[llvm] [LV] Update remaining tests to use VPlan cost output (NFC). (PR #190038)
Florian Hahn via llvm-commits
llvm-commits at lists.llvm.org
Tue Apr 7 08:06:58 PDT 2026
================
@@ -14,126 +14,32 @@ target triple = "x86_64-unknown-linux-gnu"
define void @test() {
; SSE2-LABEL: 'test'
; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load double, ptr %in0, align 8
-; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load double, ptr %in1, align 8
-; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load double, ptr %in2, align 8
-; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load double, ptr %in3, align 8
-; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load double, ptr %in4, align 8
-; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load double, ptr %in0, align 8
-; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load double, ptr %in1, align 8
-; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load double, ptr %in2, align 8
-; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load double, ptr %in3, align 8
-; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load double, ptr %in4, align 8
-; SSE2: LV: Found an estimated cost of 3 for VF 2 For instruction: %v0 = load double, ptr %in0, align 8
-; SSE2: LV: Found an estimated cost of 3 for VF 2 For instruction: %v1 = load double, ptr %in1, align 8
-; SSE2: LV: Found an estimated cost of 3 for VF 2 For instruction: %v2 = load double, ptr %in2, align 8
-; SSE2: LV: Found an estimated cost of 3 for VF 2 For instruction: %v3 = load double, ptr %in3, align 8
-; SSE2: LV: Found an estimated cost of 3 for VF 2 For instruction: %v4 = load double, ptr %in4, align 8
-; SSE2: LV: Found an estimated cost of 6 for VF 4 For instruction: %v0 = load double, ptr %in0, align 8
-; SSE2: LV: Found an estimated cost of 6 for VF 4 For instruction: %v1 = load double, ptr %in1, align 8
-; SSE2: LV: Found an estimated cost of 6 for VF 4 For instruction: %v2 = load double, ptr %in2, align 8
-; SSE2: LV: Found an estimated cost of 6 for VF 4 For instruction: %v3 = load double, ptr %in3, align 8
-; SSE2: LV: Found an estimated cost of 6 for VF 4 For instruction: %v4 = load double, ptr %in4, align 8
+; SSE2: Cost of 3 for VF 2: REPLICATE ir<%v0> = load ir<%in0>
+; SSE2: Cost of 6 for VF 4: REPLICATE ir<%v0> = load ir<%in0>
+; SSE2: Cost of 12 for VF 8: REPLICATE ir<%v0> = load ir<%in0>
+; SSE2: Cost of 24 for VF 16: REPLICATE ir<%v0> = load ir<%in0>
;
; AVX1-LABEL: 'test'
; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load double, ptr %in0, align 8
-; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load double, ptr %in1, align 8
-; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load double, ptr %in2, align 8
-; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load double, ptr %in3, align 8
-; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load double, ptr %in4, align 8
-; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load double, ptr %in0, align 8
-; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load double, ptr %in1, align 8
-; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load double, ptr %in2, align 8
-; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load double, ptr %in3, align 8
-; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load double, ptr %in4, align 8
-; AVX1: LV: Found an estimated cost of 3 for VF 2 For instruction: %v0 = load double, ptr %in0, align 8
-; AVX1: LV: Found an estimated cost of 3 for VF 2 For instruction: %v1 = load double, ptr %in1, align 8
-; AVX1: LV: Found an estimated cost of 3 for VF 2 For instruction: %v2 = load double, ptr %in2, align 8
-; AVX1: LV: Found an estimated cost of 3 for VF 2 For instruction: %v3 = load double, ptr %in3, align 8
-; AVX1: LV: Found an estimated cost of 3 for VF 2 For instruction: %v4 = load double, ptr %in4, align 8
-; AVX1: LV: Found an estimated cost of 7 for VF 4 For instruction: %v0 = load double, ptr %in0, align 8
-; AVX1: LV: Found an estimated cost of 7 for VF 4 For instruction: %v1 = load double, ptr %in1, align 8
-; AVX1: LV: Found an estimated cost of 7 for VF 4 For instruction: %v2 = load double, ptr %in2, align 8
-; AVX1: LV: Found an estimated cost of 7 for VF 4 For instruction: %v3 = load double, ptr %in3, align 8
-; AVX1: LV: Found an estimated cost of 7 for VF 4 For instruction: %v4 = load double, ptr %in4, align 8
