[llvm] def6269 - [CostModel][X86] Improve accuracy of 256-bit non-uniform vector shifts on AVX1
Simon Pilgrim via llvm-commits
llvm-commits at lists.llvm.org
Tue May 25 09:32:04 PDT 2021
Author: Simon Pilgrim
Date: 2021-05-25T17:31:45+01:00
New Revision: def6269779aff0d4f1994cf73f7910add044ff72
URL: https://github.com/llvm/llvm-project/commit/def6269779aff0d4f1994cf73f7910add044ff72
DIFF: https://github.com/llvm/llvm-project/commit/def6269779aff0d4f1994cf73f7910add044ff72.diff
LOG: [CostModel][X86] Improve accuracy of 256-bit non-uniform vector shifts on AVX1
Determined from llvm-mca analysis, AVX1 capable targets have a higher throughput for VPBLENDVB and shuffle ops, making it cheaper to perform shift+shuffle/select shift patterns.
Added:
Modified:
llvm/lib/Target/X86/X86TargetTransformInfo.cpp
llvm/test/Analysis/CostModel/X86/div.ll
llvm/test/Analysis/CostModel/X86/fshl.ll
llvm/test/Analysis/CostModel/X86/fshr.ll
llvm/test/Analysis/CostModel/X86/rem.ll
llvm/test/Analysis/CostModel/X86/vshift-ashr-cost-inseltpoison.ll
llvm/test/Analysis/CostModel/X86/vshift-ashr-cost.ll
llvm/test/Analysis/CostModel/X86/vshift-lshr-cost-inseltpoison.ll
llvm/test/Analysis/CostModel/X86/vshift-lshr-cost.ll
llvm/test/Analysis/CostModel/X86/vshift-shl-cost-inseltpoison.ll
llvm/test/Analysis/CostModel/X86/vshift-shl-cost.ll
llvm/test/Transforms/SLPVectorizer/X86/alternate-int-inseltpoison.ll
llvm/test/Transforms/SLPVectorizer/X86/alternate-int.ll
llvm/test/Transforms/SLPVectorizer/X86/powof2div.ll
llvm/test/Transforms/SLPVectorizer/X86/shift-ashr.ll
llvm/test/Transforms/SLPVectorizer/X86/shift-lshr.ll
llvm/test/Transforms/SLPVectorizer/X86/shift-shl.ll
Removed:
################################################################################
diff --git a/llvm/lib/Target/X86/X86TargetTransformInfo.cpp b/llvm/lib/Target/X86/X86TargetTransformInfo.cpp
index b0370bfae9707..838a1baa15a5a 100644
--- a/llvm/lib/Target/X86/X86TargetTransformInfo.cpp
+++ b/llvm/lib/Target/X86/X86TargetTransformInfo.cpp
@@ -845,6 +845,33 @@ InstructionCost X86TTIImpl::getArithmeticInstrCost(
{ ISD::SUB, MVT::v4i64, 4 },
{ ISD::ADD, MVT::v4i64, 4 },
+ { ISD::SHL, MVT::v16i8, 10 }, // pblendvb sequence .
+ { ISD::SHL, MVT::v32i8, 22 }, // pblendvb sequence + split.
+ { ISD::SHL, MVT::v8i16, 6 }, // pblendvb sequence.
+ { ISD::SHL, MVT::v16i16, 13 }, // pblendvb sequence + split.
+ { ISD::SHL, MVT::v4i32, 3 }, // pslld/paddd/cvttps2dq/pmulld
+ { ISD::SHL, MVT::v8i32, 9 }, // pslld/paddd/cvttps2dq/pmulld + split
+ { ISD::SHL, MVT::v2i64, 2 }, // Shift each lane + blend.
+ { ISD::SHL, MVT::v4i64, 6 }, // Shift each lane + blend + split.
+
+ { ISD::SRL, MVT::v16i8, 11 }, // pblendvb sequence.
+ { ISD::SRL, MVT::v32i8, 23 }, // pblendvb sequence + split.
+ { ISD::SRL, MVT::v8i16, 13 }, // pblendvb sequence.
+ { ISD::SRL, MVT::v16i16, 28 }, // pblendvb sequence + split.
+ { ISD::SRL, MVT::v4i32, 6 }, // Shift each lane + blend.
+ { ISD::SRL, MVT::v8i32, 14 }, // Shift each lane + blend + split.
+ { ISD::SRL, MVT::v2i64, 2 }, // Shift each lane + blend.
+ { ISD::SRL, MVT::v4i64, 6 }, // Shift each lane + blend + split.
+
+ { ISD::SRA, MVT::v16i8, 21 }, // pblendvb sequence.
+ { ISD::SRA, MVT::v32i8, 44 }, // pblendvb sequence + split.
+ { ISD::SRA, MVT::v8i16, 13 }, // pblendvb sequence.
+ { ISD::SRA, MVT::v16i16, 28 }, // pblendvb sequence + split.
+ { ISD::SRA, MVT::v4i32, 6 }, // Shift each lane + blend.
+ { ISD::SRA, MVT::v8i32, 14 }, // Shift each lane + blend + split.
+ { ISD::SRA, MVT::v2i64, 5 }, // Shift each lane + blend.
+ { ISD::SRA, MVT::v4i64, 12 }, // Shift each lane + blend + split.
+
{ ISD::FNEG, MVT::v4f64, 2 }, // BTVER2 from http://www.agner.org/
{ ISD::FNEG, MVT::v8f32, 2 }, // BTVER2 from http://www.agner.org/
@@ -894,25 +921,16 @@ InstructionCost X86TTIImpl::getArithmeticInstrCost(
static const CostTblEntry SSE41CostTable[] = {
{ ISD::SHL, MVT::v16i8, 11 }, // pblendvb sequence.
- { ISD::SHL, MVT::v32i8, 2*11+2 }, // pblendvb sequence + split.
{ ISD::SHL, MVT::v8i16, 14 }, // pblendvb sequence.
- { ISD::SHL, MVT::v16i16, 2*14+2 }, // pblendvb sequence + split.
{ ISD::SHL, MVT::v4i32, 4 }, // pslld/paddd/cvttps2dq/pmulld
- { ISD::SHL, MVT::v8i32, 2*4+2 }, // pslld/paddd/cvttps2dq/pmulld + split
{ ISD::SRL, MVT::v16i8, 12 }, // pblendvb sequence.
- { ISD::SRL, MVT::v32i8, 2*12+2 }, // pblendvb sequence + split.
{ ISD::SRL, MVT::v8i16, 14 }, // pblendvb sequence.
- { ISD::SRL, MVT::v16i16, 2*14+2 }, // pblendvb sequence + split.
{ ISD::SRL, MVT::v4i32, 11 }, // Shift each lane + blend.
- { ISD::SRL, MVT::v8i32, 2*11+2 }, // Shift each lane + blend + split.
{ ISD::SRA, MVT::v16i8, 24 }, // pblendvb sequence.
- { ISD::SRA, MVT::v32i8, 2*24+2 }, // pblendvb sequence + split.
{ ISD::SRA, MVT::v8i16, 14 }, // pblendvb sequence.
- { ISD::SRA, MVT::v16i16, 2*14+2 }, // pblendvb sequence + split.
{ ISD::SRA, MVT::v4i32, 12 }, // Shift each lane + blend.
- { ISD::SRA, MVT::v8i32, 2*12+2 }, // Shift each lane + blend + split.
{ ISD::MUL, MVT::v4i32, 2 } // pmulld (Nehalem from agner.org)
};
@@ -928,19 +946,16 @@ InstructionCost X86TTIImpl::getArithmeticInstrCost(
{ ISD::SHL, MVT::v8i16, 32 }, // cmpgtb sequence.
{ ISD::SHL, MVT::v4i32, 2*5 }, // We optimized this using mul.
{ ISD::SHL, MVT::v2i64, 4 }, // splat+shuffle sequence.
- { ISD::SHL, MVT::v4i64, 2*4+2 }, // splat+shuffle sequence + split.
{ ISD::SRL, MVT::v16i8, 26 }, // cmpgtb sequence.
{ ISD::SRL, MVT::v8i16, 32 }, // cmpgtb sequence.
{ ISD::SRL, MVT::v4i32, 16 }, // Shift each lane + blend.
{ ISD::SRL, MVT::v2i64, 4 }, // splat+shuffle sequence.
- { ISD::SRL, MVT::v4i64, 2*4+2 }, // splat+shuffle sequence + split.
{ ISD::SRA, MVT::v16i8, 54 }, // unpacked cmpgtb sequence.
{ ISD::SRA, MVT::v8i16, 32 }, // cmpgtb sequence.
{ ISD::SRA, MVT::v4i32, 16 }, // Shift each lane + blend.
{ ISD::SRA, MVT::v2i64, 12 }, // srl/xor/sub sequence.
- { ISD::SRA, MVT::v4i64, 2*12+2 }, // srl/xor/sub sequence+split.
