Linux kernel mirror (for testing)
git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
kernel
os
linux
accel: ethosu: Handle possible underflow in IFM size calculations
If the command stream has larger padding sizes than the IFM and OFM
diminsions, then the calculations will underflow to a negative value.
The result is a very large region bounds which is caught on submit, but
it's better to catch it earlier.
Current mesa ethosu driver has a signedness bug which resulted in
padding of 127 (the max) and triggers this issue.
Reviewed-and-Tested-by: Anders Roxell <anders.roxell@linaro.org>
Link: https://patch.msgid.link/20260218-ethos-fixes-v1-3-be3fa3ea9a30@kernel.org
Signed-off-by: Rob Herring (Arm) <robh@kernel.org>
+5
-2
+5
-2
drivers/accel/ethosu/ethosu_gem.c
+5
-2
drivers/accel/ethosu/ethosu_gem.c
···
245
245
((st->ifm.stride_kernel >> 1) & 0x1) + 1;
246
246
u32 stride_x = ((st->ifm.stride_kernel >> 5) & 0x2) +
247
247
(st->ifm.stride_kernel & 0x1) + 1;
248
-
u32 ifm_height = st->ofm.height[2] * stride_y +
248
+
s32 ifm_height = st->ofm.height[2] * stride_y +
249
249
st->ifm.height[2] - (st->ifm.pad_top + st->ifm.pad_bottom);
250
-
u32 ifm_width = st->ofm.width * stride_x +
250
+
s32 ifm_width = st->ofm.width * stride_x +
251
251
st->ifm.width - (st->ifm.pad_left + st->ifm.pad_right);
252
+
253
+
if (ifm_height < 0 || ifm_width < 0)
254
+
return -EINVAL;
252
255
253
256
len = feat_matrix_length(info, &st->ifm, ifm_width,
254
257
ifm_height, st->ifm.depth);