-; AVX1: LV: Found an estimated cost of 14 for VF 8 For instruction: %v0 = load double, ptr %in0, align 8
-; AVX1: LV: Found an estimated cost of 14 for VF 8 For instruction: %v1 = load double, ptr %in1, align 8
-; AVX1: LV: Found an estimated cost of 14 for VF 8 For instruction: %v2 = load double, ptr %in2, align 8
-; AVX1: LV: Found an estimated cost of 14 for VF 8 For instruction: %v3 = load double, ptr %in3, align 8
-; AVX1: LV: Found an estimated cost of 14 for VF 8 For instruction: %v4 = load double, ptr %in4, align 8
+; AVX1: Cost of 3 for VF 2: REPLICATE ir<%v0> = load ir<%in0>
+; AVX1: Cost of 7 for VF 4: REPLICATE ir<%v0> = load ir<%in0>
+; AVX1: Cost of 14 for VF 8: REPLICATE ir<%v0> = load ir<%in0>
+; AVX1: Cost of 28 for VF 16: REPLICATE ir<%v0> = load ir<%in0>
+; AVX1: Cost of 56 for VF 32: REPLICATE ir<%v0> = load ir<%in0>
;
; AVX2-LABEL: 'test'
; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load double, ptr %in0, align 8
-; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load double, ptr %in1, align 8
-; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load double, ptr %in2, align 8
-; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load double, ptr %in3, align 8
-; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load double, ptr %in4, align 8
-; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load double, ptr %in0, align 8
-; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load double, ptr %in1, align 8
-; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load double, ptr %in2, align 8
-; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load double, ptr %in3, align 8
-; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load double, ptr %in4, align 8
-; AVX2: LV: Found an estimated cost of 3 for VF 2 For instruction: %v0 = load double, ptr %in0, align 8
-; AVX2: LV: Found an estimated cost of 3 for VF 2 For instruction: %v1 = load double, ptr %in1, align 8
-; AVX2: LV: Found an estimated cost of 3 for VF 2 For instruction: %v2 = load double, ptr %in2, align 8
-; AVX2: LV: Found an estimated cost of 3 for VF 2 For instruction: %v3 = load double, ptr %in3, align 8
-; AVX2: LV: Found an estimated cost of 3 for VF 2 For instruction: %v4 = load double, ptr %in4, align 8
-; AVX2: LV: Found an estimated cost of 7 for VF 4 For instruction: %v0 = load double, ptr %in0, align 8
-; AVX2: LV: Found an estimated cost of 7 for VF 4 For instruction: %v1 = load double, ptr %in1, align 8
-; AVX2: LV: Found an estimated cost of 7 for VF 4 For instruction: %v2 = load double, ptr %in2, align 8
-; AVX2: LV: Found an estimated cost of 7 for VF 4 For instruction: %v3 = load double, ptr %in3, align 8
-; AVX2: LV: Found an estimated cost of 7 for VF 4 For instruction: %v4 = load double, ptr %in4, align 8
-; AVX2: LV: Found an estimated cost of 14 for VF 8 For instruction: %v0 = load double, ptr %in0, align 8
-; AVX2: LV: Found an estimated cost of 14 for VF 8 For instruction: %v1 = load double, ptr %in1, align 8
-; AVX2: LV: Found an estimated cost of 14 for VF 8 For instruction: %v2 = load double, ptr %in2, align 8
-; AVX2: LV: Found an estimated cost of 14 for VF 8 For instruction: %v3 = load double, ptr %in3, align 8
-; AVX2: LV: Found an estimated cost of 14 for VF 8 For instruction: %v4 = load double, ptr %in4, align 8
-; AVX2: LV: Found an estimated cost of 28 for VF 16 For instruction: %v0 = load double, ptr %in0, align 8
-; AVX2: LV: Found an estimated cost of 28 for VF 16 For instruction: %v1 = load double, ptr %in1, align 8
-; AVX2: LV: Found an estimated cost of 28 for VF 16 For instruction: %v2 = load double, ptr %in2, align 8
-; AVX2: LV: Found an estimated cost of 28 for VF 16 For instruction: %v3 = load double, ptr %in3, align 8
-; AVX2: LV: Found an estimated cost of 28 for VF 16 For instruction: %v4 = load double, ptr %in4, align 8
-; AVX2: LV: Found an estimated cost of 56 for VF 32 For instruction: %v0 = load double, ptr %in0, align 8
-; AVX2: LV: Found an estimated cost of 56 for VF 32 For instruction: %v1 = load double, ptr %in1, align 8
-; AVX2: LV: Found an estimated cost of 56 for VF 32 For instruction: %v2 = load double, ptr %in2, align 8
-; AVX2: LV: Found an estimated cost of 56 for VF 32 For instruction: %v3 = load double, ptr %in3, align 8
-; AVX2: LV: Found an estimated cost of 56 for VF 32 For instruction: %v4 = load double, ptr %in4, align 8
