Add target hook to allow merging stores of nonzero constants

On GPU targets, materializing constants is cheap and stores are
expensive, so only doing this for zero vectors was silly.

Most of the new testcases aren't optimally merged, and are for
later improvements.

llvm-svn: 238108

1 files changed, 6 insertions, 0 deletions

diff --git a/llvm/lib/Target/R600/AMDGPUISelLowering.cpp b/llvm/lib/Target/R600/AMDGPUISelLowering.cpp
index b304488142a..880240c51a8 100644
--- a/llvm/lib/Target/R600/AMDGPUISelLowering.cpp
+++ b/llvm/lib/Target/R600/AMDGPUISelLowering.cpp
@@ -509,6 +509,12 @@ bool AMDGPUTargetLowering::isFNegFree(EVT VT) const {
   return VT == MVT::f32 || VT == MVT::f64;
 }
 
+bool AMDGPUTargetLowering:: storeOfVectorConstantIsCheap(EVT MemVT,
+                                                         unsigned NumElem,
+                                                         unsigned AS) const {
+  return true;
+}
+
 bool AMDGPUTargetLowering::isTruncateFree(EVT Source, EVT Dest) const {
   // Truncate is just accessing a subregister.
   return Dest.bitsLT(Source) && (Dest.getSizeInBits() % 32 == 0);