[llvm] [VectorCombine] Generalize foldBitOpOfBitcasts to support more cast operations (PR #148350)

[Simon Pilgrim via llvm-commits]

================
@@ -808,46 +808,105 @@ bool VectorCombine::foldInsExtBinop(Instruction &I) {
   return true;
 }
 
-bool VectorCombine::foldBitOpOfBitcasts(Instruction &I) {
-  // Match: bitop(bitcast(x), bitcast(y)) -> bitcast(bitop(x, y))
-  Value *LHSSrc, *RHSSrc;
-  if (!match(&I, m_BitwiseLogic(m_BitCast(m_Value(LHSSrc)),
-                                m_BitCast(m_Value(RHSSrc)))))
+bool VectorCombine::foldBitOpOfCastops(Instruction &I) {
+  // Match: bitop(castop(x), castop(y)) -> castop(bitop(x, y))
+  // Supports: bitcast, trunc, sext, zext
----------------
RKSimon wrote:

(style) Move the description comment up a few lines to outside the function def

https://github.com/llvm/llvm-project/pull/148350