Re: [PATCH 2.6.28 1/2] memory: improve find_vma

[Ingo Molnar <mingo@elte.hu>]

* Daniel Lowengrub <lowdanie@gmail.com> wrote:

> On Thu, Jan 22, 2009 at 7:22 PM, Hugh Dickins <hugh@veritas.com> wrote:
> > Do you have some performance figures to support this patch?
> > Some of the lmbench tests may be appropriate.
> >
> > The thing is, expanding vm_area_struct to include another pointer
> > will have its own cost, which may well outweigh the efficiency
> > (in one particular case) which you're adding.  Expanding mm_struct
> > for this would be much more palatable, but I don't think that flies.
> >
> > And it seems a little greedy to require both an rbtree and a doubly
> > linked list for working our way around the vmas.
> >
> > I suspect that originally your enhancement would only have hit when
> > extending the stack: which I guess isn't enough to justify the cost.
> > But it could well be that unmapped area handling has grown more
> > complex down the years, and you get some hits now from that.
> >
> Thanks for the reply.
> I ran an lmbench test on the 2.6.28 kernel and on the same kernel
> after applying the patch.  Here's a portion of the results with the
> format of
> test : standard kernel / kernel after patch
> 
> Simple syscall: 0.7419 / 0.4244 microseconds
> Simple read: 1.2071 / 0.7270 microseconds

there must be a significant measurement mistake here: none of your patches 
affect the 'simple syscall' path, nor the sys_read() path.

	Ingo