From 9a17eb199cdc853ef3803f9d93260e6d0c328254 Mon Sep 17 00:00:00 2001
From: Simon Pilgrim <llvm-dev@redking.me.uk>
Date: Thu, 10 Aug 2017 19:02:51 +0000
Subject: [PATCH] [CostModel][X86] Add avx1 two-src shuffle costs

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@310650 91177308-0d34-0410-b5e6-96231b3b80d8
---
 lib/Target/X86/X86TargetTransformInfo.cpp     |  9 +++++
 .../CostModel/X86/shuffle-single-src.ll       | 18 +++++-----
 .../Analysis/CostModel/X86/shuffle-two-src.ll | 34 +++++++++----------
 3 files changed, 35 insertions(+), 26 deletions(-)

diff --git a/lib/Target/X86/X86TargetTransformInfo.cpp b/lib/Target/X86/X86TargetTransformInfo.cpp
index 54074f970f5..03dff0fb895 100644
--- a/lib/Target/X86/X86TargetTransformInfo.cpp
+++ b/lib/Target/X86/X86TargetTransformInfo.cpp
@@ -893,6 +893,15 @@ int X86TTIImpl::getShuffleCost(TTI::ShuffleKind Kind, Type *Tp, int Index,
                                                   // + 2*por + vinsertf128
     { TTI::SK_PermuteSingleSrc, MVT::v32i8,  8 }, // vextractf128 + 4*pshufb
                                                   // + 2*por + vinsertf128
+
+    { TTI::SK_PermuteTwoSrc,    MVT::v4f64,   4 }, // 2*vperm2f128 + 2*vshufpd
+    { TTI::SK_PermuteTwoSrc,    MVT::v8f32,   4 }, // 2*vperm2f128 + 2*vshufps
+    { TTI::SK_PermuteTwoSrc,    MVT::v4i64,   4 }, // 2*vperm2f128 + 2*vshufpd
+    { TTI::SK_PermuteTwoSrc,    MVT::v8i32,   4 }, // 2*vperm2f128 + 2*vshufps
+    { TTI::SK_PermuteTwoSrc,    MVT::v16i16, 15 }, // 2*vextractf128 + 8*pshufb
+                                                   // + 4*por + vinsertf128
+    { TTI::SK_PermuteTwoSrc,    MVT::v32i8,  15 }, // 2*vextractf128 + 8*pshufb
+                                                   // + 4*por + vinsertf128
   };
 
   if (ST->hasAVX())
diff --git a/test/Analysis/CostModel/X86/shuffle-single-src.ll b/test/Analysis/CostModel/X86/shuffle-single-src.ll
index 15aa3f7aabf..842c72a335a 100644
--- a/test/Analysis/CostModel/X86/shuffle-single-src.ll
+++ b/test/Analysis/CostModel/X86/shuffle-single-src.ll
@@ -33,7 +33,7 @@ define void @test_vXf64(<2 x double> %src128, <4 x double> %src256, <8 x double>
   ; SSE2: cost of 12 {{.*}} %V512 = shufflevector
   ; SSSE3: cost of 12 {{.*}} %V512 = shufflevector
   ; SSE42: cost of 12 {{.*}} %V512 = shufflevector
-  ; AVX1: cost of 12 {{.*}} %V512 = shufflevector
+  ; AVX1: cost of 8 {{.*}} %V512 = shufflevector
   ; AVX2: cost of 6 {{.*}} %V512 = shufflevector
   ; AVX512: cost of 1 {{.*}} %V512 = shufflevector
   %V512 = shufflevector <8 x double> %src512, <8 x double> undef, <8 x i32> <i32 7, i32 6, i32 6, i32 4, i32 3, i32 2, i32 1, i32 0>
