[PATCH] D155884: [Attributor][AMDGPU] Improve indirect call support in closed modules

[Johannes Doerfert via Phabricator via llvm-commits]

jdoerfert added inline comments.


================
Comment at: llvm/lib/Target/AMDGPU/AMDGPUAttributor.cpp:953
     AC.DefaultInitializeLiveInternals = false;
+    AC.IsClosedWorldModule = true;
     AC.IPOAmendableCB = [](const Function &F) {
----------------
arsenm wrote:
> This should probably feed from a pass option like other end-of-LTO things
Those do not feed into the backend passes, as far as I can tell. I am tempted to add the Module flag back.


================
Comment at: llvm/lib/Transforms/IPO/AttributorAttributes.cpp:10416
+        return true;
+      if (T1.isFloatTy() && T2.isFloatTy())
+        return true;
----------------
arsenm wrote:
> The float and double cases are covered by T1 == T2?
Right.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D155884/new/

https://reviews.llvm.org/D155884