AMDGPU: Fold readlane intrinsics of constants
I'm not 100% sure about this, since I'm worried about IR transforms
that might end up introducing divergence downstream once replaced with
a constant, but I haven't come up with an example yet.
llvm-svn: 363406
diff --git a/llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp b/llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp
index f167762..347aeb6 100644
--- a/llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp
+++ b/llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp
@@ -3776,6 +3776,13 @@
II->setOperand(0, UndefValue::get(Old->getType()));
return II;
}
+ case Intrinsic::amdgcn_readfirstlane:
+ case Intrinsic::amdgcn_readlane: {
+ // A constant value is trivially uniform.
+ if (Constant *C = dyn_cast<Constant>(II->getArgOperand(0)))
+ return replaceInstUsesWith(*II, C);
+ break;
+ }
case Intrinsic::stackrestore: {
// If the save is right next to the restore, remove the restore. This can
// happen when variable allocas are DCE'd.