[x86] add cost model special-case for insert/extract from element 0

This is a follow-up to D70607 where we made any
extract element on SLM more costly than default. But that is
pessimistic for extract from element 0 because that corresponds
to x86 movd/movq instructions. These generally have >1 cycle
latency, but they are probably implemented as single uop
instructions.

Note that no vectorization tests are affected by this change.
Also, no targets besides SLM are affected because those are
falling through to the default cost of 1 anyway. But this will
become visible/important if we add more specializations via cost
tables.

Differential Revision: https://reviews.llvm.org/D71023

4 files changed, 62 insertions, 62 deletions

diff --git a/llvm/test/Analysis/CostModel/X86/fptosi.ll b/llvm/test/Analysis/CostModel/X86/fptosi.ll
index bb03b56e48f..35a86fb131c 100644
--- a/llvm/test/Analysis/CostModel/X86/fptosi.ll
+++ b/llvm/test/Analysis/CostModel/X86/fptosi.ll
@@ -41,9 +41,9 @@ define i32 @fptosi_double_i64(i32 %arg) {
 ;
 ; SLM-LABEL: 'fptosi_double_i64'
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I64 = fptosi double undef to i64
-; SLM-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %V2I64 = fptosi <2 x double> undef to <2 x i64>
-; SLM-NEXT:  Cost Model: Found an estimated cost of 37 for instruction: %V4I64 = fptosi <4 x double> undef to <4 x i64>
-; SLM-NEXT:  Cost Model: Found an estimated cost of 75 for instruction: %V8I64 = fptosi <8 x double> undef to <8 x i64>
+; SLM-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V2I64 = fptosi <2 x double> undef to <2 x i64>
+; SLM-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %V4I64 = fptosi <4 x double> undef to <4 x i64>
+; SLM-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %V8I64 = fptosi <8 x double> undef to <8 x i64>
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; BTVER2-LABEL: 'fptosi_double_i64'
@@ -170,9 +170,9 @@ define i32 @fptosi_double_i8(i32 %arg) {
 ;
 ; SLM-LABEL: 'fptosi_double_i8'
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = fptosi double undef to i8
-; SLM-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V2I8 = fptosi <2 x double> undef to <2 x i8>
-; SLM-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %V4I8 = fptosi <4 x double> undef to <4 x i8>
-; SLM-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %V8I8 = fptosi <8 x double> undef to <8 x i8>
+; SLM-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V2I8 = fptosi <2 x double> undef to <2 x i8>
+; SLM-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4I8 = fptosi <4 x double> undef to <4 x i8>
+; SLM-NEXT:  Cost Model: Found an estimated cost of 39 for instruction: %V8I8 = fptosi <8 x double> undef to <8 x i8>
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; BTVER2-LABEL: 'fptosi_double_i8'
@@ -224,10 +224,10 @@ define i32 @fptosi_float_i64(i32 %arg) {
 ;
 ; SLM-LABEL: 'fptosi_float_i64'
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I64 = fptosi float undef to i64
-; SLM-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %V2I64 = fptosi <2 x float> undef to <2 x i64>
-; SLM-NEXT:  Cost Model: Found an estimated cost of 37 for instruction: %V4I64 = fptosi <4 x float> undef to <4 x i64>
-; SLM-NEXT:  Cost Model: Found an estimated cost of 75 for instruction: %V8I64 = fptosi <8 x float> undef to <8 x i64>
-; SLM-NEXT:  Cost Model: Found an estimated cost of 151 for instruction: %V16I64 = fptosi <16 x float> undef to <16 x i64>
+; SLM-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V2I64 = fptosi <2 x float> undef to <2 x i64>
+; SLM-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %V4I64 = fptosi <4 x float> undef to <4 x i64>
+; SLM-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %V8I64 = fptosi <8 x float> undef to <8 x i64>
+; SLM-NEXT:  Cost Model: Found an estimated cost of 103 for instruction: %V16I64 = fptosi <16 x float> undef to <16 x i64>
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; BTVER2-LABEL: 'fptosi_float_i64'
@@ -342,9 +342,9 @@ define i32 @fptosi_float_i8(i32 %arg) {
 ;
 ; SLM-LABEL: 'fptosi_float_i8'
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = fptosi float undef to i8
-; SLM-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V4I8 = fptosi <4 x float> undef to <4 x i8>
