[PATCH] D97392: [amdgpu] Atomic should be source of divergence.

[Konstantin Zhuravlyov via Phabricator via llvm-commits]

kzhuravl added inline comments.


================
Comment at: llvm/lib/Target/AMDGPU/SIISelLowering.cpp:11841
+    // TODO: May need to check operand values to determine divergence. Besides
+    // some atomic ops like CLR are alway divergent. The divergence of some
+    // atomic ops is value dependant. For instance, atomic<op> p, v is always
----------------
always


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D97392/new/

https://reviews.llvm.org/D97392