From: Nick Piggin <npiggin@suse.de>
To: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ingo Molnar <mingo@elte.hu>, David Howells <dhowells@redhat.com>,
Ulrich Drepper <drepper@redhat.com>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
Andrew Morton <akpm@linux-foundation.org>
Subject: Re: [PATCH 0/3] 64-bit futexes: Intro
Date: Thu, 5 Jun 2008 03:58:57 +0200 [thread overview]
Message-ID: <20080605015857.GD4327@wotan.suse.de> (raw)
In-Reply-To: <20080605015634.GC4327@wotan.suse.de>
On Thu, Jun 05, 2008 at 03:56:34AM +0200, Nick Piggin wrote:
>
> The other thing that could be possible, and I'd imagine maybe more likely
> to be implemented in a real CPU because it should give more imrpovement
> (and which does break my algorithm) is just that the load to the cacheline
> may get to execute first, then if the cacheline gets evicted and
> modified by another CPU before our store completes, we effectively see
> store/load reordering again.
Well, I should qualify that: it doesn't actually break my algorithm
because you can still implement the unlock without atomic RMWs.
This may require you to have an mfence there, but we can still get
away without atomics (whether that's much cheaper or not, I haven't
measured recently!)
next prev parent reply other threads:[~2008-06-05 1:59 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-05-31 1:27 [PATCH 0/3] 64-bit futexes: Intro Ulrich Drepper
2008-05-31 2:13 ` Linus Torvalds
2008-05-31 3:14 ` Ulrich Drepper
2008-05-31 3:44 ` Linus Torvalds
2008-05-31 4:04 ` Ulrich Drepper
2008-05-31 4:16 ` Linus Torvalds
2008-05-31 4:23 ` Linus Torvalds
2008-05-31 4:38 ` Ulrich Drepper
2008-05-31 4:58 ` Linus Torvalds
2008-05-31 22:25 ` Linus Torvalds
2008-05-31 22:32 ` Linus Torvalds
2008-06-02 18:54 ` Ingo Molnar
2008-06-02 20:22 ` Linus Torvalds
2008-06-02 23:03 ` Ingo Molnar
2008-06-03 3:24 ` Nick Piggin
2008-06-04 19:57 ` Linus Torvalds
2008-06-04 20:38 ` Linus Torvalds
2008-06-05 1:56 ` Nick Piggin
2008-06-05 1:58 ` Nick Piggin [this message]
2008-06-05 3:08 ` Linus Torvalds
2008-06-05 4:29 ` Nick Piggin
2008-06-05 1:45 ` Nick Piggin
2008-06-06 1:27 ` Nick Piggin
2008-06-06 3:37 ` Linus Torvalds
2008-06-06 11:53 ` Nick Piggin
2008-06-06 15:01 ` Linus Torvalds
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=20080605015857.GD4327@wotan.suse.de \
--to=npiggin@suse.de \
--cc=akpm@linux-foundation.org \
--cc=dhowells@redhat.com \
--cc=drepper@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--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 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.