@@ -41,7 +41,7 @@ define void @test_vXf64(<2 x double> %src128, <4 x double> %src256, <8 x double>
   ; SSE2: cost of 56 {{.*}} %V1024 = shufflevector
   ; SSSE3: cost of 56 {{.*}} %V1024 = shufflevector
   ; SSE42: cost of 56 {{.*}} %V1024 = shufflevector
-  ; AVX1: cost of 72 {{.*}} %V1024 = shufflevector
+  ; AVX1: cost of 48 {{.*}} %V1024 = shufflevector
   ; AVX2: cost of 36 {{.*}} %V1024 = shufflevector
   ; AVX512: cost of 2 {{.*}} %V1024 = shufflevector
   %V1024 = shufflevector <16 x double> %src1024, <16 x double> undef, <16 x i32> <i32 15, i32 14, i32 13, i32 13, i32 11, i32 10, i32 8, i32 8, i32 7, i32 6, i32 5, i32 4, i32 3, i32 2, i32 1, i32 0>
@@ -71,7 +71,7 @@ define void @test_vXi64(<2 x i64> %src128, <4 x i64> %src256, <8 x i64> %src512)
   ; SSE2: cost of 12 {{.*}} %V512 = shufflevector
   ; SSSE3: cost of 12 {{.*}} %V512 = shufflevector
   ; SSE42: cost of 12 {{.*}} %V512 = shufflevector
-  ; AVX1: cost of 16 {{.*}} %V512 = shufflevector
+  ; AVX1: cost of 8 {{.*}} %V512 = shufflevector
   ; AVX2: cost of 6 {{.*}} %V512 = shufflevector
   ; AVX512: cost of 1 {{.*}} %V512 = shufflevector
   %V512 = shufflevector <8 x i64> %src512, <8 x i64> undef, <8 x i32> <i32 7, i32 6, i32 6, i32 4, i32 3, i32 2, i32 1, i32 0>
@@ -101,7 +101,7 @@ define void @test_vXf32(<4 x float> %src128, <8 x float> %src256, <16 x float> %
   ; SSE2: cost of 24 {{.*}} %V512 = shufflevector
   ; SSSE3: cost of 24 {{.*}} %V512 = shufflevector
   ; SSE42: cost of 24 {{.*}} %V512 = shufflevector
-  ; AVX1: cost of 28 {{.*}} %V512 = shufflevector
+  ; AVX1: cost of 8 {{.*}} %V512 = shufflevector
   ; AVX2: cost of 6 {{.*}} %V512 = shufflevector
   ; AVX512: cost of 1 {{.*}} %V512 = shufflevector
   %V512 = shufflevector <16 x float> %src512, <16 x float> undef, <16 x i32> <i32 15, i32 14, i32 13, i32 13, i32 11, i32 10, i32 8, i32 8, i32 7, i32 6, i32 5, i32 4, i32 3, i32 2, i32 1, i32 0>
@@ -131,7 +131,7 @@ define void @test_vXi32(<4 x i32> %src128, <8 x i32> %src256, <16 x i32> %src512
   ; SSE2: cost of 24 {{.*}} %V512 = shufflevector
   ; SSSE3: cost of 24 {{.*}} %V512 = shufflevector
   ; SSE42: cost of 24 {{.*}} %V512 = shufflevector
-  ; AVX1: cost of 32 {{.*}} %V512 = shufflevector
+  ; AVX1: cost of 8 {{.*}} %V512 = shufflevector
   ; AVX2: cost of 6 {{.*}} %V512 = shufflevector
   ; AVX512: cost of 1 {{.*}} %V512 = shufflevector
   %V512 = shufflevector <16 x i32> %src512, <16 x i32> undef, <16 x i32> <i32 15, i32 14, i32 13, i32 12, i32 13, i32 10, i32 9, i32 8, i32 8, i32 6, i32 5, i32 4, i32 3, i32 2, i32 1, i32 0>
