RE: [PATCH] SCSI: aacraid: potential integer overflow in aac_get_containers()

[Mark Salyzyn]

NAK

I dispute that it is necessary or worth the additional abstraction,
since MAXIMUM_NUM_CONTAINERS is in the order of 32, and
sizeof(*fsa_dev_ptr) is in the range of 80 bytes ... There is a LONG
road to hoe to get to the point of overload!

kcalloc -> __kmalloc(size_t, flags | __GFP_ZERO);
kzalloc -> kmalloc(size_t, flags | __GFP_ZERO);
kmalloc -> __kmalloc(size_t, flags);

kcalloc can not allocate a larger entity than kzalloc. But alas it can
report that the size has exceeded ULONG_MAX so that part of the patch is
VERY sound.

Sincerely -- Mark Salyzyn

-----Original Message-----
From: linux-scsi-owner@xxxxxxxxxxxxxxx
[mailto:linux-scsi-owner@xxxxxxxxxxxxxxx] On Behalf Of Haogang Chen
Sent: Wednesday, November 30, 2011 9:30 PM
To: aacraid@xxxxxxxxxxx
Cc: JBottomley@xxxxxxxxxxxxx; linux-scsi@xxxxxxxxxxxxxxx;
linux-kernel@xxxxxxxxxxxxxxx; haogangchen@xxxxxxxxx
Subject: [PATCH] SCSI: aacraid: potential integer overflow in
aac_get_containers()

There is a potential integer overflow in aac_get_containers(). When
maximum_num_containers is large, the subsequent call to kzalloc() will
allocate a buffer smaller than expected, which leads to memory
corruption in the for loop.

The patch replaces kzalloc with kcalloc.

Signed-off-by: Haogang Chen <haogangchen@xxxxxxxxx>
---
 drivers/scsi/aacraid/aachba.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/drivers/scsi/aacraid/aachba.c
b/drivers/scsi/aacraid/aachba.c
index 409f580..440b84d 100644
--- a/drivers/scsi/aacraid/aachba.c
+++ b/drivers/scsi/aacraid/aachba.c
@@ -381,7 +381,7 @@ int aac_get_containers(struct aac_dev *dev)
 
 	if (maximum_num_containers < MAXIMUM_NUM_CONTAINERS)
 		maximum_num_containers = MAXIMUM_NUM_CONTAINERS;
-	fsa_dev_ptr = kzalloc(sizeof(*fsa_dev_ptr) *
maximum_num_containers,
+	fsa_dev_ptr = kcalloc(maximum_num_containers,
sizeof(*fsa_dev_ptr), 
 			GFP_KERNEL);
 	if (!fsa_dev_ptr)
 		return -ENOMEM;
-- 
1.7.5.4
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/