linux-arch.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Will Deacon <will.deacon@arm.com>
To: Peter Zijlstra <peterz@infradead.org>
Cc: Waiman Long <waiman.long@hp.com>, Ingo Molnar <mingo@redhat.com>,
	Arnd Bergmann <arnd@arndb.de>,
	Thomas Gleixner <tglx@linutronix.de>,
	"linux-arch@vger.kernel.org" <linux-arch@vger.kernel.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	Scott J Norton <scott.norton@hp.com>,
	Douglas Hatch <doug.hatch@hp.com>
Subject: Re: [PATCH 2/4] locking/qrwlock: Reduce reader/writer to reader lock transfer latency
Date: Wed, 8 Jul 2015 18:19:49 +0100	[thread overview]
Message-ID: <20150708171949.GE6348@arm.com> (raw)
In-Reply-To: <20150708095248.GZ3644@twins.programming.kicks-ass.net>

On Wed, Jul 08, 2015 at 10:52:48AM +0100, Peter Zijlstra wrote:
> On Tue, Jul 07, 2015 at 05:29:50PM -0400, Waiman Long wrote:
> > I prefer the current setting as you won't know if the writer has the
> > lock or not when you take a snapshot of the value of the lock. You
> > need the whole time sequence in this case to figure it out and so will
> > be more prone to error.
> 
> I still need to wake up, but I suspect we need to change
> queue_read_{try,}lock() to use cmpxchg/inc_not_zero like things, which
> is fine for ARM, but not so much for x86.
> 
> So I think I agree with Waiman, but am willing to be shown differently.

That's fine; I just wanted to make sure I wasn't going round the twist!

Will

  parent reply	other threads:[~2015-07-08 17:19 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-07-06 15:43 [PATCH 0/4] locking/qrwlock: Improve qrwlock performance Waiman Long
2015-07-06 15:43 ` [PATCH 1/4] locking/qrwlock: Better optimization for interrupt context readers Waiman Long
2015-07-06 15:43 ` [PATCH 2/4] locking/qrwlock: Reduce reader/writer to reader lock transfer latency Waiman Long
2015-07-06 18:23   ` Will Deacon
2015-07-06 19:49     ` Waiman Long
2015-07-06 19:49       ` Waiman Long
2015-07-07  9:17       ` Will Deacon
2015-07-07  9:17         ` Will Deacon
2015-07-07 11:17         ` Peter Zijlstra
2015-07-07 11:49           ` Will Deacon
2015-07-07 14:30             ` Waiman Long
2015-07-07 17:27               ` Will Deacon
2015-07-07 18:10                 ` Will Deacon
2015-07-07 21:29                   ` Waiman Long
2015-07-08  9:52                     ` Peter Zijlstra
2015-07-08  9:52                       ` Peter Zijlstra
2015-07-08 17:19                       ` Will Deacon [this message]
2015-07-06 15:43 ` [PATCH 3/4] locking/qrwlock: Reduce writer to writer " Waiman Long
2015-07-06 15:43 ` [PATCH 4/4] locking/qrwlock: Use direct MCS lock/unlock in slowpath Waiman Long
2015-07-07 11:24   ` Peter Zijlstra
2015-07-07 21:59     ` Waiman Long
2015-07-07 21:59       ` Waiman Long
2015-07-07 22:13       ` 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=20150708171949.GE6348@arm.com \
    --to=will.deacon@arm.com \
    --cc=arnd@arndb.de \
    --cc=doug.hatch@hp.com \
    --cc=linux-arch@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@redhat.com \
    --cc=peterz@infradead.org \
    --cc=scott.norton@hp.com \
    --cc=tglx@linutronix.de \
    --cc=waiman.long@hp.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).