From: Peter Zijlstra <peterz@infradead.org>
To: Michael Ellerman <mpe@ellerman.id.au>
Cc: paulmck@linux.vnet.ibm.com, Will Deacon <will.deacon@arm.com>,
linux-arch@vger.kernel.org, linux-kernel@vger.kernel.org,
Boqun Feng <boqun.feng@gmail.com>,
Anton Blanchard <anton@samba.org>,
Benjamin Herrenschmidt <benh@kernel.crashing.org>,
Paul Mackerras <paulus@samba.org>,
linuxppc-dev@lists.ozlabs.org
Subject: Re: [PATCH v2] barriers: introduce smp_mb__release_acquire and update documentation
Date: Thu, 8 Oct 2015 13:16:38 +0200 [thread overview]
Message-ID: <20151008111638.GL3816@twins.programming.kicks-ass.net> (raw)
In-Reply-To: <1444276236.9940.5.camel@ellerman.id.au>
On Thu, Oct 08, 2015 at 02:50:36PM +1100, Michael Ellerman wrote:
> On Wed, 2015-10-07 at 08:25 -0700, Paul E. McKenney wrote:
> > Currently, we do need smp_mb__after_unlock_lock() to be after the
> > acquisition on PPC -- putting it between the unlock and the lock
> > of course doesn't cut it for the cross-thread unlock/lock case.
This ^, that makes me think I don't understand
smp_mb__after_unlock_lock.
How is:
UNLOCK x
smp_mb__after_unlock_lock()
LOCK y
a problem? That's still a full barrier.
> > I am with Peter -- we do need the benchmark results for PPC.
>
> Urgh, sorry guys. I have been slowly doing some benchmarks, but time is not
> plentiful at the moment.
>
> If we do a straight lwsync -> sync conversion for unlock it looks like that
> will cost us ~4.2% on Anton's standard context switch benchmark.
And that does not seem to agree with Paul's smp_mb__after_unlock_lock()
usage and would not be sufficient for the same (as of yet unexplained)
reason.
Why does it matter which of the LOCK or UNLOCK gets promoted to full
barrier on PPC in order to become RCsc?
next prev parent reply other threads:[~2015-10-08 11:16 UTC|newest]
Thread overview: 48+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-10-07 10:59 [PATCH v2] barriers: introduce smp_mb__release_acquire and update documentation Will Deacon
2015-10-07 11:19 ` Peter Zijlstra
2015-10-07 13:23 ` Will Deacon
2015-10-07 13:23 ` Will Deacon
2015-10-07 14:53 ` Peter Zijlstra
2015-10-07 15:25 ` Paul E. McKenney
2015-10-08 3:50 ` Michael Ellerman
2015-10-08 3:50 ` Michael Ellerman
2015-10-08 11:16 ` Peter Zijlstra [this message]
2015-10-08 12:59 ` Will Deacon
2015-10-08 22:17 ` Paul E. McKenney
2015-10-09 9:51 ` Will Deacon
2015-10-09 9:51 ` Will Deacon
2015-10-09 11:25 ` Peter Zijlstra
2015-10-09 11:25 ` Peter Zijlstra
2015-10-09 17:44 ` Paul E. McKenney
2015-10-09 17:44 ` Paul E. McKenney
2015-10-09 17:43 ` Paul E. McKenney
2015-10-09 18:33 ` Will Deacon
2015-10-12 23:30 ` Paul E. McKenney
2015-10-20 14:20 ` Boqun Feng
2015-10-08 21:44 ` Paul E. McKenney
2015-10-09 7:29 ` Peter Zijlstra
2015-10-09 8:31 ` Peter Zijlstra
2015-10-09 9:40 ` Will Deacon
2015-10-09 9:40 ` Will Deacon
2015-10-09 11:02 ` Peter Zijlstra
2015-10-09 12:41 ` Will Deacon
2015-10-09 11:12 ` Peter Zijlstra
2015-10-09 12:51 ` Will Deacon
2015-10-09 13:06 ` Peter Zijlstra
2015-10-09 11:13 ` Peter Zijlstra
2015-10-09 17:21 ` Paul E. McKenney
2015-10-19 1:17 ` Boqun Feng
2015-10-19 1:17 ` Boqun Feng
2015-10-19 10:23 ` Peter Zijlstra
2015-10-19 10:23 ` Peter Zijlstra
2015-10-20 7:35 ` Boqun Feng
2015-10-20 7:35 ` Boqun Feng
2015-10-20 23:34 ` Paul E. McKenney
2015-10-21 8:24 ` Peter Zijlstra
2015-10-21 19:29 ` Paul E. McKenney
2015-10-21 19:36 ` Peter Zijlstra
2015-10-21 19:36 ` Peter Zijlstra
2015-10-21 19:56 ` Paul E. McKenney
2015-10-21 16:04 ` David Laight
2015-10-21 16:04 ` David Laight
2015-10-21 19:34 ` Paul E. McKenney
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=20151008111638.GL3816@twins.programming.kicks-ass.net \
--to=peterz@infradead.org \
--cc=anton@samba.org \
--cc=benh@kernel.crashing.org \
--cc=boqun.feng@gmail.com \
--cc=linux-arch@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linuxppc-dev@lists.ozlabs.org \
--cc=mpe@ellerman.id.au \
--cc=paulmck@linux.vnet.ibm.com \
--cc=paulus@samba.org \
--cc=will.deacon@arm.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).