[CostModel][X86] Add some specific cpu targets to the cost models

We're mostly testing with generic isa attributes, but PR36550 will require testing of specific target's scheduler models as well.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@330056 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/test/Analysis/CostModel/X86/alternate-shuffle-cost.ll b/test/Analysis/CostModel/X86/alternate-shuffle-cost.ll
index b498e42..ee22675 100644 (file)
--- a/test/Analysis/CostModel/X86/alternate-shuffle-cost.ll
+++ b/test/Analysis/CostModel/X86/alternate-shuffle-cost.ll
@@ -1,11 +1,15 @@
 ; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+sse2 | FileCheck %s -check-prefixes=CHECK,SSE,SSE2
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+ssse3 | FileCheck %s -check-prefixes=CHECK,SSE,SSSE3
-; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+sse4.2 | FileCheck %s -check-prefixes=CHECK,SSE,SSE41
+; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+sse4.2 | FileCheck %s -check-prefixes=CHECK,SSE,SSE42
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+avx | FileCheck %s -check-prefixes=CHECK,AVX,AVX1
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+avx2 | FileCheck %s -check-prefixes=CHECK,AVX,AVX2
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+xop,+avx | FileCheck %s -check-prefixes=CHECK,AVX,AVX1
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+xop,+avx2 | FileCheck %s -check-prefixes=CHECK,AVX,AVX2
+;
+; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mcpu=slm | FileCheck %s --check-prefixes=CHECK,SSE,SSE42
+; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mcpu=goldmont | FileCheck %s --check-prefixes=CHECK,SSE,SSE42
+; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mcpu=btver2 | FileCheck %s --check-prefixes=CHECK,AVX,AVX1
 
 ; Verify the cost model for alternate shuffles.
 
