[llvm] [RISCV] Combine vslidedown_vl with known VL and offset to a smaller LMUL (PR #66267)

[Philip Reames via llvm-commits]

================
@@ -875,11 +875,15 @@ define i64 @explode_8xi64(<8 x i64> %v) {
 ; RV32-NEXT:    vsrl.vx v12, v8, a0
 ; RV32-NEXT:    vmv.x.s a1, v12
 ; RV32-NEXT:    vmv.x.s a2, v8
+; RV32-NEXT:    vsetivli zero, 1, e64, m2, ta, ma
 ; RV32-NEXT:    vslidedown.vi v12, v8, 1
-; RV32-NEXT:    vsrl.vx v16, v12, a0
-; RV32-NEXT:    vmv.x.s a3, v16
+; RV32-NEXT:    vmv.x.s a3, v12
+; RV32-NEXT:    vsetivli zero, 1, e64, m4, ta, ma
----------------
preames wrote:

Same as previous

https://github.com/llvm/llvm-project/pull/66267