All of lore.kernel.org
 help / color / mirror / Atom feed
From: Andrew Morton <akpm@digeo.com>
To: William Lee Irwin III <wli@holomorphy.com>
Cc: lkml <linux-kernel@vger.kernel.org>,
	"linux-mm@kvack.org" <linux-mm@kvack.org>
Subject: Re: [patch] remove page->virtual
Date: Thu, 19 Sep 2002 22:41:44 -0700	[thread overview]
Message-ID: <3D8AB518.218F8503@digeo.com> (raw)
In-Reply-To: 20020920050320.GH3530@holomorphy.com

William Lee Irwin III wrote:
> 
> On Thu, Sep 19, 2002 at 09:55:52PM -0700, Andrew Morton wrote:
> > set_page_address() and page_address() implementations consume 0.4% and
> > 1.3% of CPU time respectively.   I think that's OK. (Plus the tested code
> > was doing an unneeded lookup in set_page_address(), for debug purposes)
> 
> Looks yummy. I'll take it for a spin tonight on my benchmark-o-matic.
> Clearing some more air in ZONE_NORMAL is always welcome here.

Ta.

> On Thu, Sep 19, 2002 at 09:55:52PM -0700, Andrew Morton wrote:
> > c01884f2 6914     10.5108     .text.lock.dir
> > c01546b3 5847     8.88872     .text.lock.namei
> > c01eb99e 3811     5.79355     .text.lock.dec_and_lock
> > c01515dc 3775     5.73883     link_path_walk
> > c015207c 3567     5.42262     path_lookup
> > c015aba4 3194     4.85558     __d_lookup
> > c01eb690 2814     4.2779      __generic_copy_to_user
> > c01eb950 2562     3.8948      atomic_dec_and_lock
> > c0187580 2473     3.7595      ext2_readdir
> > c0155d3c 2172     3.30192     filldir64
> > c0151114 1786     2.71511     path_release
> > c0145b6a 1753     2.66494     .text.lock.open
> 
> What's going on here? fs stuff is really hurting. At any rate, the
> overhead of the address calculation and hashtable lookup is microscopic
> according to this, and I want space.


c0182f90 5949     14.5552     ext2_readdir

           lock_kernel()         

c014f65c 5790     14.1662     path_lookup            

    Confused.  oprofile claims read_lock(&current->fs->lock) but
    it's surely dcache_lock.

c01e6e80 4275     10.4595     atomic_dec_and_lock     

    dput/iput

c014eba4 2264     5.53924     link_path_walk          
c0158050 1988     4.86397     __d_lookup              
c01426e8 1903     4.656       sys_chdir               
c01e6bc0 1735     4.24496     __generic_copy_to_user  
c015319c 1344     3.28831     filldir64               
c014e6b4 1205     2.94823     path_release            
c0109070 1065     2.6057      system_call             
c014b934 929      2.27295     cp_new_stat64           
c014b380 822      2.01116     generic_fillattr        
c0157250 712      1.74202     dput                    
c014b3fc 622      1.52182     vfs_getattr             
c0157468 556      1.36034     dget_locked

WARNING: multiple messages have this Message-ID (diff)
From: Andrew Morton <akpm@digeo.com>
To: William Lee Irwin III <wli@holomorphy.com>
Cc: lkml <linux-kernel@vger.kernel.org>,
	"linux-mm@kvack.org" <linux-mm@kvack.org>
Subject: Re: [patch] remove page->virtual
Date: Thu, 19 Sep 2002 22:41:44 -0700	[thread overview]
Message-ID: <3D8AB518.218F8503@digeo.com> (raw)
In-Reply-To: 20020920050320.GH3530@holomorphy.com

William Lee Irwin III wrote:
> 
> On Thu, Sep 19, 2002 at 09:55:52PM -0700, Andrew Morton wrote:
> > set_page_address() and page_address() implementations consume 0.4% and
> > 1.3% of CPU time respectively.   I think that's OK. (Plus the tested code
> > was doing an unneeded lookup in set_page_address(), for debug purposes)
> 
> Looks yummy. I'll take it for a spin tonight on my benchmark-o-matic.
> Clearing some more air in ZONE_NORMAL is always welcome here.

Ta.

> On Thu, Sep 19, 2002 at 09:55:52PM -0700, Andrew Morton wrote:
> > c01884f2 6914     10.5108     .text.lock.dir
> > c01546b3 5847     8.88872     .text.lock.namei
> > c01eb99e 3811     5.79355     .text.lock.dec_and_lock
> > c01515dc 3775     5.73883     link_path_walk
> > c015207c 3567     5.42262     path_lookup
> > c015aba4 3194     4.85558     __d_lookup
> > c01eb690 2814     4.2779      __generic_copy_to_user
> > c01eb950 2562     3.8948      atomic_dec_and_lock
> > c0187580 2473     3.7595      ext2_readdir
> > c0155d3c 2172     3.30192     filldir64
> > c0151114 1786     2.71511     path_release
> > c0145b6a 1753     2.66494     .text.lock.open
> 
> What's going on here? fs stuff is really hurting. At any rate, the
> overhead of the address calculation and hashtable lookup is microscopic
> according to this, and I want space.


c0182f90 5949     14.5552     ext2_readdir

           lock_kernel()         

c014f65c 5790     14.1662     path_lookup            

    Confused.  oprofile claims read_lock(&current->fs->lock) but
    it's surely dcache_lock.

c01e6e80 4275     10.4595     atomic_dec_and_lock     

    dput/iput

c014eba4 2264     5.53924     link_path_walk          
c0158050 1988     4.86397     __d_lookup              
c01426e8 1903     4.656       sys_chdir               
c01e6bc0 1735     4.24496     __generic_copy_to_user  
c015319c 1344     3.28831     filldir64               
c014e6b4 1205     2.94823     path_release            
c0109070 1065     2.6057      system_call             
c014b934 929      2.27295     cp_new_stat64           
c014b380 822      2.01116     generic_fillattr        
c0157250 712      1.74202     dput                    
c014b3fc 622      1.52182     vfs_getattr             
c0157468 556      1.36034     dget_locked
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/

  reply	other threads:[~2002-09-20  5:36 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2002-09-20  4:55 [patch] remove page->virtual Andrew Morton
2002-09-20  4:55 ` Andrew Morton
2002-09-20  5:03 ` William Lee Irwin III
2002-09-20  5:03   ` William Lee Irwin III
2002-09-20  5:41   ` Andrew Morton [this message]
2002-09-20  5:41     ` Andrew Morton

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=3D8AB518.218F8503@digeo.com \
    --to=akpm@digeo.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=wli@holomorphy.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.