                   ` Jens Axboe
2024-01-12 14:25                     ` Christoph Hellwig
2024-01-12 16:10                       ` Jens Axboe
2024-01-15 11:20                     ` Ulf Hansson
2024-01-22  7:34                       ` mmc vs highmem, was: " Christoph Hellwig
2024-01-22  9:26                         ` Arnd Bergmann
2024-01-22 13:39                           ` Christoph Hellwig
2024-01-22 14:57                             ` Arnd Bergmann
2024-01-23  9:11                               ` Christoph Hellwig
2024-01-24 11:59                                 ` Arnd Bergmann
2024-01-24 12:33                                   ` Linus Walleij
2024-01-24 12:54                                     ` Arnd Bergmann
2024-01-24 13:16                                       ` Linus Walleij
2024-01-24 14:14                                         ` Arnd Bergmann
2024-01-24 12:45                               ` Arnd Bergmann
2024-01-24 13:49                           ` Linus Walleij
2024-01-24 16:35                             ` Arnd Bergmann
2024-01-11 22:22   ` Ming Lei
2024-01-12  5:46     ` Christoph Hellwig
2024-01-11 16:03 ` ensure q_usage_counter is held over bio splits Jens Axboe
2024-01-14 14:38 ` (subset) " Jens Axboe

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=d91cb6ee-50bd-4397-b82f-d34dfc135b4a@kernel.dk \
    --to=axboe@kernel.dk \
    --cc=hch@lst.de \
    --cc=linux-block@vger.kernel.org \
    --cc=ming.lei@redhat.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