@@ -139,7 +139,7 @@ define void @test_vXi32(<4 x i32> %src128, <8 x i32> %src256, <16 x i32> %src512
   ; SSE2: cost of 112 {{.*}} %V1024 = shufflevector
   ; SSSE3: cost of 112 {{.*}} %V1024 = shufflevector
   ; SSE42: cost of 112 {{.*}} %V1024 = shufflevector
-  ; AVX1: cost of 192 {{.*}} %V1024 = shufflevector
+  ; AVX1: cost of 48 {{.*}} %V1024 = shufflevector
   ; AVX2: cost of 36 {{.*}} %V1024 = shufflevector
   ; AVX512: cost of 2 {{.*}} %V1024 = shufflevector
   %V1024 = shufflevector <32 x i32> %src1024, <32 x i32> undef, <32 x i32> <i32 31, i32 30, i32 20, i32 28, i32 27, i32 26, i32 25, i32 24, i32 23, i32 22, i32 21, i32 20, i32 19, i32 18, i32 17, i32 16, i32 15, i32 14, i32 13, i32 12, i32 11, i32 11, i32 9, i32 8, i32 7, i32 11, i32 5, i32 4, i32 3, i32 2, i32 1, i32 0>
@@ -172,7 +172,7 @@ define void @test_vXi16(<8 x i16> %src128, <16 x i16> %src256, <32 x i16> %src51
   ; SSE2: cost of 192 {{.*}} %V512 = shufflevector
   ; SSSE3: cost of 36 {{.*}} %V512 = shufflevector
   ; SSE42: cost of 36 {{.*}} %V512 = shufflevector
-  ; AVX1: cost of 64 {{.*}} %V512 = shufflevector
+  ; AVX1: cost of 30 {{.*}} %V512 = shufflevector
   ; AVX2: cost of 14 {{.*}} %V512 = shufflevector
   ; AVX512F: cost of 14 {{.*}} %V512 = shufflevector
   ; AVX512BW: cost of 1 {{.*}} %V512 = shufflevector
@@ -182,7 +182,7 @@ define void @test_vXi16(<8 x i16> %src128, <16 x i16> %src256, <32 x i16> %src51
   ; SSE2: cost of 896 {{.*}} %V1024 = shufflevector
   ; SSSE3: cost of 168 {{.*}} %V1024 = shufflevector
   ; SSE42: cost of 168 {{.*}} %V1024 = shufflevector
-  ; AVX1: cost of 384 {{.*}} %V1024 = shufflevector
+  ; AVX1: cost of 180 {{.*}} %V1024 = shufflevector
   ; AVX2: cost of 84 {{.*}} %V1024 = shufflevector
   ; AVX512F: cost of 84 {{.*}} %V1024 = shufflevector
   ; AVX512BW: cost of 2 {{.*}} %V1024 = shufflevector
@@ -214,7 +214,7 @@ define void @test_vXi8(<16 x i8> %src128, <32 x i8> %src256, <64 x i8> %src512)
   ; SSE2: cost of 384 {{.*}} %V512 = shufflevector
   ; SSSE3: cost of 36 {{.*}} %V512 = shufflevector
   ; SSE42: cost of 36 {{.*}} %V512 = shufflevector
-  ; AVX1: cost of 128 {{.*}} %V512 = shufflevector
+  ; AVX1: cost of 30 {{.*}} %V512 = shufflevector
   ; AVX2: cost of 14 {{.*}} %V512 = shufflevector
   ; AVX512F: cost of 14 {{.*}} %V512 = shufflevector
   ; AVX512BW: cost of 8 {{.*}} %V512 = shufflevector
diff --git a/test/Analysis/CostModel/X86/shuffle-two-src.ll b/test/Analysis/CostModel/X86/shuffle-two-src.ll
index 33e08dcb9f4..dc6f5ceff50 100644
--- a/test/Analysis/CostModel/X86/shuffle-two-src.ll
+++ b/test/Analysis/CostModel/X86/shuffle-two-src.ll
