[PATCH] D146737: [AMDGPU] Default component broadcast store

Mateja Marjanovic via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Thu Apr 27 08:13:21 PDT 2023


matejam marked an inline comment as not done.
matejam added inline comments.


================
Comment at: llvm/lib/Target/AMDGPU/AMDGPUInstCombineIntrinsic.cpp:365
+
+  while (!isa<UndefValue>(FirstArg)) {
+    CurInst = dyn_cast<Instruction>(FirstArg);
----------------
foad wrote:
> matejam wrote:
> > foad wrote:
> > > I wonder if the whole of this loop could be replaced by calling llvm::computeKnownBits with an appropriate DemandedElts mask to test each element from the last to the first.
> > I looked that up.
> > computeKnownBits works only for integers, pointers and vector of integers.
> You might be able to use the brand new computeKnownFPClass instead, and check if the result is exactly fcPosZero?
Thanks.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D146737/new/

https://reviews.llvm.org/D146737



More information about the llvm-commits mailing list