[PATCH] D107345: Introduce LowerGCLeafIntrinsics pass

[Artur Pilipenko via Phabricator via llvm-commits]

apilipenko added a comment.

There is a target-specific hook to emit code for regular (non-atomic) memcpys: EmitTargetCodeForMemcpy. Maybe we should just implement a similar hook for element-atomic copy?



================
Comment at: llvm/lib/Transforms/Scalar/LowerGCLeafIntrinsics.cpp:67-68
+  if (!LenType->isIntegerTy(64)) {
+    assert(LenInBytes->getType()->isIntegerTy(32) &&
+           "Only 32 and 64-bit lengths are allowed!");
+    LenInBytes = Builder.CreateZExt(LenInBytes, Type::getInt64Ty(C),
----------------
LangRef doesn't prohibit other length types.


================
Comment at: llvm/lib/Transforms/Scalar/LowerGCLeafIntrinsics.cpp:128-131
+    LoadInst *Val = Builder.CreateLoad(ValueType, SrcBC, "vec");
+    Val->setAlignment(Align(ElementSizeInBytes));
+    Value *DestBC = Builder.CreateBitCast(Dest, DestValuePtrType, "dst.vec");
+    StoreInst *Store = Builder.CreateStore(Val, DestBC);
----------------
These loads and stores must be at least ElementSizeInBytes-atomic. I'm not sure you can express this in the IR at the moment. 


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D107345/new/

https://reviews.llvm.org/D107345