[llvm] [DAG][RISCV] Use vp_reduce_fadd/fmul when widening types for FP reductions (PR #105840)

[Philip Reames via llvm-commits]

================
@@ -791,12 +791,7 @@ define float @reduce_fadd_16xi32_prefix5(ptr %p) {
 ; CHECK-NEXT:    vle32.v v8, (a0)
 ; CHECK-NEXT:    lui a0, 524288
 ; CHECK-NEXT:    vmv.s.x v10, a0
-; CHECK-NEXT:    vsetivli zero, 6, e32, m2, tu, ma
-; CHECK-NEXT:    vslideup.vi v8, v10, 5
-; CHECK-NEXT:    vsetivli zero, 7, e32, m2, tu, ma
-; CHECK-NEXT:    vslideup.vi v8, v10, 6
-; CHECK-NEXT:    vsetivli zero, 8, e32, m2, ta, ma
-; CHECK-NEXT:    vslideup.vi v8, v10, 7
+; CHECK-NEXT:    vsetivli zero, 5, e32, m2, ta, ma
----------------
preames wrote:

Once we add reductions to tryToReduceVL, this VL toggle will be removed as we'll narrow the load above.  I'm deliberately leaving out that part of the change as a post commit concern was raised on that review, and I want that to be settled before extending it.

https://github.com/llvm/llvm-project/pull/105840