From: Jeremy Fitzhardinge <jeremy@goop.org>
To: Rusty Russell <rusty@rustcorp.com.au>
Cc: Ingo Molnar <mingo@elte.hu>,
lkml - Kernel Mailing List <linux-kernel@vger.kernel.org>,
Zachary Amsden <zach@vmware.com>,
Jeremy Fitzhardinge <jeremy@xensource.com>,
Andrew Morton <akpm@linux-foundation.org>,
Andi Kleen <ak@suse.de>
Subject: Re: [PATCH 8/8] Convert PDA into the percpu section
Date: Tue, 06 Mar 2007 16:35:06 -0800 [thread overview]
Message-ID: <45EE08BA.6080309@goop.org> (raw)
In-Reply-To: <1173226350.4644.54.camel@localhost.localdomain>
Rusty Russell wrote:
> If we used __thread, then gcc could do this optimization for us when it
> knows an rvalue is needed, however:
>
> 1) gcc wants to use %gs, not %fs, which is measurably slower for the
> kernel,
> 2) gcc wants to use huge offsets to store the address of the per-cpu
> space, and this breaks Xen (and current lguest, but new lguest no longer
> uses segments for protection)
>
Well, if we go to the effort of teaching gcc how to use %fs, we can
probably convince it to generate positive offset TLS relocs too.
J
next prev parent reply other threads:[~2007-03-07 0:35 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-03-06 12:39 [PATCH 0/8] x86 boot, pda and gdt cleanups Rusty Russell
2007-03-06 12:53 ` [PATCH 1/8] Remove cpu_gdt_table: use boot_gdt_table until migration to per-cpu Rusty Russell
2007-03-06 12:54 ` [PATCH 2/8] Remove NR_CPUS from asm-generic/percpu.h Rusty Russell
2007-03-06 12:55 ` [PATCH 3/8] Use per-cpu variables for GDT, PDA Rusty Russell
2007-03-06 12:57 ` [PATCH 4/8] Cleanup setup_pda Rusty Russell
2007-03-06 12:58 ` [PATCH 5/8] Cleanup GDT access Rusty Russell
2007-03-06 13:00 ` [PATCH 6/8] Allow per-cpu variables to be page-aligned Rusty Russell
2007-03-06 13:01 ` [PATCH 7/8] Page-align the GDT Rusty Russell
2007-03-06 13:03 ` [PATCH 8/8] Convert PDA into the percpu section Rusty Russell
2007-03-06 13:10 ` Ingo Molnar
2007-03-07 0:12 ` Rusty Russell
2007-03-07 0:35 ` Jeremy Fitzhardinge [this message]
2007-03-06 18:28 ` Jeremy Fitzhardinge
2007-03-06 19:34 ` Andi Kleen
2007-03-06 18:37 ` Jeremy Fitzhardinge
2007-03-07 0:33 ` Rusty Russell
2007-03-07 11:55 ` Rusty Russell
2007-03-13 17:15 ` Jeremy Fitzhardinge
2007-03-14 2:27 ` Rusty Russell
2007-03-06 13:15 ` [PATCH 6/8] Allow per-cpu variables to be page-aligned Ingo Molnar
2007-03-07 0:16 ` Rusty Russell
2007-03-07 0:44 ` H. Peter Anvin
2007-03-06 18:17 ` Jeremy Fitzhardinge
2007-03-07 0:29 ` Rusty Russell
2007-03-06 18:16 ` [PATCH 5/8] Cleanup GDT access Jeremy Fitzhardinge
2007-03-06 18:14 ` [PATCH 3/8] Use per-cpu variables for GDT, PDA Jeremy Fitzhardinge
2007-03-06 13:21 ` [PATCH 2/8] Remove NR_CPUS from asm-generic/percpu.h Ingo Molnar
2007-03-06 13:20 ` [PATCH 1/8] Remove cpu_gdt_table: use boot_gdt_table until migration to per-cpu Ingo Molnar
2007-03-06 13:26 ` Ingo Molnar
2007-03-07 0:22 ` Rusty Russell
2007-03-13 20:48 ` [PATCH 0/8] x86 boot, pda and gdt cleanups Jeremy Fitzhardinge
2007-03-14 2:25 ` Rusty Russell
2007-03-14 4:39 ` Jeremy Fitzhardinge
2007-03-14 6:54 ` Rusty Russell
2007-03-14 23:55 ` Rusty Russell
2007-03-15 1:57 ` Jeremy Fitzhardinge
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=45EE08BA.6080309@goop.org \
--to=jeremy@goop.org \
--cc=ak@suse.de \
--cc=akpm@linux-foundation.org \
--cc=jeremy@xensource.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=rusty@rustcorp.com.au \
--cc=zach@vmware.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;
as well as URLs for NNTP newsgroup(s).