Extracts coming from phis were being hoisted, while all others were
sunk to their uses. This was inconsistent and didn't seem to serve a
purpose. Changing all extracts to be sunk to uses is a prerequisite
for adding block frequency to the SLP vectorizer's cost model.
I benchmarked the change in isolation (without block frequency). I
only saw noise on x86 and some potentially significant improvements on
ARM. No major regressions is good enough for me.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204699
91177308-0d34-0410-b5e6-
96231b3b80d8
Value *Lane = Builder.getInt32(it->Lane);
// Generate extracts for out-of-tree users.
// Find the insertion point for the extractelement lane.
- if (PHINode *PN = dyn_cast<PHINode>(Vec)) {
- Builder.SetInsertPoint(PN->getParent()->getFirstInsertionPt());
- Value *Ex = Builder.CreateExtractElement(Vec, Lane);
- CSEBlocks.insert(PN->getParent());
- User->replaceUsesOfWith(Scalar, Ex);
- } else if (isa<Instruction>(Vec)){
+ if (isa<Instruction>(Vec)){
if (PHINode *PH = dyn_cast<PHINode>(User)) {
for (int i = 0, e = PH->getNumIncomingValues(); i != e; ++i) {
if (PH->getIncomingValue(i) == Scalar) {