[llvm] [RISCV][VLOPT] Add support for checkUsers when UserMI is a Single-Width Integer Reduction (PR #120345)

Philip Reames via llvm-commits llvm-commits at lists.llvm.org
Thu Jan 2 11:52:56 PST 2025


================
@@ -930,26 +960,64 @@ bool RISCVVLOptimizer::isCandidate(const MachineInstr &MI) const {
   return true;
 }
 
-bool RISCVVLOptimizer::checkUsers(const MachineOperand *&CommonVL,
-                                  MachineInstr &MI) {
+std::optional<const MachineOperand>
+RISCVVLOptimizer::getVLForUser(MachineOperand &UserOp) {
+  const MachineInstr &UserMI = *UserOp.getParent();
+  const MCInstrDesc &Desc = UserMI.getDesc();
+
+  // Instructions like reductions may use a vector register as a scalar
+  // register. In this case, we should treat it like a scalar register which
+  // does not impact the decision on whether to optimize VL. But if there is
+  // another user of MI and it may have VL=0, we need to be sure not to reduce
+  // the VL of MI to zero when the VLOp of UserOp is may be non-zero. The most
+  // we can reduce it to is one.
+  if (isVectorOpUsedAsScalarOp(UserOp)) {
+    [[maybe_unused]] Register R = UserOp.getReg();
+    [[maybe_unused]] const TargetRegisterClass *RC = MRI->getRegClass(R);
+    assert(RISCV::VRRegClass.hasSubClassEq(RC) &&
+           "Expect LMUL 1 register class for vector as scalar operands!");
+    LLVM_DEBUG(dbgs() << "    Used this operand as a scalar operand\n");
+    // VMV_X_S and VFMV_F_S do not have a VL opt which would cause an assert
+    // assert failure if we called getVLOpNum. Therefore, we will set the
+    // CommonVL in that case as 1, even if it could have been set to 0.
----------------
preames wrote:

Please update the comment.  (i.e. no CommonVL)

https://github.com/llvm/llvm-project/pull/120345


More information about the llvm-commits mailing list