[PATCH] D45219: [AMDGPU] performMinMaxCombine should not optimize patterns of vectors to min3/max3

[Matt Arsenault via Phabricator via llvm-commits]

arsenm added a comment.

LGTM except needs check lines



================
Comment at: test/CodeGen/AMDGPU/fmax3.ll:89
+; since there are no pack instructions for fmax3.
+define <2 x half> @no_fmax3_v2f16(<2 x half> %a, <2 x half> %b, <2 x half> %c, <2 x half> %d) {
+entry:
----------------
Needs check lines


================
Comment at: test/CodeGen/AMDGPU/fmin3.ll:87
+; since there are no pack instructions for fmin3.
+define <2 x half> @no_fmin3_v2f16(<2 x half> %a, <2 x half> %b, <2 x half> %c, <2 x half> %d) {
+entry:
----------------
Ditto


https://reviews.llvm.org/D45219