[PATCH] D103348: [AMDGPU] Add maximum NSA size limit ISA feature

[Matt Arsenault via Phabricator via llvm-commits]

arsenm added inline comments.


================
Comment at: llvm/lib/Target/AMDGPU/AMDGPULegalizerInfo.cpp:4125
+  if (NumAddrRegs > 4 && !isPowerOf2_32(NumAddrRegs)) {
+    // Round up elements to power of 2
+    const int RoundedNumRegs = NextPowerOf2(NumAddrRegs);
----------------
Why does this need to round up? We should be able to directly handle non powers of 2


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D103348/new/

https://reviews.llvm.org/D103348