removing clean mapped pages

removing clean mapped pages

[Raghu R. Arur]

 Hi,

   How does a frame, mapped to a disk file gets released to the free list.
I do not see any place in shrink_cache() nor in try_to_swap_out() such a
page getting released.
What am i missing over here ?

 Thanks,
 Raghu
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"aart@kvack.org"> aart@kvack.org </a>

Re: removing clean mapped pages

[Raghu R. Arur]

 I meant a clean mapped frame.

Thanks,
Raghu

On Tue, 12 Aug 2003, Raghu R. Arur wrote:

>  Hi,
>
>    How does a frame, mapped to a disk file gets released to the free list.
> I do not see any place in shrink_cache() nor in try_to_swap_out() such a
> page getting released.
> What am i missing over here ?
>
>  Thanks,
>  Raghu
> --
> To unsubscribe, send a message with 'unsubscribe linux-mm' in
> the body to majordomo@kvack.org.  For more info on Linux MM,
> see: http://www.linux-mm.org/ .
> Don't email: <a href=mailto:"aart@kvack.org"> aart@kvack.org </a>
>
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"aart@kvack.org"> aart@kvack.org </a>

Re: removing clean mapped pages

[Mel Gorman]

On Tue, 12 Aug 2003, Raghu R. Arur wrote:

>
>  I meant a clean mapped frame.
>

Freed in try_to_swap_out() . There is no need to do anything with a clean
page, so it is just dropped from the page tables and page_cache_release()
is called. When the reference reaches 0, the page is reclaimed

-- 
Mel Gorman
http://www.csn.ul.ie/~mel
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"aart@kvack.org"> aart@kvack.org </a>

Re: removing clean mapped pages

[Mel Gorman]

On Wed, 13 Aug 2003, Mel Gorman wrote:

> Freed in try_to_swap_out() . There is no need to do anything with a clean
> page, so it is just dropped from the page tables and page_cache_release()
> is called. When the reference reaches 0, the page is reclaimed
>

Bah, this is wrong, shouldn't be let near e-mail in the morning. The
freeing of a file-mapped page is actually two-stage.

try_to_swap_out() will unmap the file-backed page from a process page
table. Once there are no processes mapping the page, the only user in
page->count will be the page cache. It stays in the page cache until it is
reclaimed later by shrink_cache().

-- 
Mel Gorman
http://www.csn.ul.ie/~mel
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"aart@kvack.org"> aart@kvack.org </a>