[llvm] [RISCV][TTI] Properly model odd vector sized LD/ST operations (PR #100436)

[Philip Reames via llvm-commits]

================
@@ -7,24 +7,22 @@ define void @small_trip_count_min_vlen_128(ptr nocapture %a) nounwind vscale_ran
 ; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
 ; CHECK:       vector.ph:
 ; CHECK-NEXT:    [[TMP0:%.*]] = call i32 @llvm.vscale.i32()
-; CHECK-NEXT:    [[TMP1:%.*]] = mul i32 [[TMP0]], 2
-; CHECK-NEXT:    [[TMP4:%.*]] = sub i32 [[TMP1]], 1
-; CHECK-NEXT:    [[N_RND_UP:%.*]] = add i32 4, [[TMP4]]
-; CHECK-NEXT:    [[N_MOD_VF:%.*]] = urem i32 [[N_RND_UP]], [[TMP1]]
+; CHECK-NEXT:    [[TMP1:%.*]] = sub i32 [[TMP0]], 1
+; CHECK-NEXT:    [[N_RND_UP:%.*]] = add i32 4, [[TMP1]]
+; CHECK-NEXT:    [[N_MOD_VF:%.*]] = urem i32 [[N_RND_UP]], [[TMP0]]
 ; CHECK-NEXT:    [[N_VEC:%.*]] = sub i32 [[N_RND_UP]], [[N_MOD_VF]]
-; CHECK-NEXT:    [[TMP5:%.*]] = call i32 @llvm.vscale.i32()
-; CHECK-NEXT:    [[TMP6:%.*]] = mul i32 [[TMP5]], 2
+; CHECK-NEXT:    [[TMP2:%.*]] = call i32 @llvm.vscale.i32()
 ; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
 ; CHECK:       vector.body:
 ; CHECK-NEXT:    [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
-; CHECK-NEXT:    [[TMP7:%.*]] = add i32 [[INDEX]], 0
----------------
preames wrote:

For this test, key details are:
<vscale x 2 x i32> is legal
<vscale x 1 x i32> is not
scalarization (the prior costing approach) returns Invalid for scalable vectors
our actual lowering uses the VL predicated <vscale x 2 x i32>

So this is a valid and correct change.  I make no comment on whether the vectorizer chose the cheapest option.  :)

https://github.com/llvm/llvm-project/pull/100436