@@ -31,9 +35,9 @@ define <2 x float> @test_v2f32(<2 x float> %a, <2 x float> %b) {
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %1 = shufflevector <2 x float> %a, <2 x float> %b, <2 x i32> <i32 0, i32 3>
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <2 x float> %1
 ;
-; SSE41-LABEL: 'test_v2f32'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = shufflevector <2 x float> %a, <2 x float> %b, <2 x i32> <i32 0, i32 3>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <2 x float> %1
+; SSE42-LABEL: 'test_v2f32'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = shufflevector <2 x float> %a, <2 x float> %b, <2 x i32> <i32 0, i32 3>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <2 x float> %1
 ;
 ; AVX-LABEL: 'test_v2f32'
 ; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = shufflevector <2 x float> %a, <2 x float> %b, <2 x i32> <i32 0, i32 3>
@@ -61,9 +65,9 @@ define <2 x float> @test_v2f32_2(<2 x float> %a, <2 x float> %b) {
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %1 = shufflevector <2 x float> %a, <2 x float> %b, <2 x i32> <i32 2, i32 1>
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <2 x float> %1
 ;
-; SSE41-LABEL: 'test_v2f32_2'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = shufflevector <2 x float> %a, <2 x float> %b, <2 x i32> <i32 2, i32 1>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <2 x float> %1
+; SSE42-LABEL: 'test_v2f32_2'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = shufflevector <2 x float> %a, <2 x float> %b, <2 x i32> <i32 2, i32 1>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <2 x float> %1
 ;
 ; AVX-LABEL: 'test_v2f32_2'
 ; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = shufflevector <2 x float> %a, <2 x float> %b, <2 x i32> <i32 2, i32 1>
@@ -122,9 +126,9 @@ define <4 x i32> @test_v4i32(<4 x i32> %a, <4 x i32> %b) {
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %1 = shufflevector <4 x i32> %a, <4 x i32> %b, <4 x i32> <i32 0, i32 5, i32 2, i32 7>
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %1
 ;
-; SSE41-LABEL: 'test_v4i32'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = shufflevector <4 x i32> %a, <4 x i32> %b, <4 x i32> <i32 0, i32 5, i32 2, i32 7>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %1
+; SSE42-LABEL: 'test_v4i32'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = shufflevector <4 x i32> %a, <4 x i32> %b, <4 x i32> <i32 0, i32 5, i32 2, i32 7>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %1
 ;
 ; AVX-LABEL: 'test_v4i32'
 ; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = shufflevector <4 x i32> %a, <4 x i32> %b, <4 x i32> <i32 0, i32 5, i32 2, i32 7>
@@ -143,9 +147,9 @@ define <4 x i32> @test_v4i32_2(<4 x i32> %a, <4 x i32> %b) {
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %1 = shufflevector <4 x i32> %a, <4 x i32> %b, <4 x i32> <i32 4, i32 1, i32 6, i32 3>
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %1
 ;
-; SSE41-LABEL: 'test_v4i32_2'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = shufflevector <4 x i32> %a, <4 x i32> %b, <4 x i32> <i32 4, i32 1, i32 6, i32 3>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %1
+; SSE42-LABEL: 'test_v4i32_2'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = shufflevector <4 x i32> %a, <4 x i32> %b, <4 x i32> <i32 4, i32 1, i32 6, i32 3>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %1
 ;
 ; AVX-LABEL: 'test_v4i32_2'
 ; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = shufflevector <4 x i32> %a, <4 x i32> %b, <4 x i32> <i32 4, i32 1, i32 6, i32 3>
@@ -164,9 +168,9 @@ define <4 x float> @test_v4f32(<4 x float> %a, <4 x float> %b) {
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %1 = shufflevector <4 x float> %a, <4 x float> %b, <4 x i32> <i32 0, i32 5, i32 2, i32 7>
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x float> %1
 ;
-; SSE41-LABEL: 'test_v4f32'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = shufflevector <4 x float> %a, <4 x float> %b, <4 x i32> <i32 0, i32 5, i32 2, i32 7>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x float> %1
+; SSE42-LABEL: 'test_v4f32'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = shufflevector <4 x float> %a, <4 x float> %b, <4 x i32> <i32 0, i32 5, i32 2, i32 7>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x float> %1
 ;
 ; AVX-LABEL: 'test_v4f32'
 ; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = shufflevector <4 x float> %a, <4 x float> %b, <4 x i32> <i32 0, i32 5, i32 2, i32 7>
@@ -185,9 +189,9 @@ define <4 x float> @test_v4f32_2(<4 x float> %a, <4 x float> %b) {
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %1 = shufflevector <4 x float> %a, <4 x float> %b, <4 x i32> <i32 4, i32 1, i32 6, i32 3>
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x float> %1
 ;
-; SSE41-LABEL: 'test_v4f32_2'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = shufflevector <4 x float> %a, <4 x float> %b, <4 x i32> <i32 4, i32 1, i32 6, i32 3>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x float> %1
+; SSE42-LABEL: 'test_v4f32_2'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = shufflevector <4 x float> %a, <4 x float> %b, <4 x i32> <i32 4, i32 1, i32 6, i32 3>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x float> %1
 ;
 ; AVX-LABEL: 'test_v4f32_2'
 ; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = shufflevector <4 x float> %a, <4 x float> %b, <4 x i32> <i32 4, i32 1, i32 6, i32 3>
@@ -260,9 +264,9 @@ define <8 x i16> @test_v8i16(<8 x i16> %a, <8 x i16> %b) {
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %1 = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 0, i32 9, i32 2, i32 11, i32 4, i32 13, i32 6, i32 15>
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %1
 ;
-; SSE41-LABEL: 'test_v8i16'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 0, i32 9, i32 2, i32 11, i32 4, i32 13, i32 6, i32 15>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %1
+; SSE42-LABEL: 'test_v8i16'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 0, i32 9, i32 2, i32 11, i32 4, i32 13, i32 6, i32 15>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %1
 ;
 ; AVX-LABEL: 'test_v8i16'
 ; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 0, i32 9, i32 2, i32 11, i32 4, i32 13, i32 6, i32 15>
@@ -281,9 +285,9 @@ define <8 x i16> @test_v8i16_2(<8 x i16> %a, <8 x i16> %b) {
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %1 = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 8, i32 1, i32 10, i32 3, i32 12, i32 5, i32 14, i32 7>
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %1
 ;
-; SSE41-LABEL: 'test_v8i16_2'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 8, i32 1, i32 10, i32 3, i32 12, i32 5, i32 14, i32 7>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %1
+; SSE42-LABEL: 'test_v8i16_2'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 8, i32 1, i32 10, i32 3, i32 12, i32 5, i32 14, i32 7>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %1
 ;
 ; AVX-LABEL: 'test_v8i16_2'
 ; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 8, i32 1, i32 10, i32 3, i32 12, i32 5, i32 14, i32 7>
@@ -302,9 +306,9 @@ define <8 x i32> @test_v8i32(<8 x i32> %a, <8 x i32> %b) {
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %1 = shufflevector <8 x i32> %a, <8 x i32> %b, <8 x i32> <i32 0, i32 9, i32 2, i32 11, i32 4, i32 13, i32 6, i32 15>
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %1
 ;
-; SSE41-LABEL: 'test_v8i32'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %1 = shufflevector <8 x i32> %a, <8 x i32> %b, <8 x i32> <i32 0, i32 9, i32 2, i32 11, i32 4, i32 13, i32 6, i32 15>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %1
+; SSE42-LABEL: 'test_v8i32'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %1 = shufflevector <8 x i32> %a, <8 x i32> %b, <8 x i32> <i32 0, i32 9, i32 2, i32 11, i32 4, i32 13, i32 6, i32 15>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %1
 ;
 ; AVX-LABEL: 'test_v8i32'
 ; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = shufflevector <8 x i32> %a, <8 x i32> %b, <8 x i32> <i32 0, i32 9, i32 2, i32 11, i32 4, i32 13, i32 6, i32 15>
@@ -323,9 +327,9 @@ define <8 x i32> @test_v8i32_2(<8 x i32> %a, <8 x i32> %b) {
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %1 = shufflevector <8 x i32> %a, <8 x i32> %b, <8 x i32> <i32 8, i32 1, i32 10, i32 3, i32 12, i32 5, i32 14, i32 7>
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %1
 ;
-; SSE41-LABEL: 'test_v8i32_2'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %1 = shufflevector <8 x i32> %a, <8 x i32> %b, <8 x i32> <i32 8, i32 1, i32 10, i32 3, i32 12, i32 5, i32 14, i32 7>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %1
+; SSE42-LABEL: 'test_v8i32_2'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %1 = shufflevector <8 x i32> %a, <8 x i32> %b, <8 x i32> <i32 8, i32 1, i32 10, i32 3, i32 12, i32 5, i32 14, i32 7>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %1
 ;
 ; AVX-LABEL: 'test_v8i32_2'
 ; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = shufflevector <8 x i32> %a, <8 x i32> %b, <8 x i32> <i32 8, i32 1, i32 10, i32 3, i32 12, i32 5, i32 14, i32 7>
@@ -344,9 +348,9 @@ define <8 x float> @test_v8f32(<8 x float> %a, <8 x float> %b) {
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %1 = shufflevector <8 x float> %a, <8 x float> %b, <8 x i32> <i32 0, i32 9, i32 2, i32 11, i32 4, i32 13, i32 6, i32 15>
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x float> %1
 ;
-; SSE41-LABEL: 'test_v8f32'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %1 = shufflevector <8 x float> %a, <8 x float> %b, <8 x i32> <i32 0, i32 9, i32 2, i32 11, i32 4, i32 13, i32 6, i32 15>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x float> %1
+; SSE42-LABEL: 'test_v8f32'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %1 = shufflevector <8 x float> %a, <8 x float> %b, <8 x i32> <i32 0, i32 9, i32 2, i32 11, i32 4, i32 13, i32 6, i32 15>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x float> %1
 ;
 ; AVX-LABEL: 'test_v8f32'
 ; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = shufflevector <8 x float> %a, <8 x float> %b, <8 x i32> <i32 0, i32 9, i32 2, i32 11, i32 4, i32 13, i32 6, i32 15>
@@ -365,9 +369,9 @@ define <8 x float> @test_v8f32_2(<8 x float> %a, <8 x float> %b) {
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %1 = shufflevector <8 x float> %a, <8 x float> %b, <8 x i32> <i32 8, i32 1, i32 10, i32 3, i32 12, i32 5, i32 14, i32 7>
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x float> %1
 ;
-; SSE41-LABEL: 'test_v8f32_2'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %1 = shufflevector <8 x float> %a, <8 x float> %b, <8 x i32> <i32 8, i32 1, i32 10, i32 3, i32 12, i32 5, i32 14, i32 7>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x float> %1
+; SSE42-LABEL: 'test_v8f32_2'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %1 = shufflevector <8 x float> %a, <8 x float> %b, <8 x i32> <i32 8, i32 1, i32 10, i32 3, i32 12, i32 5, i32 14, i32 7>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x float> %1
 ;
 ; AVX-LABEL: 'test_v8f32_2'
 ; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = shufflevector <8 x float> %a, <8 x float> %b, <8 x i32> <i32 8, i32 1, i32 10, i32 3, i32 12, i32 5, i32 14, i32 7>
@@ -388,9 +392,9 @@ define <16 x i8> @test_v16i8(<16 x i8> %a, <16 x i8> %b) {
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %1 = shufflevector <16 x i8> %a, <16 x i8> %b, <16 x i32> <i32 0, i32 17, i32 2, i32 19, i32 4, i32 21, i32 6, i32 23, i32 8, i32 25, i32 10, i32 27, i32 12, i32 29, i32 14, i32 31>
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %1
 ;
-; SSE41-LABEL: 'test_v16i8'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = shufflevector <16 x i8> %a, <16 x i8> %b, <16 x i32> <i32 0, i32 17, i32 2, i32 19, i32 4, i32 21, i32 6, i32 23, i32 8, i32 25, i32 10, i32 27, i32 12, i32 29, i32 14, i32 31>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %1
+; SSE42-LABEL: 'test_v16i8'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = shufflevector <16 x i8> %a, <16 x i8> %b, <16 x i32> <i32 0, i32 17, i32 2, i32 19, i32 4, i32 21, i32 6, i32 23, i32 8, i32 25, i32 10, i32 27, i32 12, i32 29, i32 14, i32 31>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %1
 ;
 ; AVX-LABEL: 'test_v16i8'
 ; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = shufflevector <16 x i8> %a, <16 x i8> %b, <16 x i32> <i32 0, i32 17, i32 2, i32 19, i32 4, i32 21, i32 6, i32 23, i32 8, i32 25, i32 10, i32 27, i32 12, i32 29, i32 14, i32 31>
@@ -409,9 +413,9 @@ define <16 x i8> @test_v16i8_2(<16 x i8> %a, <16 x i8> %b) {
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %1 = shufflevector <16 x i8> %a, <16 x i8> %b, <16 x i32> <i32 16, i32 1, i32 18, i32 3, i32 20, i32 5, i32 22, i32 7, i32 24, i32 9, i32 26, i32 11, i32 28, i32 13, i32 30, i32 15>
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %1
 ;
-; SSE41-LABEL: 'test_v16i8_2'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = shufflevector <16 x i8> %a, <16 x i8> %b, <16 x i32> <i32 16, i32 1, i32 18, i32 3, i32 20, i32 5, i32 22, i32 7, i32 24, i32 9, i32 26, i32 11, i32 28, i32 13, i32 30, i32 15>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %1
+; SSE42-LABEL: 'test_v16i8_2'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = shufflevector <16 x i8> %a, <16 x i8> %b, <16 x i32> <i32 16, i32 1, i32 18, i32 3, i32 20, i32 5, i32 22, i32 7, i32 24, i32 9, i32 26, i32 11, i32 28, i32 13, i32 30, i32 15>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %1
 ;
 ; AVX-LABEL: 'test_v16i8_2'
 ; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = shufflevector <16 x i8> %a, <16 x i8> %b, <16 x i32> <i32 16, i32 1, i32 18, i32 3, i32 20, i32 5, i32 22, i32 7, i32 24, i32 9, i32 26, i32 11, i32 28, i32 13, i32 30, i32 15>
@@ -430,9 +434,9 @@ define <16 x i16> @test_v16i16(<16 x i16> %a, <16 x i16> %b) {
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %1 = shufflevector <16 x i16> %a, <16 x i16> %b, <16 x i32> <i32 0, i32 17, i32 2, i32 19, i32 4, i32 21, i32 6, i32 23, i32 8, i32 25, i32 10, i32 27, i32 12, i32 29, i32 14, i32 31>
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %1
 ;
-; SSE41-LABEL: 'test_v16i16'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %1 = shufflevector <16 x i16> %a, <16 x i16> %b, <16 x i32> <i32 0, i32 17, i32 2, i32 19, i32 4, i32 21, i32 6, i32 23, i32 8, i32 25, i32 10, i32 27, i32 12, i32 29, i32 14, i32 31>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %1
+; SSE42-LABEL: 'test_v16i16'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %1 = shufflevector <16 x i16> %a, <16 x i16> %b, <16 x i32> <i32 0, i32 17, i32 2, i32 19, i32 4, i32 21, i32 6, i32 23, i32 8, i32 25, i32 10, i32 27, i32 12, i32 29, i32 14, i32 31>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %1
 ;
 ; AVX1-LABEL: 'test_v16i16'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %1 = shufflevector <16 x i16> %a, <16 x i16> %b, <16 x i32> <i32 0, i32 17, i32 2, i32 19, i32 4, i32 21, i32 6, i32 23, i32 8, i32 25, i32 10, i32 27, i32 12, i32 29, i32 14, i32 31>
@@ -455,9 +459,9 @@ define <16 x i16> @test_v16i16_2(<16 x i16> %a, <16 x i16> %b) {
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %1 = shufflevector <16 x i16> %a, <16 x i16> %b, <16 x i32> <i32 16, i32 1, i32 18, i32 3, i32 20, i32 5, i32 22, i32 7, i32 24, i32 9, i32 26, i32 11, i32 28, i32 13, i32 30, i32 15>
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %1
 ;
-; SSE41-LABEL: 'test_v16i16_2'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %1 = shufflevector <16 x i16> %a, <16 x i16> %b, <16 x i32> <i32 16, i32 1, i32 18, i32 3, i32 20, i32 5, i32 22, i32 7, i32 24, i32 9, i32 26, i32 11, i32 28, i32 13, i32 30, i32 15>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %1
+; SSE42-LABEL: 'test_v16i16_2'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %1 = shufflevector <16 x i16> %a, <16 x i16> %b, <16 x i32> <i32 16, i32 1, i32 18, i32 3, i32 20, i32 5, i32 22, i32 7, i32 24, i32 9, i32 26, i32 11, i32 28, i32 13, i32 30, i32 15>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %1
 ;
 ; AVX1-LABEL: 'test_v16i16_2'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %1 = shufflevector <16 x i16> %a, <16 x i16> %b, <16 x i32> <i32 16, i32 1, i32 18, i32 3, i32 20, i32 5, i32 22, i32 7, i32 24, i32 9, i32 26, i32 11, i32 28, i32 13, i32 30, i32 15>
@@ -480,9 +484,9 @@ define <32 x i8> @test_v32i8(<32 x i8> %a, <32 x i8> %b) {
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %1 = shufflevector <32 x i8> %a, <32 x i8> %b, <32 x i32> <i32 0, i32 33, i32 2, i32 35, i32 4, i32 37, i32 6, i32 39, i32 8, i32 41, i32 10, i32 43, i32 12, i32 45, i32 14, i32 47, i32 16, i32 49, i32 18, i32 51, i32 20, i32 53, i32 22, i32 55, i32 24, i32 57, i32 26, i32 59, i32 28, i32 61, i32 30, i32 63>
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %1
 ;
-; SSE41-LABEL: 'test_v32i8'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %1 = shufflevector <32 x i8> %a, <32 x i8> %b, <32 x i32> <i32 0, i32 33, i32 2, i32 35, i32 4, i32 37, i32 6, i32 39, i32 8, i32 41, i32 10, i32 43, i32 12, i32 45, i32 14, i32 47, i32 16, i32 49, i32 18, i32 51, i32 20, i32 53, i32 22, i32 55, i32 24, i32 57, i32 26, i32 59, i32 28, i32 61, i32 30, i32 63>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %1
+; SSE42-LABEL: 'test_v32i8'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %1 = shufflevector <32 x i8> %a, <32 x i8> %b, <32 x i32> <i32 0, i32 33, i32 2, i32 35, i32 4, i32 37, i32 6, i32 39, i32 8, i32 41, i32 10, i32 43, i32 12, i32 45, i32 14, i32 47, i32 16, i32 49, i32 18, i32 51, i32 20, i32 53, i32 22, i32 55, i32 24, i32 57, i32 26, i32 59, i32 28, i32 61, i32 30, i32 63>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %1
 ;
 ; AVX1-LABEL: 'test_v32i8'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %1 = shufflevector <32 x i8> %a, <32 x i8> %b, <32 x i32> <i32 0, i32 33, i32 2, i32 35, i32 4, i32 37, i32 6, i32 39, i32 8, i32 41, i32 10, i32 43, i32 12, i32 45, i32 14, i32 47, i32 16, i32 49, i32 18, i32 51, i32 20, i32 53, i32 22, i32 55, i32 24, i32 57, i32 26, i32 59, i32 28, i32 61, i32 30, i32 63>
@@ -505,9 +509,9 @@ define <32 x i8> @test_v32i8_2(<32 x i8> %a, <32 x i8> %b) {
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %1 = shufflevector <32 x i8> %a, <32 x i8> %b, <32 x i32> <i32 32, i32 1, i32 34, i32 3, i32 36, i32 5, i32 38, i32 7, i32 40, i32 9, i32 42, i32 11, i32 44, i32 13, i32 46, i32 15, i32 48, i32 17, i32 50, i32 19, i32 52, i32 21, i32 54, i32 23, i32 56, i32 25, i32 58, i32 27, i32 60, i32 29, i32 62, i32 31>
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %1
 ;
-; SSE41-LABEL: 'test_v32i8_2'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %1 = shufflevector <32 x i8> %a, <32 x i8> %b, <32 x i32> <i32 32, i32 1, i32 34, i32 3, i32 36, i32 5, i32 38, i32 7, i32 40, i32 9, i32 42, i32 11, i32 44, i32 13, i32 46, i32 15, i32 48, i32 17, i32 50, i32 19, i32 52, i32 21, i32 54, i32 23, i32 56, i32 25, i32 58, i32 27, i32 60, i32 29, i32 62, i32 31>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %1
+; SSE42-LABEL: 'test_v32i8_2'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %1 = shufflevector <32 x i8> %a, <32 x i8> %b, <32 x i32> <i32 32, i32 1, i32 34, i32 3, i32 36, i32 5, i32 38, i32 7, i32 40, i32 9, i32 42, i32 11, i32 44, i32 13, i32 46, i32 15, i32 48, i32 17, i32 50, i32 19, i32 52, i32 21, i32 54, i32 23, i32 56, i32 25, i32 58, i32 27, i32 60, i32 29, i32 62, i32 31>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %1
 ;
 ; AVX1-LABEL: 'test_v32i8_2'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %1 = shufflevector <32 x i8> %a, <32 x i8> %b, <32 x i32> <i32 32, i32 1, i32 34, i32 3, i32 36, i32 5, i32 38, i32 7, i32 40, i32 9, i32 42, i32 11, i32 44, i32 13, i32 46, i32 15, i32 48, i32 17, i32 50, i32 19, i32 52, i32 21, i32 54, i32 23, i32 56, i32 25, i32 58, i32 27, i32 60, i32 29, i32 62, i32 31>

