From: Darren Hart <dvhltc@us.ibm.com>
To: Gregory Haskins <ghaskins@novell.com>
Cc: Thomas Gleixner <tglx@linutronix.de>, Ingo Molnar <mingo@elte.hu>,
Eric Dumazet <eric.dumazet@gmail.com>,
Steven Rostedt <rostedt@goodmis.org>,
Peter Zijlstra <peterz@infradead.org>,
Peter Morreale <PMorreale@novell.com>,
Sven Dietrich <SDietrich@novell.com>,
Chris Mason <chris.mason@oracle.com>, Avi Kivity <avi@redhat.com>,
Rik van Riel <riel@redhat.com>,
Chris Wright <chrisw@sous-sol.org>,
John Cooper <john.cooper@third-harmonic.com>,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH 5/6] futex: handle timeout inside adaptive lock spin
Date: Wed, 07 Apr 2010 16:15:05 -0700 [thread overview]
Message-ID: <4BBD11F9.1050705@us.ibm.com> (raw)
In-Reply-To: <4BBC9A3B0200005A00063E34@soto.provo.novell.com>
Gregory Haskins wrote:
>>>> On 4/7/2010 at 01:31 PM, in message <4BBCC174.7020409@us.ibm.com>,
Darren Hart
> <dvhltc@us.ibm.com> wrote:
>> Thomas Gleixner wrote:
>>> On Mon, 5 Apr 2010, Darren Hart wrote:
>>>
>>>> Signed-off-by: Darren Hart <dvhltc@us.ibm.com>
>
>>>> + if (timeout) {
>>>> + now = ktime_get();
>>> Hmm. Calling that in every iteration might hurt especially on non
>>> TSC systems, but well...
>> I haven't come across a better alternative since arming the timer
before
>> setting TASK_INTERRUPTIBLE isn't appropriate.
>
> Hey Darren,
>
> I remember we tried something similar in early versions of the
> adaptive locks and this was definitely bad. :(
>
> It ended up putting so much contention on the xtime_lock (IIRC) that
> it resulted in adaptive locks hurting overall performance verses not
> using adaptive at all. Alternative mechanisms employed a hybrid where
> the inner loops used a pseudo calibrated counter loop, and the outer
> loop checks periodically against a real clock. It all plays into "you
> are burning CPU cycles anyway, so might as well put them to use"
> theory. Hacky, yes, but it did relieve the pressure on the time
> subsystem locks and freed up a _lot_ of performance. Without this,
> the concept of timeouts+adaptive was unworkable. I think Steven
> ultimately rejected the timeout related patches outright when he
> merged adaptive to -rt, but I think Sven pulled them into SLERT if you
> would like a potential code reference to a working solution.
>
Hi Greg,
Thanks for the info! I haven't tested with timeouts yet as I'm still
struggling to get decent performance out of just plain old adaptive
right now. I'll keep that in mind when I get to that, and yeah, if the
plan is to burn CPU cycles, might as well do something constructive :-)
I do feel the timeouts are a necessary feature. Interruptibility may be
as well, but I'm going to ignore it for the time being...
