The "num_syncobj_handles" is a u32 value that comes from the user via the
ioctl.  On 32bit systems the "sizeof(uint32_t) * num_syncobj_handles"
multiplication can have an integer overflow.  Use size_mul() to fix that.

Fixes: 38c67ec9aa4b ("drm/amdgpu: Add input fence to sync bo map/unmap")
Cc: sta...@vger.kernel.org
Signed-off-by: Dan Carpenter <dan.carpen...@linaro.org>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c 
b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
index 2c68118fe9fd..0ecc88df7208 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
@@ -58,7 +58,7 @@ amdgpu_gem_add_input_fence(struct drm_file *filp,
                return 0;
 
        syncobj_handles = memdup_user(u64_to_user_ptr(syncobj_handles_array),
-                                     sizeof(uint32_t) * num_syncobj_handles);
+                                     size_mul(sizeof(uint32_t), 
num_syncobj_handles));
        if (IS_ERR(syncobj_handles))
                return PTR_ERR(syncobj_handles);
 
-- 
2.47.2

Reply via email to