diff --git a/test/Analysis/CostModel/X86/arith-fp.ll b/test/Analysis/CostModel/X86/arith-fp.ll
index bfef397..4ca0648 100644 (file)
--- a/test/Analysis/CostModel/X86/arith-fp.ll
+++ b/test/Analysis/CostModel/X86/arith-fp.ll
@@ -5,8 +5,10 @@
 ; RUN: opt < %s -enable-no-nans-fp-math  -cost-model -analyze -mtriple=x86_64-apple-macosx10.8.0 -mattr=+avx2 | FileCheck %s --check-prefixes=CHECK,AVX,AVX2
 ; RUN: opt < %s -enable-no-nans-fp-math  -cost-model -analyze -mtriple=x86_64-apple-macosx10.8.0 -mattr=+avx512f | FileCheck %s --check-prefixes=CHECK,AVX512,AVX512F
 ; RUN: opt < %s -enable-no-nans-fp-math  -cost-model -analyze -mtriple=x86_64-apple-macosx10.8.0 -mattr=+avx512f,+avx512bw | FileCheck %s --check-prefixes=CHECK,AVX512,AVX512BW
+;
 ; RUN: opt < %s -enable-no-nans-fp-math  -cost-model -analyze -mtriple=x86_64-apple-macosx10.8.0 -mcpu=slm | FileCheck %s --check-prefixes=CHECK,SLM
 ; RUN: opt < %s -enable-no-nans-fp-math  -cost-model -analyze -mtriple=x86_64-apple-macosx10.8.0 -mcpu=goldmont | FileCheck %s --check-prefixes=CHECK,GLM
+; RUN: opt < %s -enable-no-nans-fp-math  -cost-model -analyze -mtriple=x86_64-apple-macosx10.8.0 -mcpu=btver2 | FileCheck %s --check-prefixes=CHECK,AVX,AVX1
 
 target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128"
 target triple = "x86_64-apple-macosx10.8.0"

diff --git a/test/Analysis/CostModel/X86/arith.ll b/test/Analysis/CostModel/X86/arith.ll
index b52e937..07445b6 100644 (file)
--- a/test/Analysis/CostModel/X86/arith.ll
+++ b/test/Analysis/CostModel/X86/arith.ll
@@ -6,8 +6,10 @@
 ; RUN: opt < %s -cost-model -analyze -mtriple=x86_64-apple-macosx10.8.0 -mattr=+avx512f | FileCheck %s --check-prefixes=CHECK,AVX512,AVX512F
 ; RUN: opt < %s -cost-model -analyze -mtriple=x86_64-apple-macosx10.8.0 -mattr=+avx512f,+avx512bw | FileCheck %s --check-prefixes=CHECK,AVX512,AVX512BW
 ; RUN: opt < %s -cost-model -analyze -mtriple=x86_64-apple-macosx10.8.0 -mattr=+avx512f,+avx512dq | FileCheck %s --check-prefixes=CHECK,AVX512,AVX512DQ
+;
 ; RUN: opt < %s -cost-model -analyze -mtriple=x86_64-apple-macosx10.8.0 -mcpu=slm | FileCheck %s --check-prefixes=CHECK,SLM
 ; RUN: opt < %s -cost-model -analyze -mtriple=x86_64-apple-macosx10.8.0 -mcpu=goldmont | FileCheck %s --check-prefixes=CHECK,GLM
+; RUN: opt < %s -cost-model -analyze -mtriple=x86_64-apple-macosx10.8.0 -mcpu=btver2 | FileCheck %s --check-prefixes=CHECK,AVX,AVX1
 
 target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128"
 target triple = "x86_64-apple-macosx10.8.0"

diff --git a/test/Analysis/CostModel/X86/fptosi.ll b/test/Analysis/CostModel/X86/fptosi.ll
index ce08cf0..2b18d12 100644 (file)
--- a/test/Analysis/CostModel/X86/fptosi.ll
+++ b/test/Analysis/CostModel/X86/fptosi.ll
@@ -5,6 +5,10 @@
 ; RUN: opt < %s -mtriple=x86_64-apple-darwin -cost-model -analyze -mattr=+avx2 | FileCheck %s --check-prefixes=CHECK,AVX,AVX2
 ; RUN: opt < %s -mtriple=x86_64-apple-darwin -cost-model -analyze -mattr=+avx512f | FileCheck %s --check-prefixes=CHECK,AVX512,AVX512F
 ; RUN: opt < %s -mtriple=x86_64-apple-darwin -cost-model -analyze -mattr=+avx512f,+avx512dq | FileCheck %s --check-prefixes=CHECK,AVX512,AVX512DQ
+;
+; RUN: opt < %s -mtriple=x86_64-apple-darwin -cost-model -analyze -mcpu=slm | FileCheck %s --check-prefixes=CHECK,SSE,SSE42
+; RUN: opt < %s -mtriple=x86_64-apple-darwin -cost-model -analyze -mcpu=goldmont | FileCheck %s --check-prefixes=CHECK,SSE,SSE42
+; RUN: opt < %s -mtriple=x86_64-apple-darwin -cost-model -analyze -mcpu=btver2 | FileCheck %s --check-prefixes=CHECK,AVX,AVX1
 
 define i32 @fptosi_double_i64(i32 %arg) {
 ; SSE-LABEL: 'fptosi_double_i64'

diff --git a/test/Analysis/CostModel/X86/fptoui.ll b/test/Analysis/CostModel/X86/fptoui.ll
index af22f78..1d2331c 100644 (file)
--- a/test/Analysis/CostModel/X86/fptoui.ll
+++ b/test/Analysis/CostModel/X86/fptoui.ll
@@ -5,6 +5,10 @@
 ; RUN: opt < %s -mtriple=x86_64-apple-darwin -cost-model -analyze -mattr=+avx2 | FileCheck %s --check-prefixes=CHECK,AVX,AVX2
 ; RUN: opt < %s -mtriple=x86_64-apple-darwin -cost-model -analyze -mattr=+avx512f | FileCheck %s --check-prefixes=CHECK,AVX512,AVX512F
 ; RUN: opt < %s -mtriple=x86_64-apple-darwin -cost-model -analyze -mattr=+avx512f,+avx512dq | FileCheck %s --check-prefixes=CHECK,AVX512,AVX512DQ
+;
+; RUN: opt < %s -mtriple=x86_64-apple-darwin -cost-model -analyze -mcpu=slm | FileCheck %s --check-prefixes=CHECK,SSE,SSE42
+; RUN: opt < %s -mtriple=x86_64-apple-darwin -cost-model -analyze -mcpu=goldmont | FileCheck %s --check-prefixes=CHECK,SSE,SSE42
+; RUN: opt < %s -mtriple=x86_64-apple-darwin -cost-model -analyze -mcpu=btver2 | FileCheck %s --check-prefixes=CHECK,AVX,AVX1
 
 define i32 @fptoui_double_i64(i32 %arg) {
 ; SSE-LABEL: 'fptoui_double_i64'

diff --git a/test/Analysis/CostModel/X86/shuffle-broadcast.ll b/test/Analysis/CostModel/X86/shuffle-broadcast.ll
index d6fc593..0257913 100644 (file)
--- a/test/Analysis/CostModel/X86/shuffle-broadcast.ll
+++ b/test/Analysis/CostModel/X86/shuffle-broadcast.ll
@@ -7,6 +7,10 @@
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+avx512f | FileCheck %s --check-prefixes=CHECK,AVX512,AVX512F
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+avx512f,+avx512bw | FileCheck %s --check-prefixes=CHECK,AVX512
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+avx512f,+avx512bw,+avx512vbmi | FileCheck %s --check-prefixes=CHECK,AVX512
+;
+; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mcpu=slm | FileCheck %s --check-prefixes=CHECK,SSE,SSE42
+; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mcpu=goldmont | FileCheck %s --check-prefixes=CHECK,SSE,SSE42
+; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mcpu=btver2 | FileCheck %s --check-prefixes=CHECK,AVX,AVX1
 
 ;
 ; Verify the cost model for broadcast shuffles.

diff --git a/test/Analysis/CostModel/X86/shuffle-reverse.ll b/test/Analysis/CostModel/X86/shuffle-reverse.ll
index c12f53c..5768238 100644 (file)
--- a/test/Analysis/CostModel/X86/shuffle-reverse.ll
+++ b/test/Analysis/CostModel/X86/shuffle-reverse.ll
@@ -7,6 +7,10 @@
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+avx512f | FileCheck %s --check-prefixes=CHECK,AVX512,AVX512F
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+avx512f,+avx512bw | FileCheck %s --check-prefixes=CHECK,AVX512,AVX512BW
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+avx512f,+avx512bw,+avx512vbmi | FileCheck %s --check-prefixes=CHECK,AVX512,AVX512VBMI
+;
+; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mcpu=slm | FileCheck %s --check-prefixes=CHECK,SSE,SSE42
+; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mcpu=goldmont | FileCheck %s --check-prefixes=CHECK,SSE,SSE42
+; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mcpu=btver2 | FileCheck %s --check-prefixes=CHECK,AVX,AVX1
 
 ;
 ; Verify the cost model for reverse shuffles.

diff --git a/test/Analysis/CostModel/X86/shuffle-single-src.ll b/test/Analysis/CostModel/X86/shuffle-single-src.ll
index f429650..f937945 100644 (file)
--- a/test/Analysis/CostModel/X86/shuffle-single-src.ll
+++ b/test/Analysis/CostModel/X86/shuffle-single-src.ll
@@ -8,6 +8,10 @@
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+avx512f | FileCheck %s --check-prefixes=CHECK,AVX512,AVX512F
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+avx512f,+avx512bw | FileCheck %s --check-prefixes=CHECK,AVX512,AVX512BW
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+avx512f,+avx512bw,+avx512vbmi | FileCheck %s --check-prefixes=CHECK,AVX512,AVX512VBMI
+;
+; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mcpu=slm | FileCheck %s --check-prefixes=CHECK,SSE,SSE42
+; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mcpu=goldmont | FileCheck %s --check-prefixes=CHECK,SSE,SSE42
+; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mcpu=btver2 | FileCheck %s --check-prefixes=CHECK,AVX,AVX1
 
 ;
 ; Verify the cost model for 1 src shuffles

diff --git a/test/Analysis/CostModel/X86/shuffle-two-src.ll b/test/Analysis/CostModel/X86/shuffle-two-src.ll
index 33eef4e..a6529a8 100644 (file)
--- a/test/Analysis/CostModel/X86/shuffle-two-src.ll
+++ b/test/Analysis/CostModel/X86/shuffle-two-src.ll
@@ -8,6 +8,10 @@
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+avx512f | FileCheck %s --check-prefixes=CHECK,AVX512,AVX512F
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+avx512f,+avx512bw | FileCheck %s --check-prefixes=CHECK,AVX512,AVX512BW
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+avx512f,+avx512bw,+avx512vbmi | FileCheck %s --check-prefixes=CHECK,AVX512,AVX512VBMI
+;
+; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mcpu=slm | FileCheck %s --check-prefixes=CHECK,SSE,SSE42
+; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mcpu=goldmont | FileCheck %s --check-prefixes=CHECK,SSE,SSE42
+; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mcpu=btver2 | FileCheck %s --check-prefixes=CHECK,AVX,AVX1
 