--
Darren
next prev parent reply other threads:[~2010-04-07 23:15 UTC|newest]
Thread overview: 62+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-04-05 20:23 [PATCH V2 0/6][RFC] futex: FUTEX_LOCK with optional adaptive spinning Darren Hart
2010-04-05 20:23 ` [PATCH 1/6] futex: replace fshared and clockrt with combined flags Darren Hart
2010-04-05 20:23 ` [PATCH 2/6] futex: add futex_q static initializer Darren Hart
2010-04-05 20:23 ` [PATCH 3/6] futex: refactor futex_lock_pi_atomic Darren Hart
2010-04-05 20:23 ` [PATCH 4/6] futex: Add FUTEX_LOCK with optional adaptive spinning Darren Hart
2010-04-06 16:55 ` Thomas Gleixner
2010-04-07 17:26 ` Darren Hart
2010-04-07 19:59 ` Thomas Gleixner
2010-04-08 3:25 ` Darren Hart
2010-04-08 23:10 ` Peter W. Morreale
2010-04-09 5:41 ` Darren Hart
2010-04-09 13:13 ` Peter W. Morreale
2010-04-05 20:23 ` [PATCH 5/6] futex: handle timeout inside adaptive lock spin Darren Hart
2010-04-06 8:27 ` Thomas Gleixner
2010-04-07 17:31 ` Darren Hart
2010-04-07 18:44 ` Gregory Haskins
2010-04-07 23:15 ` Darren Hart [this message]
2010-04-05 20:23 ` [PATCH 6/6] futex: Add aggressive adaptive spinning argument to FUTEX_LOCK Darren Hart
2010-04-08 5:58 ` Darren Hart
2010-04-05 20:48 ` [PATCH V2^W V4 0/6][RFC] futex: FUTEX_LOCK with optional adaptive spinning Darren Hart
2010-04-05 21:15 ` [PATCH V2 " Avi Kivity
2010-04-05 21:54 ` Darren Hart
2010-04-05 22:21 ` Avi Kivity
2010-04-05 22:59 ` Darren Hart
2010-04-06 13:28 ` Avi Kivity
2010-04-06 13:35 ` Peter Zijlstra
2010-04-06 13:41 ` Avi Kivity
2010-04-06 14:09 ` Peter Zijlstra
2010-04-06 16:10 ` Avi Kivity
2010-04-06 16:53 ` Alan Cox
2010-04-06 13:51 ` Alan Cox
2010-04-06 15:28 ` Darren Hart
2010-04-06 16:06 ` Avi Kivity
2010-04-06 16:14 ` Thomas Gleixner
2010-04-06 16:20 ` Avi Kivity
2010-04-07 6:18 ` john cooper
2010-04-08 3:33 ` Darren Hart
2010-04-09 5:52 ` john cooper
2010-04-06 16:54 ` Alan Cox
2010-04-06 18:15 ` Thomas Gleixner
2010-04-06 16:44 ` Alan Cox
2010-04-06 17:34 ` Ulrich Drepper
2010-04-10 23:35 ` Alan Cox
2010-04-10 23:53 ` Ulrich Drepper
2010-04-06 19:31 ` Thomas Gleixner
2010-04-06 20:02 ` Ulrich Drepper
2010-04-06 23:16 ` Thomas Gleixner
2010-04-06 23:36 ` Darren Hart
2010-04-07 6:08 ` drepper
2010-04-08 3:41 ` Darren Hart
2010-04-08 4:29 ` drepper
2010-04-07 5:33 ` Avi Kivity
2010-04-06 21:22 ` Darren Hart
2010-04-05 23:15 ` Darren Hart
2010-04-05 23:29 ` Chris Wright
2010-04-06 13:30 ` Avi Kivity
2010-04-06 8:48 ` Peter Zijlstra
2010-04-06 14:47 ` Ulrich Drepper
2010-04-06 14:51 ` Peter Zijlstra
2010-04-06 15:33 ` Darren Hart
2010-04-06 15:37 ` Peter Zijlstra
2010-04-06 15:29 ` Peter Zijlstra
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=4BBD11F9.1050705@us.ibm.com \
--to=dvhltc@us.ibm.com \
--cc=PMorreale@novell.com \
--cc=SDietrich@novell.com \
--cc=avi@redhat.com \
--cc=chris.mason@oracle.com \
--cc=chrisw@sous-sol.org \
--cc=eric.dumazet@gmail.com \
--cc=ghaskins@novell.com \
--cc=john.cooper@third-harmonic.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=peterz@infradead.org \
--cc=riel@redhat.com \
--cc=rostedt@goodmis.org \
--cc=tglx@linutronix.de \
/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
Powered by JetHome