From: Peter Zijlstra <peterz@infradead.org>
To: Waiman Long <waiman.long@hpe.com>
Cc: Ingo Molnar <mingo@redhat.com>,
linux-kernel@vger.kernel.org, Davidlohr Bueso <dave@stgolabs.net>,
Jason Low <jason.low2@hp.com>, Dave Chinner <david@fromorbit.com>,
Scott J Norton <scott.norton@hpe.com>,
Douglas Hatch <doug.hatch@hpe.com>
Subject: Re: [PATCH v2] locking/rwsem: Add reader-owned state to the owner field
Date: Tue, 10 May 2016 09:02:19 +0200 [thread overview]
Message-ID: <20160510070219.GG3408@twins.programming.kicks-ass.net> (raw)
In-Reply-To: <57314650.5050206@hpe.com>
On Mon, May 09, 2016 at 10:24:16PM -0400, Waiman Long wrote:
> >>+static inline void rwsem_set_reader_owned(struct rw_semaphore *sem)
> >>+{
> >>+ /*
> >>+ * We check the owner value first to make sure that we will only
> >>+ * do a write to the rwsem cacheline when it is really necessary
> >>+ * to minimize cacheline contention.
> >>+ */
> >>+ if (sem->owner != (struct task_struct *)RWSEM_READER_OWNED)
> >>+ sem->owner = (struct task_struct *)RWSEM_READER_OWNED;
> >How much if anything did this optimization matter?
>
> I hadn't run any performance test to verify the effective of this change.
> For a reader-heavy rwsem, this change should be able to save quite a lot of
> needless write to the rwsem cacheline.
Right; I was just wondering.
> >>+static inline bool rwsem_is_writer_owned(struct task_struct *owner)
> >>+{
> >>+ return (unsigned long)owner> RWSEM_READER_OWNED;
> >>+}
> >Tad too clever that; what does GCC generate if you write the obvious:
> >
> > return owner&& owner != RWSEM_READER_OWNER;
>
> You are right. GCC is intelligent enough to make the necessary optimization.
> I will revert it to this form which is more obvious.
Yay! thanks for checking. Sometimes GCC throws a wobbly and does
something unexpected, but it tends to get these 'simple' things right.
next prev parent reply other threads:[~2016-05-10 7:02 UTC|newest]
Thread overview: 33+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-05-07 0:20 [PATCH v2] locking/rwsem: Add reader-owned state to the owner field Waiman Long
2016-05-07 4:56 ` Ingo Molnar
2016-05-08 3:04 ` Waiman Long
2016-05-09 8:27 ` Peter Zijlstra
2016-05-10 2:24 ` Waiman Long
2016-05-10 7:02 ` Peter Zijlstra [this message]
2016-05-09 18:44 ` Jason Low
2016-05-10 13:03 ` Davidlohr Bueso
2016-05-11 22:04 ` Peter Hurley
2016-05-12 20:15 ` Waiman Long
2016-05-12 21:27 ` Peter Hurley
2016-05-12 23:13 ` Waiman Long
2016-05-13 15:07 ` Peter Zijlstra
2016-05-13 17:58 ` Peter Hurley
2016-05-15 14:47 ` Waiman Long
2016-05-16 11:09 ` Peter Zijlstra
2016-05-16 12:17 ` Paul E. McKenney
2016-05-16 14:17 ` Peter Hurley
2016-05-16 17:22 ` Paul E. McKenney
2016-05-17 19:46 ` Peter Hurley
2016-05-17 19:53 ` Peter Hurley
2016-05-16 17:50 ` Peter Zijlstra
2016-05-17 19:15 ` Peter Hurley
2016-05-17 19:46 ` Paul E. McKenney
2016-05-18 11:05 ` Peter Zijlstra
2016-05-18 15:56 ` Waiman Long
2016-05-18 17:28 ` Paul E. McKenney
2016-05-18 17:26 ` Paul E. McKenney
2016-05-19 9:00 ` Peter Zijlstra
2016-05-19 13:43 ` Paul E. McKenney
2016-05-19 1:37 ` Dave Chinner
2016-05-19 8:32 ` Peter Zijlstra
2016-05-20 22:56 ` Waiman Long
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=20160510070219.GG3408@twins.programming.kicks-ass.net \
--to=peterz@infradead.org \
--cc=dave@stgolabs.net \
--cc=david@fromorbit.com \
--cc=doug.hatch@hpe.com \
--cc=jason.low2@hp.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=scott.norton@hpe.com \
--cc=waiman.long@hpe.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).