From: Richard Henderson <rth@twiddle.net>
To: "Emilio G. Cota" <cota@braap.org>,
QEMU Developers <qemu-devel@nongnu.org>,
MTTCG Devel <mttcg@greensocs.com>
Cc: "Alex Bennée" <alex.bennee@linaro.org>,
"Paolo Bonzini" <pbonzini@redhat.com>,
"Sergey Fedorov" <serge.fdrv@gmail.com>
Subject: Re: [Qemu-devel] [PATCH 1/2] atomics: do not use __atomic primitives for RCU atomics
Date: Mon, 23 May 2016 09:53:00 -0700 [thread overview]
Message-ID: <e48bb769-9339-7c3f-6ec4-496609bc2bca@twiddle.net> (raw)
In-Reply-To: <1463863336-28760-2-git-send-email-cota@braap.org>
On 05/21/2016 01:42 PM, Emilio G. Cota wrote:
> In the process, the atomic_rcu_read/set were converted to implement
> consume/release semantics, respectively. This is inefficient; for
> correctness and maximum performance we only need an smp_barrier_depends
> for reads, and an smp_wmb for writes. Fix it by using the original
> definition of these two primitives for all compilers.
For what host do you think this is inefficient?
In particular, what you've done is going to be less efficient for e.g. armv8,
where the __atomic formulation is going to produce load-acquire and
store-release instructions. Whereas the separate barriers are going to produce
two insns.
As for the common case of x86_64, what you're doing is going to make no
difference at all.
So what are you trying to improve?
r~
next prev parent reply other threads:[~2016-05-23 16:53 UTC|newest]
Thread overview: 14+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-05-21 20:42 [Qemu-devel] [PATCH 0/2] atomics: fix small RCU perf. regression + update documentation Emilio G. Cota
2016-05-21 20:42 ` [Qemu-devel] [PATCH 1/2] atomics: do not use __atomic primitives for RCU atomics Emilio G. Cota
2016-05-22 7:58 ` Alex Bennée
2016-05-24 18:42 ` Emilio G. Cota
2016-05-23 14:21 ` Paolo Bonzini
2016-05-23 15:55 ` Emilio G. Cota
2016-05-23 16:53 ` Richard Henderson [this message]
2016-05-23 17:09 ` Emilio G. Cota
2016-05-24 7:08 ` Paolo Bonzini
2016-05-24 19:56 ` Emilio G. Cota
2016-05-24 19:59 ` Sergey Fedorov
2016-05-25 8:52 ` Alex Bennée
2016-05-25 11:02 ` Sergey Fedorov
2016-05-21 20:42 ` [Qemu-devel] [PATCH 2/2] docs/atomics: update atomic_read/set comparison with Linux Emilio G. Cota
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=e48bb769-9339-7c3f-6ec4-496609bc2bca@twiddle.net \
--to=rth@twiddle.net \
--cc=alex.bennee@linaro.org \
--cc=cota@braap.org \
--cc=mttcg@greensocs.com \
--cc=pbonzini@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=serge.fdrv@gmail.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.