[x86] enable CGP memcmp() expansion for 2/4/8 byte sizes

There are a couple of potential improvements as seen in the IR and asm:
1. We're unnecessarily extending to a larger type to compare values.
2. The codegen for (select cond, 1, -1) could avoid a cmov.
(or we could change the order of the compares, so we have a select with 0 operand)

llvm-svn: 305802

1 files changed, 6 insertions, 0 deletions

diff --git a/llvm/lib/Target/X86/X86ISelLowering.cpp b/llvm/lib/Target/X86/X86ISelLowering.cpp
index eacf2e55143..8dfaf3f080e 100644
--- a/llvm/lib/Target/X86/X86ISelLowering.cpp
+++ b/llvm/lib/Target/X86/X86ISelLowering.cpp
@@ -1662,6 +1662,12 @@ X86TargetLowering::X86TargetLowering(const X86TargetMachine &TM,
   MaxStoresPerMemcpyOptSize = 4;
   MaxStoresPerMemmove = 8; // For @llvm.memmove -> sequence of stores
   MaxStoresPerMemmoveOptSize = 4;
+
+  // TODO: These control memcmp expansion in CGP and are set low to prevent
+  // altering the vector expansion for 16/32 byte memcmp in SelectionDAGBuilder.
+  MaxLoadsPerMemcmp = 1;
+  MaxLoadsPerMemcmpOptSize = 1;
+
   // Set loop alignment to 2^ExperimentalPrefLoopAlignment bytes (default: 2^4).
   setPrefLoopAlignment(ExperimentalPrefLoopAlignment);