{ ISD::MUL, MVT::v8i16, 1 }, // pmullw
{ ISD::MUL, MVT::v4i32, 6 }, // 3*pmuludq/4*shuffle
diff --git a/llvm/test/Analysis/CostModel/X86/div.ll b/llvm/test/Analysis/CostModel/X86/div.ll
index 90c731690e8c6..4464c384a6707 100644
--- a/llvm/test/Analysis/CostModel/X86/div.ll
+++ b/llvm/test/Analysis/CostModel/X86/div.ll
@@ -766,21 +766,21 @@ define i32 @sdiv_constpow2() {
;
; AVX1-LABEL: 'sdiv_constpow2'
; AVX1-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %I64 = sdiv i64 undef, 16
-; AVX1-NEXT: Cost Model: Found an estimated cost of 29 for instruction: %V2i64 = sdiv <2 x i64> undef, <i64 8, i64 16>
-; AVX1-NEXT: Cost Model: Found an estimated cost of 66 for instruction: %V4i64 = sdiv <4 x i64> undef, <i64 2, i64 4, i64 8, i64 16>
-; AVX1-NEXT: Cost Model: Found an estimated cost of 132 for instruction: %V8i64 = sdiv <8 x i64> undef, <i64 2, i64 4, i64 8, i64 16, i64 32, i64 64, i64 128, i64 256>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %V2i64 = sdiv <2 x i64> undef, <i64 8, i64 16>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 34 for instruction: %V4i64 = sdiv <4 x i64> undef, <i64 2, i64 4, i64 8, i64 16>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 68 for instruction: %V8i64 = sdiv <8 x i64> undef, <i64 2, i64 4, i64 8, i64 16, i64 32, i64 64, i64 128, i64 256>
; AVX1-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %I32 = sdiv i32 undef, 16
-; AVX1-NEXT: Cost Model: Found an estimated cost of 36 for instruction: %V4i32 = sdiv <4 x i32> undef, <i32 2, i32 4, i32 8, i32 16>
-; AVX1-NEXT: Cost Model: Found an estimated cost of 80 for instruction: %V8i32 = sdiv <8 x i32> undef, <i32 2, i32 4, i32 8, i32 16, i32 32, i32 64, i32 128, i32 256>
-; AVX1-NEXT: Cost Model: Found an estimated cost of 160 for instruction: %V16i32 = sdiv <16 x i32> undef, <i32 2, i32 4, i32 8, i32 16, i32 32, i32 64, i32 128, i32 256, i32 2, i32 4, i32 8, i32 16, i32 32, i32 64, i32 128, i32 256>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %V4i32 = sdiv <4 x i32> undef, <i32 2, i32 4, i32 8, i32 16>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 46 for instruction: %V8i32 = sdiv <8 x i32> undef, <i32 2, i32 4, i32 8, i32 16, i32 32, i32 64, i32 128, i32 256>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 92 for instruction: %V16i32 = sdiv <16 x i32> undef, <i32 2, i32 4, i32 8, i32 16, i32 32, i32 64, i32 128, i32 256, i32 2, i32 4, i32 8, i32 16, i32 32, i32 64, i32 128, i32 256>
; AVX1-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %I16 = sdiv i16 undef, 16
-; AVX1-NEXT: Cost Model: Found an estimated cost of 43 for instruction: %V8i16 = sdiv <8 x i16> undef, <i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256>
-; AVX1-NEXT: Cost Model: Found an estimated cost of 94 for instruction: %V16i16 = sdiv <16 x i16> undef, <i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256>
-; AVX1-NEXT: Cost Model: Found an estimated cost of 188 for instruction: %V32i16 = sdiv <32 x i16> undef, <i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 40 for instruction: %V8i16 = sdiv <8 x i16> undef, <i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 88 for instruction: %V16i16 = sdiv <16 x i16> undef, <i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 176 for instruction: %V32i16 = sdiv <32 x i16> undef, <i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256>
; AVX1-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %I8 = sdiv i8 undef, 16
-; AVX1-NEXT: Cost Model: Found an estimated cost of 61 for instruction: %V16i8 = sdiv <16 x i8> undef, <i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16>
-; AVX1-NEXT: Cost Model: Found an estimated cost of 130 for instruction: %V32i8 = sdiv <32 x i8> undef, <i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16>
-; AVX1-NEXT: Cost Model: Found an estimated cost of 260 for instruction: %V64i8 = sdiv <64 x i8> undef, <i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 54 for instruction: %V16i8 = sdiv <16 x i8> undef, <i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 115 for instruction: %V32i8 = sdiv <32 x i8> undef, <i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 230 for instruction: %V64i8 = sdiv <64 x i8> undef, <i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; AVX2-LABEL: 'sdiv_constpow2'
@@ -861,21 +861,21 @@ define i32 @sdiv_constpow2() {
;
; BTVER2-LABEL: 'sdiv_constpow2'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %I64 = sdiv i64 undef, 16
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 29 for instruction: %V2i64 = sdiv <2 x i64> undef, <i64 8, i64 16>
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 66 for instruction: %V4i64 = sdiv <4 x i64> undef, <i64 2, i64 4, i64 8, i64 16>
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 132 for instruction: %V8i64 = sdiv <8 x i64> undef, <i64 2, i64 4, i64 8, i64 16, i64 32, i64 64, i64 128, i64 256>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %V2i64 = sdiv <2 x i64> undef, <i64 8, i64 16>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 34 for instruction: %V4i64 = sdiv <4 x i64> undef, <i64 2, i64 4, i64 8, i64 16>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 68 for instruction: %V8i64 = sdiv <8 x i64> undef, <i64 2, i64 4, i64 8, i64 16, i64 32, i64 64, i64 128, i64 256>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %I32 = sdiv i32 undef, 16
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 36 for instruction: %V4i32 = sdiv <4 x i32> undef, <i32 2, i32 4, i32 8, i32 16>
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 80 for instruction: %V8i32 = sdiv <8 x i32> undef, <i32 2, i32 4, i32 8, i32 16, i32 32, i32 64, i32 128, i32 256>
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 160 for instruction: %V16i32 = sdiv <16 x i32> undef, <i32 2, i32 4, i32 8, i32 16, i32 32, i32 64, i32 128, i32 256, i32 2, i32 4, i32 8, i32 16, i32 32, i32 64, i32 128, i32 256>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %V4i32 = sdiv <4 x i32> undef, <i32 2, i32 4, i32 8, i32 16>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 46 for instruction: %V8i32 = sdiv <8 x i32> undef, <i32 2, i32 4, i32 8, i32 16, i32 32, i32 64, i32 128, i32 256>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 92 for instruction: %V16i32 = sdiv <16 x i32> undef, <i32 2, i32 4, i32 8, i32 16, i32 32, i32 64, i32 128, i32 256, i32 2, i32 4, i32 8, i32 16, i32 32, i32 64, i32 128, i32 256>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %I16 = sdiv i16 undef, 16
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 43 for instruction: %V8i16 = sdiv <8 x i16> undef, <i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256>
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 94 for instruction: %V16i16 = sdiv <16 x i16> undef, <i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256>
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 188 for instruction: %V32i16 = sdiv <32 x i16> undef, <i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 40 for instruction: %V8i16 = sdiv <8 x i16> undef, <i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 88 for instruction: %V16i16 = sdiv <16 x i16> undef, <i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 176 for instruction: %V32i16 = sdiv <32 x i16> undef, <i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %I8 = sdiv i8 undef, 16
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 61 for instruction: %V16i8 = sdiv <16 x i8> undef, <i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16>
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 130 for instruction: %V32i8 = sdiv <32 x i8> undef, <i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16>
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 260 for instruction: %V64i8 = sdiv <64 x i8> undef, <i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 54 for instruction: %V16i8 = sdiv <16 x i8> undef, <i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 115 for instruction: %V32i8 = sdiv <32 x i8> undef, <i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 230 for instruction: %V64i8 = sdiv <64 x i8> undef, <i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
%I64 = sdiv i64 undef, 16
@@ -961,21 +961,21 @@ define i32 @udiv_constpow2() {
;
; AVX1-LABEL: 'udiv_constpow2'
; AVX1-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %I64 = udiv i64 undef, 16
-; AVX1-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V2i64 = udiv <2 x i64> undef, <i64 8, i64 16>
-; AVX1-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %V4i64 = udiv <4 x i64> undef, <i64 2, i64 4, i64 8, i64 16>
-; AVX1-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %V8i64 = udiv <8 x i64> undef, <i64 2, i64 4, i64 8, i64 16, i64 32, i64 64, i64 128, i64 256>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2i64 = udiv <2 x i64> undef, <i64 8, i64 16>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V4i64 = udiv <4 x i64> undef, <i64 2, i64 4, i64 8, i64 16>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %V8i64 = udiv <8 x i64> undef, <i64 2, i64 4, i64 8, i64 16, i64 32, i64 64, i64 128, i64 256>
; AVX1-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %I32 = udiv i32 undef, 16
-; AVX1-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %V4i32 = udiv <4 x i32> undef, <i32 2, i32 4, i32 8, i32 16>
-; AVX1-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %V8i32 = udiv <8 x i32> undef, <i32 2, i32 4, i32 8, i32 16, i32 32, i32 64, i32 128, i32 256>
-; AVX1-NEXT: Cost Model: Found an estimated cost of 48 for instruction: %V16i32 = udiv <16 x i32> undef, <i32 2, i32 4, i32 8, i32 16, i32 32, i32 64, i32 128, i32 256, i32 2, i32 4, i32 8, i32 16, i32 32, i32 64, i32 128, i32 256>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V4i32 = udiv <4 x i32> undef, <i32 2, i32 4, i32 8, i32 16>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %V8i32 = udiv <8 x i32> undef, <i32 2, i32 4, i32 8, i32 16, i32 32, i32 64, i32 128, i32 256>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 28 for instruction: %V16i32 = udiv <16 x i32> undef, <i32 2, i32 4, i32 8, i32 16, i32 32, i32 64, i32 128, i32 256, i32 2, i32 4, i32 8, i32 16, i32 32, i32 64, i32 128, i32 256>
; AVX1-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %I16 = udiv i16 undef, 16
-; AVX1-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %V8i16 = udiv <8 x i16> undef, <i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256>
-; AVX1-NEXT: Cost Model: Found an estimated cost of 30 for instruction: %V16i16 = udiv <16 x i16> undef, <i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256>
-; AVX1-NEXT: Cost Model: Found an estimated cost of 60 for instruction: %V32i16 = udiv <32 x i16> undef, <i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %V8i16 = udiv <8 x i16> undef, <i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 28 for instruction: %V16i16 = udiv <16 x i16> undef, <i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 56 for instruction: %V32i16 = udiv <32 x i16> undef, <i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256>
; AVX1-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %I8 = udiv i8 undef, 16
-; AVX1-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %V16i8 = udiv <16 x i8> undef, <i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16>
-; AVX1-NEXT: Cost Model: Found an estimated cost of 26 for instruction: %V32i8 = udiv <32 x i8> undef, <i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16>
-; AVX1-NEXT: Cost Model: Found an estimated cost of 52 for instruction: %V64i8 = udiv <64 x i8> undef, <i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %V16i8 = udiv <16 x i8> undef, <i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 23 for instruction: %V32i8 = udiv <32 x i8> undef, <i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 46 for instruction: %V64i8 = udiv <64 x i8> undef, <i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; AVX2-LABEL: 'udiv_constpow2'
@@ -1056,21 +1056,21 @@ define i32 @udiv_constpow2() {
;
; BTVER2-LABEL: 'udiv_constpow2'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %I64 = udiv i64 undef, 16
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V2i64 = udiv <2 x i64> undef, <i64 8, i64 16>
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %V4i64 = udiv <4 x i64> undef, <i64 2, i64 4, i64 8, i64 16>
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %V8i64 = udiv <8 x i64> undef, <i64 2, i64 4, i64 8, i64 16, i64 32, i64 64, i64 128, i64 256>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2i64 = udiv <2 x i64> undef, <i64 8, i64 16>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V4i64 = udiv <4 x i64> undef, <i64 2, i64 4, i64 8, i64 16>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %V8i64 = udiv <8 x i64> undef, <i64 2, i64 4, i64 8, i64 16, i64 32, i64 64, i64 128, i64 256>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %I32 = udiv i32 undef, 16