-; SLM-NEXT:  Cost Model: Found an estimated cost of 49 for instruction: %V8I8 = fptosi <8 x float> undef to <8 x i8>
-; SLM-NEXT:  Cost Model: Found an estimated cost of 99 for instruction: %V16I8 = fptosi <16 x float> undef to <16 x i8>
+; SLM-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V4I8 = fptosi <4 x float> undef to <4 x i8>
+; SLM-NEXT:  Cost Model: Found an estimated cost of 43 for instruction: %V8I8 = fptosi <8 x float> undef to <8 x i8>
+; SLM-NEXT:  Cost Model: Found an estimated cost of 87 for instruction: %V16I8 = fptosi <16 x float> undef to <16 x i8>
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; BTVER2-LABEL: 'fptosi_float_i8'
diff --git a/llvm/test/Analysis/CostModel/X86/fptoui.ll b/llvm/test/Analysis/CostModel/X86/fptoui.ll
index cdb3e548660..c3b2cb513c7 100644
--- a/llvm/test/Analysis/CostModel/X86/fptoui.ll
+++ b/llvm/test/Analysis/CostModel/X86/fptoui.ll
@@ -41,9 +41,9 @@ define i32 @fptoui_double_i64(i32 %arg) {
 ;
 ; SLM-LABEL: 'fptoui_double_i64'
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %I64 = fptoui double undef to i64
-; SLM-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V2I64 = fptoui <2 x double> undef to <2 x i64>
-; SLM-NEXT:  Cost Model: Found an estimated cost of 49 for instruction: %V4I64 = fptoui <4 x double> undef to <4 x i64>
-; SLM-NEXT:  Cost Model: Found an estimated cost of 99 for instruction: %V8I64 = fptoui <8 x double> undef to <8 x i64>
+; SLM-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %V2I64 = fptoui <2 x double> undef to <2 x i64>
+; SLM-NEXT:  Cost Model: Found an estimated cost of 37 for instruction: %V4I64 = fptoui <4 x double> undef to <4 x i64>
+; SLM-NEXT:  Cost Model: Found an estimated cost of 75 for instruction: %V8I64 = fptoui <8 x double> undef to <8 x i64>
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; BTVER2-LABEL: 'fptoui_double_i64'
@@ -84,9 +84,9 @@ define i32 @fptoui_double_i32(i32 %arg) {
 ;
 ; SLM-LABEL: 'fptoui_double_i32'
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I32 = fptoui double undef to i32
-; SLM-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V2I32 = fptoui <2 x double> undef to <2 x i32>
-; SLM-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %V4I32 = fptoui <4 x double> undef to <4 x i32>
-; SLM-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %V8I32 = fptoui <8 x double> undef to <8 x i32>
+; SLM-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V2I32 = fptoui <2 x double> undef to <2 x i32>
+; SLM-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4I32 = fptoui <4 x double> undef to <4 x i32>
+; SLM-NEXT:  Cost Model: Found an estimated cost of 39 for instruction: %V8I32 = fptoui <8 x double> undef to <8 x i32>
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; BTVER2-LABEL: 'fptoui_double_i32'
@@ -127,9 +127,9 @@ define i32 @fptoui_double_i16(i32 %arg) {
 ;
 ; SLM-LABEL: 'fptoui_double_i16'
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = fptoui double undef to i16
-; SLM-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V2I16 = fptoui <2 x double> undef to <2 x i16>
-; SLM-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %V4I16 = fptoui <4 x double> undef to <4 x i16>
-; SLM-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %V8I16 = fptoui <8 x double> undef to <8 x i16>
+; SLM-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V2I16 = fptoui <2 x double> undef to <2 x i16>
+; SLM-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4I16 = fptoui <4 x double> undef to <4 x i16>
+; SLM-NEXT:  Cost Model: Found an estimated cost of 39 for instruction: %V8I16 = fptoui <8 x double> undef to <8 x i16>
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; BTVER2-LABEL: 'fptoui_double_i16'
@@ -170,9 +170,9 @@ define i32 @fptoui_double_i8(i32 %arg) {
 ;
 ; SLM-LABEL: 'fptoui_double_i8'
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = fptoui double undef to i8
-; SLM-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V2I8 = fptoui <2 x double> undef to <2 x i8>
-; SLM-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %V4I8 = fptoui <4 x double> undef to <4 x i8>
-; SLM-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %V8I8 = fptoui <8 x double> undef to <8 x i8>
+; SLM-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V2I8 = fptoui <2 x double> undef to <2 x i8>
