This patch adds patterns to teach the AArch64 backend to merge [US]MULL instructions and adds/subs of half the size into [US]ML[AS]L where we don't use the top half of the result. Differential Revision: https://reviews.llvm.org/D95218
Andre Vieira