[PATCH] D127041: [LLVM] Add the support for fmax and fmin in atomicrmw instruction

[Matt Arsenault via Phabricator via llvm-commits]

arsenm added inline comments.


================
Comment at: llvm/lib/Transforms/InstCombine/InstCombineAtomicRMW.cpp:66-67
+    case AtomicRMWInst::FMax:
+      // maxnum(x, +inf) -> +inf
+      return !CF->isNegative() && CF->isInfinity();
+    case AtomicRMWInst::FMin:
----------------
This isn't correct for signaling nans


================
Comment at: llvm/test/Bitcode/compatibility.ll:870
+; CHECK: %atomicrmw.fmin = atomicrmw fmin float* %word, float 1.000000e+00 monotonic
+  %atomicrmw.fmin = atomicrmw fmin float* %word, float 1.0 monotonic
+
----------------
Since you also touched dxil, is an additional test needed for dxil encoding?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D127041/new/

https://reviews.llvm.org/D127041