virtualization.lists.linux-foundation.org archive mirror
 help / color / mirror / Atom feed
From: Rusty Russell <rusty@rustcorp.com.au>
To: Jeremy Fitzhardinge <jeremy@goop.org>
Cc: Andi Kleen <ak@muc.de>,
	lkml - Kernel Mailing List <linux-kernel@vger.kernel.org>,
	virtualization <virtualization@lists.osdl.org>
Subject: Re: [PATCH 5/7] Use %gs for per-cpu sections in kernel
Date: Mon, 25 Sep 2006 12:51:16 +1000	[thread overview]
Message-ID: <1159152678.26986.38.camel@localhost.localdomain> (raw)
In-Reply-To: <45173287.8070204@goop.org>

On Sun, 2006-09-24 at 18:36 -0700, Jeremy Fitzhardinge wrote:
> Rusty Russell wrote:
> > 	You're thinking of it in a convoluted way, by converting to offsets
> > from the per-cpu section, then converting it back.  How about this
> > explanation: the local cpu's versions are offset from where the compiler
> > thinks they are by __per_cpu_offset[cpu].  We set the segment base to
> > __per_cpu_offset[cpu], so "%gs:per_cpu__foo" gets us straight to the
> > local cpu version.  __per_cpu_offset[cpu] is always positive (kernel
> > image sits at bottom of kernel address space).
> >   
> 
> We're talking kernel virtual addresses, so the physical load address 
> doesn't matter, of course.
> 
> So, take this kernel I have here as an explicit example:
> 
> $ nm -n vmlinux
> [...]
> c0431100 A __per_cpu_start
> [...]
> c0433800 D per_cpu__cpu_gdt_descr
> c0433880 D per_cpu__cpu_tlbstate
> 
> 
> And say that this CPU has its percpu data allocated at 0xc100000.

That can't happen, since 0xc100000 is not in the kernel address space.
0xc1000000 is though, perhaps that's what you meant?

> So, in this case the %gs base will be loaded with 0xc100000-0xc0431100 = 
> 0x4bccef00

A negative offset, exactly, which can't happen, as I said.

Hope that clarifies?

Confused,
Rusty.
-- 
Help! Save Australia from the worst of the DMCA: http://linux.org.au/law

  reply	other threads:[~2006-09-25  2:51 UTC|newest]

Thread overview: 32+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-09-22 11:51 [PATCH 0/7] Using %gs for per-cpu areas on x86 Rusty Russell
2006-09-22 11:53 ` [PATCH 1/7] Use per-cpu GDT tables from early in boot Rusty Russell
2006-09-22 11:55   ` [PATCH 2/7] Rusty Russell
2006-09-22 11:56     ` [PATCH 3/7] Update sys_vm86 to cope with changed pt_regs and %gs usage Rusty Russell
2006-09-22 11:58       ` [PATCH 4/7] Fix places where using %gs changes the usermode ABI Rusty Russell
2006-09-22 11:59         ` [PATCH 5/7] Use %gs for per-cpu sections in kernel Rusty Russell
2006-09-22 12:00           ` [PATCH 6/7] (Optional) implement smp_processor_id() as a per-cpu var Rusty Russell
2006-09-22 12:01             ` [PATCH 7/7] (Optional) implement current " Rusty Russell
2006-09-25  5:29               ` Rusty Russell
2006-09-25  5:27             ` [PATCH 6/7] (Optional) implement smp_processor_id() " Rusty Russell
2006-09-22 12:32           ` [PATCH 5/7] Use %gs for per-cpu sections in kernel Andi Kleen
2006-09-22 22:43             ` Jeremy Fitzhardinge
2006-09-22 23:52               ` Andi Kleen
2006-09-23  4:51             ` Rusty Russell
2006-09-23  8:17               ` Andi Kleen
2006-09-23  8:55                 ` Rusty Russell
2006-09-22 22:39           ` Jeremy Fitzhardinge
2006-09-23  4:31             ` Rusty Russell
2006-09-25  1:03               ` Jeremy Fitzhardinge
2006-09-25  1:16                 ` Rusty Russell
2006-09-25  1:36                   ` Jeremy Fitzhardinge
2006-09-25  2:51                     ` Rusty Russell [this message]
2006-09-25  5:25                       ` Jeremy Fitzhardinge
2006-09-25  6:03                         ` Rusty Russell
2006-09-25  6:25                           ` Jeremy Fitzhardinge
2006-09-25 23:33                             ` Rusty Russell
2006-09-23  8:13             ` Andi Kleen
2006-09-25  1:07               ` Jeremy Fitzhardinge
2006-09-25  1:20                 ` Rusty Russell
2006-09-25  5:26                   ` Rusty Russell
2006-09-22 22:24     ` [PATCH 2/7] Jeremy Fitzhardinge
2006-09-23  4:36       ` Rusty Russell

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=1159152678.26986.38.camel@localhost.localdomain \
    --to=rusty@rustcorp.com.au \
    --cc=ak@muc.de \
    --cc=jeremy@goop.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=virtualization@lists.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).