[PATCH] ext4: fix wrong verification in ext4_ext_insert_index

[PATCH] ext4: fix wrong verification in ext4_ext_insert_index

[Yongqiang Yang]

After inserting an new index, current number of indexes should
be greater than original number by 1.  So if the new index is
less or equal than LAST_INDEX + 1, then indexes are continugous.
If new index will be placed on the end, then ix will equals
LAST_INDEX + 1. Index entries has been verifiyed in pervious code
in ext4_ext_insert_index.

Signed-off-by: Yongqiang Yang <xiaoqiangnk@gmail.com>
---
 fs/ext4/extents.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
index 2dff31e..322398e 100644
--- a/fs/ext4/extents.c
+++ b/fs/ext4/extents.c
@@ -779,7 +779,7 @@ static int ext4_ext_insert_index(handle_t *handle, struct inode *inode,
 		ix = curp->p_idx;
 	}
 
-	if (unlikely(ix > EXT_LAST_INDEX(curp->p_hdr))) {
+	if (unlikely(ix > EXT_LAST_INDEX(curp->p_hdr) + 1)) {
 		EXT4_ERROR_INODE(inode, "ix > EXT_LAST_INDEX!");
 		return -EIO;
 	}
-- 
1.7.5.1

Re: [PATCH] ext4: fix wrong verification in ext4_ext_insert_index

[Tao Ma]

Hi yongqiang,
On 10/17/2011 02:11 PM, Yongqiang Yang wrote:
> After inserting an new index, current number of indexes should
> be greater than original number by 1.  So if the new index is
> less or equal than LAST_INDEX + 1, then indexes are continugous.
> If new index will be placed on the end, then ix will equals
> LAST_INDEX + 1. Index entries has been verifiyed in pervious code
> in ext4_ext_insert_index.
Thanks for the effort. But actually this is caused by my original patch
4fd30c033, and your fix doesn't resolve the problem I want to fix. So
consider ix is already overflow, the check can't prevent it from
stamping on the memory after this extent block. I have sent another fix
for it.

Thanks
Tao
> 
> Signed-off-by: Yongqiang Yang <xiaoqiangnk@gmail.com>
> ---
>  fs/ext4/extents.c |    2 +-
>  1 files changed, 1 insertions(+), 1 deletions(-)
> 
> diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
> index 2dff31e..322398e 100644
> --- a/fs/ext4/extents.c
> +++ b/fs/ext4/extents.c
> @@ -779,7 +779,7 @@ static int ext4_ext_insert_index(handle_t *handle, struct inode *inode,
>  		ix = curp->p_idx;
>  	}
>  
> -	if (unlikely(ix > EXT_LAST_INDEX(curp->p_hdr))) {
> +	if (unlikely(ix > EXT_LAST_INDEX(curp->p_hdr) + 1)) {
>  		EXT4_ERROR_INODE(inode, "ix > EXT_LAST_INDEX!");
>  		return -EIO;
>  	}

Re: [PATCH] ext4: fix wrong verification in ext4_ext_insert_index

[Yongqiang Yang]

On Mon, Oct 17, 2011 at 4:49 PM, Tao Ma <tm@tao.ma> wrote:
> Hi yongqiang,
> On 10/17/2011 02:11 PM, Yongqiang Yang wrote:
>> After inserting an new index, current number of indexes should
>> be greater than original number by 1.  So if the new index is
>> less or equal than LAST_INDEX + 1, then indexes are continugous.
>> If new index will be placed on the end, then ix will equals
>> LAST_INDEX + 1. Index entries has been verifiyed in pervious code
>> in ext4_ext_insert_index.
> Thanks for the effort. But actually this is caused by my original patch
> 4fd30c033, and your fix doesn't resolve the problem I want to fix. So
> consider ix is already overflow, the check can't prevent it from
> stamping on the memory after this extent block. I have sent another fix
> for it.
Thanks!

Sorry!  My working tree is not up to date:-)

Yongqiang.
>
> Thanks
> Tao
>>
>> Signed-off-by: Yongqiang Yang <xiaoqiangnk@gmail.com>
>> ---
>>  fs/ext4/extents.c |    2 +-
>>  1 files changed, 1 insertions(+), 1 deletions(-)
>>
>> diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
>> index 2dff31e..322398e 100644
>> --- a/fs/ext4/extents.c
>> +++ b/fs/ext4/extents.c
>> @@ -779,7 +779,7 @@ static int ext4_ext_insert_index(handle_t *handle, struct inode *inode,
>>               ix = curp->p_idx;
>>       }
>>
>> -     if (unlikely(ix > EXT_LAST_INDEX(curp->p_hdr))) {
>> +     if (unlikely(ix > EXT_LAST_INDEX(curp->p_hdr) + 1)) {
>>               EXT4_ERROR_INODE(inode, "ix > EXT_LAST_INDEX!");
>>               return -EIO;
>>       }
>
>



-- 
Best Wishes
Yongqiang Yang
--
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