+; SLM-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4I8 = fptoui <4 x double> undef to <4 x i8>
+; SLM-NEXT:  Cost Model: Found an estimated cost of 39 for instruction: %V8I8 = fptoui <8 x double> undef to <8 x i8>
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; BTVER2-LABEL: 'fptoui_double_i8'
@@ -224,10 +224,10 @@ define i32 @fptoui_float_i64(i32 %arg) {
 ;
 ; SLM-LABEL: 'fptoui_float_i64'
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %I64 = fptoui float undef to i64
-; SLM-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V2I64 = fptoui <2 x float> undef to <2 x i64>
-; SLM-NEXT:  Cost Model: Found an estimated cost of 49 for instruction: %V4I64 = fptoui <4 x float> undef to <4 x i64>
-; SLM-NEXT:  Cost Model: Found an estimated cost of 99 for instruction: %V8I64 = fptoui <8 x float> undef to <8 x i64>
-; SLM-NEXT:  Cost Model: Found an estimated cost of 199 for instruction: %V16I64 = fptoui <16 x float> undef to <16 x i64>
+; SLM-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %V2I64 = fptoui <2 x float> undef to <2 x i64>
+; SLM-NEXT:  Cost Model: Found an estimated cost of 37 for instruction: %V4I64 = fptoui <4 x float> undef to <4 x i64>
+; SLM-NEXT:  Cost Model: Found an estimated cost of 75 for instruction: %V8I64 = fptoui <8 x float> undef to <8 x i64>
+; SLM-NEXT:  Cost Model: Found an estimated cost of 151 for instruction: %V16I64 = fptoui <16 x float> undef to <16 x i64>
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; BTVER2-LABEL: 'fptoui_float_i64'
@@ -270,9 +270,9 @@ define i32 @fptoui_float_i32(i32 %arg) {
 ;
 ; SLM-LABEL: 'fptoui_float_i32'
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I32 = fptoui float undef to i32
-; SLM-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V4I32 = fptoui <4 x float> undef to <4 x i32>
-; SLM-NEXT:  Cost Model: Found an estimated cost of 49 for instruction: %V8I32 = fptoui <8 x float> undef to <8 x i32>
-; SLM-NEXT:  Cost Model: Found an estimated cost of 99 for instruction: %V16I32 = fptoui <16 x float> undef to <16 x i32>
+; SLM-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V4I32 = fptoui <4 x float> undef to <4 x i32>
+; SLM-NEXT:  Cost Model: Found an estimated cost of 43 for instruction: %V8I32 = fptoui <8 x float> undef to <8 x i32>
+; SLM-NEXT:  Cost Model: Found an estimated cost of 87 for instruction: %V16I32 = fptoui <16 x float> undef to <16 x i32>
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; BTVER2-LABEL: 'fptoui_float_i32'
@@ -313,9 +313,9 @@ define i32 @fptoui_float_i16(i32 %arg) {
 ;
 ; SLM-LABEL: 'fptoui_float_i16'
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = fptoui float undef to i16
-; SLM-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V4I16 = fptoui <4 x float> undef to <4 x i16>
-; SLM-NEXT:  Cost Model: Found an estimated cost of 49 for instruction: %V8I16 = fptoui <8 x float> undef to <8 x i16>
-; SLM-NEXT:  Cost Model: Found an estimated cost of 99 for instruction: %V16I16 = fptoui <16 x float> undef to <16 x i16>
+; SLM-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V4I16 = fptoui <4 x float> undef to <4 x i16>
+; SLM-NEXT:  Cost Model: Found an estimated cost of 43 for instruction: %V8I16 = fptoui <8 x float> undef to <8 x i16>
+; SLM-NEXT:  Cost Model: Found an estimated cost of 87 for instruction: %V16I16 = fptoui <16 x float> undef to <16 x i16>
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; BTVER2-LABEL: 'fptoui_float_i16'
@@ -356,9 +356,9 @@ define i32 @fptoui_float_i8(i32 %arg) {
 ;
 ; SLM-LABEL: 'fptoui_float_i8'
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = fptoui float undef to i8
-; SLM-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V4I8 = fptoui <4 x float> undef to <4 x i8>
-; SLM-NEXT:  Cost Model: Found an estimated cost of 49 for instruction: %V8I8 = fptoui <8 x float> undef to <8 x i8>
-; SLM-NEXT:  Cost Model: Found an estimated cost of 99 for instruction: %V16I8 = fptoui <16 x float> undef to <16 x i8>
+; SLM-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V4I8 = fptoui <4 x float> undef to <4 x i8>
+; SLM-NEXT:  Cost Model: Found an estimated cost of 43 for instruction: %V8I8 = fptoui <8 x float> undef to <8 x i8>
+; SLM-NEXT:  Cost Model: Found an estimated cost of 87 for instruction: %V16I8 = fptoui <16 x float> undef to <16 x i8>
