lkml.org 
[lkml]   [2011]   [May]   [31]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRE: [PATCH 1/14] mm: invalidate_mapping_pages flush cleancache
> From: Hugh Dickins [mailto:hughd@google.com]
> Sent: Monday, May 30, 2011 6:36 PM
> To: Andrew Morton
> Cc: Dan Magenheimer; linux-kernel@vger.kernel.org; linux-mm@kvack.org
> Subject: [PATCH 1/14] mm: invalidate_mapping_pages flush cleancache
>
> truncate_inode_pages_range() and invalidate_inode_pages2_range()
> call cleancache_flush_inode(mapping) before and after: shouldn't
> invalidate_mapping_pages() be doing the same?
>
> Signed-off-by: Hugh Dickins <hughd@google.com>
> Cc: Dan Magenheimer <dan.magenheimer@oracle.com>
> ---
> mm/truncate.c | 2 ++
> 1 file changed, 2 insertions(+)
>
> --- linux.orig/mm/truncate.c 2011-05-30 13:56:10.416798124 -0700
> +++ linux/mm/truncate.c 2011-05-30 14:08:46.612547848 -0700
> @@ -333,6 +333,7 @@ unsigned long invalidate_mapping_pages(s
> unsigned long count = 0;
> int i;
>
> + cleancache_flush_inode(mapping);
> pagevec_init(&pvec, 0);
> while (next <= end &&
> pagevec_lookup(&pvec, mapping, next, PAGEVEC_SIZE)) {
> @@ -373,6 +374,7 @@ unsigned long invalidate_mapping_pages(s
> mem_cgroup_uncharge_end();
> cond_resched();
> }
> + cleancache_flush_inode(mapping);
> return count;
> }
> EXPORT_SYMBOL(invalidate_mapping_pages);

Hi Hugh --

I don't claim to be an expert on VFS, and so I have cc'ed
Chris Mason who originally placed the cleancache hooks
in VFS, but I think this patch is unnecessary. Instead
of flushing ALL of the cleancache pages belonging to
the inode with cleancache_flush_inode, the existing code
eventually calls __delete_from_page_cache on EACH page
that is being invalidated. And since __delete_from_page_cache
calls cleancache_flush_page, only that subset of pages
in the mapping that invalidate_mapping_pages() would
invalidate (which, from the comment above the routine
indicates, is only *unlocked* pages) is removed from
cleancache.

However, there may be some path through VFS I am missing
or something else subtle I am misunderstanding so please
clarify if either is true.

Thanks,
Dan
---
Thanks... for the memory!
I really could use more / my throughput's on the floor
The balloon is flat / my swap disk's fat / I've OOM's in store
Overcommitted so much
(with apologies to Bob Hope)


\
 
 \ /
  Last update: 2011-05-31 17:57    [from the cache]
©2003-2011 Jasper Spaans