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %V4i32 = udiv <4 x i32> undef, <i32 2, i32 4, i32 8, i32 16>
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %V8i32 = udiv <8 x i32> undef, <i32 2, i32 4, i32 8, i32 16, i32 32, i32 64, i32 128, i32 256>
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 48 for instruction: %V16i32 = udiv <16 x i32> undef, <i32 2, i32 4, i32 8, i32 16, i32 32, i32 64, i32 128, i32 256, i32 2, i32 4, i32 8, i32 16, i32 32, i32 64, i32 128, i32 256>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V4i32 = udiv <4 x i32> undef, <i32 2, i32 4, i32 8, i32 16>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %V8i32 = udiv <8 x i32> undef, <i32 2, i32 4, i32 8, i32 16, i32 32, i32 64, i32 128, i32 256>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 28 for instruction: %V16i32 = udiv <16 x i32> undef, <i32 2, i32 4, i32 8, i32 16, i32 32, i32 64, i32 128, i32 256, i32 2, i32 4, i32 8, i32 16, i32 32, i32 64, i32 128, i32 256>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %I16 = udiv i16 undef, 16
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %V8i16 = udiv <8 x i16> undef, <i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256>
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 30 for instruction: %V16i16 = udiv <16 x i16> undef, <i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256>
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 60 for instruction: %V32i16 = udiv <32 x i16> undef, <i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %V8i16 = udiv <8 x i16> undef, <i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 28 for instruction: %V16i16 = udiv <16 x i16> undef, <i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 56 for instruction: %V32i16 = udiv <32 x i16> undef, <i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %I8 = udiv i8 undef, 16
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %V16i8 = udiv <16 x i8> undef, <i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16>
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 26 for instruction: %V32i8 = udiv <32 x i8> undef, <i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16>
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 52 for instruction: %V64i8 = udiv <64 x i8> undef, <i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %V16i8 = udiv <16 x i8> undef, <i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 23 for instruction: %V32i8 = udiv <32 x i8> undef, <i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 46 for instruction: %V64i8 = udiv <64 x i8> undef, <i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
%I64 = udiv i64 undef, 16
diff --git a/llvm/test/Analysis/CostModel/X86/fshl.ll b/llvm/test/Analysis/CostModel/X86/fshl.ll
index f1f0370638950..e43569c8124be 100644
--- a/llvm/test/Analysis/CostModel/X86/fshl.ll
+++ b/llvm/test/Analysis/CostModel/X86/fshl.ll
@@ -71,9 +71,9 @@ define void @var_funnel_i64(i64 %a64, <2 x i64> %a128, <4 x i64> %a256, <8 x i64
;
; BTVER2-LABEL: 'var_funnel_i64'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %I64 = call i64 @llvm.fshl.i64(i64 %a64, i64 %b64, i64 %c64)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %V2I64 = call <2 x i64> @llvm.fshl.v2i64(<2 x i64> %a128, <2 x i64> %b128, <2 x i64> %c128)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 31 for instruction: %V4I64 = call <4 x i64> @llvm.fshl.v4i64(<4 x i64> %a256, <4 x i64> %b256, <4 x i64> %c256)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 62 for instruction: %V8I64 = call <8 x i64> @llvm.fshl.v8i64(<8 x i64> %a512, <8 x i64> %b512, <8 x i64> %c512)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V2I64 = call <2 x i64> @llvm.fshl.v2i64(<2 x i64> %a128, <2 x i64> %b128, <2 x i64> %c128)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 23 for instruction: %V4I64 = call <4 x i64> @llvm.fshl.v4i64(<4 x i64> %a256, <4 x i64> %b256, <4 x i64> %c256)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 46 for instruction: %V8I64 = call <8 x i64> @llvm.fshl.v8i64(<8 x i64> %a512, <8 x i64> %b512, <8 x i64> %c512)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%I64 = call i64 @llvm.fshl.i64(i64 %a64, i64 %b64, i64 %c64)
@@ -135,9 +135,9 @@ define void @var_funnel_i32(i32 %a32, <4 x i32> %a128, <8 x i32> %a256, <16 x i3
;
; BTVER2-LABEL: 'var_funnel_i32'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %I32 = call i32 @llvm.fshl.i32(i32 %a32, i32 %b32, i32 %c32)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %V2I32 = call <4 x i32> @llvm.fshl.v4i32(<4 x i32> %a128, <4 x i32> %b128, <4 x i32> %c128)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 45 for instruction: %V4I32 = call <8 x i32> @llvm.fshl.v8i32(<8 x i32> %a256, <8 x i32> %b256, <8 x i32> %c256)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 90 for instruction: %V8I32 = call <16 x i32> @llvm.fshl.v16i32(<16 x i32> %a512, <16 x i32> %b512, <16 x i32> %c512)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %V2I32 = call <4 x i32> @llvm.fshl.v4i32(<4 x i32> %a128, <4 x i32> %b128, <4 x i32> %c128)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 34 for instruction: %V4I32 = call <8 x i32> @llvm.fshl.v8i32(<8 x i32> %a256, <8 x i32> %b256, <8 x i32> %c256)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 68 for instruction: %V8I32 = call <16 x i32> @llvm.fshl.v16i32(<16 x i32> %a512, <16 x i32> %b512, <16 x i32> %c512)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%I32 = call i32 @llvm.fshl.i32(i32 %a32, i32 %b32, i32 %c32)
@@ -213,9 +213,9 @@ define void @var_funnel_i16(i16 %a16, <8 x i16> %a128, <16 x i16> %a256, <32 x i
;
; BTVER2-LABEL: 'var_funnel_i16'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %I16 = call i16 @llvm.fshl.i16(i16 %a16, i16 %b16, i16 %c16)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 33 for instruction: %V8I16 = call <8 x i16> @llvm.fshl.v8i16(<8 x i16> %a128, <8 x i16> %b128, <8 x i16> %c128)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 73 for instruction: %V16I16 = call <16 x i16> @llvm.fshl.v16i16(<16 x i16> %a256, <16 x i16> %b256, <16 x i16> %c256)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 146 for instruction: %V32I16 = call <32 x i16> @llvm.fshl.v32i16(<32 x i16> %a512, <32 x i16> %b512, <32 x i16> %c512)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %V8I16 = call <8 x i16> @llvm.fshl.v8i16(<8 x i16> %a128, <8 x i16> %b128, <8 x i16> %c128)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 54 for instruction: %V16I16 = call <16 x i16> @llvm.fshl.v16i16(<16 x i16> %a256, <16 x i16> %b256, <16 x i16> %c256)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 108 for instruction: %V32I16 = call <32 x i16> @llvm.fshl.v32i16(<32 x i16> %a512, <32 x i16> %b512, <32 x i16> %c512)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%I16 = call i16 @llvm.fshl.i16(i16 %a16, i16 %b16, i16 %c16)
@@ -291,9 +291,9 @@ define void @var_funnel_i8(i8 %a8, <16 x i8> %a128, <32 x i8> %a256, <64 x i8> %
;
; BTVER2-LABEL: 'var_funnel_i8'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %I8 = call i8 @llvm.fshl.i8(i8 %a8, i8 %b8, i8 %c8)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 28 for instruction: %V16I8 = call <16 x i8> @llvm.fshl.v16i8(<16 x i8> %a128, <16 x i8> %b128, <16 x i8> %c128)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 63 for instruction: %V32I8 = call <32 x i8> @llvm.fshl.v32i8(<32 x i8> %a256, <32 x i8> %b256, <32 x i8> %c256)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 126 for instruction: %V64I8 = call <64 x i8> @llvm.fshl.v64i8(<64 x i8> %a512, <64 x i8> %b512, <64 x i8> %c512)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 26 for instruction: %V16I8 = call <16 x i8> @llvm.fshl.v16i8(<16 x i8> %a128, <16 x i8> %b128, <16 x i8> %c128)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 58 for instruction: %V32I8 = call <32 x i8> @llvm.fshl.v32i8(<32 x i8> %a256, <32 x i8> %b256, <32 x i8> %c256)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 116 for instruction: %V64I8 = call <64 x i8> @llvm.fshl.v64i8(<64 x i8> %a512, <64 x i8> %b512, <64 x i8> %c512)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%I8 = call i8 @llvm.fshl.i8(i8 %a8, i8 %b8, i8 %c8)
@@ -657,9 +657,9 @@ define void @splatvar_funnel_i8(i8 %a8, <16 x i8> %a128, <32 x i8> %a256, <64 x
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %u128 = shufflevector <16 x i8> %c128, <16 x i8> undef, <16 x i32> zeroinitializer
; BTVER2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %u256 = shufflevector <32 x i8> %c256, <32 x i8> undef, <32 x i32> zeroinitializer
; BTVER2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %u512 = shufflevector <64 x i8> %c512, <64 x i8> undef, <64 x i32> zeroinitializer
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 28 for instruction: %V16I8 = call <16 x i8> @llvm.fshl.v16i8(<16 x i8> %a128, <16 x i8> %b128, <16 x i8> %u128)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 63 for instruction: %V32I8 = call <32 x i8> @llvm.fshl.v32i8(<32 x i8> %a256, <32 x i8> %b256, <32 x i8> %u256)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 126 for instruction: %V64I8 = call <64 x i8> @llvm.fshl.v64i8(<64 x i8> %a512, <64 x i8> %b512, <64 x i8> %u512)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 26 for instruction: %V16I8 = call <16 x i8> @llvm.fshl.v16i8(<16 x i8> %a128, <16 x i8> %b128, <16 x i8> %u128)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 58 for instruction: %V32I8 = call <32 x i8> @llvm.fshl.v32i8(<32 x i8> %a256, <32 x i8> %b256, <32 x i8> %u256)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 116 for instruction: %V64I8 = call <64 x i8> @llvm.fshl.v64i8(<64 x i8> %a512, <64 x i8> %b512, <64 x i8> %u512)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%u128 = shufflevector <16 x i8> %c128, <16 x i8> undef, <16 x i32> zeroinitializer
@@ -727,9 +727,9 @@ define void @constant_funnel_i64(i64 %a64, <2 x i64> %a128, <4 x i64> %a256, <8
;
; BTVER2-LABEL: 'constant_funnel_i64'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %I64 = call i64 @llvm.fshl.i64(i64 %a64, i64 %b64, i64 7)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %V2I64 = call <2 x i64> @llvm.fshl.v2i64(<2 x i64> %a128, <2 x i64> %b128, <2 x i64> <i64 1, i64 7>)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 30 for instruction: %V4I64 = call <4 x i64> @llvm.fshl.v4i64(<4 x i64> %a256, <4 x i64> %b256, <4 x i64> <i64 1, i64 7, i64 15, i64 31>)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 60 for instruction: %V8I64 = call <8 x i64> @llvm.fshl.v8i64(<8 x i64> %a512, <8 x i64> %b512, <8 x i64> <i64 1, i64 7, i64 15, i64 31, i64 1, i64 7, i64 15, i64 31>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V2I64 = call <2 x i64> @llvm.fshl.v2i64(<2 x i64> %a128, <2 x i64> %b128, <2 x i64> <i64 1, i64 7>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 22 for instruction: %V4I64 = call <4 x i64> @llvm.fshl.v4i64(<4 x i64> %a256, <4 x i64> %b256, <4 x i64> <i64 1, i64 7, i64 15, i64 31>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 44 for instruction: %V8I64 = call <8 x i64> @llvm.fshl.v8i64(<8 x i64> %a512, <8 x i64> %b512, <8 x i64> <i64 1, i64 7, i64 15, i64 31, i64 1, i64 7, i64 15, i64 31>)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%I64 = call i64 @llvm.fshl.i64(i64 %a64, i64 %b64, i64 7)
@@ -791,9 +791,9 @@ define void @constant_funnel_i32(i32 %a32, <4 x i32> %a128, <8 x i32> %a256, <16
;
; BTVER2-LABEL: 'constant_funnel_i32'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %I32 = call i32 @llvm.fshl.i32(i32 %a32, i32 %b32, i32 7)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 17 for instruction: %V2I32 = call <4 x i32> @llvm.fshl.v4i32(<4 x i32> %a128, <4 x i32> %b128, <4 x i32> <i32 4, i32 5, i32 6, i32 7>)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 39 for instruction: %V4I32 = call <8 x i32> @llvm.fshl.v8i32(<8 x i32> %a256, <8 x i32> %b256, <8 x i32> <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 78 for instruction: %V8I32 = call <16 x i32> @llvm.fshl.v16i32(<16 x i32> %a512, <16 x i32> %b512, <16 x i32> <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>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %V2I32 = call <4 x i32> @llvm.fshl.v4i32(<4 x i32> %a128, <4 x i32> %b128, <4 x i32> <i32 4, i32 5, i32 6, i32 7>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 29 for instruction: %V4I32 = call <8 x i32> @llvm.fshl.v8i32(<8 x i32> %a256, <8 x i32> %b256, <8 x i32> <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 58 for instruction: %V8I32 = call <16 x i32> @llvm.fshl.v16i32(<16 x i32> %a512, <16 x i32> %b512, <16 x i32> <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>)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%I32 = call i32 @llvm.fshl.i32(i32 %a32, i32 %b32, i32 7)
