From: Peter Zijlstra <peterz@infradead.org>
To: David Howells <dhowells@redhat.com>
Cc: linux-arch@vger.kernel.org, x86@kernel.org, will.deacon@arm.com,
linux-kernel@vger.kernel.org, ramana.radhakrishnan@arm.com,
paulmck@linux.vnet.ibm.com, dwmw2@infradead.org
Subject: Re: [RFC PATCH 00/15] Provide atomics and bitops implemented with ISO C++11 atomics
Date: Wed, 18 May 2016 19:22:58 +0200 [thread overview]
Message-ID: <20160518172258.GB3206@twins.programming.kicks-ass.net> (raw)
In-Reply-To: <146358423711.8596.9104061348359986393.stgit@warthog.procyon.org.uk>
On Wed, May 18, 2016 at 04:10:37PM +0100, David Howells wrote:
> There are some advantages to using ISO C++11 atomics:
>
> (1) The compiler can make use of extra information, such as condition
> flags, that are tricky to get out of inline assembly in an efficient
> manner. This should reduce the number of instructions required in
> some cases - such as in x86 where we use SETcc to store the condition
> inside the inline asm and then CMP outside to put it back again.
>
> Whilst this can be alleviated by the use of asm-goto constructs, this
> adds mandatory conditional jumps where the use of CMOVcc and SETcc
> might be better.
Supposedly gcc-6.1 has cc-output which makes this go away.
next prev parent reply other threads:[~2016-05-18 17:23 UTC|newest]
Thread overview: 40+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-05-18 15:10 [RFC PATCH 00/15] Provide atomics and bitops implemented with ISO C++11 atomics David Howells
2016-05-18 15:10 ` [RFC PATCH 01/15] cmpxchg_local() is not signed-value safe, so fix generic atomics David Howells
2016-05-18 15:29 ` Arnd Bergmann
2016-05-18 15:10 ` [RFC PATCH 02/15] tty: ldsem_cmpxchg() should use cmpxchg() not atomic_long_cmpxchg() David Howells
2016-05-18 15:10 ` [RFC PATCH 03/15] Provide atomic_t functions implemented with ISO-C++11 atomics David Howells
2016-05-18 17:31 ` Peter Zijlstra
2016-05-18 17:32 ` Peter Zijlstra
2016-05-19 7:36 ` David Woodhouse
2016-05-19 7:45 ` Peter Zijlstra
2016-05-18 17:33 ` Peter Zijlstra
2016-05-19 9:52 ` David Howells
2016-05-19 10:50 ` Peter Zijlstra
2016-05-19 11:31 ` Peter Zijlstra
2016-05-19 11:33 ` Peter Zijlstra
2016-05-19 14:22 ` Paul E. McKenney
2016-05-19 14:41 ` Peter Zijlstra
2016-05-19 15:00 ` Paul E. McKenney
2016-05-20 9:32 ` Michael Ellerman
2016-05-23 18:39 ` Paul E. McKenney
2016-06-01 14:16 ` Will Deacon
2016-05-18 15:11 ` [RFC PATCH 04/15] Convert 32-bit ISO atomics into a template David Howells
2016-05-18 15:11 ` [RFC PATCH 05/15] Provide atomic64_t and atomic_long_t using ISO atomics David Howells
2016-05-18 15:11 ` [RFC PATCH 06/15] Provide 16-bit " David Howells
2016-05-18 17:28 ` Peter Zijlstra
2016-05-18 15:11 ` [RFC PATCH 07/15] Provide cmpxchg(), xchg(), xadd() and __add() based on ISO C++11 intrinsics David Howells
2016-05-18 15:11 ` [RFC PATCH 08/15] Provide an implementation of bitops using C++11 atomics David Howells
2016-05-18 15:11 ` [RFC PATCH 09/15] Make the ISO bitops use 32-bit values internally David Howells
2016-05-18 15:11 ` [RFC PATCH 10/15] x86: Use ISO atomics David Howells
2016-05-18 15:12 ` [RFC PATCH 11/15] x86: Use ISO bitops David Howells
2016-05-18 15:12 ` [RFC PATCH 12/15] x86: Use ISO xchg(), cmpxchg() and friends David Howells
2016-05-18 15:12 ` [RFC PATCH 13/15] x86: Improve spinlocks using ISO C++11 intrinsic atomics David Howells
2016-05-18 17:37 ` Peter Zijlstra
2016-05-18 15:12 ` [RFC PATCH 14/15] x86: Make the mutex implementation use ISO atomic ops David Howells
2016-05-18 15:12 ` [RFC PATCH 15/15] x86: Fix misc cmpxchg() and atomic_cmpxchg() calls to use try/return variants David Howells
2016-05-18 17:22 ` Peter Zijlstra [this message]
2016-05-18 17:45 ` [RFC PATCH 00/15] Provide atomics and bitops implemented with ISO C++11 atomics Peter Zijlstra
2016-05-18 18:05 ` Peter Zijlstra
2016-05-19 0:23 ` Paul E. McKenney
2016-06-01 14:45 ` Will Deacon
2016-06-08 20:01 ` 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=20160518172258.GB3206@twins.programming.kicks-ass.net \
--to=peterz@infradead.org \
--cc=dhowells@redhat.com \
--cc=dwmw2@infradead.org \
--cc=linux-arch@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=paulmck@linux.vnet.ibm.com \
--cc=ramana.radhakrishnan@arm.com \
--cc=will.deacon@arm.com \
--cc=x86@kernel.org \
/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).