From: Andi Kleen <ak@suse.de>
To: Ingo Molnar <mingo@elte.hu>
Cc: Andi Kleen <ak@suse.de>, Andrew Morton <akpm@osdl.org>,
Zwane Mwaikambo <zwane@fsmlabs.com>,
linux-kernel@vger.kernel.org, wli@holomorphy.com
Subject: Re: [PATCH] remove LOCK_SECTION from x86_64 spin_lock asm
Date: Thu, 16 Sep 2004 08:53:42 +0200 [thread overview]
Message-ID: <20040916065342.GE12915@wotan.suse.de> (raw)
In-Reply-To: <20040916065101.GA11726@elte.hu>
On Thu, Sep 16, 2004 at 08:51:01AM +0200, Ingo Molnar wrote:
>
> * Andi Kleen <ak@suse.de> wrote:
>
> > I think the idea was that the spinlock functions should be small
> > enough that they don't have any stack local variables. [...]
>
> this might work for x64, but even there, are you sure it works even with
> CONFIG_PREEMPT enabled (there the spinlocks are more complex)? It sure
I would expect so. x86-64 should have enough callee clobbered registers
by default to handle it. Unless someone makes them complex enough that it
needs more than 4 variables or so or adds another function call. But I hope
this won't happen.
> wont work on x86 so i think we need a generic 'soft PC' solution.
>
> the alternative would be to unwind the stack - quite some task on some
> platforms ...
Sometimes call graph profiling would be very useful, but I wouldn't
want the profiler to do it by default, especially not for this silly
simple case. dwarf2 unwinding is complex enough that just requiring
frame pointers for the CG case would look attractive.
-Andi
next prev parent reply other threads:[~2004-09-16 6:53 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2004-09-15 16:01 Zwane Mwaikambo
2004-09-15 21:45 ` Andrew Morton
2004-09-15 17:55 ` Zwane Mwaikambo
2004-09-15 21:47 ` Ingo Molnar
2004-09-16 6:13 ` Andi Kleen
2004-09-16 6:27 ` Ingo Molnar
2004-09-16 6:44 ` Andi Kleen
2004-09-16 6:51 ` Ingo Molnar
2004-09-16 6:53 ` Andi Kleen [this message]
2004-09-16 6:58 ` Ingo Molnar
2004-09-16 7:09 ` Andi Kleen
2004-09-16 7:19 ` Ingo Molnar
2004-09-16 7:29 ` Andi Kleen
2004-09-16 7:44 ` Ingo Molnar
2004-09-16 7:53 ` Andi Kleen
2004-09-16 9:01 ` Andi Kleen
2004-09-16 12:44 ` Zwane Mwaikambo
2004-09-16 19:30 ` Ingo Molnar
2004-09-15 22:42 Andrew Chew
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=20040916065342.GE12915@wotan.suse.de \
--to=ak@suse.de \
--cc=akpm@osdl.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=wli@holomorphy.com \
--cc=zwane@fsmlabs.com \
/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
Powered by JetHome