From: Gabriel Krisman Bertazi <krisman@suse.de>
To: Jan Kara <jack@suse.cz>
Cc: axboe@kernel.dk, linux-kernel@vger.kernel.org,
linux-block@vger.kernel.org, Hugh Dickins <hughd@google.com>,
Keith Busch <kbusch@kernel.org>,
Liu Song <liusong@linux.alibaba.com>
Subject: Re: [PATCH] sbitmap: Use single per-bitmap counting to wake up queued tags
Date: Mon, 14 Nov 2022 22:52:32 -0500 [thread overview]
Message-ID: <87cz9odgcf.fsf@suse.de> (raw)
In-Reply-To: <20221114132313.5cqhvzxarm7rwvmt@quack3> (Jan Kara's message of "Mon, 14 Nov 2022 14:23:13 +0100")
Jan Kara <jack@suse.cz> writes:
> Now this may be also problematic - when we were checking the number of woken
> waiters in the older version of the patch (for others: internal version of
> the patch) this was fine but now it may happen that the 'ws' we have
> selected has no waiters anymore. And in that case we need to find another
> waitqueue because otherwise we'd be loosing too many wakeups and we could
> deadlock. So I think this rather needs to be something like:
>
> do {
> if (atomic_read(&sbq->completion_cnt) - wakeups < wake_batch)
> return;
> } while (!atomic_try_cmpxchg(&sbq->wakeup_cnt,
> &wakeups, wakeups + wake_batch));
>
> do {
> ws = sbq_wake_ptr(sbq);
> if (!ws)
> return;
Jan,
Does this really solve it? There is no guarantee there will be another
waiter in the queues when we check here. So, once again we could not
wake up anyone and return it this if leg. If that is the case, don't we
end up overshooting wakeups and end up again with less completions than
required to wake up an incoming io?
--
Gabriel Krisman Bertazi
next prev parent reply other threads:[~2022-11-15 3:52 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-11-05 23:10 [PATCH] sbitmap: Use single per-bitmap counting to wake up queued tags Gabriel Krisman Bertazi
2022-11-08 23:28 ` Chaitanya Kulkarni
2022-11-09 3:03 ` Gabriel Krisman Bertazi
2022-11-09 3:35 ` Chaitanya Kulkarni
2022-11-09 22:06 ` Jens Axboe
2022-11-09 22:48 ` Gabriel Krisman Bertazi
2022-11-10 3:25 ` Jens Axboe
2022-11-10 9:42 ` Yu Kuai
2022-11-10 11:16 ` Jan Kara
2022-11-10 13:18 ` Yu Kuai
2022-11-10 15:35 ` Jan Kara
2022-11-11 0:59 ` Yu Kuai
2022-11-11 15:38 ` Jens Axboe
2022-11-14 13:23 ` Jan Kara
2022-11-14 14:20 ` [PATCH] sbitmap: Advance the queue index before waking up the queue Gabriel Krisman Bertazi
2022-11-14 14:34 ` Jan Kara
2022-11-15 3:52 ` Gabriel Krisman Bertazi [this message]
2022-11-15 10:24 ` [PATCH] sbitmap: Use single per-bitmap counting to wake up queued tags Jan Kara
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=87cz9odgcf.fsf@suse.de \
--to=krisman@suse.de \
--cc=axboe@kernel.dk \
--cc=hughd@google.com \
--cc=jack@suse.cz \
--cc=kbusch@kernel.org \
--cc=linux-block@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=liusong@linux.alibaba.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;
as well as URLs for NNTP newsgroup(s).