lkml.org 
[lkml]   [2010]   [Jul]   [6]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH]shmem: reduce one time of locking in pagefault
On Wed, Jul 07, 2010 at 09:32:54AM +0800, Andrew Morton wrote:
> On Wed, 07 Jul 2010 09:15:46 +0800 Shaohua Li <shaohua.li@intel.com> wrote:
>
> > I'm running a shmem pagefault test case (see attached file) under a 64 CPU
> > system. Profile shows shmem_inode_info->lock is heavily contented and 100%
> > CPUs time are trying to get the lock.
>
> I seem to remember complaining about that in 2002 ;) Faulting in a
> mapping of /dev/zero is just awful on a 4-way(!).
>
> > In the pagefault (no swap) case,
> > shmem_getpage gets the lock twice, the last one is avoidable if we prealloc a
> > page so we could reduce one time of locking. This is what below patch does.
> >
> > The result of the test case:
> > 2.6.35-rc3: ~20s
> > 2.6.35-rc3 + patch: ~12s
> > so this is 40% improvement.
> >
> > One might argue if we could have better locking for shmem. But even shmem is lockless,
> > the pagefault will soon have pagecache lock heavily contented because shmem must add
> > new page to pagecache. So before we have better locking for pagecache, improving shmem
> > locking doesn't have too much improvement. I did a similar pagefault test against
> > a ramfs file, the test result is ~10.5s.
> >
> > Signed-off-by: Shaohua Li <shaohua.li@intel.com>
> >
> > diff --git a/mm/shmem.c b/mm/shmem.c
> > index f65f840..c5f2939 100644
> > --- a/mm/shmem.c
> > +++ b/mm/shmem.c
>
> The patch doesn't make shmem_getpage() any clearer :(
>
> shmem_inode_info.lock appears to be held too much. Surely
> lookup_swap_cache() didn't need it (for example).
>
> What data does shmem_inode_info.lock actually protect?
As far as my understanding, it protects shmem swp_entry, which is most used
to support swap. It also protects some accounting. If no swap, the lock almost
can be removed like tiny-shmem.

Thanks,
Shaohua


\
 
 \ /
  Last update: 2010-07-07 03:41    [W:0.744 / U:0.020 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site