 ;
 ; Verify the cost model for 2 src shuffles

diff --git a/test/Analysis/CostModel/X86/sitofp.ll b/test/Analysis/CostModel/X86/sitofp.ll
index 08fbb2b..2e545f5 100644 (file)
--- a/test/Analysis/CostModel/X86/sitofp.ll
+++ b/test/Analysis/CostModel/X86/sitofp.ll
@@ -1,9 +1,14 @@
 ; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py
 ; RUN: opt < %s -mtriple=x86_64-apple-darwin -cost-model -analyze -mattr=+sse2 | FileCheck %s --check-prefixes=CHECK,SSE,SSE2
+; RUN: opt < %s -mtriple=x86_64-apple-darwin -cost-model -analyze -mattr=+sse4.2 | FileCheck %s --check-prefixes=CHECK,SSE,SSE42
 ; RUN: opt < %s -mtriple=x86_64-apple-darwin -cost-model -analyze -mattr=+avx  | FileCheck %s --check-prefixes=CHECK,AVX,AVX1
 ; RUN: opt < %s -mtriple=x86_64-apple-darwin -cost-model -analyze -mattr=+avx2 | FileCheck %s --check-prefixes=CHECK,AVX,AVX2
 ; RUN: opt < %s -mtriple=x86_64-apple-darwin -cost-model -analyze -mattr=+avx512f | FileCheck %s --check-prefixes=CHECK,AVX512,AVX512F
 ; RUN: opt < %s -mtriple=x86_64-apple-darwin -cost-model -analyze -mattr=+avx512f,+avx512dq | FileCheck %s --check-prefixes=CHECK,AVX512,AVX512DQ
+;
+; RUN: opt < %s -mtriple=x86_64-apple-darwin -cost-model -analyze -mcpu=slm | FileCheck %s --check-prefixes=CHECK,SSE,SSE42
+; RUN: opt < %s -mtriple=x86_64-apple-darwin -cost-model -analyze -mcpu=goldmont | FileCheck %s --check-prefixes=CHECK,SSE,SSE42
+; RUN: opt < %s -mtriple=x86_64-apple-darwin -cost-model -analyze -mcpu=btver2 | FileCheck %s --check-prefixes=CHECK,AVX,AVX1
 
 define i32 @sitofp_i8_double() {
 ; SSE-LABEL: 'sitofp_i8_double'

diff --git a/test/Analysis/CostModel/X86/trunc.ll b/test/Analysis/CostModel/X86/trunc.ll
index bcb86b7..bf42582 100644 (file)
--- a/test/Analysis/CostModel/X86/trunc.ll
+++ b/test/Analysis/CostModel/X86/trunc.ll
@@ -6,6 +6,10 @@
 ; RUN: opt < %s  -cost-model -analyze -mtriple=x86_64-apple-macosx10.8.0 -mattr=+avx2 | FileCheck %s --check-prefixes=CHECK,AVX,AVX2
 ; RUN: opt < %s  -cost-model -analyze -mtriple=x86_64-apple-macosx10.8.0 -mattr=+avx512f | FileCheck %s --check-prefixes=CHECK,AVX512,AVX512F
 ; RUN: opt < %s  -cost-model -analyze -mtriple=x86_64-apple-macosx10.8.0 -mattr=+avx512f,+avx512bw | FileCheck %s --check-prefixes=CHECK,AVX512,AVX512BW
+;
+; RUN: opt < %s -cost-model -analyze -mtriple=x86_64-apple-macosx10.8.0 -mcpu=slm | FileCheck %s --check-prefixes=CHECK,SSE,SSE42
+; RUN: opt < %s -cost-model -analyze -mtriple=x86_64-apple-macosx10.8.0 -mcpu=goldmont | FileCheck %s --check-prefixes=CHECK,SSE,SSE42
+; RUN: opt < %s -cost-model -analyze -mtriple=x86_64-apple-macosx10.8.0 -mcpu=btver2 | FileCheck %s --check-prefixes=CHECK,AVX,AVX1
 
 define i32 @trunc_vXi32() {
 ; SSE-LABEL: 'trunc_vXi32'

diff --git a/test/Analysis/CostModel/X86/uitofp.ll b/test/Analysis/CostModel/X86/uitofp.ll
index 91fc123..3920764 100644 (file)
--- a/test/Analysis/CostModel/X86/uitofp.ll
+++ b/test/Analysis/CostModel/X86/uitofp.ll
@@ -1,9 +1,14 @@
 ; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py
 ; RUN: opt < %s -mtriple=x86_64-apple-darwin -cost-model -analyze -mattr=+sse2 | FileCheck %s --check-prefixes=CHECK,SSE,SSE2
+; RUN: opt < %s -mtriple=x86_64-apple-darwin -cost-model -analyze -mattr=+sse4.2 | FileCheck %s --check-prefixes=CHECK,SSE,SSE42
 ; RUN: opt < %s -mtriple=x86_64-apple-darwin -cost-model -analyze -mattr=+avx  | FileCheck %s --check-prefixes=CHECK,AVX,AVX1
 ; RUN: opt < %s -mtriple=x86_64-apple-darwin -cost-model -analyze -mattr=+avx2 | FileCheck %s --check-prefixes=CHECK,AVX,AVX2
 ; RUN: opt < %s -mtriple=x86_64-apple-darwin -cost-model -analyze -mattr=+avx512f | FileCheck %s --check-prefixes=CHECK,AVX512,AVX512F
 ; RUN: opt < %s -mtriple=x86_64-apple-darwin -cost-model -analyze -mattr=+avx512f,+avx512dq | FileCheck %s --check-prefixes=CHECK,AVX512,AVX512DQ
+;
+; RUN: opt < %s -mtriple=x86_64-apple-darwin -cost-model -analyze -mcpu=slm | FileCheck %s --check-prefixes=CHECK,SSE,SSE42
+; RUN: opt < %s -mtriple=x86_64-apple-darwin -cost-model -analyze -mcpu=goldmont | FileCheck %s --check-prefixes=CHECK,SSE,SSE42
+; RUN: opt < %s -mtriple=x86_64-apple-darwin -cost-model -analyze -mcpu=btver2 | FileCheck %s --check-prefixes=CHECK,AVX,AVX1
 
 define i32 @uitofp_i8_double() {
 ; SSE-LABEL: 'uitofp_i8_double'

diff --git a/test/Analysis/CostModel/X86/vshift-ashr-cost.ll b/test/Analysis/CostModel/X86/vshift-ashr-cost.ll
index 8c94d31..2c328ba 100644 (file)
--- a/test/Analysis/CostModel/X86/vshift-ashr-cost.ll
+++ b/test/Analysis/CostModel/X86/vshift-ashr-cost.ll
@@ -1,6 +1,6 @@
 ; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+sse2 | FileCheck %s --check-prefixes=CHECK,SSE,SSE2
-; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+sse4.1 | FileCheck %s --check-prefixes=CHECK,SSE,SSE41
+; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+sse4.2 | FileCheck %s --check-prefixes=CHECK,SSE,SSE42
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+avx | FileCheck %s --check-prefixes=CHECK,AVX,AVX1
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+avx2 | FileCheck %s --check-prefixes=CHECK,AVX,AVX2
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+xop,+avx | FileCheck %s --check-prefixes=CHECK,XOP,XOPAVX1
@@ -11,6 +11,10 @@
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+avx512f,+avx512vl | FileCheck %s --check-prefixes=CHECK,AVX512,AVX512F
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+avx512f,+avx512dq,+avx512vl | FileCheck %s --check-prefixes=CHECK,AVX512,AVX512VL
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+avx512f,+avx512bw,+avx512vl | FileCheck %s --check-prefixes=CHECK,AVX512,AVX512BWVL
+;
+; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mcpu=slm | FileCheck %s --check-prefixes=CHECK,SSE,SSE42
+; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mcpu=goldmont | FileCheck %s --check-prefixes=CHECK,SSE,SSE42
+; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mcpu=btver2 | FileCheck %s --check-prefixes=CHECK,AVX,AVX1
 
 ; Verify the cost of vector arithmetic shift right instructions.
 
