From b59c2d9d736b8e5c7b25b8d9aff04437d3d2e17e Mon Sep 17 00:00:00 2001
From: Simon Pilgrim <llvm-dev@redking.me.uk>
Date: Thu, 10 Aug 2017 19:32:35 +0000
Subject: [PATCH] [CostModel][X86] Add SSE2 two-src shuffle costs

llvm-svn: 310654
---
 llvm/lib/Target/X86/X86TargetTransformInfo.cpp     |  2 ++
 .../Analysis/CostModel/X86/shuffle-single-src.ll   | 10 +++++-----
 .../test/Analysis/CostModel/X86/shuffle-two-src.ll | 14 +++++++-------
 3 files changed, 14 insertions(+), 12 deletions(-)

diff --git a/llvm/lib/Target/X86/X86TargetTransformInfo.cpp b/llvm/lib/Target/X86/X86TargetTransformInfo.cpp
index 03dff0fb89553c..ed56e4512f68e0 100644
--- a/llvm/lib/Target/X86/X86TargetTransformInfo.cpp
+++ b/llvm/lib/Target/X86/X86TargetTransformInfo.cpp
@@ -973,6 +973,8 @@ int X86TTIImpl::getShuffleCost(TTI::ShuffleKind Kind, Type *Tp, int Index,
     { TTI::SK_PermuteTwoSrc,    MVT::v2f64,  1 }, // shufpd
     { TTI::SK_PermuteTwoSrc,    MVT::v2i64,  1 }, // shufpd
     { TTI::SK_PermuteTwoSrc,    MVT::v4i32,  2 }, // 2*{unpck,movsd,pshufd}
+    { TTI::SK_PermuteTwoSrc,    MVT::v8i16,  8 }, // blend+permute
+    { TTI::SK_PermuteTwoSrc,    MVT::v16i8, 13 }, // blend+permute
   };
 
   if (ST->hasSSE2())
diff --git a/llvm/test/Analysis/CostModel/X86/shuffle-single-src.ll b/llvm/test/Analysis/CostModel/X86/shuffle-single-src.ll
index 842c72a335ae30..590f80d42539fc 100644
--- a/llvm/test/Analysis/CostModel/X86/shuffle-single-src.ll
+++ b/llvm/test/Analysis/CostModel/X86/shuffle-single-src.ll
@@ -159,7 +159,7 @@ define void @test_vXi16(<8 x i16> %src128, <16 x i16> %src256, <32 x i16> %src51
   ; AVX512VBMI: cost of 1 {{.*}} %V128 = shufflevector
   %V128 = shufflevector <8 x i16> %src128, <8 x i16> undef, <8 x i32> <i32 7, i32 6, i32 6, i32 4, i32 3, i32 2, i32 1, i32 0>
 
-  ; SSE2: cost of 32 {{.*}} %V256 = shufflevector
+  ; SSE2: cost of 16 {{.*}} %V256 = shufflevector
   ; SSSE3: cost of 6 {{.*}} %V256 = shufflevector
   ; SSE42: cost of 6 {{.*}} %V256 = shufflevector
   ; AVX1: cost of 8 {{.*}} %V256 = shufflevector
@@ -169,7 +169,7 @@ define void @test_vXi16(<8 x i16> %src128, <16 x i16> %src256, <32 x i16> %src51
   ; AVX512VBMI cost of 1 {{.*}} %V256 = shufflevector
   %V256 = shufflevector <16 x i16> %src256, <16 x i16> undef, <16 x i32> <i32 15, i32 14, i32 13, i32 13, i32 11, i32 10, i32 9, i32 8, i32 7, i32 6, i32 5, i32 4, i32 3, i32 2, i32 1, i32 0>
 
-  ; SSE2: cost of 192 {{.*}} %V512 = shufflevector
+  ; SSE2: cost of 96 {{.*}} %V512 = shufflevector
   ; SSSE3: cost of 36 {{.*}} %V512 = shufflevector
   ; SSE42: cost of 36 {{.*}} %V512 = shufflevector
   ; AVX1: cost of 30 {{.*}} %V512 = shufflevector
@@ -179,7 +179,7 @@ define void @test_vXi16(<8 x i16> %src128, <16 x i16> %src256, <32 x i16> %src51
   ; AVX512VBMI: cost of 1 {{.*}} %V512 = shufflevector
   %V512 = shufflevector <32 x i16> %src512, <32 x i16> undef, <32 x i32> <i32 31, i32 30, i32 20, i32 28, i32 27, i32 26, i32 25, i32 24, i32 23, i32 22, i32 21, i32 20, i32 19, i32 18, i32 17, i32 16, i32 15, i32 14, i32 13, i32 12, i32 11, i32 11, i32 9, i32 8, i32 7, i32 11, i32 5, i32 4, i32 3, i32 2, i32 1, i32 0>
 