@@ -25,7 +25,7 @@ define void @test_vXf64(<2 x double> %src128, <4 x double> %src256, <8 x double>
   ; SSE2: cost of 6 {{.*}} %V256 = shufflevector
   ; SSSE3: cost of 6 {{.*}} %V256 = shufflevector
   ; SSE42: cost of 6 {{.*}} %V256 = shufflevector
-  ; AVX1: cost of 6 {{.*}} %V256 = shufflevector
+  ; AVX1: cost of 4 {{.*}} %V256 = shufflevector
   ; AVX2: cost of 3 {{.*}} %V256 = shufflevector
   ; AVX512: cost of 1 {{.*}} %V256 = shufflevector
   %V256 = shufflevector <4 x double> %src256, <4 x double> %src256_1, <4 x i32> <i32 3, i32 3, i32 7, i32 6>
@@ -33,7 +33,7 @@ define void @test_vXf64(<2 x double> %src128, <4 x double> %src256, <8 x double>
   ; SSE2: cost of 28 {{.*}} %V512 = shufflevector
   ; SSSE3: cost of 28 {{.*}} %V512 = shufflevector
   ; SSE42: cost of 28 {{.*}} %V512 = shufflevector
-  ; AVX1: cost of 12 {{.*}} %V512 = shufflevector
+  ; AVX1: cost of 24 {{.*}} %V512 = shufflevector
   ; AVX2: cost of 18 {{.*}} %V512 = shufflevector
   ; AVX512: cost of 1 {{.*}} %V512 = shufflevector
   %V512 = shufflevector <8 x double> %src512, <8 x double> %src512_1, <8 x i32> <i32 7, i32 6, i32 12, i32 4, i32 3, i32 2, i32 1, i32 15>
@@ -41,7 +41,7 @@ define void @test_vXf64(<2 x double> %src128, <4 x double> %src256, <8 x double>
   ; SSE2: cost of 120 {{.*}} %V1024 = shufflevector
   ; SSSE3: cost of 120 {{.*}} %V1024 = shufflevector
   ; SSE42: cost of 120 {{.*}} %V1024 = shufflevector
-  ; AVX1: cost of 24 {{.*}} %V1024 = shufflevector
+  ; AVX1: cost of 112 {{.*}} %V1024 = shufflevector
   ; AVX2: cost of 84 {{.*}} %V1024 = shufflevector
   ; AVX512: cost of 6 {{.*}} %V1024 = shufflevector
   %V1024 = shufflevector <16 x double> %src1024, <16 x double> %src1024_1, <16 x i32> <i32 30, i32 14, i32 13, i32 12, i32 13, i32 10, i32 18, i32 8, i32 8, i32 6, i32 5, i32 4, i32 3, i32 2, i32 1, i32 0>
@@ -63,7 +63,7 @@ define void @test_vXi64(<2 x i64> %src128, <4 x i64> %src256, <8 x i64> %src512,
   ; SSE2: cost of 6 {{.*}} %V256 = shufflevector
   ; SSSE3: cost of 6 {{.*}} %V256 = shufflevector
   ; SSE42: cost of 6 {{.*}} %V256 = shufflevector
-  ; AVX1: cost of 8 {{.*}} %V256 = shufflevector
+  ; AVX1: cost of 4 {{.*}} %V256 = shufflevector
   ; AVX2: cost of 3 {{.*}} %V256 = shufflevector
   ; AVX512: cost of 1 {{.*}} %V256 = shufflevector
   %V256 = shufflevector <4 x i64> %src256, <4 x i64> %src256_1, <4 x i32> <i32 3, i32 3, i32 7, i32 6>
@@ -71,7 +71,7 @@ define void @test_vXi64(<2 x i64> %src128, <4 x i64> %src256, <8 x i64> %src512,
   ; SSE2: cost of 28 {{.*}} %V512 = shufflevector
   ; SSSE3: cost of 28 {{.*}} %V512 = shufflevector
