Previously this combine would undo AMDGPU's new custom legalization of wide vector shuffles into 2 element pieces. The comment also states that this combine is only done before legalization, but the case with a build_vector source was unconditional. We probably don't want to do this if the multiple uses are full scalarization of the vector, but this seems to work well enough. Scalarizing extracts should have folded out pre-legalize.
1.1 MiB
1.1 MiB