Patchwork [v2] ext4: Fix a BUG in mb_mark_used during trim.

login
register
mail settings
Submitter Tao Ma
Date March 3, 2011, 1:59 p.m.
Message ID <1299160774-2337-1-git-send-email-tm@tao.ma>
Download mbox | patch
Permalink /patch/85288/
State Accepted
Headers show

Comments

Tao Ma - March 3, 2011, 1:59 p.m.
From: Tao Ma <boyu.mt@taobao.com>

In a bs=4096 volume, if we call FITRIM with the following parameter as
fstrim_range(start = 102400, len = 134144000, minlen = 10240),
we will trigger a BUG_ON.
BUG_ON(start + len > (e4b->bd_sb->s_blocksize << 3));

Mar  4 00:55:52 boyu-tm kernel: ------------[ cut here ]------------
Mar  4 00:55:52 boyu-tm kernel: kernel BUG at fs/ext4/mballoc.c:1506!
Mar  4 01:21:09 boyu-tm kernel: Code: d4 00 00 00 00 49 89 fe 8b 56 0c 44 8b 7e 04 89 55 c4 48 8b 4f 28 89 d6 44 01 fe 48 63 d6 48 8b 41 18 48 c1 e0 03 48 39 c2 76 04 <0f> 0b eb fe 48 8b 55 b0 8b 47 34 3b 42 08 74 04 0f 0b eb fe 48
Mar  4 01:21:09 boyu-tm kernel: RIP  [<ffffffffa053eb42>] mb_mark_used+0x47/0x26c [ext4]
Mar  4 01:21:09 boyu-tm kernel:  RSP <ffff880121e45c38>
Mar  4 01:21:09 boyu-tm kernel: ---[ end trace 9f461696f6a9dcf2 ]---

The reason is that in ext4_trim_fs, the last_block is checked wrongly.
if (len >= EXT4_BLOCKS_PER_GROUP(sb))
	len -= (EXT4_BLOCKS_PER_GROUP(sb) - first_block);
else
	last_block = first_block + len;

So if len < EXT4_BLOCKS_PER_GROUP while first_block + len > EXT4_BLOCKS_PER_GROUP,
last_block will be set to a overflow value which exceeds EXT4_BLOCKS_PER_GROUP.

This patch fixes it and adjusts len accordingly.

Cc: Lukas Czerner <lczerner@redhat.com>
Cc: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Tao Ma <boyu.mt@taobao.com>
---
 fs/ext4/mballoc.c |   11 ++++++++---
 1 files changed, 8 insertions(+), 3 deletions(-)
Lukas Czerner - March 3, 2011, 2:20 p.m.
On Thu, 3 Mar 2011, Tao Ma wrote:

> From: Tao Ma <boyu.mt@taobao.com>
> 
> In a bs=4096 volume, if we call FITRIM with the following parameter as
> fstrim_range(start = 102400, len = 134144000, minlen = 10240),
> we will trigger a BUG_ON.
> BUG_ON(start + len > (e4b->bd_sb->s_blocksize << 3));
> 
> Mar  4 00:55:52 boyu-tm kernel: ------------[ cut here ]------------
> Mar  4 00:55:52 boyu-tm kernel: kernel BUG at fs/ext4/mballoc.c:1506!
> Mar  4 01:21:09 boyu-tm kernel: Code: d4 00 00 00 00 49 89 fe 8b 56 0c 44 8b 7e 04 89 55 c4 48 8b 4f 28 89 d6 44 01 fe 48 63 d6 48 8b 41 18 48 c1 e0 03 48 39 c2 76 04 <0f> 0b eb fe 48 8b 55 b0 8b 47 34 3b 42 08 74 04 0f 0b eb fe 48
> Mar  4 01:21:09 boyu-tm kernel: RIP  [<ffffffffa053eb42>] mb_mark_used+0x47/0x26c [ext4]
> Mar  4 01:21:09 boyu-tm kernel:  RSP <ffff880121e45c38>
> Mar  4 01:21:09 boyu-tm kernel: ---[ end trace 9f461696f6a9dcf2 ]---
> 
> The reason is that in ext4_trim_fs, the last_block is checked wrongly.
> if (len >= EXT4_BLOCKS_PER_GROUP(sb))
> 	len -= (EXT4_BLOCKS_PER_GROUP(sb) - first_block);
> else
> 	last_block = first_block + len;
> 
> So if len < EXT4_BLOCKS_PER_GROUP while first_block + len > EXT4_BLOCKS_PER_GROUP,
> last_block will be set to a overflow value which exceeds EXT4_BLOCKS_PER_GROUP.
> 
> This patch fixes it and adjusts len accordingly.
> 
> Cc: Lukas Czerner <lczerner@redhat.com>
> Cc: "Theodore Ts'o" <tytso@mit.edu>
> Signed-off-by: Tao Ma <boyu.mt@taobao.com>
> ---
>  fs/ext4/mballoc.c |   11 ++++++++---
>  1 files changed, 8 insertions(+), 3 deletions(-)
> 
> diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
> index 29d7d17..290b36c 100644
> --- a/fs/ext4/mballoc.c
> +++ b/fs/ext4/mballoc.c
> @@ -4889,10 +4889,15 @@ int ext4_trim_fs(struct super_block *sb, struct fstrim_range *range)
>  			break;
>  		}
>  
> -		if (len >= EXT4_BLOCKS_PER_GROUP(sb))
> -			len -= (EXT4_BLOCKS_PER_GROUP(sb) - first_block);
> -		else
> +		/*
> +		 * For all the groups except the last one, last block will
> +		 * always be EXT4_BLOCKS_PER_GROUP(sb), so we only need to
> +		 * change it for the last group in which case start +
> +		 * len < EXT4_BLOCKS_PER_GROUP(sb).
> +		 */
> +		if (first_block + len < EXT4_BLOCKS_PER_GROUP(sb))
>  			last_block = first_block + len;
> +		len -= last_block - first_block;
>  
>  		if (e4b.bd_info->bb_free >= minlen) {
>  			cnt = ext4_trim_all_free(sb, &e4b, first_block,
> 

Looks good.

Thanks!
-Lukas
--
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
Tao Ma - March 3, 2011, 3:33 p.m.
Hi Christoph,
On 03/03/2011 10:11 PM, Christoph Hellwig wrote:
> On Thu, Mar 03, 2011 at 09:59:34PM +0800, Tao Ma wrote:
>> From: Tao Ma<boyu.mt@taobao.com>
>>
>> In a bs=4096 volume, if we call FITRIM with the following parameter as
>> fstrim_range(start = 102400, len = 134144000, minlen = 10240),
> Can you add a xfstests regression test for this particular invocation?
> We already have Lukas' fstrim.c tester in the tree, so it could be a
> simple wrapper around it.
Actually these numbers are related to the block size of a ext4 volume.
To be more specific, with the above number, we have
first_block + block_len > EXT4_BLOCKS_PER_GROUP while
block_len < EXT4_BLOCKS_PER_GROUP.
So do you think it is generic enough for a test case in xfstests since 
now this
tool kit is also used to test xfs and btrfs?

Regards,
Tao




--
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
Christoph Hellwig - March 3, 2011, 3:38 p.m.
On Thu, Mar 03, 2011 at 11:33:35PM +0800, Tao Ma wrote:
> Actually these numbers are related to the block size of a ext4 volume.
> To be more specific, with the above number, we have
> first_block + block_len > EXT4_BLOCKS_PER_GROUP while
> block_len < EXT4_BLOCKS_PER_GROUP.
> So do you think it is generic enough for a test case in xfstests
> since now this
> tool kit is also used to test xfs and btrfs?

xfstests also allows filesystem-specific tests.  Most of the specific
tests are for xfs, but there are few for udf right now as well.

--
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
Tao Ma - March 16, 2011, 3:20 p.m.
Hi Ted,
	This is a fix for kernel panic in ext4. So any comment for it?