   ; SSE42: cost of 28 {{.*}} %V512 = shufflevector
-  ; AVX1: cost of 16 {{.*}} %V512 = shufflevector
+  ; AVX1: cost of 24 {{.*}} %V512 = shufflevector
   ; AVX2: cost of 18 {{.*}} %V512 = shufflevector
   ; AVX512: cost of 1 {{.*}} %V512 = shufflevector
   %V512 = shufflevector <8 x i64> %src512, <8 x i64> %src512_1, <8 x i32> <i32 7, i32 6, i32 12, i32 4, i32 3, i32 2, i32 1, i32 15>
@@ -79,7 +79,7 @@ define void @test_vXi64(<2 x i64> %src128, <4 x i64> %src256, <8 x i64> %src512,
   ; SSE2: cost of 120 {{.*}} %V1024 = shufflevector
   ; SSSE3: cost of 120 {{.*}} %V1024 = shufflevector
   ; SSE42: cost of 120 {{.*}} %V1024 = shufflevector
-  ; AVX1: cost of 32 {{.*}} %V1024 = shufflevector
+  ; AVX1: cost of 112 {{.*}} %V1024 = shufflevector
   ; AVX2: cost of 84 {{.*}} %V1024 = shufflevector
   ; AVX512: cost of 6 {{.*}} %V1024 = shufflevector
   %V1024 = shufflevector <16 x i64> %src1024, <16 x i64> %src1024_1, <16 x i32> <i32 30, i32 14, i32 13, i32 12, i32 13, i32 10, i32 18, i32 8, i32 8, i32 6, i32 5, i32 4, i32 3, i32 2, i32 1, i32 0>
@@ -101,7 +101,7 @@ define void @test_vXf32(<4 x float> %src128, <8 x float> %src256, <16 x float> %
   ; SSE2: cost of 12 {{.*}} %V256 = shufflevector
   ; SSSE3: cost of 12 {{.*}} %V256 = shufflevector
   ; SSE42: cost of 12 {{.*}} %V256 = shufflevector
-  ; AVX1: cost of 14 {{.*}} %V256 = shufflevector
+  ; AVX1: cost of 4 {{.*}} %V256 = shufflevector
   ; AVX2: cost of 3 {{.*}} %V256 = shufflevector
   ; AVX512: cost of 1 {{.*}} %V256 = shufflevector
   %V256 = shufflevector <8 x float> %src256, <8 x float> %src256_1, <8 x i32> <i32 7, i32 6, i32 8, i32 4, i32 3, i32 2, i32 12, i32 0>
@@ -109,7 +109,7 @@ define void @test_vXf32(<4 x float> %src128, <8 x float> %src256, <16 x float> %
   ; SSE2: cost of 56 {{.*}} %V512 = shufflevector
   ; SSSE3: cost of 56 {{.*}} %V512 = shufflevector
   ; SSE42: cost of 56 {{.*}} %V512 = shufflevector
-  ; AVX1: cost of 28 {{.*}} %V512 = shufflevector
+  ; AVX1: cost of 24 {{.*}} %V512 = shufflevector
   ; AVX2: cost of 18 {{.*}} %V512 = shufflevector
   ; AVX512: cost of 1 {{.*}} %V512 = shufflevector
   %V512 = shufflevector <16 x float> %src512, <16 x float> %src512_1, <16 x i32> <i32 15, i32 17, i32 13, i32 20, i32 11, i32 10, i32 8, i32 8, i32 7, i32 22, i32 5, i32 4, i32 3, i32 2, i32 1, i32 0>
@@ -117,7 +117,7 @@ define void @test_vXf32(<4 x float> %src128, <8 x float> %src256, <16 x float> %
   ; SSE2: cost of 240 {{.*}} %V1024 = shufflevector
   ; SSSE3: cost of 240 {{.*}} %V1024 = shufflevector
   ; SSE42: cost of 240 {{.*}} %V1024 = shufflevector
