[ARM][CGP] Fix scalar condition selects for MVE

The arm backend does not handle select/select_cc on vectors with scalar
conditions, preferring to expand them in codegenprepare instead. This
usually works except when optimizing for size, where the optsize check
would end up overruling the backend isSelectSupported check.

We could handle the selects in ISel too, but this seems like smaller
code than trying to splat the condition to all lanes.

Differential Revision: https://reviews.llvm.org/D86433

1 files changed, 3 insertions, 4 deletions

diff --git a/llvm/lib/CodeGen/CodeGenPrepare.cpp b/llvm/lib/CodeGen/CodeGenPrepare.cpp
index a85ac80..86b5d20 100644
--- a/llvm/lib/CodeGen/CodeGenPrepare.cpp
+++ b/llvm/lib/CodeGen/CodeGenPrepare.cpp
@@ -6482,9 +6482,7 @@ bool CodeGenPrepare::optimizeFunnelShift(IntrinsicInst *Fsh) {
 /// If we have a SelectInst that will likely profit from branch prediction,
 /// turn it into a branch.
 bool CodeGenPrepare::optimizeSelectInst(SelectInst *SI) {
-  // If branch conversion isn't desirable, exit early.
-  if (DisableSelectToBranch || OptSize ||
-      llvm::shouldOptimizeForSize(SI->getParent(), PSI, BFI.get()))
+  if (DisableSelectToBranch)
     return false;
 
   // Find all consecutive select instructions that share the same condition.
@@ -6520,7 +6518,8 @@ bool CodeGenPrepare::optimizeSelectInst(SelectInst *SI) {
     SelectKind = TargetLowering::ScalarValSelect;
 
   if (TLI->isSelectSupported(SelectKind) &&
-      !isFormingBranchFromSelectProfitable(TTI, TLI, SI))
+      (!isFormingBranchFromSelectProfitable(TTI, TLI, SI) || OptSize ||
+       llvm::shouldOptimizeForSize(SI->getParent(), PSI, BFI.get())))
     return false;
 
   // The DominatorTree needs to be rebuilt by any consumers after this