@@ -869,9 +869,9 @@ define void @constant_funnel_i16(i16 %a16, <8 x i16> %a128, <16 x i16> %a256, <3
;
; BTVER2-LABEL: 'constant_funnel_i16'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %I16 = call i16 @llvm.fshl.i16(i16 %a16, i16 %b16, i16 7)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %V8I16 = call <8 x i16> @llvm.fshl.v8i16(<8 x i16> %a128, <8 x i16> %b128, <8 x i16> <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 46 for instruction: %V16I16 = call <16 x i16> @llvm.fshl.v16i16(<16 x i16> %a256, <16 x i16> %b256, <16 x i16> <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>)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 92 for instruction: %V32I16 = call <32 x i16> @llvm.fshl.v32i16(<32 x i16> %a512, <32 x i16> %b512, <32 x i16> <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>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %V8I16 = call <8 x i16> @llvm.fshl.v8i16(<8 x i16> %a128, <8 x i16> %b128, <8 x i16> <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 44 for instruction: %V16I16 = call <16 x i16> @llvm.fshl.v16i16(<16 x i16> %a256, <16 x i16> %b256, <16 x i16> <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>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 88 for instruction: %V32I16 = call <32 x i16> @llvm.fshl.v32i16(<32 x i16> %a512, <32 x i16> %b512, <32 x i16> <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>)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%I16 = call i16 @llvm.fshl.i16(i16 %a16, i16 %b16, i16 7)
@@ -947,9 +947,9 @@ define void @constant_funnel_i8(i8 %a8, <16 x i8> %a128, <32 x i8> %a256, <64 x
;
; BTVER2-LABEL: 'constant_funnel_i8'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %I8 = call i8 @llvm.fshl.i8(i8 %a8, i8 %b8, i8 7)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 27 for instruction: %V16I8 = call <16 x i8> @llvm.fshl.v16i8(<16 x i8> %a128, <16 x i8> %b128, <16 x i8> <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>)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 62 for instruction: %V32I8 = call <32 x i8> @llvm.fshl.v32i8(<32 x i8> %a256, <32 x i8> %b256, <32 x i8> <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>)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 124 for instruction: %V64I8 = call <64 x i8> @llvm.fshl.v64i8(<64 x i8> %a512, <64 x i8> %b512, <64 x i8> <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>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 25 for instruction: %V16I8 = call <16 x i8> @llvm.fshl.v16i8(<16 x i8> %a128, <16 x i8> %b128, <16 x i8> <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>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 57 for instruction: %V32I8 = call <32 x i8> @llvm.fshl.v32i8(<32 x i8> %a256, <32 x i8> %b256, <32 x i8> <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>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 114 for instruction: %V64I8 = call <64 x i8> @llvm.fshl.v64i8(<64 x i8> %a512, <64 x i8> %b512, <64 x i8> <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>)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%I8 = call i8 @llvm.fshl.i8(i8 %a8, i8 %b8, i8 7)
@@ -1275,9 +1275,9 @@ define void @var_rotate_i64(i64 %a64, <2 x i64> %a128, <4 x i64> %a256, <8 x i64
;
; BTVER2-LABEL: 'var_rotate_i64'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %I64 = call i64 @llvm.fshl.i64(i64 %a64, i64 %a64, i64 %c64)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %V2I64 = call <2 x i64> @llvm.fshl.v2i64(<2 x i64> %a128, <2 x i64> %a128, <2 x i64> %c128)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 26 for instruction: %V4I64 = call <4 x i64> @llvm.fshl.v4i64(<4 x i64> %a256, <4 x i64> %a256, <4 x i64> %c256)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 52 for instruction: %V8I64 = call <8 x i64> @llvm.fshl.v8i64(<8 x i64> %a512, <8 x i64> %a512, <8 x i64> %c512)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V2I64 = call <2 x i64> @llvm.fshl.v2i64(<2 x i64> %a128, <2 x i64> %a128, <2 x i64> %c128)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %V4I64 = call <4 x i64> @llvm.fshl.v4i64(<4 x i64> %a256, <4 x i64> %a256, <4 x i64> %c256)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 36 for instruction: %V8I64 = call <8 x i64> @llvm.fshl.v8i64(<8 x i64> %a512, <8 x i64> %a512, <8 x i64> %c512)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%I64 = call i64 @llvm.fshl.i64(i64 %a64, i64 %a64, i64 %c64)
@@ -1339,9 +1339,9 @@ define void @var_rotate_i32(i32 %a32, <4 x i32> %a128, <8 x i32> %a256, <16 x i3
;
; BTVER2-LABEL: 'var_rotate_i32'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %I32 = call i32 @llvm.fshl.i32(i32 %a32, i32 %a32, i32 %c32)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %V2I32 = call <4 x i32> @llvm.fshl.v4i32(<4 x i32> %a128, <4 x i32> %a128, <4 x i32> %c128)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 40 for instruction: %V4I32 = call <8 x i32> @llvm.fshl.v8i32(<8 x i32> %a256, <8 x i32> %a256, <8 x i32> %c256)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 80 for instruction: %V8I32 = call <16 x i32> @llvm.fshl.v16i32(<16 x i32> %a512, <16 x i32> %a512, <16 x i32> %c512)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %V2I32 = call <4 x i32> @llvm.fshl.v4i32(<4 x i32> %a128, <4 x i32> %a128, <4 x i32> %c128)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 29 for instruction: %V4I32 = call <8 x i32> @llvm.fshl.v8i32(<8 x i32> %a256, <8 x i32> %a256, <8 x i32> %c256)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 58 for instruction: %V8I32 = call <16 x i32> @llvm.fshl.v16i32(<16 x i32> %a512, <16 x i32> %a512, <16 x i32> %c512)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%I32 = call i32 @llvm.fshl.i32(i32 %a32, i32 %a32, i32 %c32)
@@ -1417,9 +1417,9 @@ define void @var_rotate_i16(i16 %a16, <8 x i16> %a128, <16 x i16> %a256, <32 x i
;
; BTVER2-LABEL: 'var_rotate_i16'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.fshl.i16(i16 %a16, i16 %a16, i16 %c16)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 31 for instruction: %V8I16 = call <8 x i16> @llvm.fshl.v8i16(<8 x i16> %a128, <8 x i16> %a128, <8 x i16> %c128)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 66 for instruction: %V16I16 = call <16 x i16> @llvm.fshl.v16i16(<16 x i16> %a256, <16 x i16> %a256, <16 x i16> %c256)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 132 for instruction: %V32I16 = call <32 x i16> @llvm.fshl.v32i16(<32 x i16> %a512, <32 x i16> %a512, <32 x i16> %c512)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 22 for instruction: %V8I16 = call <8 x i16> @llvm.fshl.v8i16(<8 x i16> %a128, <8 x i16> %a128, <8 x i16> %c128)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 47 for instruction: %V16I16 = call <16 x i16> @llvm.fshl.v16i16(<16 x i16> %a256, <16 x i16> %a256, <16 x i16> %c256)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 94 for instruction: %V32I16 = call <32 x i16> @llvm.fshl.v32i16(<32 x i16> %a512, <32 x i16> %a512, <32 x i16> %c512)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%I16 = call i16 @llvm.fshl.i16(i16 %a16, i16 %a16, i16 %c16)
@@ -1495,9 +1495,9 @@ define void @var_rotate_i8(i8 %a8, <16 x i8> %a128, <32 x i8> %a256, <64 x i8> %
;
; BTVER2-LABEL: 'var_rotate_i8'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.fshl.i8(i8 %a8, i8 %a8, i8 %c8)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 26 for instruction: %V16I8 = call <16 x i8> @llvm.fshl.v16i8(<16 x i8> %a128, <16 x i8> %a128, <16 x i8> %c128)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 56 for instruction: %V32I8 = call <32 x i8> @llvm.fshl.v32i8(<32 x i8> %a256, <32 x i8> %a256, <32 x i8> %c256)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 112 for instruction: %V64I8 = call <64 x i8> @llvm.fshl.v64i8(<64 x i8> %a512, <64 x i8> %a512, <64 x i8> %c512)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %V16I8 = call <16 x i8> @llvm.fshl.v16i8(<16 x i8> %a128, <16 x i8> %a128, <16 x i8> %c128)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 51 for instruction: %V32I8 = call <32 x i8> @llvm.fshl.v32i8(<32 x i8> %a256, <32 x i8> %a256, <32 x i8> %c256)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 102 for instruction: %V64I8 = call <64 x i8> @llvm.fshl.v64i8(<64 x i8> %a512, <64 x i8> %a512, <64 x i8> %c512)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%I8 = call i8 @llvm.fshl.i8(i8 %a8, i8 %a8, i8 %c8)
@@ -1834,9 +1834,9 @@ define void @splatvar_rotate_i8(i8 %a8, <16 x i8> %a128, <32 x i8> %a256, <64 x
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %u128 = shufflevector <16 x i8> %c128, <16 x i8> undef, <16 x i32> zeroinitializer
; BTVER2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %u256 = shufflevector <32 x i8> %c256, <32 x i8> undef, <32 x i32> zeroinitializer
; BTVER2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %u512 = shufflevector <64 x i8> %c512, <64 x i8> undef, <64 x i32> zeroinitializer
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 26 for instruction: %V16I8 = call <16 x i8> @llvm.fshl.v16i8(<16 x i8> %a128, <16 x i8> %a128, <16 x i8> %u128)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 56 for instruction: %V32I8 = call <32 x i8> @llvm.fshl.v32i8(<32 x i8> %a256, <32 x i8> %a256, <32 x i8> %u256)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 112 for instruction: %V64I8 = call <64 x i8> @llvm.fshl.v64i8(<64 x i8> %a512, <64 x i8> %a512, <64 x i8> %u512)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %V16I8 = call <16 x i8> @llvm.fshl.v16i8(<16 x i8> %a128, <16 x i8> %a128, <16 x i8> %u128)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 51 for instruction: %V32I8 = call <32 x i8> @llvm.fshl.v32i8(<32 x i8> %a256, <32 x i8> %a256, <32 x i8> %u256)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 102 for instruction: %V64I8 = call <64 x i8> @llvm.fshl.v64i8(<64 x i8> %a512, <64 x i8> %a512, <64 x i8> %u512)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%u128 = shufflevector <16 x i8> %c128, <16 x i8> undef, <16 x i32> zeroinitializer
@@ -1897,9 +1897,9 @@ define void @constant_rotate_i64(i64 %a64, <2 x i64> %a128, <4 x i64> %a256, <8
;
; BTVER2-LABEL: 'constant_rotate_i64'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %I64 = call i64 @llvm.fshl.i64(i64 %a64, i64 %a64, i64 7)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %V2I64 = call <2 x i64> @llvm.fshl.v2i64(<2 x i64> %a128, <2 x i64> %a128, <2 x i64> <i64 1, i64 7>)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 25 for instruction: %V4I64 = call <4 x i64> @llvm.fshl.v4i64(<4 x i64> %a256, <4 x i64> %a256, <4 x i64> <i64 1, i64 7, i64 15, i64 31>)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 50 for instruction: %V8I64 = call <8 x i64> @llvm.fshl.v8i64(<8 x i64> %a512, <8 x i64> %a512, <8 x i64> <i64 1, i64 7, i64 15, i64 31, i64 1, i64 7, i64 15, i64 31>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V2I64 = call <2 x i64> @llvm.fshl.v2i64(<2 x i64> %a128, <2 x i64> %a128, <2 x i64> <i64 1, i64 7>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 17 for instruction: %V4I64 = call <4 x i64> @llvm.fshl.v4i64(<4 x i64> %a256, <4 x i64> %a256, <4 x i64> <i64 1, i64 7, i64 15, i64 31>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 34 for instruction: %V8I64 = call <8 x i64> @llvm.fshl.v8i64(<8 x i64> %a512, <8 x i64> %a512, <8 x i64> <i64 1, i64 7, i64 15, i64 31, i64 1, i64 7, i64 15, i64 31>)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%I64 = call i64 @llvm.fshl.i64(i64 %a64, i64 %a64, i64 7)
@@ -1961,9 +1961,9 @@ define void @constant_rotate_i32(i32 %a32, <4 x i32> %a128, <8 x i32> %a256, <16
;