-  ; AVX1: cost of 56 {{.*}} %V1024 = shufflevector
+  ; AVX1: cost of 112 {{.*}} %V1024 = shufflevector
   ; AVX2: cost of 84 {{.*}} %V1024 = shufflevector
   ; AVX512: cost of 6 {{.*}} %V1024 = shufflevector
   %V1024 = shufflevector <32 x float> %src1024, <32 x float> %src1024_1, <32 x i32> <i32 31, i32 33, i32 20, i32 28, i32 27, i32 26, i32 25, i32 24, i32 23, i32 22, i32 21, i32 20, i32 19, i32 18, i32 17, i32 16, i32 15, i32 48, i32 13, i32 12, i32 11, i32 11, i32 9, i32 45, i32 7, i32 11, i32 5, i32 4, i32 3, i32 2, i32 1, i32 0>
@@ -139,7 +139,7 @@ define void @test_vXi32(<4 x i32> %src128, <8 x i32> %src256, <16 x i32> %src512
   ; SSE2: cost of 12 {{.*}} %V256 = shufflevector
   ; SSSE3: cost of 12 {{.*}} %V256 = shufflevector
   ; SSE42: cost of 12 {{.*}} %V256 = shufflevector
-  ; AVX1: cost of 16 {{.*}} %V256 = shufflevector
+  ; AVX1: cost of 4 {{.*}} %V256 = shufflevector
   ; AVX2: cost of 3 {{.*}} %V256 = shufflevector
   ; AVX512: cost of 1 {{.*}} %V256 = shufflevector
   %V256 = shufflevector <8 x i32> %src256, <8 x i32> %src256_1, <8 x i32> <i32 7, i32 6, i32 8, i32 4, i32 3, i32 2, i32 12, i32 0>
@@ -147,7 +147,7 @@ define void @test_vXi32(<4 x i32> %src128, <8 x i32> %src256, <16 x i32> %src512
   ; SSE2: cost of 56 {{.*}} %V512 = shufflevector
   ; SSSE3: cost of 56 {{.*}} %V512 = shufflevector
   ; SSE42: cost of 56 {{.*}} %V512 = shufflevector
-  ; AVX1: cost of 32 {{.*}} %V512 = shufflevector
+  ; AVX1: cost of 24 {{.*}} %V512 = shufflevector
   ; AVX2: cost of 18 {{.*}} %V512 = shufflevector
   ; AVX512: cost of 1 {{.*}} %V512 = shufflevector
   %V512 = shufflevector <16 x i32> %src512, <16 x i32> %src512_1, <16 x i32> <i32 15, i32 17, i32 13, i32 20, i32 11, i32 10, i32 8, i32 8, i32 7, i32 22, i32 5, i32 4, i32 3, i32 2, i32 1, i32 0>
@@ -155,7 +155,7 @@ define void @test_vXi32(<4 x i32> %src128, <8 x i32> %src256, <16 x i32> %src512
   ; SSE2: cost of 240 {{.*}} %V1024 = shufflevector
   ; SSSE3: cost of 240 {{.*}} %V1024 = shufflevector
   ; SSE42: cost of 240 {{.*}} %V1024 = shufflevector
-  ; AVX1: cost of 64 {{.*}} %V1024 = shufflevector
+  ; AVX1: cost of 112 {{.*}} %V1024 = shufflevector
   ; AVX2: cost of 84 {{.*}} %V1024 = shufflevector
   ; AVX512: cost of 6 {{.*}} %V1024 = shufflevector
   %V1024 = shufflevector <32 x i32> %src1024, <32 x i32> %src1024_1, <32 x i32> <i32 31, i32 33, i32 20, i32 28, i32 27, i32 26, i32 25, i32 24, i32 23, i32 22, i32 21, i32 20, i32 19, i32 18, i32 17, i32 16, i32 15, i32 48, i32 13, i32 12, i32 11, i32 11, i32 9, i32 45, i32 7, i32 11, i32 5, i32 4, i32 3, i32 2, i32 1, i32 0>