@@ -98,9 +102,9 @@ define <4 x i32> @var_shift_v4i32(<4 x i32> %a, <4 x i32> %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %shift = ashr <4 x i32> %a, %b
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
 ;
-; SSE41-LABEL: 'var_shift_v4i32'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %shift = ashr <4 x i32> %a, %b
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
+; SSE42-LABEL: 'var_shift_v4i32'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %shift = ashr <4 x i32> %a, %b
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
 ;
 ; AVX1-LABEL: 'var_shift_v4i32'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %shift = ashr <4 x i32> %a, %b
@@ -134,9 +138,9 @@ define <8 x i32> @var_shift_v8i32(<8 x i32> %a, <8 x i32> %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %shift = ashr <8 x i32> %a, %b
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
 ;
-; SSE41-LABEL: 'var_shift_v8i32'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %shift = ashr <8 x i32> %a, %b
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
+; SSE42-LABEL: 'var_shift_v8i32'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %shift = ashr <8 x i32> %a, %b
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
 ;
 ; AVX1-LABEL: 'var_shift_v8i32'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %shift = ashr <8 x i32> %a, %b
@@ -170,9 +174,9 @@ define <16 x i32> @var_shift_v16i32(<16 x i32> %a, <16 x i32> %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %shift = ashr <16 x i32> %a, %b
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
 ;
-; SSE41-LABEL: 'var_shift_v16i32'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %shift = ashr <16 x i32> %a, %b
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
+; SSE42-LABEL: 'var_shift_v16i32'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %shift = ashr <16 x i32> %a, %b
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
 ;
 ; AVX1-LABEL: 'var_shift_v16i32'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %shift = ashr <16 x i32> %a, %b
@@ -206,9 +210,9 @@ define <8 x i16> @var_shift_v8i16(<8 x i16> %a, <8 x i16> %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %shift = ashr <8 x i16> %a, %b
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
 ;
-; SSE41-LABEL: 'var_shift_v8i16'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %shift = ashr <8 x i16> %a, %b
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
+; SSE42-LABEL: 'var_shift_v8i16'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %shift = ashr <8 x i16> %a, %b
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
 ;
 ; AVX-LABEL: 'var_shift_v8i16'
 ; AVX-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %shift = ashr <8 x i16> %a, %b
@@ -243,9 +247,9 @@ define <16 x i16> @var_shift_v16i16(<16 x i16> %a, <16 x i16> %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %shift = ashr <16 x i16> %a, %b
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
 ;
-; SSE41-LABEL: 'var_shift_v16i16'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %shift = ashr <16 x i16> %a, %b
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
+; SSE42-LABEL: 'var_shift_v16i16'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %shift = ashr <16 x i16> %a, %b
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
 ;
 ; AVX1-LABEL: 'var_shift_v16i16'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 30 for instruction: %shift = ashr <16 x i16> %a, %b
@@ -284,9 +288,9 @@ define <32 x i16> @var_shift_v32i16(<32 x i16> %a, <32 x i16> %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 128 for instruction: %shift = ashr <32 x i16> %a, %b
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
 ;
-; SSE41-LABEL: 'var_shift_v32i16'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 56 for instruction: %shift = ashr <32 x i16> %a, %b
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
+; SSE42-LABEL: 'var_shift_v32i16'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 56 for instruction: %shift = ashr <32 x i16> %a, %b
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
 ;
 ; AVX1-LABEL: 'var_shift_v32i16'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 60 for instruction: %shift = ashr <32 x i16> %a, %b
@@ -325,9 +329,9 @@ define <16 x i8> @var_shift_v16i8(<16 x i8> %a, <16 x i8> %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 54 for instruction: %shift = ashr <16 x i8> %a, %b
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
 ;
-; SSE41-LABEL: 'var_shift_v16i8'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %shift = ashr <16 x i8> %a, %b
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
+; SSE42-LABEL: 'var_shift_v16i8'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %shift = ashr <16 x i8> %a, %b
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
 ;
 ; AVX-LABEL: 'var_shift_v16i8'
 ; AVX-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %shift = ashr <16 x i8> %a, %b
@@ -350,9 +354,9 @@ define <32 x i8> @var_shift_v32i8(<32 x i8> %a, <32 x i8> %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 108 for instruction: %shift = ashr <32 x i8> %a, %b
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
 ;
-; SSE41-LABEL: 'var_shift_v32i8'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %shift = ashr <32 x i8> %a, %b
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
+; SSE42-LABEL: 'var_shift_v32i8'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %shift = ashr <32 x i8> %a, %b
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
 ;
 ; AVX1-LABEL: 'var_shift_v32i8'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 50 for instruction: %shift = ashr <32 x i8> %a, %b
@@ -379,9 +383,9 @@ define <64 x i8> @var_shift_v64i8(<64 x i8> %a, <64 x i8> %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 216 for instruction: %shift = ashr <64 x i8> %a, %b
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
 ;
-; SSE41-LABEL: 'var_shift_v64i8'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %shift = ashr <64 x i8> %a, %b
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
+; SSE42-LABEL: 'var_shift_v64i8'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %shift = ashr <64 x i8> %a, %b
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
 ;
 ; AVX1-LABEL: 'var_shift_v64i8'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 100 for instruction: %shift = ashr <64 x i8> %a, %b
@@ -662,11 +666,11 @@ define <8 x i16> @splatvar_shift_v8i16(<8 x i16> %a, i16 %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %shift = ashr <8 x i16> %a, %splat
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
 ;
-; SSE41-LABEL: 'splatvar_shift_v8i16'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <8 x i16> undef, i16 %b, i32 0
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <8 x i16> %insert, <8 x i16> undef, <8 x i32> zeroinitializer
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %shift = ashr <8 x i16> %a, %splat
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
+; SSE42-LABEL: 'splatvar_shift_v8i16'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <8 x i16> undef, i16 %b, i32 0
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <8 x i16> %insert, <8 x i16> undef, <8 x i32> zeroinitializer
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %shift = ashr <8 x i16> %a, %splat
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
 ;
 ; AVX-LABEL: 'splatvar_shift_v8i16'
 ; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <8 x i16> undef, i16 %b, i32 0
@@ -699,11 +703,11 @@ define <16 x i16> @splatvar_shift_v16i16(<16 x i16> %a, i16 %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %shift = ashr <16 x i16> %a, %splat
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
 ;
-; SSE41-LABEL: 'splatvar_shift_v16i16'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <16 x i16> undef, i16 %b, i32 0
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <16 x i16> %insert, <16 x i16> undef, <16 x i32> zeroinitializer
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %shift = ashr <16 x i16> %a, %splat
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
+; SSE42-LABEL: 'splatvar_shift_v16i16'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <16 x i16> undef, i16 %b, i32 0
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <16 x i16> %insert, <16 x i16> undef, <16 x i32> zeroinitializer
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %shift = ashr <16 x i16> %a, %splat
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
 ;
 ; AVX1-LABEL: 'splatvar_shift_v16i16'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <16 x i16> undef, i16 %b, i32 0
@@ -753,11 +757,11 @@ define <32 x i16> @splatvar_shift_v32i16(<32 x i16> %a, i16 %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %shift = ashr <32 x i16> %a, %splat
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
 ;
-; SSE41-LABEL: 'splatvar_shift_v32i16'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <32 x i16> undef, i16 %b, i32 0
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <32 x i16> %insert, <32 x i16> undef, <32 x i32> zeroinitializer
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %shift = ashr <32 x i16> %a, %splat
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
+; SSE42-LABEL: 'splatvar_shift_v32i16'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <32 x i16> undef, i16 %b, i32 0
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <32 x i16> %insert, <32 x i16> undef, <32 x i32> zeroinitializer
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %shift = ashr <32 x i16> %a, %splat
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
 ;
 ; AVX1-LABEL: 'splatvar_shift_v32i16'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <32 x i16> undef, i16 %b, i32 0
@@ -825,11 +829,11 @@ define <16 x i8> @splatvar_shift_v16i8(<16 x i8> %a, i8 %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 54 for instruction: %shift = ashr <16 x i8> %a, %splat
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
 ;
-; SSE41-LABEL: 'splatvar_shift_v16i8'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <16 x i8> undef, i8 %b, i32 0
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <16 x i8> %insert, <16 x i8> undef, <16 x i32> zeroinitializer
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %shift = ashr <16 x i8> %a, %splat
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
+; SSE42-LABEL: 'splatvar_shift_v16i8'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <16 x i8> undef, i8 %b, i32 0
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <16 x i8> %insert, <16 x i8> undef, <16 x i32> zeroinitializer
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %shift = ashr <16 x i8> %a, %splat
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
 ;
 ; AVX-LABEL: 'splatvar_shift_v16i8'
 ; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <16 x i8> undef, i8 %b, i32 0
@@ -862,11 +866,11 @@ define <32 x i8> @splatvar_shift_v32i8(<32 x i8> %a, i8 %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 108 for instruction: %shift = ashr <32 x i8> %a, %splat
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
 ;
-; SSE41-LABEL: 'splatvar_shift_v32i8'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <32 x i8> undef, i8 %b, i32 0
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <32 x i8> %insert, <32 x i8> undef, <32 x i32> zeroinitializer
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %shift = ashr <32 x i8> %a, %splat
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
+; SSE42-LABEL: 'splatvar_shift_v32i8'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <32 x i8> undef, i8 %b, i32 0
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <32 x i8> %insert, <32 x i8> undef, <32 x i32> zeroinitializer
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %shift = ashr <32 x i8> %a, %splat
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
 ;
 ; AVX1-LABEL: 'splatvar_shift_v32i8'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <32 x i8> undef, i8 %b, i32 0
@@ -916,11 +920,11 @@ define <64 x i8> @splatvar_shift_v64i8(<64 x i8> %a, i8 %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 216 for instruction: %shift = ashr <64 x i8> %a, %splat
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
 ;
-; SSE41-LABEL: 'splatvar_shift_v64i8'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <64 x i8> undef, i8 %b, i32 0
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <64 x i8> %insert, <64 x i8> undef, <64 x i32> zeroinitializer
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %shift = ashr <64 x i8> %a, %splat
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
+; SSE42-LABEL: 'splatvar_shift_v64i8'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <64 x i8> undef, i8 %b, i32 0
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <64 x i8> %insert, <64 x i8> undef, <64 x i32> zeroinitializer
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %shift = ashr <64 x i8> %a, %splat
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
 ;
 ; AVX1-LABEL: 'splatvar_shift_v64i8'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <64 x i8> undef, i8 %b, i32 0
@@ -1065,9 +1069,9 @@ define <4 x i32> @constant_shift_v4i32(<4 x i32> %a) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %shift = ashr <4 x i32> %a, <i32 4, i32 5, i32 6, i32 7>
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
 ;
-; SSE41-LABEL: 'constant_shift_v4i32'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %shift = ashr <4 x i32> %a, <i32 4, i32 5, i32 6, i32 7>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
+; SSE42-LABEL: 'constant_shift_v4i32'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %shift = ashr <4 x i32> %a, <i32 4, i32 5, i32 6, i32 7>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
 ;
 ; AVX1-LABEL: 'constant_shift_v4i32'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %shift = ashr <4 x i32> %a, <i32 4, i32 5, i32 6, i32 7>
@@ -1101,9 +1105,9 @@ define <8 x i32> @constant_shift_v8i32(<8 x i32> %a) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %shift = ashr <8 x i32> %a, <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
 ;
-; SSE41-LABEL: 'constant_shift_v8i32'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %shift = ashr <8 x i32> %a, <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
+; SSE42-LABEL: 'constant_shift_v8i32'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %shift = ashr <8 x i32> %a, <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
 ;
 ; AVX1-LABEL: 'constant_shift_v8i32'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %shift = ashr <8 x i32> %a, <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>
@@ -1137,9 +1141,9 @@ define <16 x i32> @constant_shift_v16i32(<16 x i32> %a) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %shift = ashr <16 x i32> %a, <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
 ;
-; SSE41-LABEL: 'constant_shift_v16i32'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %shift = ashr <16 x i32> %a, <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
+; SSE42-LABEL: 'constant_shift_v16i32'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %shift = ashr <16 x i32> %a, <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
 ;
 ; AVX1-LABEL: 'constant_shift_v16i32'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %shift = ashr <16 x i32> %a, <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>
@@ -1173,9 +1177,9 @@ define <8 x i16> @constant_shift_v8i16(<8 x i16> %a) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %shift = ashr <8 x i16> %a, <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
 ;
-; SSE41-LABEL: 'constant_shift_v8i16'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %shift = ashr <8 x i16> %a, <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
+; SSE42-LABEL: 'constant_shift_v8i16'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %shift = ashr <8 x i16> %a, <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
 ;
 ; AVX-LABEL: 'constant_shift_v8i16'
 ; AVX-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %shift = ashr <8 x i16> %a, <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>
@@ -1210,9 +1214,9 @@ define <16 x i16> @constant_shift_v16i16(<16 x i16> %a) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %shift = ashr <16 x i16> %a, <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7, i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
 ;
-; SSE41-LABEL: 'constant_shift_v16i16'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %shift = ashr <16 x i16> %a, <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7, i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
+; SSE42-LABEL: 'constant_shift_v16i16'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %shift = ashr <16 x i16> %a, <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7, i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
 ;
 ; AVX1-LABEL: 'constant_shift_v16i16'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 30 for instruction: %shift = ashr <16 x i16> %a, <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7, i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>
@@ -1251,9 +1255,9 @@ define <32 x i16> @constant_shift_v32i16(<32 x i16> %a) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 128 for instruction: %shift = ashr <32 x i16> %a, <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7, i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7, i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7, i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
 ;
-; SSE41-LABEL: 'constant_shift_v32i16'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 56 for instruction: %shift = ashr <32 x i16> %a, <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7, i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7, i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7, i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
+; SSE42-LABEL: 'constant_shift_v32i16'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 56 for instruction: %shift = ashr <32 x i16> %a, <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7, i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7, i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7, i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
 ;
 ; AVX1-LABEL: 'constant_shift_v32i16'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 60 for instruction: %shift = ashr <32 x i16> %a, <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7, i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7, i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7, i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>
@@ -1292,9 +1296,9 @@ define <16 x i8> @constant_shift_v16i8(<16 x i8> %a) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 54 for instruction: %shift = ashr <16 x i8> %a, <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0>
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
 ;
-; SSE41-LABEL: 'constant_shift_v16i8'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %shift = ashr <16 x i8> %a, <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
+; SSE42-LABEL: 'constant_shift_v16i8'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %shift = ashr <16 x i8> %a, <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
 ;
 ; AVX-LABEL: 'constant_shift_v16i8'
 ; AVX-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %shift = ashr <16 x i8> %a, <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0>
@@ -1317,9 +1321,9 @@ define <32 x i8> @constant_shift_v32i8(<32 x i8> %a) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 108 for instruction: %shift = ashr <32 x i8> %a, <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0>
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
 ;
-; SSE41-LABEL: 'constant_shift_v32i8'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %shift = ashr <32 x i8> %a, <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
+; SSE42-LABEL: 'constant_shift_v32i8'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %shift = ashr <32 x i8> %a, <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
 ;
 ; AVX1-LABEL: 'constant_shift_v32i8'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 50 for instruction: %shift = ashr <32 x i8> %a, <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0>
@@ -1346,9 +1350,9 @@ define <64 x i8> @constant_shift_v64i8(<64 x i8> %a) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 216 for instruction: %shift = ashr <64 x i8> %a, <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0>
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
 ;
-; SSE41-LABEL: 'constant_shift_v64i8'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %shift = ashr <64 x i8> %a, <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
+; SSE42-LABEL: 'constant_shift_v64i8'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %shift = ashr <64 x i8> %a, <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
 ;
 ; AVX1-LABEL: 'constant_shift_v64i8'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 100 for instruction: %shift = ashr <64 x i8> %a, <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0>

diff --git a/test/Analysis/CostModel/X86/vshift-lshr-cost.ll b/test/Analysis/CostModel/X86/vshift-lshr-cost.ll
index c60a8dc..23a9c3a 100644 (file)
--- a/test/Analysis/CostModel/X86/vshift-lshr-cost.ll
+++ b/test/Analysis/CostModel/X86/vshift-lshr-cost.ll
@@ -1,6 +1,6 @@
 ; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+sse2 | FileCheck %s --check-prefixes=CHECK,SSE,SSE2
-; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+sse4.1 | FileCheck %s --check-prefixes=CHECK,SSE,SSE41
+; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+sse4.2 | FileCheck %s --check-prefixes=CHECK,SSE,SSE42
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+avx | FileCheck %s --check-prefixes=CHECK,AVX,AVX1
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+avx2 | FileCheck %s --check-prefixes=CHECK,AVX,AVX2
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+xop,+avx | FileCheck %s --check-prefixes=CHECK,XOP,XOPAVX1
@@ -11,6 +11,10 @@
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+avx512f,+avx512vl | FileCheck %s --check-prefixes=CHECK,AVX512,AVX512F
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+avx512f,+avx512dq,+avx512vl | FileCheck %s --check-prefixes=CHECK,AVX512,AVX512VL
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+avx512f,+avx512bw,+avx512vl | FileCheck %s --check-prefixes=CHECK,AVX512,AVX512BWVL
+;
+; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mcpu=slm | FileCheck %s --check-prefixes=CHECK,SSE,SSE42
+; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mcpu=goldmont | FileCheck %s --check-prefixes=CHECK,SSE,SSE42
+; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mcpu=btver2 | FileCheck %s --check-prefixes=CHECK,AVX,AVX1
 
 ; Verify the cost of vector logical shift right instructions.
 
