3.16.57-rc1 review patch. If anyone has any objections, please let me know.
------------------
From: Dan Carpenter <dan.carpenter@xxxxxxxxxx>
commit 7f7aeea7cf30368b9fdb86dcc9d2c8a3ebc65dfb upstream.
I get some static checker warnings like this:
drivers/staging/rts5208/ms.c:2607 ms_build_l2p_tbl()
error: buffer underflow 'ms_card->segment' (-1)-16
The problem is that we memset "ms_card" to zero at the start of the
reset_ms_card() function. That means that when we try to calculate
"ms_card->total_block / 512 - 1" then it's just always -1. The fix is
to calculate "seg_no" before doing the memset().
This is a static checker fix, and I am not able to test it. My theory
is that reset_ms_card() gets very little testing which is why this bug
exists.
Fixes: fa590c222fba ("staging: rts5208: add support for rts5208 and rts5288")
Signed-off-by: Dan Carpenter <dan.carpenter@xxxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
[bwh: Backported to 3.16: adjust context]
Signed-off-by: Ben Hutchings <ben@xxxxxxxxxxxxxxx>
---
drivers/staging/rts5208/ms.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
--- a/drivers/staging/rts5208/ms.c
+++ b/drivers/staging/rts5208/ms.c
@@ -2397,6 +2397,7 @@ BUILD_FAIL:
int reset_ms_card(struct rtsx_chip *chip)
{
struct ms_info *ms_card = &(chip->ms_card);
+ int seg_no = ms_card->total_block / 512 - 1;
int retval;
memset(ms_card, 0, sizeof(struct ms_info));
@@ -2430,7 +2431,7 @@ int reset_ms_card(struct rtsx_chip *chip
/* Build table for the last segment,
* to check if L2P table block exists, erasing it
*/
- retval = ms_build_l2p_tbl(chip, ms_card->total_block / 512 - 1);
+ retval = ms_build_l2p_tbl(chip, seg_no);
if (retval != STATUS_SUCCESS)
TRACE_RET(chip, STATUS_FAIL);
}
