search for: atomicmemcpyinst

...itance. There is precedent for something like this: OverflowingBinaryOperator will return true for both isa<Instruction> and isa<ConstantExpr>. I also think this combined with (a) is probably the cleanest path forward: AtomicMemIntrinsic * AtomicMemSetInst * AtomicMemTransferInst ** AtomicMemCpyInst ** AtomicMemMoveInst PlainMemIntrinsic * PlainMemSetInst * PlainMemTransferInst ** PlainMemCpyInst ** PlainMemMoveInst MemIntrinsic : PlainMemIntrinsic, AtomicMemIntrinsic MemSetInst : PlainMemSetInst, AtomicMemSetInst MemTransferInst : PlainMemTransferInst, AtomicMemTransferInst MemCpyInst : Plai...

Hi all, We somewhat recently created/updated element-wise unordered-atomic versions of the memcpy, memmove, and memset memory intrinsics: Memcpy: https://reviews.llvm.org/rL305558 Memmove: https://reviews.llvm.org/rL307796 Memset: https://reviews.llvm.org/rL307854 These intrinsics are semantically similar to the regular versions. The main difference is that the underlying operation is performed