@@ -110,9 +114,9 @@ define <4 x i32> @var_shift_v4i32(<4 x i32> %a, <4 x i32> %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %shift = lshr <4 x i32> %a, %b
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
 ;
-; SSE41-LABEL: 'var_shift_v4i32'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %shift = lshr <4 x i32> %a, %b
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
+; SSE42-LABEL: 'var_shift_v4i32'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %shift = lshr <4 x i32> %a, %b
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
 ;
 ; AVX1-LABEL: 'var_shift_v4i32'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %shift = lshr <4 x i32> %a, %b
@@ -143,9 +147,9 @@ define <8 x i32> @var_shift_v8i32(<8 x i32> %a, <8 x i32> %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %shift = lshr <8 x i32> %a, %b
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
 ;
-; SSE41-LABEL: 'var_shift_v8i32'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %shift = lshr <8 x i32> %a, %b
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
+; SSE42-LABEL: 'var_shift_v8i32'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %shift = lshr <8 x i32> %a, %b
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
 ;
 ; AVX1-LABEL: 'var_shift_v8i32'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %shift = lshr <8 x i32> %a, %b
@@ -176,9 +180,9 @@ define <16 x i32> @var_shift_v16i32(<16 x i32> %a, <16 x i32> %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %shift = lshr <16 x i32> %a, %b
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
 ;
-; SSE41-LABEL: 'var_shift_v16i32'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 44 for instruction: %shift = lshr <16 x i32> %a, %b
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
+; SSE42-LABEL: 'var_shift_v16i32'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 44 for instruction: %shift = lshr <16 x i32> %a, %b
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
 ;
 ; AVX1-LABEL: 'var_shift_v16i32'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %shift = lshr <16 x i32> %a, %b
@@ -209,9 +213,9 @@ define <8 x i16> @var_shift_v8i16(<8 x i16> %a, <8 x i16> %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %shift = lshr <8 x i16> %a, %b
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
 ;
-; SSE41-LABEL: 'var_shift_v8i16'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %shift = lshr <8 x i16> %a, %b
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
+; SSE42-LABEL: 'var_shift_v8i16'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %shift = lshr <8 x i16> %a, %b
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
 ;
 ; AVX-LABEL: 'var_shift_v8i16'
 ; AVX-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %shift = lshr <8 x i16> %a, %b
@@ -246,9 +250,9 @@ define <16 x i16> @var_shift_v16i16(<16 x i16> %a, <16 x i16> %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %shift = lshr <16 x i16> %a, %b
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
 ;
-; SSE41-LABEL: 'var_shift_v16i16'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %shift = lshr <16 x i16> %a, %b
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
+; SSE42-LABEL: 'var_shift_v16i16'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %shift = lshr <16 x i16> %a, %b
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
 ;
 ; AVX1-LABEL: 'var_shift_v16i16'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 30 for instruction: %shift = lshr <16 x i16> %a, %b
@@ -287,9 +291,9 @@ define <32 x i16> @var_shift_v32i16(<32 x i16> %a, <32 x i16> %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 128 for instruction: %shift = lshr <32 x i16> %a, %b
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
 ;
-; SSE41-LABEL: 'var_shift_v32i16'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 56 for instruction: %shift = lshr <32 x i16> %a, %b
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
+; SSE42-LABEL: 'var_shift_v32i16'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 56 for instruction: %shift = lshr <32 x i16> %a, %b
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
 ;
 ; AVX1-LABEL: 'var_shift_v32i16'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 60 for instruction: %shift = lshr <32 x i16> %a, %b
@@ -328,9 +332,9 @@ define <16 x i8> @var_shift_v16i8(<16 x i8> %a, <16 x i8> %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %shift = lshr <16 x i8> %a, %b
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
 ;
-; SSE41-LABEL: 'var_shift_v16i8'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %shift = lshr <16 x i8> %a, %b
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
+; SSE42-LABEL: 'var_shift_v16i8'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %shift = lshr <16 x i8> %a, %b
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
 ;
 ; AVX-LABEL: 'var_shift_v16i8'
 ; AVX-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %shift = lshr <16 x i8> %a, %b
@@ -353,9 +357,9 @@ define <32 x i8> @var_shift_v32i8(<32 x i8> %a, <32 x i8> %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %shift = lshr <32 x i8> %a, %b
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
 ;
-; SSE41-LABEL: 'var_shift_v32i8'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %shift = lshr <32 x i8> %a, %b
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
+; SSE42-LABEL: 'var_shift_v32i8'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %shift = lshr <32 x i8> %a, %b
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
 ;
 ; AVX1-LABEL: 'var_shift_v32i8'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %shift = lshr <32 x i8> %a, %b
@@ -382,9 +386,9 @@ define <64 x i8> @var_shift_v64i8(<64 x i8> %a, <64 x i8> %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 104 for instruction: %shift = lshr <64 x i8> %a, %b
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
 ;
-; SSE41-LABEL: 'var_shift_v64i8'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %shift = lshr <64 x i8> %a, %b
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
+; SSE42-LABEL: 'var_shift_v64i8'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %shift = lshr <64 x i8> %a, %b
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
 ;
 ; AVX1-LABEL: 'var_shift_v64i8'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %shift = lshr <64 x i8> %a, %b
@@ -627,11 +631,11 @@ define <8 x i16> @splatvar_shift_v8i16(<8 x i16> %a, i16 %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %shift = lshr <8 x i16> %a, %splat
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
 ;
-; SSE41-LABEL: 'splatvar_shift_v8i16'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <8 x i16> undef, i16 %b, i32 0
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <8 x i16> %insert, <8 x i16> undef, <8 x i32> zeroinitializer
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %shift = lshr <8 x i16> %a, %splat
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
+; SSE42-LABEL: 'splatvar_shift_v8i16'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <8 x i16> undef, i16 %b, i32 0
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <8 x i16> %insert, <8 x i16> undef, <8 x i32> zeroinitializer
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %shift = lshr <8 x i16> %a, %splat
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
 ;
 ; AVX-LABEL: 'splatvar_shift_v8i16'
 ; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <8 x i16> undef, i16 %b, i32 0
@@ -664,11 +668,11 @@ define <16 x i16> @splatvar_shift_v16i16(<16 x i16> %a, i16 %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %shift = lshr <16 x i16> %a, %splat
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
 ;
-; SSE41-LABEL: 'splatvar_shift_v16i16'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <16 x i16> undef, i16 %b, i32 0
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <16 x i16> %insert, <16 x i16> undef, <16 x i32> zeroinitializer
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %shift = lshr <16 x i16> %a, %splat
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
+; SSE42-LABEL: 'splatvar_shift_v16i16'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <16 x i16> undef, i16 %b, i32 0
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <16 x i16> %insert, <16 x i16> undef, <16 x i32> zeroinitializer
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %shift = lshr <16 x i16> %a, %splat
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
 ;
 ; AVX1-LABEL: 'splatvar_shift_v16i16'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <16 x i16> undef, i16 %b, i32 0
@@ -713,11 +717,11 @@ define <32 x i16> @splatvar_shift_v32i16(<32 x i16> %a, i16 %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %shift = lshr <32 x i16> %a, %splat
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
 ;
-; SSE41-LABEL: 'splatvar_shift_v32i16'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <32 x i16> undef, i16 %b, i32 0
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <32 x i16> %insert, <32 x i16> undef, <32 x i32> zeroinitializer
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %shift = lshr <32 x i16> %a, %splat
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
+; SSE42-LABEL: 'splatvar_shift_v32i16'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <32 x i16> undef, i16 %b, i32 0
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <32 x i16> %insert, <32 x i16> undef, <32 x i32> zeroinitializer
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %shift = lshr <32 x i16> %a, %splat
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
 ;
 ; AVX1-LABEL: 'splatvar_shift_v32i16'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <32 x i16> undef, i16 %b, i32 0
@@ -780,11 +784,11 @@ define <16 x i8> @splatvar_shift_v16i8(<16 x i8> %a, i8 %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %shift = lshr <16 x i8> %a, %splat
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
 ;
-; SSE41-LABEL: 'splatvar_shift_v16i8'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <16 x i8> undef, i8 %b, i32 0
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <16 x i8> %insert, <16 x i8> undef, <16 x i32> zeroinitializer
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %shift = lshr <16 x i8> %a, %splat
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
+; SSE42-LABEL: 'splatvar_shift_v16i8'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <16 x i8> undef, i8 %b, i32 0
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <16 x i8> %insert, <16 x i8> undef, <16 x i32> zeroinitializer
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %shift = lshr <16 x i8> %a, %splat
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
 ;
 ; AVX-LABEL: 'splatvar_shift_v16i8'
 ; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <16 x i8> undef, i8 %b, i32 0
@@ -817,11 +821,11 @@ define <32 x i8> @splatvar_shift_v32i8(<32 x i8> %a, i8 %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %shift = lshr <32 x i8> %a, %splat
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
 ;
-; SSE41-LABEL: 'splatvar_shift_v32i8'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <32 x i8> undef, i8 %b, i32 0
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <32 x i8> %insert, <32 x i8> undef, <32 x i32> zeroinitializer
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %shift = lshr <32 x i8> %a, %splat
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
+; SSE42-LABEL: 'splatvar_shift_v32i8'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <32 x i8> undef, i8 %b, i32 0
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <32 x i8> %insert, <32 x i8> undef, <32 x i32> zeroinitializer
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %shift = lshr <32 x i8> %a, %splat
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
 ;
 ; AVX1-LABEL: 'splatvar_shift_v32i8'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <32 x i8> undef, i8 %b, i32 0
@@ -866,11 +870,11 @@ define <64 x i8> @splatvar_shift_v64i8(<64 x i8> %a, i8 %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 104 for instruction: %shift = lshr <64 x i8> %a, %splat
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
 ;
-; SSE41-LABEL: 'splatvar_shift_v64i8'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <64 x i8> undef, i8 %b, i32 0
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <64 x i8> %insert, <64 x i8> undef, <64 x i32> zeroinitializer
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %shift = lshr <64 x i8> %a, %splat
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
+; SSE42-LABEL: 'splatvar_shift_v64i8'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <64 x i8> undef, i8 %b, i32 0
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <64 x i8> %insert, <64 x i8> undef, <64 x i32> zeroinitializer
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %shift = lshr <64 x i8> %a, %splat
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
 ;
 ; AVX1-LABEL: 'splatvar_shift_v64i8'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <64 x i8> undef, i8 %b, i32 0
@@ -1022,9 +1026,9 @@ define <4 x i32> @constant_shift_v4i32(<4 x i32> %a) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %shift = lshr <4 x i32> %a, <i32 4, i32 5, i32 6, i32 7>
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
 ;
-; SSE41-LABEL: 'constant_shift_v4i32'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %shift = lshr <4 x i32> %a, <i32 4, i32 5, i32 6, i32 7>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
+; SSE42-LABEL: 'constant_shift_v4i32'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %shift = lshr <4 x i32> %a, <i32 4, i32 5, i32 6, i32 7>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
 ;
 ; AVX1-LABEL: 'constant_shift_v4i32'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %shift = lshr <4 x i32> %a, <i32 4, i32 5, i32 6, i32 7>
@@ -1055,9 +1059,9 @@ define <8 x i32> @constant_shift_v8i32(<8 x i32> %a) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %shift = lshr <8 x i32> %a, <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
 ;
-; SSE41-LABEL: 'constant_shift_v8i32'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %shift = lshr <8 x i32> %a, <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
+; SSE42-LABEL: 'constant_shift_v8i32'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %shift = lshr <8 x i32> %a, <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
 ;
 ; AVX1-LABEL: 'constant_shift_v8i32'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %shift = lshr <8 x i32> %a, <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>
@@ -1088,9 +1092,9 @@ define <16 x i32> @constant_shift_v16i32(<16 x i32> %a) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %shift = lshr <16 x i32> %a, <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
 ;
-; SSE41-LABEL: 'constant_shift_v16i32'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 44 for instruction: %shift = lshr <16 x i32> %a, <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
+; SSE42-LABEL: 'constant_shift_v16i32'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 44 for instruction: %shift = lshr <16 x i32> %a, <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
 ;
 ; AVX1-LABEL: 'constant_shift_v16i32'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %shift = lshr <16 x i32> %a, <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>
@@ -1121,9 +1125,9 @@ define <8 x i16> @constant_shift_v8i16(<8 x i16> %a) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %shift = lshr <8 x i16> %a, <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
 ;
-; SSE41-LABEL: 'constant_shift_v8i16'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %shift = lshr <8 x i16> %a, <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
+; SSE42-LABEL: 'constant_shift_v8i16'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %shift = lshr <8 x i16> %a, <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
 ;
 ; AVX-LABEL: 'constant_shift_v8i16'
 ; AVX-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %shift = lshr <8 x i16> %a, <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>
@@ -1158,9 +1162,9 @@ define <16 x i16> @constant_shift_v16i16(<16 x i16> %a) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %shift = lshr <16 x i16> %a, <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7, i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
 ;
-; SSE41-LABEL: 'constant_shift_v16i16'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %shift = lshr <16 x i16> %a, <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7, i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
+; SSE42-LABEL: 'constant_shift_v16i16'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %shift = lshr <16 x i16> %a, <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7, i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
 ;
 ; AVX1-LABEL: 'constant_shift_v16i16'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 30 for instruction: %shift = lshr <16 x i16> %a, <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7, i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>
@@ -1199,9 +1203,9 @@ define <32 x i16> @constant_shift_v32i16(<32 x i16> %a) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 128 for instruction: %shift = lshr <32 x i16> %a, <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7, i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7, i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7, i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
 ;
-; SSE41-LABEL: 'constant_shift_v32i16'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 56 for instruction: %shift = lshr <32 x i16> %a, <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7, i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7, i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7, i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
+; SSE42-LABEL: 'constant_shift_v32i16'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 56 for instruction: %shift = lshr <32 x i16> %a, <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7, i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7, i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7, i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
 ;
 ; AVX1-LABEL: 'constant_shift_v32i16'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 60 for instruction: %shift = lshr <32 x i16> %a, <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7, i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7, i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7, i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>
@@ -1240,9 +1244,9 @@ define <16 x i8> @constant_shift_v16i8(<16 x i8> %a) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %shift = lshr <16 x i8> %a, <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0>
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
 ;
-; SSE41-LABEL: 'constant_shift_v16i8'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %shift = lshr <16 x i8> %a, <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
+; SSE42-LABEL: 'constant_shift_v16i8'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %shift = lshr <16 x i8> %a, <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
 ;
 ; AVX-LABEL: 'constant_shift_v16i8'
 ; AVX-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %shift = lshr <16 x i8> %a, <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0>
@@ -1265,9 +1269,9 @@ define <32 x i8> @constant_shift_v32i8(<32 x i8> %a) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %shift = lshr <32 x i8> %a, <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0>
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
 ;
-; SSE41-LABEL: 'constant_shift_v32i8'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %shift = lshr <32 x i8> %a, <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
+; SSE42-LABEL: 'constant_shift_v32i8'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %shift = lshr <32 x i8> %a, <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
 ;
 ; AVX1-LABEL: 'constant_shift_v32i8'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %shift = lshr <32 x i8> %a, <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0>
@@ -1294,9 +1298,9 @@ define <64 x i8> @constant_shift_v64i8(<64 x i8> %a) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 104 for instruction: %shift = lshr <64 x i8> %a, <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0>
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
 ;
-; SSE41-LABEL: 'constant_shift_v64i8'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %shift = lshr <64 x i8> %a, <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
+; SSE42-LABEL: 'constant_shift_v64i8'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %shift = lshr <64 x i8> %a, <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
 ;
 ; AVX1-LABEL: 'constant_shift_v64i8'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %shift = lshr <64 x i8> %a, <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0>

