[PATCH] D120265: AMDGPU: Use the implicit kernargs for code object version 5

[Matt Arsenault via Phabricator via llvm-commits]

arsenm added inline comments.


================
Comment at: clang/lib/CodeGen/CGBuiltin.cpp:16243
+  Call->addRetAttr(
+      Attribute::getWithDereferenceableBytes(Call->getContext(), 64));
+  Call->addRetAttr(Attribute::getWithAlignment(Call->getContext(), Align(4)));
----------------
Isn't the total size 256?


================
Comment at: clang/lib/CodeGen/CGBuiltin.cpp:16244
+      Attribute::getWithDereferenceableBytes(Call->getContext(), 64));
+  Call->addRetAttr(Attribute::getWithAlignment(Call->getContext(), Align(4)));
+  return Call;
----------------
If I remember correctly we require 8 byte alignment (not that it makes much difference)


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D120265/new/

https://reviews.llvm.org/D120265