From: Paul Mackerras <paulus@samba.org>
To: David Howells <dhowells@redhat.com>
Cc: ebiederm@xmission.com (Eric W. Biederman),
akpm@osdl.org, linux-arch@vger.kernel.org,
linux-kernel@vger.kernel.org, torvalds@osdl.org,
mingo@redhat.com, alan@redhat.com, linuxppc64-dev@ozlabs.org
Subject: Re: [PATCH] Document Linux's memory barriers [try #4]
Date: Wed, 15 Mar 2006 08:48:03 +1100 [thread overview]
Message-ID: <17431.14867.211423.851470@cargo.ozlabs.ibm.com> (raw)
In-Reply-To: <32068.1142371612@warthog.cambridge.redhat.com>
David Howells writes:
> + CPU 1 CPU 2 COMMENT
> + =============== =============== =======================================
> + a == 0, b == 1 and p == &a, q == &a
> + b = 2;
> + smp_wmb(); Make sure b is changed before p
> + p = &b; q = p;
> + d = *q;
> +
> +then old data values may be used in the address calculation for the second
> +value, potentially resulting in q == &b and d == 0 being seen, which is never
> +correct. What is required is a data dependency memory barrier:
No, that's not the problem. The problem is that you can get q == &b
and d == 1, believe it or not. That is, you can see the new value of
the pointer but the old value of the thing pointed to.
Paul.
next prev parent reply other threads:[~2006-03-14 21:48 UTC|newest]
Thread overview: 42+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <20060315200956.4a9e2cb3.akpm@osdl.org>
2006-03-09 20:29 ` David Howells
2006-03-09 23:34 ` Paul Mackerras
2006-03-09 23:45 ` Michael Buesch
2006-03-09 23:56 ` Linus Torvalds
2006-03-10 0:07 ` Michael Buesch
2006-03-10 0:48 ` Alan Cox
2006-03-10 0:54 ` Paul Mackerras
2006-03-10 5:28 ` Nick Piggin
2006-03-10 15:19 ` David Howells
2006-03-11 0:01 ` Paul Mackerras
2006-03-12 17:15 ` Eric W. Biederman
2006-03-13 12:32 ` Sergei Organov
2006-03-14 20:31 ` David Howells
2006-03-14 21:11 ` linux-os (Dick Johnson)
2006-03-15 9:09 ` Sergei Organov
2006-03-15 9:04 ` Sergei Organov
2006-03-14 20:35 ` David Howells
2006-03-15 9:11 ` Sergei Organov
2006-03-14 21:26 ` David Howells
2006-03-14 21:48 ` Paul Mackerras [this message]
2006-03-14 23:59 ` David Howells
2006-03-15 0:20 ` Linus Torvalds
2006-03-15 1:25 ` Nick Piggin
2006-03-15 0:54 ` Paul Mackerras
2006-03-15 1:19 ` David Howells
2006-03-15 1:47 ` Linus Torvalds
2006-03-15 11:10 ` David Howells
2006-03-15 11:51 ` Nick Piggin
2006-03-15 13:47 ` David Howells
2006-03-15 23:21 ` Nick Piggin
2006-03-15 14:23 ` [PATCH] Document Linux's memory barriers [try #5] David Howells
2006-03-16 11:50 ` David Howells
2006-03-16 17:18 ` Linus Torvalds
2006-03-17 1:20 ` Nick Piggin
2006-03-16 23:17 ` Paul E. McKenney
2006-03-16 23:55 ` Linus Torvalds
2006-03-17 1:29 ` Paul E. McKenney
2006-03-17 5:32 ` Linus Torvalds
2006-03-17 6:23 ` Paul E. McKenney
2006-03-23 18:34 ` David Howells
2006-03-23 19:28 ` Linus Torvalds
2006-03-23 22:26 ` 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=17431.14867.211423.851470@cargo.ozlabs.ibm.com \
--to=paulus@samba.org \
--cc=akpm@osdl.org \
--cc=alan@redhat.com \
--cc=dhowells@redhat.com \
--cc=ebiederm@xmission.com \
--cc=linux-arch@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linuxppc64-dev@ozlabs.org \
--cc=mingo@redhat.com \
--cc=torvalds@osdl.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
all inboxes | Powered by JetHome®