diff --git a/test/Analysis/CostModel/X86/vshift-shl-cost.ll b/test/Analysis/CostModel/X86/vshift-shl-cost.ll
index 51575f9..4be3579 100644 (file)
--- a/test/Analysis/CostModel/X86/vshift-shl-cost.ll
+++ b/test/Analysis/CostModel/X86/vshift-shl-cost.ll
@@ -1,6 +1,6 @@
 ; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+sse2 | FileCheck %s --check-prefixes=CHECK,SSE,SSE2
-; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+sse4.1 | FileCheck %s --check-prefixes=CHECK,SSE,SSE41
+; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+sse4.2 | FileCheck %s --check-prefixes=CHECK,SSE,SSE42
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+avx | FileCheck %s --check-prefixes=CHECK,AVX,AVX1
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+avx2 | FileCheck %s --check-prefixes=CHECK,AVX,AVX2
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+xop,+avx | FileCheck %s --check-prefixes=CHECK,XOP,XOPAVX1
@@ -11,6 +11,10 @@
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+avx512f,+avx512vl | FileCheck %s --check-prefixes=CHECK,AVX512,AVX512F
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+avx512f,+avx512dq,+avx512vl | FileCheck %s --check-prefixes=CHECK,AVX512,AVX512VL
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mattr=+avx512f,+avx512bw,+avx512vl | FileCheck %s --check-prefixes=CHECK,AVX512,AVX512BWVL
+;
+; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mcpu=slm | FileCheck %s --check-prefixes=CHECK,SSE,SSE42
+; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mcpu=goldmont | FileCheck %s --check-prefixes=CHECK,SSE,SSE42
+; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -cost-model -analyze -mcpu=btver2 | FileCheck %s --check-prefixes=CHECK,AVX,AVX1
 
 ; Verify the cost of vector shift left instructions.
 
