From: Christoph Lameter <clameter@sgi.com>
To: Chris Snook <csnook@redhat.com>
Cc: linux-arch@vger.kernel.org, linux-kernel@vger.kernel.org,
Linus Torvalds <torvalds@linux-foundation.org>,
akpm@linux-foundation.org, paulmck@linux.vnet.ibm.com,
Segher Boessenkool <segher@kernel.crashing.org>,
"Luck, Tony" <tony.luck@intel.com>,
Chris Friesen <cfriesen@nortel.com>,
"Robert P. J. Day" <rpjday@mindspring.com>
Subject: Re: [PATCH 1/23] document preferred use of volatile with atomic_t
Date: Tue, 14 Aug 2007 15:45:28 -0700 (PDT) [thread overview]
Message-ID: <Pine.LNX.4.64.0708141544001.401@schroedinger.engr.sgi.com> (raw)
In-Reply-To: <20070813110415.GA24018@shell.boston.redhat.com>
On Mon, 13 Aug 2007, Chris Snook wrote:
> @@ -38,7 +45,7 @@
>
> Next, we have:
>
> - #define atomic_read(v) ((v)->counter)
> + #define atomic_read(v) (*(volatile int *)&(v)->counter)
>
> which simply reads the current value of the counter.
volatile means that there is some vague notion of "read it now". But that
really does not exist. Instead we control visibility via barriers
(smp_wmb, smp_rmb). Would it not be best to not have volatile at all in
atomic operations and let the barriers do the work?
next prev parent reply other threads:[~2007-08-14 22:45 UTC|newest]
Thread overview: 49+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-08-13 10:55 [PATCH 0/23] make atomic_read() and atomic_set() behavior consistent across all architectures Chris Snook
2007-08-13 11:04 ` [PATCH 1/23] document preferred use of volatile with atomic_t Chris Snook
2007-08-13 23:54 ` Paul E. McKenney
2007-08-14 22:45 ` Christoph Lameter [this message]
2007-08-14 22:53 ` Chris Snook
2007-08-14 22:56 ` Christoph Lameter
2007-08-14 23:25 ` Chris Snook
2007-08-14 23:28 ` Paul E. McKenney
2007-08-16 21:36 ` Segher Boessenkool
2007-08-13 11:06 ` [PATCH 2/23] make atomic_read() and atomic_set() behavior consistent on alpha Chris Snook
2007-08-13 11:09 ` [PATCH 3/23] make atomic_read() and atomic_set() behavior consistent on arm Chris Snook
2007-08-13 12:19 ` Russell King
2007-08-13 12:46 ` Chris Snook
2007-08-13 12:59 ` Russell King
2007-08-13 11:11 ` [PATCH 4/23] make atomic_read() and atomic_set() behavior consistent on avr32 Chris Snook
2007-08-13 11:12 ` [PATCH 5/23] make atomic_read() and atomic_set() behavior consistent on blackfin Chris Snook
2007-08-13 11:14 ` [PATCH 6/23] make atomic_read() and atomic_set() behavior consistent on cris Chris Snook
2007-08-13 11:15 ` [PATCH 7/23] make atomic_read() and atomic_set() behavior consistent on frv Chris Snook
2007-08-13 11:18 ` [PATCH 8/23] make atomic_read() and atomic_set() behavior consistent on h8300 Chris Snook
2007-08-13 11:21 ` [PATCH 9/23] make atomic_read() and atomic_set() behavior consistent on i386 Chris Snook
2007-08-13 11:23 ` [PATCH 10/23] make atomic_read() and atomic_set() behavior consistent on ia64 Chris Snook
2007-08-14 18:27 ` Luck, Tony
2007-08-14 18:48 ` Chris Snook
2007-08-14 22:06 ` Luck, Tony
2007-08-14 22:11 ` Christoph Lameter
2007-08-14 22:21 ` Chris Snook
2007-08-13 11:24 ` [PATCH 11/23] make atomic_read() and atomic_set() behavior consistent on m32r Chris Snook
2007-08-22 1:56 ` Hirokazu Takata
2007-08-22 5:00 ` Hirokazu Takata
2007-08-22 14:06 ` Chris Snook
2007-08-22 14:24 ` Segher Boessenkool
2007-08-22 18:20 ` Linus Torvalds
2007-08-23 19:29 ` Segher Boessenkool
2007-08-23 20:12 ` Linus Torvalds
2007-08-23 20:40 ` Valdis.Kletnieks
2007-08-23 20:05 ` David Howells
2007-08-13 11:26 ` [PATCH 12/23] make atomic_read() and atomic_set() behavior consistent on m68knommu Chris Snook
2007-08-13 11:28 ` [PATCH 13/23] make atomic_read() and atomic_set() behavior consistent on m68k Chris Snook
2007-08-13 11:29 ` [PATCH 14/23] make atomic_read() and atomic_set() behavior consistent on mips Chris Snook
2007-08-13 11:31 ` [PATCH 15/23] make atomic_read() and atomic_set() behavior consistent on parisc Chris Snook
2007-08-13 11:33 ` [PATCH 16/23] make atomic_read() and atomic_set() behavior consistent on s390 Chris Snook
2007-08-13 11:34 ` [PATCH 17/23] make atomic_read() and atomic_set() behavior consistent on sh64 Chris Snook
2007-08-13 11:36 ` [PATCH 18/23] make atomic_read() and atomic_set() behavior consistent on sh Chris Snook
2007-08-13 11:40 ` [PATCH 19/23] make atomic_read() and atomic_set() behavior consistent on sparc64 Chris Snook
2007-08-13 11:42 ` [PATCH 20/23] make atomic_read() and atomic_set() behavior consistent on sparc Chris Snook
2007-08-13 11:43 ` [PATCH 21/23] make atomic_read() and atomic_set() behavior consistent on v850 Chris Snook
2007-08-13 11:44 ` [PATCH 22/23] make atomic_read() and atomic_set() behavior consistent on x86_64 Chris Snook
2007-08-13 11:45 ` [PATCH 23/23] make atomic_read() and atomic_set() behavior consistent on xtensa Chris Snook
2007-08-14 9:42 ` [PATCH 7/23] make atomic_read() and atomic_set() behavior consistent on frv David Howells
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=Pine.LNX.4.64.0708141544001.401@schroedinger.engr.sgi.com \
--to=clameter@sgi.com \
--cc=akpm@linux-foundation.org \
--cc=cfriesen@nortel.com \
--cc=csnook@redhat.com \
--cc=linux-arch@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=paulmck@linux.vnet.ibm.com \
--cc=rpjday@mindspring.com \
--cc=segher@kernel.crashing.org \
--cc=tony.luck@intel.com \
--cc=torvalds@linux-foundation.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).