[PATCH] D41651: AMDGPU: Add 32-bit constant address space

Tony Tye via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Tue Jan 2 20:08:57 PST 2018

t-tye added inline comments.

Comment at: lib/Target/AMDGPU/SIISelLowering.cpp:153
+  // TODO: This breaks unaligned loads, so don't do it on AmdHsa.
+  if (!Subtarget->isAmdHsaOS())
+    setOperationAction(ISD::LOAD, MVT::i32, Custom);
Would it be better to add a query for whether unaligned accesses were supported? Currently it would only be for AmdHsaOS but would allow it to be easily changed in the future.


More information about the llvm-commits mailing list