; BTVER2-LABEL: 'constant_rotate_i32'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %I32 = call i32 @llvm.fshl.i32(i32 %a32, i32 %a32, i32 7)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V2I32 = call <4 x i32> @llvm.fshl.v4i32(<4 x i32> %a128, <4 x i32> %a128, <4 x i32> <i32 4, i32 5, i32 6, i32 7>)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 34 for instruction: %V4I32 = call <8 x i32> @llvm.fshl.v8i32(<8 x i32> %a256, <8 x i32> %a256, <8 x i32> <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 68 for instruction: %V8I32 = call <16 x i32> @llvm.fshl.v16i32(<16 x i32> %a512, <16 x i32> %a512, <16 x i32> <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>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %V2I32 = call <4 x i32> @llvm.fshl.v4i32(<4 x i32> %a128, <4 x i32> %a128, <4 x i32> <i32 4, i32 5, i32 6, i32 7>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %V4I32 = call <8 x i32> @llvm.fshl.v8i32(<8 x i32> %a256, <8 x i32> %a256, <8 x i32> <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 48 for instruction: %V8I32 = call <16 x i32> @llvm.fshl.v16i32(<16 x i32> %a512, <16 x i32> %a512, <16 x i32> <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>)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%I32 = call i32 @llvm.fshl.i32(i32 %a32, i32 %a32, i32 7)
@@ -2039,9 +2039,9 @@ define void @constant_rotate_i16(i16 %a16, <8 x i16> %a128, <16 x i16> %a256, <3
;
; BTVER2-LABEL: 'constant_rotate_i16'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.fshl.i16(i16 %a16, i16 %a16, i16 7)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 17 for instruction: %V8I16 = call <8 x i16> @llvm.fshl.v8i16(<8 x i16> %a128, <8 x i16> %a128, <8 x i16> <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 39 for instruction: %V16I16 = call <16 x i16> @llvm.fshl.v16i16(<16 x i16> %a256, <16 x i16> %a256, <16 x i16> <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>)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 78 for instruction: %V32I16 = call <32 x i16> @llvm.fshl.v32i16(<32 x i16> %a512, <32 x i16> %a512, <32 x i16> <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>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %V8I16 = call <8 x i16> @llvm.fshl.v8i16(<8 x i16> %a128, <8 x i16> %a128, <8 x i16> <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 37 for instruction: %V16I16 = call <16 x i16> @llvm.fshl.v16i16(<16 x i16> %a256, <16 x i16> %a256, <16 x i16> <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>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 74 for instruction: %V32I16 = call <32 x i16> @llvm.fshl.v32i16(<32 x i16> %a512, <32 x i16> %a512, <32 x i16> <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>)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%I16 = call i16 @llvm.fshl.i16(i16 %a16, i16 %a16, i16 7)
@@ -2117,9 +2117,9 @@ define void @constant_rotate_i8(i8 %a8, <16 x i8> %a128, <32 x i8> %a256, <64 x
;
; BTVER2-LABEL: 'constant_rotate_i8'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.fshl.i8(i8 %a8, i8 %a8, i8 7)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 25 for instruction: %V16I8 = call <16 x i8> @llvm.fshl.v16i8(<16 x i8> %a128, <16 x i8> %a128, <16 x i8> <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>)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 55 for instruction: %V32I8 = call <32 x i8> @llvm.fshl.v32i8(<32 x i8> %a256, <32 x i8> %a256, <32 x i8> <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>)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 110 for instruction: %V64I8 = call <64 x i8> @llvm.fshl.v64i8(<64 x i8> %a512, <64 x i8> %a512, <64 x i8> <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>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 23 for instruction: %V16I8 = call <16 x i8> @llvm.fshl.v16i8(<16 x i8> %a128, <16 x i8> %a128, <16 x i8> <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>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 50 for instruction: %V32I8 = call <32 x i8> @llvm.fshl.v32i8(<32 x i8> %a256, <32 x i8> %a256, <32 x i8> <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>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 100 for instruction: %V64I8 = call <64 x i8> @llvm.fshl.v64i8(<64 x i8> %a512, <64 x i8> %a512, <64 x i8> <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>)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%I8 = call i8 @llvm.fshl.i8(i8 %a8, i8 %a8, i8 7)
diff --git a/llvm/test/Analysis/CostModel/X86/fshr.ll b/llvm/test/Analysis/CostModel/X86/fshr.ll
index 104020742ef90..afc3defa0f01a 100644
--- a/llvm/test/Analysis/CostModel/X86/fshr.ll
+++ b/llvm/test/Analysis/CostModel/X86/fshr.ll
@@ -71,9 +71,9 @@ define void @var_funnel_i64(i64 %a64, <2 x i64> %a128, <4 x i64> %a256, <8 x i64
;
; BTVER2-LABEL: 'var_funnel_i64'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %I64 = call i64 @llvm.fshr.i64(i64 %a64, i64 %b64, i64 %c64)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %V2I64 = call <2 x i64> @llvm.fshr.v2i64(<2 x i64> %a128, <2 x i64> %b128, <2 x i64> %c128)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 31 for instruction: %V4I64 = call <4 x i64> @llvm.fshr.v4i64(<4 x i64> %a256, <4 x i64> %b256, <4 x i64> %c256)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 62 for instruction: %V8I64 = call <8 x i64> @llvm.fshr.v8i64(<8 x i64> %a512, <8 x i64> %b512, <8 x i64> %c512)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V2I64 = call <2 x i64> @llvm.fshr.v2i64(<2 x i64> %a128, <2 x i64> %b128, <2 x i64> %c128)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 23 for instruction: %V4I64 = call <4 x i64> @llvm.fshr.v4i64(<4 x i64> %a256, <4 x i64> %b256, <4 x i64> %c256)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 46 for instruction: %V8I64 = call <8 x i64> @llvm.fshr.v8i64(<8 x i64> %a512, <8 x i64> %b512, <8 x i64> %c512)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%I64 = call i64 @llvm.fshr.i64(i64 %a64, i64 %b64, i64 %c64)
@@ -135,9 +135,9 @@ define void @var_funnel_i32(i32 %a32, <4 x i32> %a128, <8 x i32> %a256, <16 x i3
;
; BTVER2-LABEL: 'var_funnel_i32'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %I32 = call i32 @llvm.fshr.i32(i32 %a32, i32 %b32, i32 %c32)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %V2I32 = call <4 x i32> @llvm.fshr.v4i32(<4 x i32> %a128, <4 x i32> %b128, <4 x i32> %c128)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 45 for instruction: %V4I32 = call <8 x i32> @llvm.fshr.v8i32(<8 x i32> %a256, <8 x i32> %b256, <8 x i32> %c256)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 90 for instruction: %V8I32 = call <16 x i32> @llvm.fshr.v16i32(<16 x i32> %a512, <16 x i32> %b512, <16 x i32> %c512)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %V2I32 = call <4 x i32> @llvm.fshr.v4i32(<4 x i32> %a128, <4 x i32> %b128, <4 x i32> %c128)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 34 for instruction: %V4I32 = call <8 x i32> @llvm.fshr.v8i32(<8 x i32> %a256, <8 x i32> %b256, <8 x i32> %c256)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 68 for instruction: %V8I32 = call <16 x i32> @llvm.fshr.v16i32(<16 x i32> %a512, <16 x i32> %b512, <16 x i32> %c512)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%I32 = call i32 @llvm.fshr.i32(i32 %a32, i32 %b32, i32 %c32)
@@ -213,9 +213,9 @@ define void @var_funnel_i16(i16 %a16, <8 x i16> %a128, <16 x i16> %a256, <32 x i
;
; BTVER2-LABEL: 'var_funnel_i16'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %I16 = call i16 @llvm.fshr.i16(i16 %a16, i16 %b16, i16 %c16)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 33 for instruction: %V8I16 = call <8 x i16> @llvm.fshr.v8i16(<8 x i16> %a128, <8 x i16> %b128, <8 x i16> %c128)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 73 for instruction: %V16I16 = call <16 x i16> @llvm.fshr.v16i16(<16 x i16> %a256, <16 x i16> %b256, <16 x i16> %c256)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 146 for instruction: %V32I16 = call <32 x i16> @llvm.fshr.v32i16(<32 x i16> %a512, <32 x i16> %b512, <32 x i16> %c512)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %V8I16 = call <8 x i16> @llvm.fshr.v8i16(<8 x i16> %a128, <8 x i16> %b128, <8 x i16> %c128)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 54 for instruction: %V16I16 = call <16 x i16> @llvm.fshr.v16i16(<16 x i16> %a256, <16 x i16> %b256, <16 x i16> %c256)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 108 for instruction: %V32I16 = call <32 x i16> @llvm.fshr.v32i16(<32 x i16> %a512, <32 x i16> %b512, <32 x i16> %c512)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%I16 = call i16 @llvm.fshr.i16(i16 %a16, i16 %b16, i16 %c16)
@@ -291,9 +291,9 @@ define void @var_funnel_i8(i8 %a8, <16 x i8> %a128, <32 x i8> %a256, <64 x i8> %
;
; BTVER2-LABEL: 'var_funnel_i8'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %I8 = call i8 @llvm.fshr.i8(i8 %a8, i8 %b8, i8 %c8)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 28 for instruction: %V16I8 = call <16 x i8> @llvm.fshr.v16i8(<16 x i8> %a128, <16 x i8> %b128, <16 x i8> %c128)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 63 for instruction: %V32I8 = call <32 x i8> @llvm.fshr.v32i8(<32 x i8> %a256, <32 x i8> %b256, <32 x i8> %c256)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 126 for instruction: %V64I8 = call <64 x i8> @llvm.fshr.v64i8(<64 x i8> %a512, <64 x i8> %b512, <64 x i8> %c512)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 26 for instruction: %V16I8 = call <16 x i8> @llvm.fshr.v16i8(<16 x i8> %a128, <16 x i8> %b128, <16 x i8> %c128)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 58 for instruction: %V32I8 = call <32 x i8> @llvm.fshr.v32i8(<32 x i8> %a256, <32 x i8> %b256, <32 x i8> %c256)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 116 for instruction: %V64I8 = call <64 x i8> @llvm.fshr.v64i8(<64 x i8> %a512, <64 x i8> %b512, <64 x i8> %c512)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%I8 = call i8 @llvm.fshr.i8(i8 %a8, i8 %b8, i8 %c8)
@@ -657,9 +657,9 @@ define void @splatvar_funnel_i8(i8 %a8, <16 x i8> %a128, <32 x i8> %a256, <64 x
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %u128 = shufflevector <16 x i8> %c128, <16 x i8> undef, <16 x i32> zeroinitializer
; BTVER2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %u256 = shufflevector <32 x i8> %c256, <32 x i8> undef, <32 x i32> zeroinitializer
; BTVER2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %u512 = shufflevector <64 x i8> %c512, <64 x i8> undef, <64 x i32> zeroinitializer
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 28 for instruction: %V16I8 = call <16 x i8> @llvm.fshr.v16i8(<16 x i8> %a128, <16 x i8> %b128, <16 x i8> %u128)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 63 for instruction: %V32I8 = call <32 x i8> @llvm.fshr.v32i8(<32 x i8> %a256, <32 x i8> %b256, <32 x i8> %u256)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 126 for instruction: %V64I8 = call <64 x i8> @llvm.fshr.v64i8(<64 x i8> %a512, <64 x i8> %b512, <64 x i8> %u512)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 26 for instruction: %V16I8 = call <16 x i8> @llvm.fshr.v16i8(<16 x i8> %a128, <16 x i8> %b128, <16 x i8> %u128)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 58 for instruction: %V32I8 = call <32 x i8> @llvm.fshr.v32i8(<32 x i8> %a256, <32 x i8> %b256, <32 x i8> %u256)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 116 for instruction: %V64I8 = call <64 x i8> @llvm.fshr.v64i8(<64 x i8> %a512, <64 x i8> %b512, <64 x i8> %u512)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%u128 = shufflevector <16 x i8> %c128, <16 x i8> undef, <16 x i32> zeroinitializer
@@ -727,9 +727,9 @@ define void @constant_funnel_i64(i64 %a64, <2 x i64> %a128, <4 x i64> %a256, <8
;
; BTVER2-LABEL: 'constant_funnel_i64'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %I64 = call i64 @llvm.fshr.i64(i64 %a64, i64 %b64, i64 7)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %V2I64 = call <2 x i64> @llvm.fshr.v2i64(<2 x i64> %a128, <2 x i64> %b128, <2 x i64> <i64 1, i64 7>)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 30 for instruction: %V4I64 = call <4 x i64> @llvm.fshr.v4i64(<4 x i64> %a256, <4 x i64> %b256, <4 x i64> <i64 1, i64 7, i64 15, i64 31>)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 60 for instruction: %V8I64 = call <8 x i64> @llvm.fshr.v8i64(<8 x i64> %a512, <8 x i64> %b512, <8 x i64> <i64 1, i64 7, i64 15, i64 31, i64 1, i64 7, i64 15, i64 31>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V2I64 = call <2 x i64> @llvm.fshr.v2i64(<2 x i64> %a128, <2 x i64> %b128, <2 x i64> <i64 1, i64 7>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 22 for instruction: %V4I64 = call <4 x i64> @llvm.fshr.v4i64(<4 x i64> %a256, <4 x i64> %b256, <4 x i64> <i64 1, i64 7, i64 15, i64 31>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 44 for instruction: %V8I64 = call <8 x i64> @llvm.fshr.v8i64(<8 x i64> %a512, <8 x i64> %b512, <8 x i64> <i64 1, i64 7, i64 15, i64 31, i64 1, i64 7, i64 15, i64 31>)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%I64 = call i64 @llvm.fshr.i64(i64 %a64, i64 %b64, i64 7)
@@ -791,9 +791,9 @@ define void @constant_funnel_i32(i32 %a32, <4 x i32> %a128, <8 x i32> %a256, <16
