On 18.05.2016 08:17, Jakob Sinclair wrote:
In the function r600_test_dma the max_width and max_height can be set to
zero.
Thanks for looking into that. Can they really? I see
max_width = MIN2(tsrc.width0, tdst.width0);
max_height = MIN2(tsrc.height0, tdst.height0);
and
tsrc.width0 = (rand() % max_tex_side_gen) + 1;
and similarly for tdst.
Given that, I'd prefer instead an assert(max_width > 0 && max_height >
0) to help Coverity out.
Thanks,
Nicolai
These variables are later used to do a modulo operation which can
lead to unexpected behavior if they are zero. This patch corrects that
by first checking if the max_width or max_height are zero and if they
are it sets width or height to one. Issue was discoverd by Coverity.
CID: 1361542, 1361543
Signed-off-by: Jakob Sinclair <sinclair.ja...@openmailbox.org>
---
src/gallium/drivers/radeon/r600_test_dma.c | 11 +++++++++--
1 file changed, 9 insertions(+), 2 deletions(-)
diff --git a/src/gallium/drivers/radeon/r600_test_dma.c
b/src/gallium/drivers/radeon/r600_test_dma.c
index c203b4d..37d6202 100644
--- a/src/gallium/drivers/radeon/r600_test_dma.c
+++ b/src/gallium/drivers/radeon/r600_test_dma.c
@@ -345,8 +345,15 @@ void r600_test_dma(struct r600_common_screen *rscreen)
dstx = rand() % (tdst.width0 - width + 1)
& ~0x7;
dsty = rand() % (tdst.height0 - height + 1)
& ~0x7;
} else {
- width = (rand() % max_width) + 1;
- height = (rand() % max_height) + 1;
+ if (max_width == 0)
+ width = 1;
+ else
+ width = (rand() % max_width) +
1;
+
+ if (max_height == 0)
+ height = 1;
+ else
+ height = (rand() % max_height)
+ 1;
srcx = rand() % (tsrc.width0 - width +
1);
srcy = rand() % (tsrc.height0 - height
+ 1);
_______________________________________________
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/mesa-dev
