[lkml]   [2002]   [Mar]   [6]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: [PATCH] struct page shrinkage
Bulent Abali wrote:
> extern struct page_state {
> unsigned long nr_dirty;
> unsigned long nr_locked;
> } ____cacheline_aligned page_states[NR_CPUS];
> This is perfect. Looks like, if a run summation over all the CPUs I will
> get the total locked and dirty pages, provided mm.h macros are respected.

That's correct. And the mm.h macros *are* respected. That patch
ensures that they are.

It goes as far as to rename PG_locked and PG_dirty to PG_locked_dontuse
and PG_dirty_dontuse.

I'll be adding page_cache_size to the above struct, at least.

The "run summation" function is already there, btw: get_page_state().

> What is the outlook for inclusion of this patch in the main kernel? Do you
> plan to submit or have been included yet?

Well it's all a part of a work to aggressively improve the efficiency
of regular file I/O. I don't know if the big grand plan will be successful
yet. At this time, it's thumbs up - way up.

Nor do I know if this is a direction in which Linus wishes to take
his kernel.

But this change, the readahead changes, the pdflush pool and a few other
pieces I have planned are probably appropriate for the base kernel
irrespective of the end outcome.

We'll see...

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to
More majordomo info at
Please read the FAQ at

 \ /
  Last update: 2005-03-22 13:21    [W:0.047 / U:2.048 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site