Regards,
Tao
On 03/03/2011 09:59 PM, Tao Ma wrote:
> From: Tao Ma <boyu.mt@taobao.com>
> 
> In a bs=4096 volume, if we call FITRIM with the following parameter as
> fstrim_range(start = 102400, len = 134144000, minlen = 10240),
> we will trigger a BUG_ON.
> BUG_ON(start + len > (e4b->bd_sb->s_blocksize << 3));
> 
> Mar  4 00:55:52 boyu-tm kernel: ------------[ cut here ]------------
> Mar  4 00:55:52 boyu-tm kernel: kernel BUG at fs/ext4/mballoc.c:1506!
> Mar  4 01:21:09 boyu-tm kernel: Code: d4 00 00 00 00 49 89 fe 8b 56 0c 44 8b 7e 04 89 55 c4 48 8b 4f 28 89 d6 44 01 fe 48 63 d6 48 8b 41 18 48 c1 e0 03 48 39 c2 76 04 <0f> 0b eb fe 48 8b 55 b0 8b 47 34 3b 42 08 74 04 0f 0b eb fe 48
> Mar  4 01:21:09 boyu-tm kernel: RIP  [<ffffffffa053eb42>] mb_mark_used+0x47/0x26c [ext4]
> Mar  4 01:21:09 boyu-tm kernel:  RSP <ffff880121e45c38>
> Mar  4 01:21:09 boyu-tm kernel: ---[ end trace 9f461696f6a9dcf2 ]---
> 
> The reason is that in ext4_trim_fs, the last_block is checked wrongly.
> if (len >= EXT4_BLOCKS_PER_GROUP(sb))
> 	len -= (EXT4_BLOCKS_PER_GROUP(sb) - first_block);
> else
> 	last_block = first_block + len;
> 
> So if len < EXT4_BLOCKS_PER_GROUP while first_block + len > EXT4_BLOCKS_PER_GROUP,
> last_block will be set to a overflow value which exceeds EXT4_BLOCKS_PER_GROUP.
> 
> This patch fixes it and adjusts len accordingly.
> 
> Cc: Lukas Czerner <lczerner@redhat.com>
> Cc: "Theodore Ts'o" <tytso@mit.edu>
> Signed-off-by: Tao Ma <boyu.mt@taobao.com>
> ---
>  fs/ext4/mballoc.c |   11 ++++++++---
>  1 files changed, 8 insertions(+), 3 deletions(-)
> 
> diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
> index 29d7d17..290b36c 100644
> --- a/fs/ext4/mballoc.c
> +++ b/fs/ext4/mballoc.c
> @@ -4889,10 +4889,15 @@ int ext4_trim_fs(struct super_block *sb, struct fstrim_range *range)
>  			break;
>  		}
>  
> -		if (len >= EXT4_BLOCKS_PER_GROUP(sb))
> -			len -= (EXT4_BLOCKS_PER_GROUP(sb) - first_block);
> -		else
> +		/*
> +		 * For all the groups except the last one, last block will
> +		 * always be EXT4_BLOCKS_PER_GROUP(sb), so we only need to
> +		 * change it for the last group in which case start +
> +		 * len < EXT4_BLOCKS_PER_GROUP(sb).
> +		 */
> +		if (first_block + len < EXT4_BLOCKS_PER_GROUP(sb))
>  			last_block = first_block + len;
> +		len -= last_block - first_block;
>  
>  		if (e4b.bd_info->bb_free >= minlen) {
>  			cnt = ext4_trim_all_free(sb, &e4b, first_block,

--
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
Theodore Ts'o - March 23, 2011, 7:47 p.m.
Thanks, added to the ext4 patch queue.  My apologies for the delay.

	      	     	  	- Ted

On Thu, Mar 03, 2011 at 09:59:34PM +0800, Tao Ma wrote:
> From: Tao Ma <boyu.mt@taobao.com>
> 
> In a bs=4096 volume, if we call FITRIM with the following parameter as
> fstrim_range(start = 102400, len = 134144000, minlen = 10240),
> we will trigger a BUG_ON.
> BUG_ON(start + len > (e4b->bd_sb->s_blocksize << 3));
> 
--
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

diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
index 29d7d17..290b36c 100644
--- a/fs/ext4/mballoc.c
+++ b/fs/ext4/mballoc.c
@@ -4889,10 +4889,15 @@  int ext4_trim_fs(struct super_block *sb, struct fstrim_range *range)
 			break;
 		}
 
-		if (len >= EXT4_BLOCKS_PER_GROUP(sb))
-			len -= (EXT4_BLOCKS_PER_GROUP(sb) - first_block);
-		else
+		/*
+		 * For all the groups except the last one, last block will
+		 * always be EXT4_BLOCKS_PER_GROUP(sb), so we only need to
+		 * change it for the last group in which case start +
+		 * len < EXT4_BLOCKS_PER_GROUP(sb).
+		 */
+		if (first_block + len < EXT4_BLOCKS_PER_GROUP(sb))
 			last_block = first_block + len;
+		len -= last_block - first_block;
 
 		if (e4b.bd_info->bb_free >= minlen) {
 			cnt = ext4_trim_all_free(sb, &e4b, first_block,