From: Peter Zijlstra <peterz@infradead.org>
To: Waiman Long <waiman.long@hp.com>
Cc: Will Deacon <will.deacon@arm.com>,
linux-arch@vger.kernel.org, linux-arm-kernel@lists.infradead.org,
mingo@redhat.com
Subject: Re: [PATCH 2/9] locking/qrwlock: avoid redundant atomic_add_return on read_lock_slowpath
Date: Tue, 7 Jul 2015 23:30:01 +0200 [thread overview]
Message-ID: <20150707213000.GL19282@twins.programming.kicks-ass.net> (raw)
In-Reply-To: <559C11BA.1070309@hp.com>
On Tue, Jul 07, 2015 at 01:51:54PM -0400, Waiman Long wrote:
> >- cnts = atomic_add_return(_QR_BIAS,&lock->cnts) - _QR_BIAS;
> >+ atomic_add(_QR_BIAS,&lock->cnts);
> >+ cnts = smp_load_acquire((u32 *)&lock->cnts);
> > rspin_until_writer_unlock(lock, cnts);
> >
> > /*
>
> Atomic add in x86 is actually a full barrier too. The performance difference
> between "lock add" and "lock xadd" should be minor. The additional load,
> however, could potentially cause an additional cacheline load on a contended
> lock. So do you see actual performance benefit of this change in ARM?
Yes, atomic_add() does not imply (and does not have) any memory barriers
on ARM.
next prev parent reply other threads:[~2015-07-07 21:30 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-07-07 17:24 [PATCH 0/9] locking/qrwlock: get qrwlocks up and running on arm64 Will Deacon
2015-07-07 17:24 ` [PATCH 1/9] locking/qrwlock: include <linux/spinlock.h> for arch_spin_{lock,unlock} Will Deacon
2015-07-07 17:24 ` [PATCH 2/9] locking/qrwlock: avoid redundant atomic_add_return on read_lock_slowpath Will Deacon
2015-07-07 17:51 ` Waiman Long
2015-07-07 18:19 ` Will Deacon
2015-07-07 19:28 ` Waiman Long
2015-07-08 9:59 ` Peter Zijlstra
2015-07-08 13:37 ` Will Deacon
2015-07-07 21:30 ` Peter Zijlstra [this message]
2015-07-07 17:24 ` [PATCH 3/9] locking/qrwlock: tidy up rspin_until_writer_unlock Will Deacon
2015-07-07 17:24 ` [PATCH 4/9] locking/qrwlock: implement queue_write_unlock using smp_store_release Will Deacon
2015-07-08 10:00 ` Peter Zijlstra
2015-07-07 17:24 ` [PATCH 5/9] locking/qrwlock: remove redundant cmpxchg barriers on writer slow-path Will Deacon
2015-07-08 10:05 ` Peter Zijlstra
2015-07-08 13:34 ` Will Deacon
2015-07-07 17:24 ` [PATCH 6/9] locking/qrwlock: allow architectures to hook in to contended paths Will Deacon
2015-07-08 10:06 ` Peter Zijlstra
2015-07-08 13:35 ` Will Deacon
2015-07-07 17:24 ` [PATCH 7/9] locking/qrwlock: expose internal lock structure in qrwlock definition Will Deacon
2015-07-07 17:24 ` [PATCH 8/9] arm64: cmpxchg: implement cmpxchg_relaxed Will Deacon
2015-07-07 17:24 ` [PATCH 9/9] arm64: locking: replace read/write locks with generic qrwlock code Will Deacon
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=20150707213000.GL19282@twins.programming.kicks-ass.net \
--to=peterz@infradead.org \
--cc=linux-arch@vger.kernel.org \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=mingo@redhat.com \
--cc=waiman.long@hp.com \
--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).