Mirror of the lttng-dev mailing list
 help / color / mirror / Atom feed
From: mingo@elte.hu (Ingo Molnar)
Subject: [ltt-dev] [PATCH] Fix dirty page accounting in	redirty_page_for_writepage()
Date: Thu, 30 Apr 2009 08:21:40 +0200	[thread overview]
Message-ID: <20090430062140.GA9559@elte.hu> (raw)
In-Reply-To: <20090430024303.GB19875@Krystal>


* Mathieu Desnoyers <mathieu.desnoyers at polymtl.ca> wrote:

> And thanks for the review! This excercise only convinced me that 
> the kernel memory accounting works as expected. All this gave me 
> the chance to have a good look at the memory accounting code. We 
> could probably benefit of Christoph Lameter's cpu ops (using 
> segment registers to address per-cpu variables with atomic 
> inc/dec) in there. Or at least removing interrupt disabling by 
> using preempt disable and local_t variables for the per-cpu 
> counters could bring some benefit.

Note, optimized per cpu ops are already implemented upstream, by 
Tejun Heo's percpu patches in .30:

 #define percpu_read(var)	percpu_from_op("mov", per_cpu__##var)
 #define percpu_write(var, val)	percpu_to_op("mov", per_cpu__##var, val)
 #define percpu_add(var, val)	percpu_to_op("add", per_cpu__##var, val)
 #define percpu_sub(var, val)	percpu_to_op("sub", per_cpu__##var, val)
 #define percpu_and(var, val)	percpu_to_op("and", per_cpu__##var, val)
 #define percpu_or(var, val)	percpu_to_op("or", per_cpu__##var, val)
 #define percpu_xor(var, val)	percpu_to_op("xor", per_cpu__##var, val)

See:

  6dbde35: percpu: add optimized generic percpu accessors

From the changelog:

    [...]
    The advantage is that for example to read a local percpu variable,
    instead of this sequence:
    
     return __get_cpu_var(var);
    
     ffffffff8102ca2b:  48 8b 14 fd 80 09 74    mov    -0x7e8bf680(,%rdi,8),%rdx
     ffffffff8102ca32:  81
     ffffffff8102ca33:  48 c7 c0 d8 59 00 00    mov    $0x59d8,%rax
     ffffffff8102ca3a:  48 8b 04 10             mov    (%rax,%rdx,1),%rax
    
    We can get a single instruction by using the optimized variants:
    
     return percpu_read(var);
    
     ffffffff8102ca3f:  65 48 8b 05 91 8f fd    mov    %gs:0x7efd8f91(%rip),%rax
    [...]

So if you want to make use of it, percpu_add()/percpu_sub() would be 
the place to start.

	Ingo




  reply	other threads:[~2009-04-30  6:21 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-04-29 23:25 Mathieu Desnoyers
2009-04-29 23:56 ` Mathieu Desnoyers
2009-04-29 23:59 ` Andrew Morton
2009-04-30  2:34   ` Mathieu Desnoyers
2009-04-30  0:06 ` Linus Torvalds
2009-04-30  2:43   ` Mathieu Desnoyers
2009-04-30  6:21     ` Ingo Molnar [this message]
2009-04-30  6:33       ` Mathieu Desnoyers
2009-04-30  6:50         ` Ingo Molnar
2009-04-30 13:38           ` Christoph Lameter
2009-04-30 14:10             ` Ingo Molnar
2009-04-30 14:12             ` Mathieu Desnoyers
2009-04-30 14:12               ` Christoph Lameter
2009-04-30 19:41                 ` Mathieu Desnoyers
2009-04-30 20:17                   ` Christoph Lameter
2009-04-30 21:17                     ` Mathieu Desnoyers
2009-04-30 13:22     ` Christoph Lameter
2009-04-30 13:38       ` Ingo Molnar
2009-04-30 13:40         ` Christoph Lameter
2009-04-30 14:14           ` Ingo Molnar
2009-04-30 14:15             ` Christoph Lameter
2009-04-30 14:38               ` Ingo Molnar
2009-04-30 14:45                 ` Christoph Lameter
2009-04-30 15:01                   ` Ingo Molnar
2009-04-30 15:25                     ` Christoph Lameter
2009-04-30 15:42                       ` Ingo Molnar
2009-04-30 15:44                         ` Christoph Lameter
2009-04-30 16:06                           ` Ingo Molnar
2009-04-30 16:11                             ` Christoph Lameter
2009-04-30 16:16                             ` Linus Torvalds
2009-04-30 17:23                               ` Ingo Molnar
2009-04-30 18:07                                 ` Christoph Lameter
2009-04-30 16:13                         ` Linus Torvalds
2009-04-30 15:54                       ` Ingo Molnar
2009-04-30 16:00                       ` Ingo Molnar
2009-04-30 16:08                         ` Christoph Lameter
2009-04-30 13:50         ` Mathieu Desnoyers
2009-04-30 13:55           ` Christoph Lameter
2009-04-30 14:32           ` Ingo Molnar
2009-04-30 14:42             ` Christoph Lameter
2009-04-30 14:59               ` Ingo Molnar
2009-04-30 16:03             ` Mathieu Desnoyers

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=20090430062140.GA9559@elte.hu \
    --to=mingo@elte.hu \
    /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