-  ; SSE2: cost of 896 {{.*}} %V1024 = shufflevector
+  ; SSE2: cost of 448 {{.*}} %V1024 = shufflevector
   ; SSSE3: cost of 168 {{.*}} %V1024 = shufflevector
   ; SSE42: cost of 168 {{.*}} %V1024 = shufflevector
   ; AVX1: cost of 180 {{.*}} %V1024 = shufflevector
@@ -201,7 +201,7 @@ define void @test_vXi8(<16 x i8> %src128, <32 x i8> %src256, <64 x i8> %src512)
   ; AVX512: cost of 1 {{.*}} %V128 = shufflevector
   %V128 = shufflevector <16 x i8> %src128, <16 x i8> undef, <16 x i32> <i32 15, i32 14, i32 13, i32 12, i32 11, i32 10, i32 11, i32 8, i32 7, i32 6, i32 5, i32 4, i32 3, i32 2, i32 1, i32 0>
 
-  ; SSE2: cost of 64 {{.*}} %V256 = shufflevector
+  ; SSE2: cost of 26 {{.*}} %V256 = shufflevector
   ; SSSE3: cost of 6 {{.*}} %V256 = shufflevector
   ; SSE42: cost of 6 {{.*}} %V256 = shufflevector
   ; AVX1: cost of 8 {{.*}} %V256 = shufflevector
@@ -211,7 +211,7 @@ define void @test_vXi8(<16 x i8> %src128, <32 x i8> %src256, <64 x i8> %src512)
   ; AVX512VBMI: cost of 1 {{.*}} %V256 = shufflevector
   %V256 = shufflevector <32 x i8> %src256, <32 x i8> undef, <32 x i32> <i32 31, i32 30, i32 29, i32 28, i32 27, i32 26, i32 25, i32 24, i32 23, i32 22, i32 21, i32 20, i32 19, i32 18, i32 17, i32 16, i32 15, i32 14, i32 13, i32 12, i32 11, i32 10, i32 8, i32 8, i32 7, i32 6, i32 8, i32 4, i32 3, i32 2, i32 1, i32 0>
 
-  ; SSE2: cost of 384 {{.*}} %V512 = shufflevector
+  ; SSE2: cost of 156 {{.*}} %V512 = shufflevector
   ; SSSE3: cost of 36 {{.*}} %V512 = shufflevector
   ; SSE42: cost of 36 {{.*}} %V512 = shufflevector
   ; AVX1: cost of 30 {{.*}} %V512 = shufflevector
diff --git a/llvm/test/Analysis/CostModel/X86/shuffle-two-src.ll b/llvm/test/Analysis/CostModel/X86/shuffle-two-src.ll
index dc6f5ceff5076e..1967397c3717bf 100644
--- a/llvm/test/Analysis/CostModel/X86/shuffle-two-src.ll
+++ b/llvm/test/Analysis/CostModel/X86/shuffle-two-src.ll
@@ -166,7 +166,7 @@ define void @test_vXi32(<4 x i32> %src128, <8 x i32> %src256, <16 x i32> %src512
 ; CHECK-LABEL: 'test_vXi16'
 define void @test_vXi16(<8 x i16> %src128, <16 x i16> %src256, <32 x i16> %src512, <64 x i16> %src1024, <8 x i16> %src128_1, <16 x i16> %src256_1, <32 x i16> %src512_1, <64 x i16> %src1024_1) {
 
-  ; SSE2: cost of 16 {{.*}} %V128 = shufflevector
+  ; SSE2: cost of 8 {{.*}} %V128 = shufflevector
   ; SSSE3: cost of 3 {{.*}} %V128 = shufflevector
   ; SSE42: cost of 3 {{.*}} %V128 = shufflevector
   ; AVX1: cost of 3 {{.*}} %V128 = shufflevector
@@ -176,7 +176,7 @@ define void @test_vXi16(<8 x i16> %src128, <16 x i16> %src256, <32 x i16> %src51
   ; AVX512VBMI: cost of 1 {{.*}} %V128 = shufflevector
   %V128 = shufflevector <8 x i16> %src128, <8 x i16> %src128_1, <8 x i32> <i32 7, i32 6, i32 6, i32 8, i32 9, i32 2, i32 1, i32 0>
 
-  ; SSE2: cost of 32 {{.*}} %V256 = shufflevector
+  ; SSE2: cost of 48 {{.*}} %V256 = shufflevector
   ; SSSE3: cost of 18 {{.*}} %V256 = shufflevector
   ; SSE42: cost of 18 {{.*}} %V256 = shufflevector
   ; AVX1: cost of 15 {{.*}} %V256 = shufflevector
@@ -186,7 +186,7 @@ define void @test_vXi16(<8 x i16> %src128, <16 x i16> %src256, <32 x i16> %src51
   ; AVX512VBMI: cost of 1 {{.*}} %V256 = shufflevector
   %V256 = shufflevector <16 x i16> %src256, <16 x i16> %src256_1, <16 x i32> <i32 15, i32 14, i32 13, i32 20, i32 21, i32 10, i32 9, i32 8, i32 7, i32 6, i32 5, i32 4, i32 3, i32 2, i32 1, i32 0>
 
