From: Anton Blanchard <anton@samba.org>
To: Christoph Hellwig <hch@lst.de>
Cc: linuxppc-dev@ozlabs.org, paulus@samba.org
Subject: Re: [patch 07/10] Page fault handler should not depend on CONFIG_KPROBES
Date: Fri, 23 Mar 2007 09:19:45 -0500 [thread overview]
Message-ID: <20070323141945.GB17198@kryten> (raw)
In-Reply-To: <20070321161342.GA17211@lst.de>
> NACK until the performace implication a fixed. Martin measured this
> costs 300 cycles on s390 even without any notifier activce, and while
> ppc might be slightly bad that's still too much to add to the pagefault
> handler unconditionally. Please add a light weight method to check
> whether there is any activce user of the notifications. Also please
> move the code into kernel/ at the same time instead of having the exact
> same thing duplicated all over, as in my notify_die patches.
Im struggling to understand how they could take 300 cycles to execute
this code sequence. We have seen similar questions on the ia64 side,
they both must have some pretty fundamental branch prediction issues.
I measured this in sim and the sequence takes under 20 cycles on a
power5 class machine, and thats a pessimistic estimate.
Anton
next prev parent reply other threads:[~2007-03-23 14:19 UTC|newest]
Thread overview: 29+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-03-21 1:38 [patch 00/10] Oops path and debugger hooks rework anton
2007-03-21 1:38 ` [patch 01/10] Add missing oops_enter/oops_exit anton
2007-03-21 1:38 ` [patch 02/10] Clean up pmac_backlight_unblank in oops path anton
2007-03-21 1:38 ` [patch 03/10] Handle recursive oopses anton
2007-03-21 1:38 ` [patch 04/10] Fix backwards ? : when printing machine type anton
2007-03-21 1:38 ` [patch 05/10] Use KERN_EMERG everywhere in oops printout anton
2007-03-21 2:23 ` Stephen Rothwell
2007-03-23 11:00 ` Paul Mackerras
2007-03-21 1:38 ` [patch 06/10] Add notify die hooks and remove some redundant debugger hooks anton
2007-03-21 16:14 ` Christoph Hellwig
2007-03-23 11:14 ` Paul Mackerras
2007-03-23 11:29 ` Christoph Hellwig
2007-03-23 12:04 ` Ananth N Mavinakayanahalli
2007-03-23 14:09 ` Anton Blanchard
2007-03-24 3:17 ` Paul Mackerras
2007-04-09 10:21 ` Paul Mackerras
2007-04-10 4:49 ` Ananth N Mavinakayanahalli
2007-03-21 1:38 ` [patch 07/10] Page fault handler should not depend on CONFIG_KPROBES anton
2007-03-21 16:13 ` Christoph Hellwig
2007-03-23 14:19 ` Anton Blanchard [this message]
2007-03-21 1:38 ` [patch 08/10] Use notifier hooks for xmon anton
2007-03-21 1:38 ` [patch 09/10] Use lowercase for hex printouts in oops messages anton
2007-03-21 2:14 ` Olof Johansson
2007-03-21 7:17 ` Geert Uytterhoeven
2007-03-21 16:44 ` Segher Boessenkool
2007-03-21 1:38 ` [patch 10/10] Make sure we only enable xmon once anton
2007-03-21 2:04 ` Anton Blanchard
2007-03-21 2:11 ` Olof Johansson
2007-03-21 2:05 ` Anton Blanchard
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=20070323141945.GB17198@kryten \
--to=anton@samba.org \
--cc=hch@lst.de \
--cc=linuxppc-dev@ozlabs.org \
--cc=paulus@samba.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).