@@ -179,7 +179,7 @@ define void @test_vXi16(<8 x i16> %src128, <16 x i16> %src256, <32 x i16> %src51
   ; SSE2: cost of 32 {{.*}} %V256 = shufflevector
   ; SSSE3: cost of 18 {{.*}} %V256 = shufflevector
   ; SSE42: cost of 18 {{.*}} %V256 = shufflevector
-  ; AVX1: cost of 32 {{.*}} %V256 = shufflevector
+  ; AVX1: cost of 15 {{.*}} %V256 = shufflevector
   ; AVX2: cost of 7 {{.*}} %V256 = shufflevector
   ; AVX512F: cost of 7 {{.*}} %V256 = shufflevector
   ; AVX512BW: cost of 1 {{.*}} %V256 = shufflevector
@@ -189,7 +189,7 @@ define void @test_vXi16(<8 x i16> %src128, <16 x i16> %src256, <32 x i16> %src51
   ; SSE2: cost of 64 {{.*}} %V512 = shufflevector
   ; SSSE3: cost of 84 {{.*}} %V512 = shufflevector
   ; SSE42: cost of 84 {{.*}} %V512 = shufflevector
-  ; AVX1: cost of 64 {{.*}} %V512 = shufflevector
+  ; AVX1: cost of 90 {{.*}} %V512 = shufflevector
   ; AVX2: cost of 42 {{.*}} %V512 = shufflevector
   ; AVX512F: cost of 42 {{.*}} %V512 = shufflevector
   ; AVX512BW: cost of 1 {{.*}} %V512 = shufflevector
@@ -199,7 +199,7 @@ define void @test_vXi16(<8 x i16> %src128, <16 x i16> %src256, <32 x i16> %src51
   ; SSE2: cost of 128 {{.*}} %V1024 = shufflevector
   ; SSSE3: cost of 360 {{.*}} %V1024 = shufflevector
   ; SSE42: cost of 360 {{.*}} %V1024 = shufflevector
-  ; AVX1: cost of 128 {{.*}} %V1024 = shufflevector
+  ; AVX1: cost of 420 {{.*}} %V1024 = shufflevector
   ; AVX2: cost of 196 {{.*}} %V1024 = shufflevector
   ; AVX512F: cost of 196 {{.*}} %V1024 = shufflevector
   ; AVX512BW: cost of 6 {{.*}} %V1024 = shufflevector
@@ -225,7 +225,7 @@ define void @test_vXi8(<16 x i8> %src128, <32 x i8> %src256, <64 x i8> %src512,
   ; SSE2: cost of 64 {{.*}} %V256 = shufflevector
   ; SSSE3: cost of 18 {{.*}} %V256 = shufflevector
   ; SSE42: cost of 18 {{.*}} %V256 = shufflevector
-  ; AVX1: cost of 64 {{.*}} %V256 = shufflevector
+  ; AVX1: cost of 15 {{.*}} %V256 = shufflevector
   ; AVX2: cost of 7 {{.*}} %V256 = shufflevector
   ; AVX512F: cost of 7 {{.*}} %V256 = shufflevector
   ; AVX512BW: cost of 3 {{.*}} %V256 = shufflevector
@@ -235,7 +235,7 @@ define void @test_vXi8(<16 x i8> %src128, <32 x i8> %src256, <64 x i8> %src512,
   ; SSE2: cost of 128 {{.*}} %V512 = shufflevector
   ; SSSE3: cost of 84 {{.*}} %V512 = shufflevector
   ; SSE42: cost of 84 {{.*}} %V512 = shufflevector
-  ; AVX1: cost of 128 {{.*}} %V512 = shufflevector
+  ; AVX1: cost of 90 {{.*}} %V512 = shufflevector
   ; AVX2: cost of 42 {{.*}} %V512 = shufflevector
   ; AVX512F: cost of 42 {{.*}} %V512 = shufflevector
   ; AVX512BW: cost of 19 {{.*}} %V512 = shufflevector
-- 
2.40.0