On 06/24/2015 06:49 AM, Emil Velikov wrote:
Hello gents,
On 19 June 2015 at 15:39, Brian Paul <bri...@vmware.com> wrote:
If the driver says PIPE_CAP_VERTEX_ELEMENT_SRC_OFFSET_4BYTE_ALIGNED_ONLY=1,
the driver should never receive a pipe_vertex_element::src_offset value
that's not a multiple of four. But the vbuf code wasn't actually adjusting
the src_offset value when creating the vertex element state object.
We just need to align the src_offset values put in the driver_attribs[]
array.
See the piglit gl-1.5-vertex-buffer-offsets test.
Is this something we want in the 10.6 stable branch ? According to the
comment in the piglit test, the spec does not explicitly state what
should happen in the particular case, but I'd assume that we want to
provide some safe behaviour ?
I think the patch fixes a pretty rare situation so I wasn't too
concerned about porting to the 10.6 branch. But I don't think there'd
be any harm in cherry-picking it to 10.6.
-Brian
_______________________________________________
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