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; BTVER2-LABEL: 'fptoui_float_i8'
diff --git a/llvm/test/Analysis/CostModel/X86/shuffle-extract_subvector.ll b/llvm/test/Analysis/CostModel/X86/shuffle-extract_subvector.ll
index 4ed509ff9db..a794a9a3d19 100644
--- a/llvm/test/Analysis/CostModel/X86/shuffle-extract_subvector.ll
+++ b/llvm/test/Analysis/CostModel/X86/shuffle-extract_subvector.ll
@@ -585,7 +585,7 @@ define void @test_vXi16(<4 x i16> %src64, <8 x i16> %src128, <16 x i16> %src256,
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V256_0123 = shufflevector <16 x i16> %src256, <16 x i16> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %V256_2345 = shufflevector <16 x i16> %src256, <16 x i16> undef, <4 x i32> <i32 2, i32 3, i32 4, i32 5>
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V256_4567 = shufflevector <16 x i16> %src256, <16 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
-; SLM-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %V256_6789 = shufflevector <16 x i16> %src256, <16 x i16> undef, <4 x i32> <i32 6, i32 7, i32 8, i32 9>
+; SLM-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V256_6789 = shufflevector <16 x i16> %src256, <16 x i16> undef, <4 x i32> <i32 6, i32 7, i32 8, i32 9>
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V256_89AB = shufflevector <16 x i16> %src256, <16 x i16> undef, <4 x i32> <i32 8, i32 9, i32 10, i32 11>
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V256_CDEF = shufflevector <16 x i16> %src256, <16 x i16> undef, <4 x i32> <i32 12, i32 13, i32 14, i32 15>
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V256_01234567 = shufflevector <16 x i16> %src256, <16 x i16> undef, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
@@ -609,7 +609,7 @@ define void @test_vXi16(<4 x i16> %src64, <8 x i16> %src128, <16 x i16> %src256,
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V512_00_01_02_03 = shufflevector <32 x i16> %src512, <32 x i16> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %V512_02_03_04_05 = shufflevector <32 x i16> %src512, <32 x i16> undef, <4 x i32> <i32 2, i32 3, i32 4, i32 5>
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V512_04_05_06_07 = shufflevector <32 x i16> %src512, <32 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
-; SLM-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %V512_06_07_08_09 = shufflevector <32 x i16> %src512, <32 x i16> undef, <4 x i32> <i32 6, i32 7, i32 8, i32 9>
+; SLM-NEXT:  Cost Model: Found an estimated cost of 53 for instruction: %V512_06_07_08_09 = shufflevector <32 x i16> %src512, <32 x i16> undef, <4 x i32> <i32 6, i32 7, i32 8, i32 9>
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V512_08_09_0A_0B = shufflevector <32 x i16> %src512, <32 x i16> undef, <4 x i32> <i32 8, i32 9, i32 10, i32 11>
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V512_0C_0D_0E_0F = shufflevector <32 x i16> %src512, <32 x i16> undef, <4 x i32> <i32 12, i32 13, i32 14, i32 15>
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V512_10_11_12_13 = shufflevector <32 x i16> %src512, <32 x i16> undef, <4 x i32> <i32 16, i32 17, i32 18, i32 19>
diff --git a/llvm/test/Analysis/CostModel/X86/vector-extract.ll b/llvm/test/Analysis/CostModel/X86/vector-extract.ll
index ddb3654fbc6..974bcc634f9 100644
--- a/llvm/test/Analysis/CostModel/X86/vector-extract.ll
+++ b/llvm/test/Analysis/CostModel/X86/vector-extract.ll
@@ -274,15 +274,15 @@ define i32 @extract_i64(i32 %arg) {
 ;
 ; SLM-LABEL: 'extract_i64'
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2i64_a = extractelement <2 x i64> undef, i32 %arg
-; SLM-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v2i64_0 = extractelement <2 x i64> undef, i32 0
+; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2i64_0 = extractelement <2 x i64> undef, i32 0
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v2i64_1 = extractelement <2 x i64> undef, i32 1
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v4i64_a = extractelement <4 x i64> undef, i32 %arg
-; SLM-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %v4i64_0 = extractelement <4 x i64> undef, i32 0
+; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v4i64_0 = extractelement <4 x i64> undef, i32 0
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %v4i64_3 = extractelement <4 x i64> undef, i32 3
