[PATCH] D76356: [AMDGPU] Introduce more scratch registers in the ABI.

[Matt Arsenault via Phabricator via llvm-commits]

arsenm accepted this revision.
arsenm added a comment.

LGTM with test nits. This isn't perfect but we can't do much better now



================
Comment at: llvm/test/CodeGen/AMDGPU/vgpr-tuple-allocation.ll:6
+
+define <4 x float> @non_preserved_vgpr_tuple8(<8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %bias, float %zcompare, float %s, float %t, float %clamp) {
+; The vgpr tuple8 operand in image_gather4_c_b_cl instruction needs not be
----------------
inreg currently doesn't do anything for non-shaders, so you should remove it here.


================
Comment at: llvm/test/CodeGen/AMDGPU/vgpr-tuple-allocation.ll:8
+; The vgpr tuple8 operand in image_gather4_c_b_cl instruction needs not be
+; presreved across the call and should get 8 scratch registers.
+
----------------
Typo presreved


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D76356/new/

https://reviews.llvm.org/D76356