@@ -106,9 +110,9 @@ define <4 x i32> @var_shift_v4i32(<4 x i32> %a, <4 x i32> %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %shift = shl <4 x i32> %a, %b
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
 ;
-; SSE41-LABEL: 'var_shift_v4i32'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %shift = shl <4 x i32> %a, %b
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
+; SSE42-LABEL: 'var_shift_v4i32'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %shift = shl <4 x i32> %a, %b
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
 ;
 ; AVX1-LABEL: 'var_shift_v4i32'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %shift = shl <4 x i32> %a, %b
@@ -135,9 +139,9 @@ define <8 x i32> @var_shift_v8i32(<8 x i32> %a, <8 x i32> %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 20 for instruction: %shift = shl <8 x i32> %a, %b
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
 ;
-; SSE41-LABEL: 'var_shift_v8i32'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %shift = shl <8 x i32> %a, %b
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
+; SSE42-LABEL: 'var_shift_v8i32'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %shift = shl <8 x i32> %a, %b
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
 ;
 ; AVX1-LABEL: 'var_shift_v8i32'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %shift = shl <8 x i32> %a, %b
@@ -168,9 +172,9 @@ define <16 x i32> @var_shift_v16i32(<16 x i32> %a, <16 x i32> %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 40 for instruction: %shift = shl <16 x i32> %a, %b
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
 ;
-; SSE41-LABEL: 'var_shift_v16i32'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %shift = shl <16 x i32> %a, %b
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
+; SSE42-LABEL: 'var_shift_v16i32'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %shift = shl <16 x i32> %a, %b
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
 ;
 ; AVX1-LABEL: 'var_shift_v16i32'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 20 for instruction: %shift = shl <16 x i32> %a, %b
@@ -201,9 +205,9 @@ define <8 x i16> @var_shift_v8i16(<8 x i16> %a, <8 x i16> %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %shift = shl <8 x i16> %a, %b
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
 ;
-; SSE41-LABEL: 'var_shift_v8i16'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %shift = shl <8 x i16> %a, %b
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
+; SSE42-LABEL: 'var_shift_v8i16'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %shift = shl <8 x i16> %a, %b
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
 ;
 ; AVX-LABEL: 'var_shift_v8i16'
 ; AVX-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %shift = shl <8 x i16> %a, %b
@@ -238,9 +242,9 @@ define <16 x i16> @var_shift_v16i16(<16 x i16> %a, <16 x i16> %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %shift = shl <16 x i16> %a, %b
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
 ;
-; SSE41-LABEL: 'var_shift_v16i16'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %shift = shl <16 x i16> %a, %b
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
+; SSE42-LABEL: 'var_shift_v16i16'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %shift = shl <16 x i16> %a, %b
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
 ;
 ; AVX1-LABEL: 'var_shift_v16i16'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 30 for instruction: %shift = shl <16 x i16> %a, %b
@@ -279,9 +283,9 @@ define <32 x i16> @var_shift_v32i16(<32 x i16> %a, <32 x i16> %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 128 for instruction: %shift = shl <32 x i16> %a, %b
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
 ;
-; SSE41-LABEL: 'var_shift_v32i16'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 56 for instruction: %shift = shl <32 x i16> %a, %b
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
+; SSE42-LABEL: 'var_shift_v32i16'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 56 for instruction: %shift = shl <32 x i16> %a, %b
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
 ;
 ; AVX1-LABEL: 'var_shift_v32i16'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 60 for instruction: %shift = shl <32 x i16> %a, %b
@@ -320,9 +324,9 @@ define <16 x i8> @var_shift_v16i8(<16 x i8> %a, <16 x i8> %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %shift = shl <16 x i8> %a, %b
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
 ;
-; SSE41-LABEL: 'var_shift_v16i8'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %shift = shl <16 x i8> %a, %b
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
+; SSE42-LABEL: 'var_shift_v16i8'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %shift = shl <16 x i8> %a, %b
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
 ;
 ; AVX-LABEL: 'var_shift_v16i8'
 ; AVX-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %shift = shl <16 x i8> %a, %b
@@ -345,9 +349,9 @@ define <32 x i8> @var_shift_v32i8(<32 x i8> %a, <32 x i8> %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %shift = shl <32 x i8> %a, %b
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
 ;
-; SSE41-LABEL: 'var_shift_v32i8'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %shift = shl <32 x i8> %a, %b
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
+; SSE42-LABEL: 'var_shift_v32i8'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %shift = shl <32 x i8> %a, %b
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
 ;
 ; AVX1-LABEL: 'var_shift_v32i8'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %shift = shl <32 x i8> %a, %b
@@ -374,9 +378,9 @@ define <64 x i8> @var_shift_v64i8(<64 x i8> %a, <64 x i8> %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 104 for instruction: %shift = shl <64 x i8> %a, %b
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
 ;
-; SSE41-LABEL: 'var_shift_v64i8'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 44 for instruction: %shift = shl <64 x i8> %a, %b
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
+; SSE42-LABEL: 'var_shift_v64i8'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 44 for instruction: %shift = shl <64 x i8> %a, %b
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
 ;
 ; AVX1-LABEL: 'var_shift_v64i8'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %shift = shl <64 x i8> %a, %b
@@ -619,11 +623,11 @@ define <8 x i16> @splatvar_shift_v8i16(<8 x i16> %a, i16 %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %shift = shl <8 x i16> %a, %splat
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
 ;
-; SSE41-LABEL: 'splatvar_shift_v8i16'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <8 x i16> undef, i16 %b, i32 0
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <8 x i16> %insert, <8 x i16> undef, <8 x i32> zeroinitializer
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %shift = shl <8 x i16> %a, %splat
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
+; SSE42-LABEL: 'splatvar_shift_v8i16'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <8 x i16> undef, i16 %b, i32 0
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <8 x i16> %insert, <8 x i16> undef, <8 x i32> zeroinitializer
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %shift = shl <8 x i16> %a, %splat
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
 ;
 ; AVX-LABEL: 'splatvar_shift_v8i16'
 ; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <8 x i16> undef, i16 %b, i32 0
@@ -656,11 +660,11 @@ define <16 x i16> @splatvar_shift_v16i16(<16 x i16> %a, i16 %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %shift = shl <16 x i16> %a, %splat
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
 ;
-; SSE41-LABEL: 'splatvar_shift_v16i16'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <16 x i16> undef, i16 %b, i32 0
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <16 x i16> %insert, <16 x i16> undef, <16 x i32> zeroinitializer
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %shift = shl <16 x i16> %a, %splat
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
+; SSE42-LABEL: 'splatvar_shift_v16i16'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <16 x i16> undef, i16 %b, i32 0
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <16 x i16> %insert, <16 x i16> undef, <16 x i32> zeroinitializer
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %shift = shl <16 x i16> %a, %splat
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
 ;
 ; AVX1-LABEL: 'splatvar_shift_v16i16'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <16 x i16> undef, i16 %b, i32 0
@@ -705,11 +709,11 @@ define <32 x i16> @splatvar_shift_v32i16(<32 x i16> %a, i16 %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %shift = shl <32 x i16> %a, %splat
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
 ;
-; SSE41-LABEL: 'splatvar_shift_v32i16'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <32 x i16> undef, i16 %b, i32 0
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <32 x i16> %insert, <32 x i16> undef, <32 x i32> zeroinitializer
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %shift = shl <32 x i16> %a, %splat
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
+; SSE42-LABEL: 'splatvar_shift_v32i16'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <32 x i16> undef, i16 %b, i32 0
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <32 x i16> %insert, <32 x i16> undef, <32 x i32> zeroinitializer
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %shift = shl <32 x i16> %a, %splat
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
 ;
 ; AVX1-LABEL: 'splatvar_shift_v32i16'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <32 x i16> undef, i16 %b, i32 0
@@ -772,11 +776,11 @@ define <16 x i8> @splatvar_shift_v16i8(<16 x i8> %a, i8 %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %shift = shl <16 x i8> %a, %splat
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
 ;
-; SSE41-LABEL: 'splatvar_shift_v16i8'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <16 x i8> undef, i8 %b, i32 0
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <16 x i8> %insert, <16 x i8> undef, <16 x i32> zeroinitializer
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %shift = shl <16 x i8> %a, %splat
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
+; SSE42-LABEL: 'splatvar_shift_v16i8'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <16 x i8> undef, i8 %b, i32 0
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <16 x i8> %insert, <16 x i8> undef, <16 x i32> zeroinitializer
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %shift = shl <16 x i8> %a, %splat
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
 ;
 ; AVX-LABEL: 'splatvar_shift_v16i8'
 ; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <16 x i8> undef, i8 %b, i32 0
@@ -809,11 +813,11 @@ define <32 x i8> @splatvar_shift_v32i8(<32 x i8> %a, i8 %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %shift = shl <32 x i8> %a, %splat
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
 ;
-; SSE41-LABEL: 'splatvar_shift_v32i8'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <32 x i8> undef, i8 %b, i32 0
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <32 x i8> %insert, <32 x i8> undef, <32 x i32> zeroinitializer
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %shift = shl <32 x i8> %a, %splat
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
+; SSE42-LABEL: 'splatvar_shift_v32i8'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <32 x i8> undef, i8 %b, i32 0
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <32 x i8> %insert, <32 x i8> undef, <32 x i32> zeroinitializer
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %shift = shl <32 x i8> %a, %splat
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
 ;
 ; AVX1-LABEL: 'splatvar_shift_v32i8'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <32 x i8> undef, i8 %b, i32 0
@@ -858,11 +862,11 @@ define <64 x i8> @splatvar_shift_v64i8(<64 x i8> %a, i8 %b) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 104 for instruction: %shift = shl <64 x i8> %a, %splat
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
 ;
-; SSE41-LABEL: 'splatvar_shift_v64i8'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <64 x i8> undef, i8 %b, i32 0
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <64 x i8> %insert, <64 x i8> undef, <64 x i32> zeroinitializer
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 44 for instruction: %shift = shl <64 x i8> %a, %splat
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
+; SSE42-LABEL: 'splatvar_shift_v64i8'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <64 x i8> undef, i8 %b, i32 0
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <64 x i8> %insert, <64 x i8> undef, <64 x i32> zeroinitializer
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 44 for instruction: %shift = shl <64 x i8> %a, %splat
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
 ;
 ; AVX1-LABEL: 'splatvar_shift_v64i8'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <64 x i8> undef, i8 %b, i32 0
@@ -1010,9 +1014,9 @@ define <4 x i32> @constant_shift_v4i32(<4 x i32> %a) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %shift = shl <4 x i32> %a, <i32 4, i32 5, i32 6, i32 7>
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
 ;
-; SSE41-LABEL: 'constant_shift_v4i32'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %shift = shl <4 x i32> %a, <i32 4, i32 5, i32 6, i32 7>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
+; SSE42-LABEL: 'constant_shift_v4i32'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %shift = shl <4 x i32> %a, <i32 4, i32 5, i32 6, i32 7>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
 ;
 ; AVX1-LABEL: 'constant_shift_v4i32'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %shift = shl <4 x i32> %a, <i32 4, i32 5, i32 6, i32 7>
@@ -1039,9 +1043,9 @@ define <8 x i32> @constant_shift_v8i32(<8 x i32> %a) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %shift = shl <8 x i32> %a, <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
 ;
-; SSE41-LABEL: 'constant_shift_v8i32'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %shift = shl <8 x i32> %a, <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
+; SSE42-LABEL: 'constant_shift_v8i32'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %shift = shl <8 x i32> %a, <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
 ;
 ; AVX1-LABEL: 'constant_shift_v8i32'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %shift = shl <8 x i32> %a, <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>
@@ -1072,9 +1076,9 @@ define <16 x i32> @constant_shift_v16i32(<16 x i32> %a) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %shift = shl <16 x i32> %a, <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
 ;
-; SSE41-LABEL: 'constant_shift_v16i32'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %shift = shl <16 x i32> %a, <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
+; SSE42-LABEL: 'constant_shift_v16i32'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %shift = shl <16 x i32> %a, <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
 ;
 ; AVX1-LABEL: 'constant_shift_v16i32'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %shift = shl <16 x i32> %a, <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>
@@ -1184,9 +1188,9 @@ define <16 x i8> @constant_shift_v16i8(<16 x i8> %a) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %shift = shl <16 x i8> %a, <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0>
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
 ;
-; SSE41-LABEL: 'constant_shift_v16i8'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %shift = shl <16 x i8> %a, <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
+; SSE42-LABEL: 'constant_shift_v16i8'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %shift = shl <16 x i8> %a, <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
 ;
 ; AVX-LABEL: 'constant_shift_v16i8'
 ; AVX-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %shift = shl <16 x i8> %a, <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0>
@@ -1209,9 +1213,9 @@ define <32 x i8> @constant_shift_v32i8(<32 x i8> %a) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %shift = shl <32 x i8> %a, <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0>
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
 ;
-; SSE41-LABEL: 'constant_shift_v32i8'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %shift = shl <32 x i8> %a, <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
+; SSE42-LABEL: 'constant_shift_v32i8'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %shift = shl <32 x i8> %a, <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
 ;
 ; AVX1-LABEL: 'constant_shift_v32i8'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %shift = shl <32 x i8> %a, <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0>
@@ -1238,9 +1242,9 @@ define <64 x i8> @constant_shift_v64i8(<64 x i8> %a) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 104 for instruction: %shift = shl <64 x i8> %a, <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0>
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
 ;
-; SSE41-LABEL: 'constant_shift_v64i8'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 44 for instruction: %shift = shl <64 x i8> %a, <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
+; SSE42-LABEL: 'constant_shift_v64i8'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 44 for instruction: %shift = shl <64 x i8> %a, <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
 ;
 ; AVX1-LABEL: 'constant_shift_v64i8'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %shift = shl <64 x i8> %a, <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0>
@@ -1617,9 +1621,9 @@ define <4 x i32> @test3(<4 x i32> %a) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %shl = shl <4 x i32> %a, <i32 1, i32 -1, i32 2, i32 -3>
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shl
 ;
-; SSE41-LABEL: 'test3'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %shl = shl <4 x i32> %a, <i32 1, i32 -1, i32 2, i32 -3>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shl
+; SSE42-LABEL: 'test3'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %shl = shl <4 x i32> %a, <i32 1, i32 -1, i32 2, i32 -3>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shl
 ;
 ; AVX1-LABEL: 'test3'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %shl = shl <4 x i32> %a, <i32 1, i32 -1, i32 2, i32 -3>
@@ -1646,9 +1650,9 @@ define <4 x i32> @test4(<4 x i32> %a) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %shl = shl <4 x i32> %a, <i32 0, i32 0, i32 1, i32 1>
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shl
 ;
-; SSE41-LABEL: 'test4'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %shl = shl <4 x i32> %a, <i32 0, i32 0, i32 1, i32 1>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shl
+; SSE42-LABEL: 'test4'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %shl = shl <4 x i32> %a, <i32 0, i32 0, i32 1, i32 1>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shl
 ;
 ; AVX1-LABEL: 'test4'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %shl = shl <4 x i32> %a, <i32 0, i32 0, i32 1, i32 1>
@@ -1750,9 +1754,9 @@ define <8 x i32> @test7(<8 x i32> %a) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %shl = shl <8 x i32> %a, <i32 1, i32 1, i32 2, i32 3, i32 1, i32 1, i32 2, i32 3>
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shl
 ;
-; SSE41-LABEL: 'test7'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %shl = shl <8 x i32> %a, <i32 1, i32 1, i32 2, i32 3, i32 1, i32 1, i32 2, i32 3>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shl
+; SSE42-LABEL: 'test7'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %shl = shl <8 x i32> %a, <i32 1, i32 1, i32 2, i32 3, i32 1, i32 1, i32 2, i32 3>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shl
 ;
 ; AVX1-LABEL: 'test7'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %shl = shl <8 x i32> %a, <i32 1, i32 1, i32 2, i32 3, i32 1, i32 1, i32 2, i32 3>
@@ -1862,9 +1866,9 @@ define <16 x i32> @test10(<16 x i32> %a) {
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %shl = shl <16 x i32> %a, <i32 1, i32 1, i32 2, i32 3, i32 1, i32 1, i32 2, i32 3, i32 1, i32 1, i32 2, i32 3, i32 1, i32 1, i32 2, i32 3>
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shl
 ;
-; SSE41-LABEL: 'test10'
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %shl = shl <16 x i32> %a, <i32 1, i32 1, i32 2, i32 3, i32 1, i32 1, i32 2, i32 3, i32 1, i32 1, i32 2, i32 3, i32 1, i32 1, i32 2, i32 3>
-; SSE41-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shl
+; SSE42-LABEL: 'test10'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %shl = shl <16 x i32> %a, <i32 1, i32 1, i32 2, i32 3, i32 1, i32 1, i32 2, i32 3, i32 1, i32 1, i32 2, i32 3, i32 1, i32 1, i32 2, i32 3>
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shl
 ;
 ; AVX1-LABEL: 'test10'
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %shl = shl <16 x i32> %a, <i32 1, i32 1, i32 2, i32 3, i32 1, i32 1, i32 2, i32 3, i32 1, i32 1, i32 2, i32 3, i32 1, i32 1, i32 2, i32 3>