;
; BTVER2-LABEL: 'constant_funnel_i32'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %I32 = call i32 @llvm.fshr.i32(i32 %a32, i32 %b32, i32 7)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 17 for instruction: %V2I32 = call <4 x i32> @llvm.fshr.v4i32(<4 x i32> %a128, <4 x i32> %b128, <4 x i32> <i32 4, i32 5, i32 6, i32 7>)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 39 for instruction: %V4I32 = call <8 x i32> @llvm.fshr.v8i32(<8 x i32> %a256, <8 x i32> %b256, <8 x i32> <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 78 for instruction: %V8I32 = call <16 x i32> @llvm.fshr.v16i32(<16 x i32> %a512, <16 x i32> %b512, <16 x i32> <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>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %V2I32 = call <4 x i32> @llvm.fshr.v4i32(<4 x i32> %a128, <4 x i32> %b128, <4 x i32> <i32 4, i32 5, i32 6, i32 7>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 29 for instruction: %V4I32 = call <8 x i32> @llvm.fshr.v8i32(<8 x i32> %a256, <8 x i32> %b256, <8 x i32> <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 58 for instruction: %V8I32 = call <16 x i32> @llvm.fshr.v16i32(<16 x i32> %a512, <16 x i32> %b512, <16 x i32> <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>)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%I32 = call i32 @llvm.fshr.i32(i32 %a32, i32 %b32, i32 7)
@@ -869,9 +869,9 @@ define void @constant_funnel_i16(i16 %a16, <8 x i16> %a128, <16 x i16> %a256, <3
;
; BTVER2-LABEL: 'constant_funnel_i16'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %I16 = call i16 @llvm.fshr.i16(i16 %a16, i16 %b16, i16 7)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %V8I16 = call <8 x i16> @llvm.fshr.v8i16(<8 x i16> %a128, <8 x i16> %b128, <8 x i16> <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 46 for instruction: %V16I16 = call <16 x i16> @llvm.fshr.v16i16(<16 x i16> %a256, <16 x i16> %b256, <16 x i16> <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>)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 92 for instruction: %V32I16 = call <32 x i16> @llvm.fshr.v32i16(<32 x i16> %a512, <32 x i16> %b512, <32 x i16> <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>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %V8I16 = call <8 x i16> @llvm.fshr.v8i16(<8 x i16> %a128, <8 x i16> %b128, <8 x i16> <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 44 for instruction: %V16I16 = call <16 x i16> @llvm.fshr.v16i16(<16 x i16> %a256, <16 x i16> %b256, <16 x i16> <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>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 88 for instruction: %V32I16 = call <32 x i16> @llvm.fshr.v32i16(<32 x i16> %a512, <32 x i16> %b512, <32 x i16> <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>)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%I16 = call i16 @llvm.fshr.i16(i16 %a16, i16 %b16, i16 7)
@@ -947,9 +947,9 @@ define void @constant_funnel_i8(i8 %a8, <16 x i8> %a128, <32 x i8> %a256, <64 x
;
; BTVER2-LABEL: 'constant_funnel_i8'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %I8 = call i8 @llvm.fshr.i8(i8 %a8, i8 %b8, i8 7)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 27 for instruction: %V16I8 = call <16 x i8> @llvm.fshr.v16i8(<16 x i8> %a128, <16 x i8> %b128, <16 x i8> <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>)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 62 for instruction: %V32I8 = call <32 x i8> @llvm.fshr.v32i8(<32 x i8> %a256, <32 x i8> %b256, <32 x i8> <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>)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 124 for instruction: %V64I8 = call <64 x i8> @llvm.fshr.v64i8(<64 x i8> %a512, <64 x i8> %b512, <64 x i8> <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>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 25 for instruction: %V16I8 = call <16 x i8> @llvm.fshr.v16i8(<16 x i8> %a128, <16 x i8> %b128, <16 x i8> <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>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 57 for instruction: %V32I8 = call <32 x i8> @llvm.fshr.v32i8(<32 x i8> %a256, <32 x i8> %b256, <32 x i8> <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>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 114 for instruction: %V64I8 = call <64 x i8> @llvm.fshr.v64i8(<64 x i8> %a512, <64 x i8> %b512, <64 x i8> <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>)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%I8 = call i8 @llvm.fshr.i8(i8 %a8, i8 %b8, i8 7)
@@ -1276,9 +1276,9 @@ define void @var_rotate_i64(i64 %a64, <2 x i64> %a128, <4 x i64> %a256, <8 x i64
;
; BTVER2-LABEL: 'var_rotate_i64'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %I64 = call i64 @llvm.fshr.i64(i64 %a64, i64 %a64, i64 %c64)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %V2I64 = call <2 x i64> @llvm.fshr.v2i64(<2 x i64> %a128, <2 x i64> %a128, <2 x i64> %c128)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 26 for instruction: %V4I64 = call <4 x i64> @llvm.fshr.v4i64(<4 x i64> %a256, <4 x i64> %a256, <4 x i64> %c256)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 52 for instruction: %V8I64 = call <8 x i64> @llvm.fshr.v8i64(<8 x i64> %a512, <8 x i64> %a512, <8 x i64> %c512)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V2I64 = call <2 x i64> @llvm.fshr.v2i64(<2 x i64> %a128, <2 x i64> %a128, <2 x i64> %c128)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %V4I64 = call <4 x i64> @llvm.fshr.v4i64(<4 x i64> %a256, <4 x i64> %a256, <4 x i64> %c256)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 36 for instruction: %V8I64 = call <8 x i64> @llvm.fshr.v8i64(<8 x i64> %a512, <8 x i64> %a512, <8 x i64> %c512)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%I64 = call i64 @llvm.fshr.i64(i64 %a64, i64 %a64, i64 %c64)
@@ -1340,9 +1340,9 @@ define void @var_rotate_i32(i32 %a32, <4 x i32> %a128, <8 x i32> %a256, <16 x i3
;
; BTVER2-LABEL: 'var_rotate_i32'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %I32 = call i32 @llvm.fshr.i32(i32 %a32, i32 %a32, i32 %c32)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %V2I32 = call <4 x i32> @llvm.fshr.v4i32(<4 x i32> %a128, <4 x i32> %a128, <4 x i32> %c128)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 40 for instruction: %V4I32 = call <8 x i32> @llvm.fshr.v8i32(<8 x i32> %a256, <8 x i32> %a256, <8 x i32> %c256)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 80 for instruction: %V8I32 = call <16 x i32> @llvm.fshr.v16i32(<16 x i32> %a512, <16 x i32> %a512, <16 x i32> %c512)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %V2I32 = call <4 x i32> @llvm.fshr.v4i32(<4 x i32> %a128, <4 x i32> %a128, <4 x i32> %c128)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 29 for instruction: %V4I32 = call <8 x i32> @llvm.fshr.v8i32(<8 x i32> %a256, <8 x i32> %a256, <8 x i32> %c256)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 58 for instruction: %V8I32 = call <16 x i32> @llvm.fshr.v16i32(<16 x i32> %a512, <16 x i32> %a512, <16 x i32> %c512)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%I32 = call i32 @llvm.fshr.i32(i32 %a32, i32 %a32, i32 %c32)
@@ -1418,9 +1418,9 @@ define void @var_rotate_i16(i16 %a16, <8 x i16> %a128, <16 x i16> %a256, <32 x i
;
; BTVER2-LABEL: 'var_rotate_i16'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.fshr.i16(i16 %a16, i16 %a16, i16 %c16)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 31 for instruction: %V8I16 = call <8 x i16> @llvm.fshr.v8i16(<8 x i16> %a128, <8 x i16> %a128, <8 x i16> %c128)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 66 for instruction: %V16I16 = call <16 x i16> @llvm.fshr.v16i16(<16 x i16> %a256, <16 x i16> %a256, <16 x i16> %c256)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 132 for instruction: %V32I16 = call <32 x i16> @llvm.fshr.v32i16(<32 x i16> %a512, <32 x i16> %a512, <32 x i16> %c512)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 22 for instruction: %V8I16 = call <8 x i16> @llvm.fshr.v8i16(<8 x i16> %a128, <8 x i16> %a128, <8 x i16> %c128)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 47 for instruction: %V16I16 = call <16 x i16> @llvm.fshr.v16i16(<16 x i16> %a256, <16 x i16> %a256, <16 x i16> %c256)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 94 for instruction: %V32I16 = call <32 x i16> @llvm.fshr.v32i16(<32 x i16> %a512, <32 x i16> %a512, <32 x i16> %c512)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%I16 = call i16 @llvm.fshr.i16(i16 %a16, i16 %a16, i16 %c16)
@@ -1496,9 +1496,9 @@ define void @var_rotate_i8(i8 %a8, <16 x i8> %a128, <32 x i8> %a256, <64 x i8> %
;
; BTVER2-LABEL: 'var_rotate_i8'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.fshr.i8(i8 %a8, i8 %a8, i8 %c8)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 26 for instruction: %V16I8 = call <16 x i8> @llvm.fshr.v16i8(<16 x i8> %a128, <16 x i8> %a128, <16 x i8> %c128)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 56 for instruction: %V32I8 = call <32 x i8> @llvm.fshr.v32i8(<32 x i8> %a256, <32 x i8> %a256, <32 x i8> %c256)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 112 for instruction: %V64I8 = call <64 x i8> @llvm.fshr.v64i8(<64 x i8> %a512, <64 x i8> %a512, <64 x i8> %c512)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %V16I8 = call <16 x i8> @llvm.fshr.v16i8(<16 x i8> %a128, <16 x i8> %a128, <16 x i8> %c128)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 51 for instruction: %V32I8 = call <32 x i8> @llvm.fshr.v32i8(<32 x i8> %a256, <32 x i8> %a256, <32 x i8> %c256)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 102 for instruction: %V64I8 = call <64 x i8> @llvm.fshr.v64i8(<64 x i8> %a512, <64 x i8> %a512, <64 x i8> %c512)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%I8 = call i8 @llvm.fshr.i8(i8 %a8, i8 %a8, i8 %c8)
@@ -1835,9 +1835,9 @@ define void @splatvar_rotate_i8(i8 %a8, <16 x i8> %a128, <32 x i8> %a256, <64 x
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %u128 = shufflevector <16 x i8> %c128, <16 x i8> undef, <16 x i32> zeroinitializer
; BTVER2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %u256 = shufflevector <32 x i8> %c256, <32 x i8> undef, <32 x i32> zeroinitializer
; BTVER2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %u512 = shufflevector <64 x i8> %c512, <64 x i8> undef, <64 x i32> zeroinitializer
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 26 for instruction: %V16I8 = call <16 x i8> @llvm.fshr.v16i8(<16 x i8> %a128, <16 x i8> %a128, <16 x i8> %u128)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 56 for instruction: %V32I8 = call <32 x i8> @llvm.fshr.v32i8(<32 x i8> %a256, <32 x i8> %a256, <32 x i8> %u256)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 112 for instruction: %V64I8 = call <64 x i8> @llvm.fshr.v64i8(<64 x i8> %a512, <64 x i8> %a512, <64 x i8> %u512)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %V16I8 = call <16 x i8> @llvm.fshr.v16i8(<16 x i8> %a128, <16 x i8> %a128, <16 x i8> %u128)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 51 for instruction: %V32I8 = call <32 x i8> @llvm.fshr.v32i8(<32 x i8> %a256, <32 x i8> %a256, <32 x i8> %u256)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 102 for instruction: %V64I8 = call <64 x i8> @llvm.fshr.v64i8(<64 x i8> %a512, <64 x i8> %a512, <64 x i8> %u512)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%u128 = shufflevector <16 x i8> %c128, <16 x i8> undef, <16 x i32> zeroinitializer
@@ -1898,9 +1898,9 @@ define void @constant_rotate_i64(i64 %a64, <2 x i64> %a128, <4 x i64> %a256, <8
;
; BTVER2-LABEL: 'constant_rotate_i64'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %I64 = call i64 @llvm.fshr.i64(i64 %a64, i64 %a64, i64 7)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %V2I64 = call <2 x i64> @llvm.fshr.v2i64(<2 x i64> %a128, <2 x i64> %a128, <2 x i64> <i64 1, i64 7>)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 25 for instruction: %V4I64 = call <4 x i64> @llvm.fshr.v4i64(<4 x i64> %a256, <4 x i64> %a256, <4 x i64> <i64 1, i64 7, i64 15, i64 31>)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 50 for instruction: %V8I64 = call <8 x i64> @llvm.fshr.v8i64(<8 x i64> %a512, <8 x i64> %a512, <8 x i64> <i64 1, i64 7, i64 15, i64 31, i64 1, i64 7, i64 15, i64 31>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V2I64 = call <2 x i64> @llvm.fshr.v2i64(<2 x i64> %a128, <2 x i64> %a128, <2 x i64> <i64 1, i64 7>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 17 for instruction: %V4I64 = call <4 x i64> @llvm.fshr.v4i64(<4 x i64> %a256, <4 x i64> %a256, <4 x i64> <i64 1, i64 7, i64 15, i64 31>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 34 for instruction: %V8I64 = call <8 x i64> @llvm.fshr.v8i64(<8 x i64> %a512, <8 x i64> %a512, <8 x i64> <i64 1, i64 7, i64 15, i64 31, i64 1, i64 7, i64 15, i64 31>)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%I64 = call i64 @llvm.fshr.i64(i64 %a64, i64 %a64, i64 7)
