Re: [PATCH 2.5.43-mm2] New shared page table patch

[Dave McCracken]

--On Tuesday, October 22, 2002 15:15:29 -0200 Rik van Riel
<[e-mail]> wrote:

>> Or large pages.  I confess to being a little perplexed as to
>> why we're pursuing both.
> 
> I guess that's due to two things.
> 
> 1) shared pagetables can speed up fork()+exec() somewhat
> 
> 2) if we have two options that fix the Oracle problem,
>    there's a better chance of getting at least one of
>    the two merged ;)

And
  3) The current large page implementation is only for applications
     that want anonymous *non-pageable* shared memory.  Shared page
     tables reduce resource usage for any shared area that's mapped
     at a common address and is large enough to span entire pte pages.
     Since all pte pages are shared on a COW basis at fork time, children
     will continue to share all large read-only areas with their
     parent, eg large executables.

Dave McCracken

======================================================================
Dave McCracken          IBM Linux Base Kernel Team      1-512-838-3059
[e-mail]                                        T/L   678-3059