[PATCH mmotm] memcg: avoid using buggy kmap at swap_cgroup

["KAMEZAWA Hiroyuki" <kamezawa.hiroyu@jp.fujitsu.com>]

[-- Attachment #1: Type: text/plain, Size: 912 bytes --]

Daisuke Nishimura said:
> On Tue, 18 Nov 2008 12:48:52 +0000 (GMT)
> Hugh Dickins <hugh@veritas.com> wrote:
>> That's a lot of files which you may not wish to update to get working
>> right now: I think page_cgroup.c can _probably_ reuse KM_PTE1 as a
>> temporary measure, but please verify that's safe first.
>>
> Thank you for your comment.
>
> Hmm, shmem_map_and_free_swp might unmap dir anyway and caller
> (shmem_trancate_range) handles the case, but I do agree it's
> not good manner to unmap other people's kmaps.
>
Hmm...Sorry for my original implementation.
Okay, how about this direction ?
 1. at first, remove kmap_atomic from page_cgroup.c and use GFP_KERNEL
    to allocate buffer.
 2. later, add kmap_atomic + HighMem buffer support in explicit style.
    maybe KM_BOUNCE_READ...can be used.....

patch for BUGFIX is attached.
(Sorry, I have to use Web-Mail and can't make it inlined)

Sorry,
-Kame

[-- Attachment #2: swapcg-kmap-fix.patch --]
[-- Type: application/octet-stream, Size: 1418 bytes --]

swap_cgroup's kmap logic conflicts shmem's kmap logic.
avoid to use HIGHMEM for now and revisit this later.

Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>

 mm/page_cgroup.c |    8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

Index: temp/mm/page_cgroup.c
===================================================================
--- temp.orig/mm/page_cgroup.c
+++ temp/mm/page_cgroup.c
@@ -306,7 +306,7 @@ static int swap_cgroup_prepare(int type)
 	ctrl = &swap_cgroup_ctrl[type];
 
 	for (idx = 0; idx < ctrl->length; idx++) {
-		page = alloc_page(GFP_KERNEL | __GFP_HIGHMEM | __GFP_ZERO);
+		page = alloc_page(GFP_KERNEL | __GFP_ZERO);
 		if (!page)
 			goto not_enough_page;
 		ctrl->map[idx] = page;
@@ -347,11 +347,10 @@ struct mem_cgroup *swap_cgroup_record(sw
 
 	mappage = ctrl->map[idx];
 	spin_lock_irqsave(&ctrl->lock, flags);
-	sc = kmap_atomic(mappage, KM_USER0);
+	sc = page_address(mappage);
 	sc += pos;
 	old = sc->val;
 	sc->val = mem;
-	kunmap_atomic((void *)sc, KM_USER0);
 	spin_unlock_irqrestore(&ctrl->lock, flags);
 	return old;
 }
@@ -382,10 +381,9 @@ struct mem_cgroup *lookup_swap_cgroup(sw
 	mappage = ctrl->map[idx];
 
 	spin_lock_irqsave(&ctrl->lock, flags);
-	sc = kmap_atomic(mappage, KM_USER0);
+	sc = page_address(mappage);
 	sc += pos;
 	ret = sc->val;
-	kunmap_atomic((void *)sc, KM_USER0);
 	spin_unlock_irqrestore(&ctrl->lock, flags);
 	return ret;
 }