LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Jens Axboe <axboe@kernel.dk>
To: Bart Van Assche <bvanassche@acm.org>
Cc: Christoph Hellwig <hch@lst.de>, Robert Elliott <Elliott@hp.com>,
	linux-kernel <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH] blk-mq: Avoid that I/O hangs in bt_get()
Date: Mon, 08 Dec 2014 09:49:52 -0700	[thread overview]
Message-ID: <5485D6B0.3060901@kernel.dk> (raw)
In-Reply-To: <5485BBD2.4040103@acm.org>

On 12/08/2014 07:55 AM, Bart Van Assche wrote:
> On 11/06/14 14:41, Bart Van Assche wrote:
>> With kernel 3.18-rc3 and with can_queue=62 I can trigger a hang in
>> bt_get() easily.
>
> (once more replying to my own e-mail)
>
> Hello Jens,
>
> Finally I found the time to look further into this. The patch below
> seems to be sufficient to prevent this hang. However, I'm not a block
> layer expert so it's not clear to me whether the patch below makes sense ?
>
> Thanks,
>
> Bart.
>
> [PATCH] blk-mq: Fix bt_get() hang
>
> Avoid that if there are fewer hardware queues than CPU threads that
> bt_get() can hang. The symptoms of the hang were as follows:
> * All tags allocated for a particular hardware queue.
> * (nr_tags) pending commands for that hardware queue.
> * No pending commands for the software queues associated with that
>    hardware queue.
> ---
>   block/blk-mq-tag.c | 2 ++
>   1 file changed, 2 insertions(+)
>
> diff --git a/block/blk-mq-tag.c b/block/blk-mq-tag.c
> index 67ab88b..e88af88 100644
> --- a/block/blk-mq-tag.c
> +++ b/block/blk-mq-tag.c
> @@ -256,6 +256,8 @@ static int bt_get(struct blk_mq_alloc_data *data,
>               break;
>           }
>
> +        blk_mq_run_hw_queue(hctx, false);
> +
>           blk_mq_put_ctx(data->ctx);
>
>           io_schedule();

Ah yes, that could be an issue for some cases, we do need to run the 
queue there. For a tag map shared across hardware queues, we might need 
to run more than just the current queue, however. For now we can safely 
assume that we allocate fairly, so it should not be an issue.

It might be worth experimenting with doing a __bt_get() after the queue 
run before going to sleep, in case the queue run found completions as well.

-- 
Jens Axboe


  reply	other threads:[~2014-12-08 16:49 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-10-06 12:27 Bart Van Assche
2014-10-06 17:40 ` Jens Axboe
2014-10-06 18:53   ` Jens Axboe
2014-10-07  8:46     ` Bart Van Assche
2014-10-07 14:44       ` Jens Axboe
2014-11-06 13:41       ` Bart Van Assche
2014-12-08 14:55         ` Bart Van Assche
2014-12-08 16:49           ` Jens Axboe [this message]
2014-12-08 17:59             ` Bart Van Assche

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=5485D6B0.3060901@kernel.dk \
    --to=axboe@kernel.dk \
    --cc=Elliott@hp.com \
    --cc=bvanassche@acm.org \
    --cc=hch@lst.de \
    --cc=linux-kernel@vger.kernel.org \
    --subject='Re: [PATCH] blk-mq: Avoid that I/O hangs in bt_get()' \
    /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

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).