Re: [PATCH 01/14] page-replace-single-batch-insert.patch

From: Peter Zijlstra <a.p.zijlstra@xxxxxxxxx>
Date: Sat, 31 Dec 2005 10:43:30 +0100

On Sat, 2005-12-31 at 05:03 -0200, Marcelo Tosatti wrote:
> Hi Peter,
>
> On Fri, Dec 30, 2005 at 11:40:24PM +0100, Peter Zijlstra wrote:
> >
> > From: Peter Zijlstra <a.p.zijlstra@xxxxxxxxx>
> >
> > page-replace interface function:
> > __page_replace_insert()
> >
> > This function inserts a page into the page replace data structure.
> >
> > Unify the active and inactive per cpu page lists. For now provide insertion
> > hints using the LRU specific page flags.
>
> Unification of active and inactive per cpu page lists is a requirement
> for CLOCK-Pro, right?
>
> Would be nicer to have unchanged functionality from vanilla VM
> (including the active/inactive per cpu lists).

I guess I could pull all that pcp stuff into page_replace if that would
make you happy ;-)

> Happy new year!

Best wishes!

--
Peter Zijlstra <a.p.zijlstra@xxxxxxxxx>

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

References:
- [PATCH] vm: page-replace and clockpro
  - From: Peter Zijlstra
- [PATCH 01/14] page-replace-single-batch-insert.patch
  - From: Peter Zijlstra
- Re: [PATCH 01/14] page-replace-single-batch-insert.patch
  - From: Marcelo Tosatti

Prev by Date: Re: [PATCH] Fix user data corrupted by old value return of sysctl
Next by Date: Re: libata support for JMicron JMB360 ?
Previous by thread: Re: [PATCH 01/14] page-replace-single-batch-insert.patch
Next by thread: Re: [PATCH 01/14] page-replace-single-batch-insert.patch
Index(es):
- Date
- Thread

Relevant Pages

Re: [PATCH 01/14] page-replace-single-batch-insert.patch
... > This function inserts a page into the page replace data structure. ...
> Unify the active and inactive per cpu page lists. ... Unification of active and inactive
per cpu page lists is a requirement ... (Linux-Kernel)
Re: very strange pthread problem
... The worker threads need access to a class member function or two. ... STL vector
so each index into the vector points to the list of pointers ... it's just storing the
pointers to the STL lists and the vector ... Then I moved the whole thing to a single
cpu Linux ... (comp.programming.threads)
Re: very strange pthread problem (solved)
... The worker threads need access to a class member function or two. ... STL vector
so each index into the vector points to the list of pointers ... it's just storing the
pointers to the STL lists and the vector ... Then I moved the whole thing to a single
cpu Linux ... (comp.programming.threads)
very strange pthread problem
... The worker threads need access to a class member function or two. ... changed
back to local variables easily by recompiling ... Then I moved the whole thing to a single
cpu Linux ... the STL lists via push_back but there's a unique ... (comp.programming.threads)
Re: [REPORT] cfs-v4 vs sd-0.44
... X should get more CPU time simply because it ... This would be seen like some
CPU ressource buckets. ... struct task_struct { ... Maybe we need two lists
in order of the structures to be unstacked ... (Linux-Kernel)