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v8i64_a = extractelement <8 x i64> undef, i32 %arg
-; SLM-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %v8i64_0 = extractelement <8 x i64> undef, i32 0
+; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v8i64_0 = extractelement <8 x i64> undef, i32 0
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %v8i64_3 = extractelement <8 x i64> undef, i32 3
-; SLM-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %v8i64_4 = extractelement <8 x i64> undef, i32 4
+; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v8i64_4 = extractelement <8 x i64> undef, i32 4
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %v8i64_7 = extractelement <8 x i64> undef, i32 7
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
@@ -448,20 +448,20 @@ define i32 @extract_i32(i32 %arg) {
 ;
 ; SLM-LABEL: 'extract_i32'
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2i32_a = extractelement <2 x i32> undef, i32 %arg
-; SLM-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v2i32_0 = extractelement <2 x i32> undef, i32 0
+; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2i32_0 = extractelement <2 x i32> undef, i32 0
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v2i32_1 = extractelement <2 x i32> undef, i32 1
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v4i32_a = extractelement <4 x i32> undef, i32 %arg
-; SLM-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v4i32_0 = extractelement <4 x i32> undef, i32 0
+; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v4i32_0 = extractelement <4 x i32> undef, i32 0
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v4i32_3 = extractelement <4 x i32> undef, i32 3
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v8i32_a = extractelement <8 x i32> undef, i32 %arg
-; SLM-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v8i32_0 = extractelement <8 x i32> undef, i32 0
+; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v8i32_0 = extractelement <8 x i32> undef, i32 0
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v8i32_3 = extractelement <8 x i32> undef, i32 3
-; SLM-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v8i32_4 = extractelement <8 x i32> undef, i32 4
+; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v8i32_4 = extractelement <8 x i32> undef, i32 4
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v8i32_7 = extractelement <8 x i32> undef, i32 7
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v16i32_a = extractelement <16 x i32> undef, i32 %arg
-; SLM-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v16i32_0 = extractelement <16 x i32> undef, i32 0
+; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v16i32_0 = extractelement <16 x i32> undef, i32 0
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v16i32_3 = extractelement <16 x i32> undef, i32 3
-; SLM-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v16i32_8 = extractelement <16 x i32> undef, i32 8
+; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v16i32_8 = extractelement <16 x i32> undef, i32 8
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v16i32_15 = extractelement <16 x i32> undef, i32 15
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
@@ -643,20 +643,20 @@ define i32 @extract_i16(i32 %arg) {
 ;
 ; SLM-LABEL: 'extract_i16'
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v8i16_a = extractelement <8 x i16> undef, i32 %arg
-; SLM-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v8i16_0 = extractelement <8 x i16> undef, i32 0
+; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v8i16_0 = extractelement <8 x i16> undef, i32 0
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v8i16_7 = extractelement <8 x i16> undef, i32 7
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v16i16_a = extractelement <16 x i16> undef, i32 %arg
-; SLM-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v16i16_0 = extractelement <16 x i16> undef, i32 0
+; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v16i16_0 = extractelement <16 x i16> undef, i32 0
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v16i16_7 = extractelement <16 x i16> undef, i32 7
-; SLM-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v16i16_8 = extractelement <16 x i16> undef, i32 8
+; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v16i16_8 = extractelement <16 x i16> undef, i32 8
