From: Chuck Ebbert <76306.1226@compuserve.com>
To: Ingo Molnar <mingo@elte.hu>
Cc: Jakub Jelinek <jakub@redhat.com>,
Roland McGrath <roland@redhat.com>,
Ulrich Drepper <drepper@redhat.com>, Andi Kleen <ak@suse.de>,
Linus Torvalds <torvalds@osdl.org>,
linux-kernel <linux-kernel@vger.kernel.org>
Subject: Re: [RFC, patch] i386: vgetcpu(), take 2
Date: Wed, 21 Jun 2006 08:24:30 -0400 [thread overview]
Message-ID: <200606210828_MC3-1-C30B-9D83@compuserve.com> (raw)
In-Reply-To: <20060621081539.GA14227@elte.hu>
On Wed, 21 Jun 2006 10:15:39 +0200, Ingo Molnar wrote:
> * Chuck Ebbert <76306.1226@compuserve.com> wrote:
>
> > Use a GDT entry's limit field to store per-cpu data for fast access
> > from userspace, and provide a vsyscall to access the current CPU
> > number stored there.
>
> very nice idea! I thought of doing sys_get_cpu() too, but my idea was to
> use the scheduler to keep a writable [and permanently pinned,
> per-thread] VDSO data page uptodate with the current CPU# [and other
> interesting data]. Btw., do we know how fast LSL is on modern CPUs?
Now that the GDT is a full page for each CPU there's plenty of space
for all kinds of per-cpu data, even if we waste 75% of it. LSL seems
pretty fast; I got 13 clocks for the whole lsl/jnz/and sequence on K8
and 21 clocks on PII. Myabe you can test P4?
/* test how fast lsl/jnz/and runs.
*/
#define _GNU_SOURCE
#include <stdio.h>
#include <stdlib.h>
#define rdtscll(t) asm volatile ("rdtsc" : "=A" (t))
#ifndef ITERS
#define ITERS 1000000
#endif
int main(int argc, char * const argv[])
{
unsigned long long tsc1, tsc2;
int count, cpu, junk;
rdtscll(tsc1);
asm (
" pushl %%ds \n"
" popl %2 \n"
"1: \n"
#ifdef DO_TEST
" lsl %2,%0 \n"
" jnz 2f \n"
" and $0xff,%0 \n"
#endif
" dec %1 \n"
" jnz 1b \n"
"2: \n"
: "=&r" (cpu), "=&r" (count), "=&r" (junk)
: "1" (ITERS), "0" (-1)
);
rdtscll(tsc2);
if (count == 0)
printf("loops: %d, avg: %llu clocks\n",
ITERS, (tsc2 - tsc1) / ITERS);
return 0;
}
> > +__vgetcpu:
> > +.LSTART_vgetcpu:
> > + movl $-EFAULT,%eax
> > + movl $((27<<3)|3),%edx
> > + lsll %edx,%eax
> > + jnz 1f
> > + andl $0xff,%eax
> > +1:
> > + ret
>
> this needs unwinder annotations as well to make this a proper DSO, so
> that for example a breakpoint here does not confuse gdb.
I can't write those.
> also, would be nice to do something like this in 64-bit mode too.
Andi has x86_64 patches in his tree and is considering this method for
ia32 support.
--
Chuck
"You can't read a newspaper if you can't read." --George W. Bush
next reply other threads:[~2006-06-21 12:32 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-06-21 12:24 Chuck Ebbert [this message]
2006-06-21 17:14 ` [RFC, patch] i386: vgetcpu(), take 2 Andi Kleen
2006-06-21 17:27 ` Linus Torvalds
2006-06-21 17:50 ` Andi Kleen
-- strict thread matches above, loose matches on Subject: below --
2006-06-22 12:23 Chuck Ebbert
2006-06-22 12:44 ` Andi Kleen
2006-06-21 7:27 Chuck Ebbert
2006-06-21 8:15 ` Ingo Molnar
2006-06-21 17:38 ` Artur Skawina
2006-06-28 5:44 ` Paul Jackson
2006-06-28 8:53 ` Andi Kleen
2006-06-28 9:00 ` Ingo Molnar
2006-06-29 8:47 ` Paul Jackson
2006-06-21 9:26 ` Andi Kleen
2006-06-21 9:35 ` Ingo Molnar
2006-06-21 21:54 ` Rohit Seth
2006-06-21 22:21 ` Andi Kleen
2006-06-21 22:59 ` Rohit Seth
2006-06-21 23:05 ` Andi Kleen
2006-06-21 23:18 ` Rohit Seth
2006-06-21 23:29 ` Andi Kleen
2006-06-22 0:55 ` Rohit Seth
2006-06-22 8:08 ` Andi Kleen
2006-06-22 21:06 ` Rohit Seth
2006-06-22 22:14 ` Andi Kleen
2006-06-22 23:10 ` Rohit Seth
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=200606210828_MC3-1-C30B-9D83@compuserve.com \
--to=76306.1226@compuserve.com \
--cc=ak@suse.de \
--cc=drepper@redhat.com \
--cc=jakub@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=roland@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;
as well as URLs for NNTP newsgroup(s).