Patchwork ext4: page_cache pages over pinned in ext4_mb_free_metadata()

login
register
mail settings
Submitter jing zhang
Date March 24, 2010, 1 p.m.
Message ID <ac8f92701003240600o5ce6a682i431de77bef66b736@mail.gmail.com>
Download mbox | patch
Permalink /patch/48404/
State New
Headers show

Comments

jing zhang - March 24, 2010, 1 p.m.
From: Jing Zhang <zj.barak@gmail.com>

Date: Wed Mar 24 20:38:48     2010

The function, ext4_mb_free_metadata(), is called after
     ext4_mb_load_buddy(sb, block_group, &e4b);
and before
     ext4_mb_release_desc(&e4b);
in the function, ext4_mb_free_blocks(), so the bd_bitmap_page and
bd_buddy_page of e4b are well managed.

If for special purpose, page_cache_get() is issued on these two pages,
there should also be page_cache_release() issued correspondingly, or
the two pages are over pinned.

Cc: Theodore Ts'o <tytso@mit.edu>
Cc: Andreas Dilger <adilger@sun.com>
Cc: Dave Kleikamp <shaggy@linux.vnet.ibm.com>
Signed-off-by: Jing Zhang <zj.barak@gmail.com>

---

--
To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Aneesh Kumar K.V - March 24, 2010, 5:16 p.m.
On Wed, 24 Mar 2010 21:00:34 +0800, jing zhang <zj.barak@gmail.com> wrote:
> From: Jing Zhang <zj.barak@gmail.com>
> 
> Date: Wed Mar 24 20:38:48     2010
> 
> The function, ext4_mb_free_metadata(), is called after
>      ext4_mb_load_buddy(sb, block_group, &e4b);
> and before
>      ext4_mb_release_desc(&e4b);
> in the function, ext4_mb_free_blocks(), so the bd_bitmap_page and
> bd_buddy_page of e4b are well managed.
> 
> If for special purpose, page_cache_get() is issued on these two pages,
> there should also be page_cache_release() issued correspondingly, or
> the two pages are over pinned.

release_blocks_on_commit does that. I also have a comment around that page_cache_release.

> 
> Cc: Theodore Ts'o <tytso@mit.edu>
> Cc: Andreas Dilger <adilger@sun.com>
> Cc: Dave Kleikamp <shaggy@linux.vnet.ibm.com>
> Signed-off-by: Jing Zhang <zj.barak@gmail.com>
> 
> ---
> 
> --- linux-2.6.32/fs/ext4/mballoc.c	2009-12-03 11:51:22.000000000 +0800
> +++ ext4_mm_leak/mballoc6.c	2010-03-24 20:31:52.000000000 +0800
> @@ -4361,15 +4361,6 @@ ext4_mb_free_metadata(handle_t *handle,
>  	new_node = &new_entry->node;
>  	block = new_entry->start_blk;
> 
> -	if (!*n) {
> -		/* first free block exent. We need to
> -		   protect buddy cache from being freed,
> -		 * otherwise we'll refresh it from
> -		 * on-disk bitmap and lose not-yet-available
> -		 * blocks */
> -		page_cache_get(e4b->bd_buddy_page);
> -		page_cache_get(e4b->bd_bitmap_page);
> -	}
>  	while (*n) {
>  		parent = *n;
>  		entry = rb_entry(parent, struct ext4_free_data, node);


-aneesh
--
To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
jing zhang - March 25, 2010, 2:23 p.m.
2010/3/25, Aneesh Kumar K. V <aneesh.kumar@linux.vnet.ibm.com>:
> On Wed, 24 Mar 2010 21:00:34 +0800, jing zhang <zj.barak@gmail.com> wrote:
>> From: Jing Zhang <zj.barak@gmail.com>
>>
>> Date: Wed Mar 24 20:38:48     2010
>>
>> The function, ext4_mb_free_metadata(), is called after
>>      ext4_mb_load_buddy(sb, block_group, &e4b);
>> and before
>>      ext4_mb_release_desc(&e4b);
>> in the function, ext4_mb_free_blocks(), so the bd_bitmap_page and
>> bd_buddy_page of e4b are well managed.
>>
>> If for special purpose, page_cache_get() is issued on these two pages,
>> there should also be page_cache_release() issued correspondingly, or
>> the two pages are over pinned.
>
> release_blocks_on_commit does that. I also have a comment around that
> page_cache_release.

Thank you, Aneesh. I will check it.

      - zj

>
>>
>> Cc: Theodore Ts'o <tytso@mit.edu>
>> Cc: Andreas Dilger <adilger@sun.com>
>> Cc: Dave Kleikamp <shaggy@linux.vnet.ibm.com>
>> Signed-off-by: Jing Zhang <zj.barak@gmail.com>
>>
>> ---
>>
>> --- linux-2.6.32/fs/ext4/mballoc.c	2009-12-03 11:51:22.000000000 +0800
>> +++ ext4_mm_leak/mballoc6.c	2010-03-24 20:31:52.000000000 +0800
>> @@ -4361,15 +4361,6 @@ ext4_mb_free_metadata(handle_t *handle,
>>  	new_node = &new_entry->node;
>>  	block = new_entry->start_blk;
>>
>> -	if (!*n) {
>> -		/* first free block exent. We need to
>> -		   protect buddy cache from being freed,
>> -		 * otherwise we'll refresh it from
>> -		 * on-disk bitmap and lose not-yet-available
>> -		 * blocks */
>> -		page_cache_get(e4b->bd_buddy_page);
>> -		page_cache_get(e4b->bd_bitmap_page);
>> -	}
>>  	while (*n) {
>>  		parent = *n;
>>  		entry = rb_entry(parent, struct ext4_free_data, node);
>
>
> -aneesh
>
--
To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Patch

--- linux-2.6.32/fs/ext4/mballoc.c	2009-12-03 11:51:22.000000000 +0800
+++ ext4_mm_leak/mballoc6.c	2010-03-24 20:31:52.000000000 +0800
@@ -4361,15 +4361,6 @@  ext4_mb_free_metadata(handle_t *handle,
 	new_node = &new_entry->node;
 	block = new_entry->start_blk;

-	if (!*n) {
-		/* first free block exent. We need to
-		   protect buddy cache from being freed,
-		 * otherwise we'll refresh it from
-		 * on-disk bitmap and lose not-yet-available
-		 * blocks */
-		page_cache_get(e4b->bd_buddy_page);
-		page_cache_get(e4b->bd_bitmap_page);
-	}
 	while (*n) {
 		parent = *n;
 		entry = rb_entry(parent, struct ext4_free_data, node);