Coverity reports (CID 1564769, 1564770) that we potentially overflow
by doing some 32x32 multiplies for something that ends up in a 64 bit
value. Fix this by casting the first input to uint64_t to ensure a 64
bit multiply is used.
While we are at it note why we split the calculation into stride and
bytes_pp parts.
Signed-off-by: Alex Bennée <alex.ben...@linaro.org>
Cc: Dmitry Osipenko <dmitry.osipe...@collabora.com>
---
hw/display/virtio-gpu.c | 9 +++++++--
1 file changed, 7 insertions(+), 2 deletions(-)
diff --git a/hw/display/virtio-gpu.c b/hw/display/virtio-gpu.c
index e7ca8fd1cf..572e4d92c6 100644
--- a/hw/display/virtio-gpu.c
+++ b/hw/display/virtio-gpu.c
@@ -741,9 +741,14 @@ bool virtio_gpu_scanout_blob_to_fb(struct
virtio_gpu_framebuffer *fb,
fb->stride = ss->strides[0];
fb->offset = ss->offsets[0] + ss->r.x * fb->bytes_pp + ss->r.y *
fb->stride;
+ /*
+ * We calculate fb->stride for every line but the last which we
+ * calculate purely by its width. The stride will often be larger
+ * than width to meet alignment requirements.
+ */
fbend = fb->offset;
- fbend += fb->stride * (ss->r.height - 1);
- fbend += fb->bytes_pp * ss->r.width;
+ fbend += (uint64_t) fb->stride * (ss->r.height - 1);
+ fbend += (uint64_t) fb->bytes_pp * ss->r.width;
if (fbend > blob_size) {
qemu_log_mask(LOG_GUEST_ERROR,
--
2.39.5
