[RISCV] Don't increase vslide or splat vl if +vl-dependent-latency is present (#147089)

If the subtarget's latency is dependent on vl, then we shouldn't try to
fold away vsetvli toggles if it means increasing vl.

Author: lukel97

llvm/lib/Target/RISCV/RISCVInsertVSETVLI.cpp

@@ -421,10 +421,12 @@ DemandedFields getDemanded(const MachineInstr &MI, const RISCVSubtarget *ST) {
     // * We can't modify SEW here since the slide amount is in units of SEW.
     // * VL=1 is special only because we have existing support for zero vs
     //   non-zero VL.  We could generalize this if we had a VL > C predicate.
-    // * The LMUL1 restriction is for machines whose latency may depend on VL.
+    // * The LMUL1 restriction is for machines whose latency may depend on LMUL.
     // * As above, this is only legal for tail "undefined" not "agnostic".
+    // * We avoid increasing vl if the subtarget has +vl-dependent-latency
     if (RISCVInstrInfo::isVSlideInstr(MI) && VLOp.isImm() &&
-        VLOp.getImm() == 1 && hasUndefinedPassthru(MI)) {
+        VLOp.getImm() == 1 && hasUndefinedPassthru(MI) &&
+        !ST->hasVLDependentLatency()) {
       Res.VLAny = false;
       Res.VLZeroness = true;
       Res.LMUL = DemandedFields::LMULLessThanOrEqualToM1;
@@ -438,7 +440,8 @@ DemandedFields getDemanded(const MachineInstr &MI, const RISCVSubtarget *ST) {
     // careful to not increase the number of active vector registers (unlike for
     // vmv.s.x.)
     if (RISCVInstrInfo::isScalarSplatInstr(MI) && VLOp.isImm() &&
-        VLOp.getImm() == 1 && hasUndefinedPassthru(MI)) {
+        VLOp.getImm() == 1 && hasUndefinedPassthru(MI) &&
+        !ST->hasVLDependentLatency()) {
       Res.LMUL = DemandedFields::LMULLessThanOrEqualToM1;
       Res.SEWLMULRatio = false;
       Res.VLAny = false;