@@ -1962,9 +1962,9 @@ define void @constant_rotate_i32(i32 %a32, <4 x i32> %a128, <8 x i32> %a256, <16
;
; BTVER2-LABEL: 'constant_rotate_i32'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %I32 = call i32 @llvm.fshr.i32(i32 %a32, i32 %a32, i32 7)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V2I32 = call <4 x i32> @llvm.fshr.v4i32(<4 x i32> %a128, <4 x i32> %a128, <4 x i32> <i32 4, i32 5, i32 6, i32 7>)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 34 for instruction: %V4I32 = call <8 x i32> @llvm.fshr.v8i32(<8 x i32> %a256, <8 x i32> %a256, <8 x i32> <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 68 for instruction: %V8I32 = call <16 x i32> @llvm.fshr.v16i32(<16 x i32> %a512, <16 x i32> %a512, <16 x i32> <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>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %V2I32 = call <4 x i32> @llvm.fshr.v4i32(<4 x i32> %a128, <4 x i32> %a128, <4 x i32> <i32 4, i32 5, i32 6, i32 7>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %V4I32 = call <8 x i32> @llvm.fshr.v8i32(<8 x i32> %a256, <8 x i32> %a256, <8 x i32> <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 48 for instruction: %V8I32 = call <16 x i32> @llvm.fshr.v16i32(<16 x i32> %a512, <16 x i32> %a512, <16 x i32> <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>)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%I32 = call i32 @llvm.fshr.i32(i32 %a32, i32 %a32, i32 7)
@@ -2040,9 +2040,9 @@ define void @constant_rotate_i16(i16 %a16, <8 x i16> %a128, <16 x i16> %a256, <3
;
; BTVER2-LABEL: 'constant_rotate_i16'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.fshr.i16(i16 %a16, i16 %a16, i16 7)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 17 for instruction: %V8I16 = call <8 x i16> @llvm.fshr.v8i16(<8 x i16> %a128, <8 x i16> %a128, <8 x i16> <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 39 for instruction: %V16I16 = call <16 x i16> @llvm.fshr.v16i16(<16 x i16> %a256, <16 x i16> %a256, <16 x i16> <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>)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 78 for instruction: %V32I16 = call <32 x i16> @llvm.fshr.v32i16(<32 x i16> %a512, <32 x i16> %a512, <32 x i16> <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>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %V8I16 = call <8 x i16> @llvm.fshr.v8i16(<8 x i16> %a128, <8 x i16> %a128, <8 x i16> <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 37 for instruction: %V16I16 = call <16 x i16> @llvm.fshr.v16i16(<16 x i16> %a256, <16 x i16> %a256, <16 x i16> <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>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 74 for instruction: %V32I16 = call <32 x i16> @llvm.fshr.v32i16(<32 x i16> %a512, <32 x i16> %a512, <32 x i16> <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>)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%I16 = call i16 @llvm.fshr.i16(i16 %a16, i16 %a16, i16 7)
@@ -2118,9 +2118,9 @@ define void @constant_rotate_i8(i8 %a8, <16 x i8> %a128, <32 x i8> %a256, <64 x
;
; BTVER2-LABEL: 'constant_rotate_i8'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.fshr.i8(i8 %a8, i8 %a8, i8 7)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 25 for instruction: %V16I8 = call <16 x i8> @llvm.fshr.v16i8(<16 x i8> %a128, <16 x i8> %a128, <16 x i8> <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>)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 55 for instruction: %V32I8 = call <32 x i8> @llvm.fshr.v32i8(<32 x i8> %a256, <32 x i8> %a256, <32 x i8> <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>)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 110 for instruction: %V64I8 = call <64 x i8> @llvm.fshr.v64i8(<64 x i8> %a512, <64 x i8> %a512, <64 x i8> <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>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 23 for instruction: %V16I8 = call <16 x i8> @llvm.fshr.v16i8(<16 x i8> %a128, <16 x i8> %a128, <16 x i8> <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>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 50 for instruction: %V32I8 = call <32 x i8> @llvm.fshr.v32i8(<32 x i8> %a256, <32 x i8> %a256, <32 x i8> <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>)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 100 for instruction: %V64I8 = call <64 x i8> @llvm.fshr.v64i8(<64 x i8> %a512, <64 x i8> %a512, <64 x i8> <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>)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%I8 = call i8 @llvm.fshr.i8(i8 %a8, i8 %a8, i8 7)
diff --git a/llvm/test/Analysis/CostModel/X86/rem.ll b/llvm/test/Analysis/CostModel/X86/rem.ll
index d7961065e8377..ec3a8cf40f325 100644
--- a/llvm/test/Analysis/CostModel/X86/rem.ll
+++ b/llvm/test/Analysis/CostModel/X86/rem.ll
@@ -785,21 +785,21 @@ define i32 @srem_constpow2() {
;
; AVX1-LABEL: 'srem_constpow2'
; AVX1-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %I64 = srem i64 undef, 16
-; AVX1-NEXT: Cost Model: Found an estimated cost of 36 for instruction: %V2i64 = srem <2 x i64> undef, <i64 8, i64 16>
-; AVX1-NEXT: Cost Model: Found an estimated cost of 82 for instruction: %V4i64 = srem <4 x i64> undef, <i64 2, i64 4, i64 8, i64 16>
-; AVX1-NEXT: Cost Model: Found an estimated cost of 164 for instruction: %V8i64 = srem <8 x i64> undef, <i64 2, i64 4, i64 8, i64 16, i64 32, i64 64, i64 128, i64 256>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %V2i64 = srem <2 x i64> undef, <i64 8, i64 16>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 50 for instruction: %V4i64 = srem <4 x i64> undef, <i64 2, i64 4, i64 8, i64 16>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 100 for instruction: %V8i64 = srem <8 x i64> undef, <i64 2, i64 4, i64 8, i64 16, i64 32, i64 64, i64 128, i64 256>
; AVX1-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %I32 = srem i32 undef, 16
-; AVX1-NEXT: Cost Model: Found an estimated cost of 39 for instruction: %V4i32 = srem <4 x i32> undef, <i32 2, i32 4, i32 8, i32 16>
-; AVX1-NEXT: Cost Model: Found an estimated cost of 89 for instruction: %V8i32 = srem <8 x i32> undef, <i32 2, i32 4, i32 8, i32 16, i32 32, i32 64, i32 128, i32 256>
-; AVX1-NEXT: Cost Model: Found an estimated cost of 178 for instruction: %V16i32 = srem <16 x i32> undef, <i32 2, i32 4, i32 8, i32 16, i32 32, i32 64, i32 128, i32 256, i32 2, i32 4, i32 8, i32 16, i32 32, i32 64, i32 128, i32 256>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 22 for instruction: %V4i32 = srem <4 x i32> undef, <i32 2, i32 4, i32 8, i32 16>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 55 for instruction: %V8i32 = srem <8 x i32> undef, <i32 2, i32 4, i32 8, i32 16, i32 32, i32 64, i32 128, i32 256>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 110 for instruction: %V16i32 = srem <16 x i32> undef, <i32 2, i32 4, i32 8, i32 16, i32 32, i32 64, i32 128, i32 256, i32 2, i32 4, i32 8, i32 16, i32 32, i32 64, i32 128, i32 256>
; AVX1-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %I16 = srem i16 undef, 16
-; AVX1-NEXT: Cost Model: Found an estimated cost of 45 for instruction: %V8i16 = srem <8 x i16> undef, <i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256>
-; AVX1-NEXT: Cost Model: Found an estimated cost of 102 for instruction: %V16i16 = srem <16 x i16> undef, <i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256>
-; AVX1-NEXT: Cost Model: Found an estimated cost of 204 for instruction: %V32i16 = srem <32 x i16> undef, <i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 42 for instruction: %V8i16 = srem <8 x i16> undef, <i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 96 for instruction: %V16i16 = srem <16 x i16> undef, <i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 192 for instruction: %V32i16 = srem <32 x i16> undef, <i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256>
; AVX1-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %I8 = srem i8 undef, 16
-; AVX1-NEXT: Cost Model: Found an estimated cost of 74 for instruction: %V16i8 = srem <16 x i8> undef, <i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16>
-; AVX1-NEXT: Cost Model: Found an estimated cost of 160 for instruction: %V32i8 = srem <32 x i8> undef, <i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16>
-; AVX1-NEXT: Cost Model: Found an estimated cost of 320 for instruction: %V64i8 = srem <64 x i8> undef, <i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 67 for instruction: %V16i8 = srem <16 x i8> undef, <i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 145 for instruction: %V32i8 = srem <32 x i8> undef, <i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 290 for instruction: %V64i8 = srem <64 x i8> undef, <i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; AVX2-LABEL: 'srem_constpow2'
@@ -899,21 +899,21 @@ define i32 @srem_constpow2() {
;
; BTVER2-LABEL: 'srem_constpow2'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %I64 = srem i64 undef, 16
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 36 for instruction: %V2i64 = srem <2 x i64> undef, <i64 8, i64 16>
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 82 for instruction: %V4i64 = srem <4 x i64> undef, <i64 2, i64 4, i64 8, i64 16>
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 164 for instruction: %V8i64 = srem <8 x i64> undef, <i64 2, i64 4, i64 8, i64 16, i64 32, i64 64, i64 128, i64 256>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %V2i64 = srem <2 x i64> undef, <i64 8, i64 16>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 50 for instruction: %V4i64 = srem <4 x i64> undef, <i64 2, i64 4, i64 8, i64 16>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 100 for instruction: %V8i64 = srem <8 x i64> undef, <i64 2, i64 4, i64 8, i64 16, i64 32, i64 64, i64 128, i64 256>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %I32 = srem i32 undef, 16
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 39 for instruction: %V4i32 = srem <4 x i32> undef, <i32 2, i32 4, i32 8, i32 16>
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 89 for instruction: %V8i32 = srem <8 x i32> undef, <i32 2, i32 4, i32 8, i32 16, i32 32, i32 64, i32 128, i32 256>
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 178 for instruction: %V16i32 = srem <16 x i32> undef, <i32 2, i32 4, i32 8, i32 16, i32 32, i32 64, i32 128, i32 256, i32 2, i32 4, i32 8, i32 16, i32 32, i32 64, i32 128, i32 256>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 22 for instruction: %V4i32 = srem <4 x i32> undef, <i32 2, i32 4, i32 8, i32 16>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 55 for instruction: %V8i32 = srem <8 x i32> undef, <i32 2, i32 4, i32 8, i32 16, i32 32, i32 64, i32 128, i32 256>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 110 for instruction: %V16i32 = srem <16 x i32> undef, <i32 2, i32 4, i32 8, i32 16, i32 32, i32 64, i32 128, i32 256, i32 2, i32 4, i32 8, i32 16, i32 32, i32 64, i32 128, i32 256>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %I16 = srem i16 undef, 16
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 45 for instruction: %V8i16 = srem <8 x i16> undef, <i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256>
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 102 for instruction: %V16i16 = srem <16 x i16> undef, <i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256>
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 204 for instruction: %V32i16 = srem <32 x i16> undef, <i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 42 for instruction: %V8i16 = srem <8 x i16> undef, <i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 96 for instruction: %V16i16 = srem <16 x i16> undef, <i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 192 for instruction: %V32i16 = srem <32 x i16> undef, <i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256, i16 2, i16 4, i16 8, i16 16, i16 32, i16 64, i16 128, i16 256>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %I8 = srem i8 undef, 16
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 74 for instruction: %V16i8 = srem <16 x i8> undef, <i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16>
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 160 for instruction: %V32i8 = srem <32 x i8> undef, <i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16>
