On Wed, 26 Sep 2007 20:31:02 +0100 (BST)
Hugh Dickins <[EMAIL PROTECTED]> wrote:
> Would that waste a little memory? I think not with SLUB,
> but perhaps with SLOB, which packs a little tighter.
>
maybe just depends on the amount of used anon_vma and page_mapping_info etc...
I don't think a
On Sat, 22 Sep 2007, KAMEZAWA Hiroyuki wrote:
> On Fri, 21 Sep 2007 18:02:47 +0100 (BST)
> Hugh Dickins <[EMAIL PROTECTED]> wrote:
>
> > Or should I now leave PG_swapcache as is,
> > given your designs on page->mapping?
> >
> will conflict with my idea ?
> ==
>
On Sat, 22 Sep 2007, KAMEZAWA Hiroyuki wrote:
On Fri, 21 Sep 2007 18:02:47 +0100 (BST)
Hugh Dickins [EMAIL PROTECTED] wrote:
Or should I now leave PG_swapcache as is,
given your designs on page-mapping?
will conflict with my idea ?
==
http://marc.info/?l=linux-mmm=118956492926821w=2
On Wed, 26 Sep 2007 20:31:02 +0100 (BST)
Hugh Dickins [EMAIL PROTECTED] wrote:
Would that waste a little memory? I think not with SLUB,
but perhaps with SLOB, which packs a little tighter.
maybe just depends on the amount of used anon_vma and page_mapping_info etc...
I don't think a system
On Fri, 21 Sep 2007 18:02:47 +0100 (BST)
Hugh Dickins <[EMAIL PROTECTED]> wrote:
> > 3. I want to *try* page->mapping overriding... store memory resource
> > controller's
> >information in page->mapping. By this, memory controller doesn't enlarge
> > sizeof
> >struct page. (works
On Fri, 21 Sep 2007, KAMEZAWA Hiroyuki wrote:
> On Thu, 20 Sep 2007 11:26:47 -0700 (PDT)
> Christoph Lameter <[EMAIL PROTECTED]> wrote:
> >
> > I am still a bit confused as to what the benefit of this is.
> >
> Honestly, I have 3 purposes, 2 for readability/clarificaton and 1 for my
> trial.
>
On Fri, 21 Sep 2007 13:48:28 +0200
Peter Zijlstra <[EMAIL PROTECTED]> wrote:
> > Followings are moved
> > * page_mapping() ... returns swapper_space or address_space a page
> > is on.
> > (from mm.h)
> > * page_index() ... returns position of a page in its
On Wed, 19 Sep 2007 16:43:08 +0900 KAMEZAWA Hiroyuki
<[EMAIL PROTECTED]> wrote:
> A clarification of page <-> fs interface (page cache).
>
> At first, each FS has to access to struct page->mapping directly.
> But it's not just pointer. (we use special 1bit enconding for anon.)
>
> Although
On Wed, 19 Sep 2007 16:43:08 +0900 KAMEZAWA Hiroyuki
[EMAIL PROTECTED] wrote:
A clarification of page - fs interface (page cache).
At first, each FS has to access to struct page-mapping directly.
But it's not just pointer. (we use special 1bit enconding for anon.)
Although there is
On Fri, 21 Sep 2007 13:48:28 +0200
Peter Zijlstra [EMAIL PROTECTED] wrote:
Followings are moved
* page_mapping() ... returns swapper_space or address_space a page
is on.
(from mm.h)
* page_index() ... returns position of a page in its inode
On Fri, 21 Sep 2007, KAMEZAWA Hiroyuki wrote:
On Thu, 20 Sep 2007 11:26:47 -0700 (PDT)
Christoph Lameter [EMAIL PROTECTED] wrote:
I am still a bit confused as to what the benefit of this is.
Honestly, I have 3 purposes, 2 for readability/clarificaton and 1 for my
trial.
1. Clarify
On Fri, 21 Sep 2007 18:02:47 +0100 (BST)
Hugh Dickins [EMAIL PROTECTED] wrote:
3. I want to *try* page-mapping overriding... store memory resource
controller's
information in page-mapping. By this, memory controller doesn't enlarge
sizeof
struct page. (works well in my small
On Thu, 20 Sep 2007 11:26:47 -0700 (PDT)
Christoph Lameter <[EMAIL PROTECTED]> wrote:
> On Wed, 19 Sep 2007, KAMEZAWA Hiroyuki wrote:
>
> > Any comments are welcome.
>
> I am still a bit confused as to what the benefit of this is.
>
Honestly, I have 3 purposes, 2 for readability/clarificaton
On Wed, 19 Sep 2007, KAMEZAWA Hiroyuki wrote:
> Any comments are welcome.
I am still a bit confused as to what the benefit of this is.
> Following functions are added
>
> * page_mapping_cache() ... returns address space if a page is page cache
> * page_mapping_anon() ... returns anon_vma if
On Wed, 19 Sep 2007, KAMEZAWA Hiroyuki wrote:
Any comments are welcome.
I am still a bit confused as to what the benefit of this is.
Following functions are added
* page_mapping_cache() ... returns address space if a page is page cache
* page_mapping_anon() ... returns anon_vma if a
On Thu, 20 Sep 2007 11:26:47 -0700 (PDT)
Christoph Lameter [EMAIL PROTECTED] wrote:
On Wed, 19 Sep 2007, KAMEZAWA Hiroyuki wrote:
Any comments are welcome.
I am still a bit confused as to what the benefit of this is.
Honestly, I have 3 purposes, 2 for readability/clarificaton and 1 for
Rebased to 2.6.23-rc6-mm1 and reflected comments.
Not so aggresive as previous version.
(I'm not in a hurry if -mm is busy.)
Any comments are welcome.
Thanks,
-Kame
==
A clarification of page <-> fs interface (page cache).
At first, each FS has to access to struct page->mapping directly.
But
Rebased to 2.6.23-rc6-mm1 and reflected comments.
Not so aggresive as previous version.
(I'm not in a hurry if -mm is busy.)
Any comments are welcome.
Thanks,
-Kame
==
A clarification of page - fs interface (page cache).
At first, each FS has to access to struct page-mapping directly.
But it's
18 matches
Mail list logo