-  ; SSE2: cost of 64 {{.*}} %V512 = shufflevector
+  ; SSE2: cost of 224 {{.*}} %V512 = shufflevector
   ; SSSE3: cost of 84 {{.*}} %V512 = shufflevector
   ; SSE42: cost of 84 {{.*}} %V512 = shufflevector
   ; AVX1: cost of 90 {{.*}} %V512 = shufflevector
@@ -196,7 +196,7 @@ define void @test_vXi16(<8 x i16> %src128, <16 x i16> %src256, <32 x i16> %src51
   ; AVX512VBMI: cost of 1 {{.*}} %V512 = shufflevector
   %V512 = shufflevector <32 x i16> %src512, <32 x i16> %src512_1, <32 x i32> <i32 31, i32 30, i32 45, i32 28, i32 27, i32 26, i32 25, i32 24, i32 23, i32 22, i32 21, i32 20, i32 19, i32 18, i32 17, i32 16, i32 15, i32 14, i32 13, i32 38, i32 11, i32 11, i32 9, i32 8, i32 7, i32 11, i32 5, i32 4, i32 3, i32 2, i32 1, i32 0>
 
-  ; SSE2: cost of 128 {{.*}} %V1024 = shufflevector
+  ; SSE2: cost of 960 {{.*}} %V1024 = shufflevector
   ; SSSE3: cost of 360 {{.*}} %V1024 = shufflevector
   ; SSE42: cost of 360 {{.*}} %V1024 = shufflevector
   ; AVX1: cost of 420 {{.*}} %V1024 = shufflevector
@@ -212,7 +212,7 @@ define void @test_vXi16(<8 x i16> %src128, <16 x i16> %src256, <32 x i16> %src51
 ; CHECK-LABEL: 'test_vXi8'
 define void @test_vXi8(<16 x i8> %src128, <32 x i8> %src256, <64 x i8> %src512, <16 x i8> %src128_1, <32 x i8> %src256_1, <64 x i8> %src512_1) {
 
-  ; SSE2: cost of 32 {{.*}} %V128 = shufflevector
+  ; SSE2: cost of 13 {{.*}} %V128 = shufflevector
   ; SSSE3: cost of 3 {{.*}} %V128 = shufflevector
   ; SSE42: cost of 3 {{.*}} %V128 = shufflevector
   ; AVX1: cost of 3 {{.*}} %V128 = shufflevector
@@ -222,7 +222,7 @@ define void @test_vXi8(<16 x i8> %src128, <32 x i8> %src256, <64 x i8> %src512,
   ; AVX512VBMI: cost of 1 {{.*}} %V128 = shufflevector
   %V128 = shufflevector <16 x i8> %src128, <16 x i8> %src128_1, <16 x i32> <i32 29, i32 14, i32 28, i32 12, i32 11, i32 10, i32 11, i32 8, i32 7, i32 6, i32 5, i32 4, i32 3, i32 2, i32 1, i32 0>
 
-  ; SSE2: cost of 64 {{.*}} %V256 = shufflevector
+  ; SSE2: cost of 78 {{.*}} %V256 = shufflevector
   ; SSSE3: cost of 18 {{.*}} %V256 = shufflevector
   ; SSE42: cost of 18 {{.*}} %V256 = shufflevector
   ; AVX1: cost of 15 {{.*}} %V256 = shufflevector
@@ -232,7 +232,7 @@ define void @test_vXi8(<16 x i8> %src128, <32 x i8> %src256, <64 x i8> %src512,
   ; AVX512VBMI: cost of 1 {{.*}} %V256 = shufflevector
   %V256 = shufflevector <32 x i8> %src256, <32 x i8> %src256_1, <32 x i32> <i32 31, i32 30, i32 45, i32 28, i32 27, i32 26, i32 25, i32 24, i32 23, i32 22, i32 21, i32 20, i32 19, i32 18, i32 17, i32 16, i32 15, i32 14, i32 13, i32 12, i32 11, i32 10, i32 8, i32 8, i32 7, i32 6, i32 8, i32 4, i32 3, i32 2, i32 1, i32 0>
 
-  ; SSE2: cost of 128 {{.*}} %V512 = shufflevector
+  ; SSE2: cost of 364 {{.*}} %V512 = shufflevector
   ; SSSE3: cost of 84 {{.*}} %V512 = shufflevector
   ; SSE42: cost of 84 {{.*}} %V512 = shufflevector
   ; AVX1: cost of 90 {{.*}} %V512 = shufflevector