[PATCH] D98491: [AMDGPU] Split GCN subtarget features for unaligned access

[Stanislav Mekhanoshin via Phabricator via llvm-commits]

rampitec added inline comments.


================
Comment at: llvm/lib/Target/AMDGPU/AMDGPUSubtarget.cpp:96
   if (isAmdHsaOS())
-    FullFS += "+flat-for-global,+unaligned-access-mode,+trap-handler,";
+    FullFS += "+flat-for-global,+unaligned-buffer-access,+trap-handler,";
 
----------------
HSA wants unaligned DS access as well. That is only underaligned ds_read/write_b128 shall not be produced for performance reasons.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D98491/new/

https://reviews.llvm.org/D98491