From: "Nicolai Hähnle" <nhaehnle@gmail.com>
To: linux-kernel@vger.kernel.org
Subject: [PATCH v2 00/11] locking/ww_mutex: Keep sorted wait list to avoid stampedes
Date: Thu, 1 Dec 2016 15:06:43 +0100 [thread overview]
Message-ID: <1480601214-26583-1-git-send-email-nhaehnle@gmail.com> (raw)
Changes to patches 1 & 5 based on feedback. I've also updated the branch
at https://cgit.freedesktop.org/~nh/linux/log/?h=mutex.
There's been the question of using a balanced tree rather than a list.
Frankly, I'd say the 99% use case doesn't need it. Also, dealing with
waiters without a context is easy in the list, but becomes trickier with a
tree. I think one can do it with an additional counter on the lock itself
to establish the FIFO order of context-less waiters, but it'd make the
code harder to follow. I doubt that it's worth it.
(original cover letter below)
The basic idea is to make sure that:
1. All waiters that have a ww_ctx appear in stamp order in the wait list.
Waiters without a ww_ctx are still supported and appear in FIFO order as
before.
2. At most one of the waiters can be in a state where it has to check for
back off (i.e., ww_ctx->acquire > 0). Technically, there are short time
windows in which more than one such waiter can be on the list, but all
but the first one are running. This happens when a new waiter with
ww_ctx->acquire > 0 adds itself at the front of the list and wakes up the
previous head of the list, and of course multiple such chained cases can
be in-flight simultaneously.
Then we only ever have to wake up one task at a time. This is _not_ always
the head of the wait list, since there may be waiters without a context. But
among waiters with a context, we only ever have to wake the first one.
To achieve all this, the series adds a new field to mutex_waiter which is
only used for the w/w lock case. As a consequence, calling mutex_lock
directly on w/w locks is now definitely incorrect. That was likely the
intention previously anyway, but grepping through the source I did find one
place that had slipped through.
I've included timings taken from a contention-heavy stress test to some of
the patches. The stress test performs actual GPU operations which take a
good chunk of the wall time, but even so, the series still manages to
improve the wall time quite a bit.
Cheers,
Nicolai
next reply other threads:[~2016-12-01 14:07 UTC|newest]
Thread overview: 39+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-12-01 14:06 Nicolai Hähnle [this message]
2016-12-01 14:06 ` [PATCH v2 01/11] drm/vgem: Use ww_mutex_(un)lock even with a NULL context Nicolai Hähnle
2016-12-01 14:18 ` Chris Wilson
2016-12-01 15:14 ` Daniel Vetter
2016-12-01 16:24 ` Peter Zijlstra
2016-12-01 14:06 ` [PATCH v2 02/11] locking/ww_mutex: Re-check ww->ctx in the inner optimistic spin loop Nicolai Hähnle
2016-12-01 14:36 ` Chris Wilson
2016-12-06 15:06 ` Peter Zijlstra
2016-12-06 16:03 ` Waiman Long
2016-12-06 18:29 ` Peter Zijlstra
2016-12-06 18:46 ` Waiman Long
2016-12-01 14:06 ` [PATCH v2 03/11] locking/ww_mutex: Extract stamp comparison to __ww_mutex_stamp_after Nicolai Hähnle
2016-12-01 14:42 ` Chris Wilson
2016-12-01 14:06 ` [PATCH v2 04/11] locking/ww_mutex: Set use_ww_ctx even when locking without a context Nicolai Hähnle
2016-12-06 15:14 ` Peter Zijlstra
2016-12-06 15:25 ` Peter Zijlstra
2016-12-16 13:17 ` Nicolai Hähnle
2016-12-17 7:53 ` Maarten Lankhorst
2016-12-17 13:49 ` Peter Zijlstra
2016-12-01 14:06 ` [PATCH v2 05/11] locking/ww_mutex: Add waiters in stamp order Nicolai Hähnle
2016-12-01 15:59 ` Chris Wilson
2016-12-16 14:21 ` Nicolai Hähnle
2016-12-06 15:36 ` Peter Zijlstra
2016-12-16 13:34 ` Nicolai Hähnle
2016-12-06 16:55 ` Peter Zijlstra
2016-12-16 14:19 ` Nicolai Hähnle
2016-12-16 14:46 ` Peter Zijlstra
2016-12-16 17:15 ` Peter Zijlstra
2016-12-16 18:11 ` Nicolai Hähnle
2016-12-16 20:00 ` Peter Zijlstra
2016-12-16 22:35 ` Nicolai Hähnle
2016-12-16 17:20 ` Peter Zijlstra
2016-12-16 18:12 ` Nicolai Hähnle
2016-12-01 14:06 ` [PATCH v2 06/11] locking/ww_mutex: Notify waiters that have to back off while adding tasks to wait list Nicolai Hähnle
2016-12-01 14:06 ` [PATCH v2 07/11] locking/ww_mutex: Wake at most one waiter for back off when acquiring the lock Nicolai Hähnle
2016-12-01 14:06 ` [PATCH v2 08/11] locking/ww_mutex: Yield to other waiters from optimistic spin Nicolai Hähnle
2016-12-01 14:06 ` [PATCH v2 09/11] locking/mutex: Initialize mutex_waiter::ww_ctx with poison when debugging Nicolai Hähnle
2016-12-01 14:06 ` [PATCH v2 10/11] Documentation/locking/ww_mutex: Update the design document Nicolai Hähnle
2016-12-01 14:06 ` [PATCH v2 11/11] [rfc] locking/ww_mutex: Always spin optimistically for the first waiter Nicolai Hähnle
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=1480601214-26583-1-git-send-email-nhaehnle@gmail.com \
--to=nhaehnle@gmail.com \
--cc=linux-kernel@vger.kernel.org \
/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®