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v16i16_15 = extractelement <16 x i16> undef, i32 15
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v32i16_a = extractelement <32 x i16> undef, i32 %arg
-; SLM-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v32i16_0 = extractelement <32 x i16> undef, i32 0
+; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v32i16_0 = extractelement <32 x i16> undef, i32 0
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v32i16_7 = extractelement <32 x i16> undef, i32 7
-; SLM-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v32i16_8 = extractelement <32 x i16> undef, i32 8
+; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v32i16_8 = extractelement <32 x i16> undef, i32 8
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v32i16_15 = extractelement <32 x i16> undef, i32 15
-; SLM-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v32i16_16 = extractelement <32 x i16> undef, i32 16
-; SLM-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v32i16_24 = extractelement <32 x i16> undef, i32 24
+; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v32i16_16 = extractelement <32 x i16> undef, i32 16
+; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v32i16_24 = extractelement <32 x i16> undef, i32 24
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v32i16_31 = extractelement <32 x i16> undef, i32 31
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
@@ -867,25 +867,25 @@ define i32 @extract_i8(i32 %arg) {
 ;
 ; SLM-LABEL: 'extract_i8'
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v16i8_a = extractelement <16 x i8> undef, i32 %arg
-; SLM-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v16i8_0 = extractelement <16 x i8> undef, i32 0
+; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v16i8_0 = extractelement <16 x i8> undef, i32 0
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v16i8_8 = extractelement <16 x i8> undef, i32 8
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v16i8_15 = extractelement <16 x i8> undef, i32 15
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v32i8_a = extractelement <32 x i8> undef, i32 %arg
-; SLM-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v32i8_0 = extractelement <32 x i8> undef, i32 0
+; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v32i8_0 = extractelement <32 x i8> undef, i32 0
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v32i8_7 = extractelement <32 x i8> undef, i32 7
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v32i8_8 = extractelement <32 x i8> undef, i32 8
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v32i8_15 = extractelement <32 x i8> undef, i32 15
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v32i8_24 = extractelement <32 x i8> undef, i32 24
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v32i8_31 = extractelement <32 x i8> undef, i32 31
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v64i8_a = extractelement <64 x i8> undef, i32 %arg
-; SLM-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v64i8_0 = extractelement <64 x i8> undef, i32 0
+; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v64i8_0 = extractelement <64 x i8> undef, i32 0
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v64i8_7 = extractelement <64 x i8> undef, i32 7
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v64i8_8 = extractelement <64 x i8> undef, i32 8
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v64i8_15 = extractelement <64 x i8> undef, i32 15
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v64i8_24 = extractelement <64 x i8> undef, i32 24
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v64i8_31 = extractelement <64 x i8> undef, i32 31
-; SLM-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v64i8_32 = extractelement <64 x i8> undef, i32 32
-; SLM-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v64i8_48 = extractelement <64 x i8> undef, i32 48
+; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v64i8_32 = extractelement <64 x i8> undef, i32 32
+; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v64i8_48 = extractelement <64 x i8> undef, i32 48
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v64i8_63 = extractelement <64 x i8> undef, i32 63
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;