[PATCH] D134967: [AMDGPU] Always lower SHUFFLE_VECTOR

[Matt Arsenault via Phabricator via llvm-commits]

arsenm accepted this revision.
arsenm added inline comments.
This revision is now accepted and ready to land.


================
Comment at: llvm/lib/Target/AMDGPU/AMDGPUInstructionSelector.cpp:3363-3365
   case TargetOpcode::G_SHUFFLE_VECTOR:
-    return selectG_SHUFFLE_VECTOR(I);
+    assert(false && "Illegal opcode");
+    return false;
----------------
Should just delete this 


================
Comment at: llvm/lib/Target/AMDGPU/SIInstructions.td:1989-1991
+def : GCNPat<(v2i16 (DivergentBinFrag<build_vector> (i16 (trunc (i32 (srl (bitconvert v2i16:$src0), (i32 16))))),
+                                                    (i16 (trunc (i32 (bitconvert v2i16:$src0)))))),
+             (V_ALIGNBIT_B32_e64 VGPR_32:$src0, VGPR_32:$src0, (i32 16))>;
----------------
Overlap with D134463


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D134967/new/

https://reviews.llvm.org/D134967