From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
To: Adrian Cox <adrian@humboldt.co.uk>
Cc: linuxppc-dev@ozlabs.org
Subject: Re: Worst case performance of up()
Date: Sun, 03 Dec 2006 07:52:53 +1100 [thread overview]
Message-ID: <1165092773.22108.47.camel@localhost.localdomain> (raw)
In-Reply-To: <1165060484.4380.18.camel@localhost.localdomain>
On Sat, 2006-12-02 at 11:54 +0000, Adrian Cox wrote:
> On Sat, 2006-12-02 at 22:15 +1100, Benjamin Herrenschmidt wrote:
> > I think we are hitting a livelock due to both CPUs trying to perform an
> > atomic operation on the same cache line (or same variable even).
>
> I agree.
>
> > Can you remind me what CPU this on precisely ? I know that for some
> > CPUs like 970's, Apple code has some weirdo workarounds around atomic
> > ops involving forcing a mispredict when the stwcx. fails ... but if both
> > CPUs are following the exact same pattern, I can't see another way out
> > but an interrupt, unless something in the bus protocol can prevent such
> > livelocks...
>
> It's a pair of 7448s in MPX bus mode, with a Tsi109 host bridge.
At this point, we might have good use of Freescale help... there might
be something we can do in the construct of the atomic ops to avoid
that...
Ben.
next prev parent reply other threads:[~2006-12-02 20:53 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-11-24 16:21 Worst case performance of up() Adrian Cox
2006-11-24 20:45 ` Benjamin Herrenschmidt
2006-11-27 21:02 ` Adrian Cox
2006-12-02 10:35 ` Adrian Cox
2006-12-02 11:15 ` Benjamin Herrenschmidt
2006-12-02 11:54 ` Adrian Cox
2006-12-02 20:52 ` Benjamin Herrenschmidt [this message]
2006-12-04 13:43 ` Adrian Cox
2006-12-04 20:25 ` Benjamin Herrenschmidt
2006-12-05 22:00 ` Benjamin Herrenschmidt
2006-12-08 2:33 ` Paul Mackerras
2006-12-08 3:10 ` Benjamin Herrenschmidt
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=1165092773.22108.47.camel@localhost.localdomain \
--to=benh@kernel.crashing.org \
--cc=adrian@humboldt.co.uk \
--cc=linuxppc-dev@ozlabs.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).