[PATCH] ext4: ext4_mb_seq_groups_show: Fix stack memory corruption

[PATCH] ext4: ext4_mb_seq_groups_show: Fix stack memory corruption

[Chandan Rajendra]

The number of 'counters' elements needed in 'struct sg' is
super_block->s_blocksize_bits + 2. Presently we have 16 'counters'
elements in the array. This is insufficient for block sizes >= 32k. In
such cases the memcpy operation performed in ext4_mb_seq_groups_show()
would cause stack memory corruption.

Signed-off-by: Chandan Rajendra <chandan@linux.vnet.ibm.com>
---
 fs/ext4/mballoc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
index a937ac7..67e6fcb 100644
--- a/fs/ext4/mballoc.c
+++ b/fs/ext4/mballoc.c
@@ -2287,7 +2287,7 @@ static int ext4_mb_seq_groups_show(struct seq_file *seq, void *v)
 	struct ext4_group_info *grinfo;
 	struct sg {
 		struct ext4_group_info info;
-		ext4_grpblk_t counters[16];
+		ext4_grpblk_t counters[16 + 2];
 	} sg;
 
 	group--;
-- 
2.5.5

Re: [PATCH] ext4: ext4_mb_seq_groups_show: Fix stack memory corruption

[Jan Kara]

On Mon 07-11-16 12:15:41, Chandan Rajendra wrote:
> The number of 'counters' elements needed in 'struct sg' is
> super_block->s_blocksize_bits + 2. Presently we have 16 'counters'
> elements in the array. This is insufficient for block sizes >= 32k. In
> such cases the memcpy operation performed in ext4_mb_seq_groups_show()
> would cause stack memory corruption.
> 
> Signed-off-by: Chandan Rajendra <chandan@linux.vnet.ibm.com>
> ---
>  fs/ext4/mballoc.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
> index a937ac7..67e6fcb 100644
> --- a/fs/ext4/mballoc.c
> +++ b/fs/ext4/mballoc.c
> @@ -2287,7 +2287,7 @@ static int ext4_mb_seq_groups_show(struct seq_file *seq, void *v)
>  	struct ext4_group_info *grinfo;
>  	struct sg {
>  		struct ext4_group_info info;
> -		ext4_grpblk_t counters[16];
> +		ext4_grpblk_t counters[16 + 2];

How about using EXT4_MAX_BLOCK_LOG_SIZE + 2? It would be somewhat clearer
what that means...  Having a function returning number of buddy counters in
a group info would be even better (and a constant for maximum number of
counters) but that's an independent cleanup...

									Honza
-- 
Jan Kara <jack@suse.com>
SUSE Labs, CR

Re: [PATCH] ext4: ext4_mb_seq_groups_show: Fix stack memory corruption

[Chandan Rajendra]

On Tuesday, November 08, 2016 10:21:15 AM Jan Kara wrote:
> On Mon 07-11-16 12:15:41, Chandan Rajendra wrote:
> > The number of 'counters' elements needed in 'struct sg' is
> > super_block->s_blocksize_bits + 2. Presently we have 16 'counters'
> > elements in the array. This is insufficient for block sizes >= 32k. In
> > such cases the memcpy operation performed in ext4_mb_seq_groups_show()
> > would cause stack memory corruption.
> > 
> > Signed-off-by: Chandan Rajendra <chandan@linux.vnet.ibm.com>
> > ---
> >  fs/ext4/mballoc.c | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> > 
> > diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
> > index a937ac7..67e6fcb 100644
> > --- a/fs/ext4/mballoc.c
> > +++ b/fs/ext4/mballoc.c
> > @@ -2287,7 +2287,7 @@ static int ext4_mb_seq_groups_show(struct seq_file *seq, void *v)
> >  	struct ext4_group_info *grinfo;
> >  	struct sg {
> >  		struct ext4_group_info info;
> > -		ext4_grpblk_t counters[16];
> > +		ext4_grpblk_t counters[16 + 2];
> 
> How about using EXT4_MAX_BLOCK_LOG_SIZE + 2? It would be somewhat clearer
> what that means...  Having a function returning number of buddy counters in
> a group info would be even better (and a constant for maximum number of
> counters) but that's an independent cleanup...

Sorry about the late reply. I will send out V2 with the fix. Also, I will work
on the cleanup patch and send it out soon.

-- 
chandan