Before calling add partition or resize partition, there is no check
on whether the length is aligned with the logical block size.
If the logical block size of the disk is larger than 512 bytes,
then the partition size maybe not the multiple of the logical block size,
and when the last sector is read, bio_truncate() will adjust the bio size,
resulting in an IO error if the size of the read command is smaller than
the logical block size.If integrity data is supported, this will also
result in a null pointer dereference when calling bio_integrity_free.
Signed-off-by: Min Li <[email protected]>
---
Changes from v1:
- Add a space after /* and before */.
- Move length alignment check before the "start = p.start >> SECTOR_SHIFT"
- Move check for p.start being aligned together with this length alignment check.
Changes from v2:
- Add the assignment on the first line and merge the two lines into one.
- Use IS_ALIGNED function to check alignment.
---
block/ioctl.c | 9 +++++----
1 file changed, 5 insertions(+), 4 deletions(-)
diff --git a/block/ioctl.c b/block/ioctl.c
index 3be11941fb2d..1a351132bf4e 100644
--- a/block/ioctl.c
+++ b/block/ioctl.c
@@ -16,9 +16,9 @@
static int blkpg_do_ioctl(struct block_device *bdev,
struct blkpg_partition __user *upart, int op)
{
+ long long blksz = bdev_logical_block_size(bdev), start, length;
struct gendisk *disk = bdev->bd_disk;
struct blkpg_partition p;
- long long start, length;
if (!capable(CAP_SYS_ADMIN))
return -EACCES;
@@ -33,14 +33,15 @@ static int blkpg_do_ioctl(struct block_device *bdev,
if (op == BLKPG_DEL_PARTITION)
return bdev_del_partition(disk, p.pno);
+ /* Check that the partition is aligned to the block size */
+ if (!IS_ALIGNED(p.start | p.length, blksz))
+ return -EINVAL;
+
start = p.start >> SECTOR_SHIFT;
length = p.length >> SECTOR_SHIFT;
switch (op) {
case BLKPG_ADD_PARTITION:
- /* check if partition is aligned to blocksize */
- if (p.start & (bdev_logical_block_size(bdev) - 1))
- return -EINVAL;
return bdev_add_partition(disk, p.pno, start, length);
case BLKPG_RESIZE_PARTITION:
return bdev_resize_partition(disk, p.pno, start, length);
--
2.34.1
On 6/28/23 22:09, Min Li wrote:
> Before calling add partition or resize partition, there is no check
> on whether the length is aligned with the logical block size.
> If the logical block size of the disk is larger than 512 bytes,
> then the partition size maybe not the multiple of the logical block size,
> and when the last sector is read, bio_truncate() will adjust the bio size,
> resulting in an IO error if the size of the read command is smaller than
> the logical block size.If integrity data is supported, this will also
> result in a null pointer dereference when calling bio_integrity_free.
>
> Signed-off-by: Min Li <[email protected]>
Reviewed-by: Damien Le Moal <[email protected]>
--
Damien Le Moal
Western Digital Research
On Wed, Jun 28, 2023 at 04:34:42PM +0000, Min Li wrote:
> I will modify the blksz to unsigned int.
> For pre-existing start/length, is it ok to change them to sector_t?
They are long long in the UAPI, which is weird but has been that way
for a long time. So I think we need checks for negative values
before they are shifted and converted to a sector_t.
On Mon, Jun 28, 2023 at 04:48:36PM +0000, Christoph Hellwig wrote:
> They are long long in the UAPI, which is weird but has been that way
> for a long time. So I think we need checks for negative values
> before they are shifted and converted to a sector_t.
Do you mean that we need to check if p.start and p.length are negative?
Thanks for your reply
Min li
On 6/28/23 10:07, Min Li wrote:
> On Mon, Jun 28, 2023 at 04:48:36PM +0000, Christoph Hellwig wrote:
>> They are long long in the UAPI, which is weird but has been that way
>> for a long time. So I think we need checks for negative values
>> before they are shifted and converted to a sector_t.
> Do you mean that we need to check if p.start and p.length are negative?
>
> Thanks for your reply
>
> Min li
I think so, from include/uapi/linux/blkpg.h:-
/* The data structure for ADD_PARTITION and DEL_PARTITION */
struct blkpg_partition {
long long start; /* starting offset in bytes */
long long length; /* length in bytes */
int pno; /* partition number */
char devname[BLKPG_DEVNAMELTH]; /* unused / ignored */
char volname[BLKPG_VOLNAMELTH]; /* unused / ignore */
};
so start and length can take -ve, hence in your patch we need to error
out instead inviting overflow bugs right after or before IS_ALIGNED
check.
-ck
Hi,
在 2023/06/29 1:07, Min Li 写道:
> On Mon, Jun 28, 2023 at 04:48:36PM +0000, Christoph Hellwig wrote:
>> They are long long in the UAPI, which is weird but has been that way
>> for a long time. So I think we need checks for negative values
>> before they are shifted and converted to a sector_t.
>
> Do you mean that we need to check if p.start and p.length are negative?
>
Perhaps this patch will make sense.
https://lore.kernel.org/all/[email protected]/
Thanks,
Kuai
> Thanks for your reply
>
> Min li
> .
>
On 6/29/23 10:04, Yu Kuai wrote:
> Hi,
>
> 在 2023/06/29 1:07, Min Li 写道:
>> On Mon, Jun 28, 2023 at 04:48:36PM +0000, Christoph Hellwig wrote:
>>> They are long long in the UAPI, which is weird but has been that way
>>> for a long time. So I think we need checks for negative values
>>> before they are shifted and converted to a sector_t.
>>
>> Do you mean that we need to check if p.start and p.length are negative?
>>
>
> Perhaps this patch will make sense.
>
> https://lore.kernel.org/all/[email protected]/
+ adding to it the logical block size alignment checks.
>
> Thanks,
> Kuai
>> Thanks for your reply
>>
>> Min li
>> .
>>
>
--
Damien Le Moal
Western Digital Research