+; AVX2: Cost of 3 for VF 2: REPLICATE ir<%v0> = load ir<%in0>
+; AVX2: Cost of 7 for VF 4: REPLICATE ir<%v0> = load ir<%in0>
+; AVX2: Cost of 14 for VF 8: REPLICATE ir<%v0> = load ir<%in0>
+; AVX2: Cost of 28 for VF 16: REPLICATE ir<%v0> = load ir<%in0>
+; AVX2: Cost of 56 for VF 32: REPLICATE ir<%v0> = load ir<%in0>
;
; AVX512-LABEL: 'test'
; AVX512: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load double, ptr %in0, align 8
-; AVX512: LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load double, ptr %in1, align 8
-; AVX512: LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load double, ptr %in2, align 8
-; AVX512: LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load double, ptr %in3, align 8
-; AVX512: LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load double, ptr %in4, align 8
-; AVX512: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load double, ptr %in0, align 8
-; AVX512: LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load double, ptr %in1, align 8
-; AVX512: LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load double, ptr %in2, align 8
-; AVX512: LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load double, ptr %in3, align 8
-; AVX512: LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load double, ptr %in4, align 8
-; AVX512: LV: Found an estimated cost of 9 for VF 2 For instruction: %v0 = load double, ptr %in0, align 8
-; AVX512: LV: Found an estimated cost of 0 for VF 2 For instruction: %v1 = load double, ptr %in1, align 8
-; AVX512: LV: Found an estimated cost of 0 for VF 2 For instruction: %v2 = load double, ptr %in2, align 8
-; AVX512: LV: Found an estimated cost of 0 for VF 2 For instruction: %v3 = load double, ptr %in3, align 8
-; AVX512: LV: Found an estimated cost of 0 for VF 2 For instruction: %v4 = load double, ptr %in4, align 8
-; AVX512: LV: Found an estimated cost of 18 for VF 4 For instruction: %v0 = load double, ptr %in0, align 8
-; AVX512: LV: Found an estimated cost of 0 for VF 4 For instruction: %v1 = load double, ptr %in1, align 8
-; AVX512: LV: Found an estimated cost of 0 for VF 4 For instruction: %v2 = load double, ptr %in2, align 8
-; AVX512: LV: Found an estimated cost of 0 for VF 4 For instruction: %v3 = load double, ptr %in3, align 8
-; AVX512: LV: Found an estimated cost of 0 for VF 4 For instruction: %v4 = load double, ptr %in4, align 8
-; AVX512: LV: Found an estimated cost of 35 for VF 8 For instruction: %v0 = load double, ptr %in0, align 8
-; AVX512: LV: Found an estimated cost of 0 for VF 8 For instruction: %v1 = load double, ptr %in1, align 8
-; AVX512: LV: Found an estimated cost of 0 for VF 8 For instruction: %v2 = load double, ptr %in2, align 8
-; AVX512: LV: Found an estimated cost of 0 for VF 8 For instruction: %v3 = load double, ptr %in3, align 8
-; AVX512: LV: Found an estimated cost of 0 for VF 8 For instruction: %v4 = load double, ptr %in4, align 8
-; AVX512: LV: Found an estimated cost of 20 for VF 16 For instruction: %v0 = load double, ptr %in0, align 8
-; AVX512: LV: Found an estimated cost of 20 for VF 16 For instruction: %v1 = load double, ptr %in1, align 8
-; AVX512: LV: Found an estimated cost of 20 for VF 16 For instruction: %v2 = load double, ptr %in2, align 8
-; AVX512: LV: Found an estimated cost of 20 for VF 16 For instruction: %v3 = load double, ptr %in3, align 8
-; AVX512: LV: Found an estimated cost of 20 for VF 16 For instruction: %v4 = load double, ptr %in4, align 8
-; AVX512: LV: Found an estimated cost of 40 for VF 32 For instruction: %v0 = load double, ptr %in0, align 8
-; AVX512: LV: Found an estimated cost of 40 for VF 32 For instruction: %v1 = load double, ptr %in1, align 8
-; AVX512: LV: Found an estimated cost of 40 for VF 32 For instruction: %v2 = load double, ptr %in2, align 8
-; AVX512: LV: Found an estimated cost of 40 for VF 32 For instruction: %v3 = load double, ptr %in3, align 8
-; AVX512: LV: Found an estimated cost of 40 for VF 32 For instruction: %v4 = load double, ptr %in4, align 8
+; AVX512: Cost of 9 for VF 2: INTERLEAVE-GROUP with factor 5 at %v0, ir<%in0>
+; AVX512: Cost of 18 for VF 4: INTERLEAVE-GROUP with factor 5 at %v0, ir<%in0>
+; AVX512: Cost of 35 for VF 8: INTERLEAVE-GROUP with factor 5 at %v0, ir<%in0>
----------------
fhahn wrote:
Should be added back, thanks
https://github.com/llvm/llvm-project/pull/190038
More information about the llvm-commits
mailing list