[PATCH] D123330: [AMDGPU] Split unaligned LDS access instead of scalarizing

[Stanislav Mekhanoshin via Phabricator via llvm-commits]

rampitec added inline comments.


================
Comment at: llvm/test/CodeGen/AMDGPU/load-local.96.ll:355
 ; GFX9-NEXT:    v_mov_b32_e32 v2, v0
-; GFX9-NEXT:    ds_read2_b32 v[0:1], v0 offset1:1
+; GFX9-NEXT:    ds_read_b64 v[0:1], v0
 ; GFX9-NEXT:    ds_read_b32 v2, v2 offset:8
----------------
This is the most essential effect. Here and in some other places.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D123330/new/

https://reviews.llvm.org/D123330