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 320 for instruction: %V64i8 = srem <64 x i8> undef, <i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 67 for instruction: %V16i8 = srem <16 x i8> undef, <i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 145 for instruction: %V32i8 = srem <32 x i8> undef, <i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 290 for instruction: %V64i8 = srem <64 x i8> undef, <i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16, i8 2, i8 4, i8 8, i8 16>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
%I64 = srem i64 undef, 16
diff --git a/llvm/test/Analysis/CostModel/X86/vshift-ashr-cost-inseltpoison.ll b/llvm/test/Analysis/CostModel/X86/vshift-ashr-cost-inseltpoison.ll
index cb70448b90ec1..8b03d6819ebf1 100644
--- a/llvm/test/Analysis/CostModel/X86/vshift-ashr-cost-inseltpoison.ll
+++ b/llvm/test/Analysis/CostModel/X86/vshift-ashr-cost-inseltpoison.ll
@@ -28,7 +28,7 @@ define <2 x i64> @var_shift_v2i64(<2 x i64> %a, <2 x i64> %b) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
; AVX1-LABEL: 'var_shift_v2i64'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = ashr <2 x i64> %a, %b
+; AVX1-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %shift = ashr <2 x i64> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
; AVX2-LABEL: 'var_shift_v2i64'
@@ -44,7 +44,7 @@ define <2 x i64> @var_shift_v2i64(<2 x i64> %a, <2 x i64> %b) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
; BTVER2-LABEL: 'var_shift_v2i64'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = ashr <2 x i64> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %shift = ashr <2 x i64> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
%shift = ashr <2 x i64> %a, %b
@@ -57,7 +57,7 @@ define <4 x i64> @var_shift_v4i64(<4 x i64> %a, <4 x i64> %b) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
; AVX1-LABEL: 'var_shift_v4i64'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 26 for instruction: %shift = ashr <4 x i64> %a, %b
+; AVX1-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = ashr <4 x i64> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
; AVX2-LABEL: 'var_shift_v4i64'
@@ -73,7 +73,7 @@ define <4 x i64> @var_shift_v4i64(<4 x i64> %a, <4 x i64> %b) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
; BTVER2-LABEL: 'var_shift_v4i64'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 26 for instruction: %shift = ashr <4 x i64> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = ashr <4 x i64> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
%shift = ashr <4 x i64> %a, %b
@@ -86,7 +86,7 @@ define <8 x i64> @var_shift_v8i64(<8 x i64> %a, <8 x i64> %b) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
; AVX1-LABEL: 'var_shift_v8i64'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 52 for instruction: %shift = ashr <8 x i64> %a, %b
+; AVX1-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %shift = ashr <8 x i64> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
; AVX2-LABEL: 'var_shift_v8i64'
@@ -102,7 +102,7 @@ define <8 x i64> @var_shift_v8i64(<8 x i64> %a, <8 x i64> %b) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
; BTVER2-LABEL: 'var_shift_v8i64'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 52 for instruction: %shift = ashr <8 x i64> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %shift = ashr <8 x i64> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
%shift = ashr <8 x i64> %a, %b
@@ -119,7 +119,7 @@ define <4 x i32> @var_shift_v4i32(<4 x i32> %a, <4 x i32> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %shift = ashr <4 x i32> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
;
; AVX2-LABEL: 'var_shift_v4i32'
@@ -135,7 +135,7 @@ define <4 x i32> @var_shift_v4i32(<4 x i32> %a, <4 x i32> %b) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
;
; BTVER2-LABEL: 'var_shift_v4i32'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = ashr <4 x i32> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %shift = ashr <4 x i32> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
;
%shift = ashr <4 x i32> %a, %b
@@ -152,7 +152,7 @@ define <8 x i32> @var_shift_v8i32(<8 x i32> %a, <8 x i32> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %shift = ashr <8 x i32> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
;
; AVX2-LABEL: 'var_shift_v8i32'
@@ -172,7 +172,7 @@ define <8 x i32> @var_shift_v8i32(<8 x i32> %a, <8 x i32> %b) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
;
; BTVER2-LABEL: 'var_shift_v8i32'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 26 for instruction: %shift = ashr <8 x i32> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %shift = ashr <8 x i32> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
;
%shift = ashr <8 x i32> %a, %b
@@ -189,7 +189,7 @@ define <16 x i32> @var_shift_v16i32(<16 x i32> %a, <16 x i32> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 28 for instruction: %shift = ashr <16 x i32> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
;
; AVX2-LABEL: 'var_shift_v16i32'
@@ -209,7 +209,7 @@ define <16 x i32> @var_shift_v16i32(<16 x i32> %a, <16 x i32> %b) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
;
; BTVER2-LABEL: 'var_shift_v16i32'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 52 for instruction: %shift = ashr <16 x i32> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 28 for instruction: %shift = ashr <16 x i32> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
;
%shift = ashr <16 x i32> %a, %b
@@ -226,7 +226,7 @@ define <8 x i16> @var_shift_v8i16(<8 x i16> %a, <8 x i16> %b) {
; SSE42-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
;
; AVX1-LABEL: 'var_shift_v8i16'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %shift = ashr <8 x i16> %a, %b
+; AVX1-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %shift = ashr <8 x i16> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
;
; AVX2-LABEL: 'var_shift_v8i16'
@@ -254,7 +254,7 @@ define <8 x i16> @var_shift_v8i16(<8 x i16> %a, <8 x i16> %b) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
;
; BTVER2-LABEL: 'var_shift_v8i16'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %shift = ashr <8 x i16> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %shift = ashr <8 x i16> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
;
%shift = ashr <8 x i16> %a, %b
@@ -271,7 +271,7 @@ define <16 x i16> @var_shift_v16i16(<16 x i16> %a, <16 x i16> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 28 for instruction: %shift = ashr <16 x i16> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
;
; AVX2-LABEL: 'var_shift_v16i16'
@@ -299,7 +299,7 @@ define <16 x i16> @var_shift_v16i16(<16 x i16> %a, <16 x i16> %b) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
;
; BTVER2-LABEL: 'var_shift_v16i16'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 30 for instruction: %shift = ashr <16 x i16> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 28 for instruction: %shift = ashr <16 x i16> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
;
%shift = ashr <16 x i16> %a, %b
@@ -316,7 +316,7 @@ define <32 x i16> @var_shift_v32i16(<32 x i16> %a, <32 x i16> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 56 for instruction: %shift = ashr <32 x i16> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
;
; AVX2-LABEL: 'var_shift_v32i16'
@@ -344,7 +344,7 @@ define <32 x i16> @var_shift_v32i16(<32 x i16> %a, <32 x i16> %b) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
;
; BTVER2-LABEL: 'var_shift_v32i16'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 60 for instruction: %shift = ashr <32 x i16> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 56 for instruction: %shift = ashr <32 x i16> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
;
%shift = ashr <32 x i16> %a, %b
@@ -361,7 +361,7 @@ define <16 x i8> @var_shift_v16i8(<16 x i8> %a, <16 x i8> %b) {
; SSE42-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; AVX1-LABEL: 'var_shift_v16i8'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %shift = ashr <16 x i8> %a, %b
+; AVX1-NEXT: Cost Model: Found an estimated cost of 21 for instruction: %shift = ashr <16 x i8> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; AVX2-LABEL: 'var_shift_v16i8'
@@ -389,7 +389,7 @@ define <16 x i8> @var_shift_v16i8(<16 x i8> %a, <16 x i8> %b) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; BTVER2-LABEL: 'var_shift_v16i8'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %shift = ashr <16 x i8> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 21 for instruction: %shift = ashr <16 x i8> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
%shift = ashr <16 x i8> %a, %b
@@ -406,7 +406,7 @@ define <32 x i8> @var_shift_v32i8(<32 x i8> %a, <32 x i8> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 44 for instruction: %shift = ashr <32 x i8> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
; AVX2-LABEL: 'var_shift_v32i8'
@@ -434,7 +434,7 @@ define <32 x i8> @var_shift_v32i8(<32 x i8> %a, <32 x i8> %b) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
; BTVER2-LABEL: 'var_shift_v32i8'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 50 for instruction: %shift = ashr <32 x i8> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 44 for instruction: %shift = ashr <32 x i8> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
%shift = ashr <32 x i8> %a, %b
@@ -451,7 +451,7 @@ define <64 x i8> @var_shift_v64i8(<64 x i8> %a, <64 x i8> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 88 for instruction: %shift = ashr <64 x i8> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
; AVX2-LABEL: 'var_shift_v64i8'
@@ -479,7 +479,7 @@ define <64 x i8> @var_shift_v64i8(<64 x i8> %a, <64 x i8> %b) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
; BTVER2-LABEL: 'var_shift_v64i8'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 100 for instruction: %shift = ashr <64 x i8> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 88 for instruction: %shift = ashr <64 x i8> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
%shift = ashr <64 x i8> %a, %b
@@ -965,7 +965,7 @@ define <16 x i8> @splatvar_shift_v16i8(<16 x i8> %a, i8 %b) {
; AVX1-LABEL: 'splatvar_shift_v16i8'
; AVX1-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <16 x i8> poison, i8 %b, i32 0
; AVX1-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <16 x i8> %insert, <16 x i8> poison, <16 x i32> zeroinitializer
-; AVX1-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %shift = ashr <16 x i8> %a, %splat
+; AVX1-NEXT: Cost Model: Found an estimated cost of 21 for instruction: %shift = ashr <16 x i8> %a, %splat
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; AVX2-LABEL: 'splatvar_shift_v16i8'
@@ -1007,7 +1007,7 @@ define <16 x i8> @splatvar_shift_v16i8(<16 x i8> %a, i8 %b) {
; BTVER2-LABEL: 'splatvar_shift_v16i8'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <16 x i8> poison, i8 %b, i32 0
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <16 x i8> %insert, <16 x i8> poison, <16 x i32> zeroinitializer
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %shift = ashr <16 x i8> %a, %splat
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 21 for instruction: %shift = ashr <16 x i8> %a, %splat
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
%insert = insertelement <16 x i8> poison, i8 %b, i32 0
@@ -1032,7 +1032,7 @@ define <32 x i8> @splatvar_shift_v32i8(<32 x i8> %a, i8 %b) {
; AVX1-LABEL: 'splatvar_shift_v32i8'
; AVX1-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <32 x i8> poison, i8 %b, i32 0
; AVX1-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %splat = shufflevector <32 x i8> %insert, <32 x i8> poison, <32 x i32> zeroinitializer
-; AVX1-NEXT: Cost Model: Found an estimated cost of 50 for instruction: %shift = ashr <32 x i8> %a, %splat
+; AVX1-NEXT: Cost Model: Found an estimated cost of 44 for instruction: %shift = ashr <32 x i8> %a, %splat
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
; AVX2-LABEL: 'splatvar_shift_v32i8'
@@ -1080,7 +1080,7 @@ define <32 x i8> @splatvar_shift_v32i8(<32 x i8> %a, i8 %b) {
; BTVER2-LABEL: 'splatvar_shift_v32i8'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <32 x i8> poison, i8 %b, i32 0
; BTVER2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %splat = shufflevector <32 x i8> %insert, <32 x i8> poison, <32 x i32> zeroinitializer
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 50 for instruction: %shift = ashr <32 x i8> %a, %splat
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 44 for instruction: %shift = ashr <32 x i8> %a, %splat
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
%insert = insertelement <32 x i8> poison, i8 %b, i32 0
@@ -1105,7 +1105,7 @@ define <64 x i8> @splatvar_shift_v64i8(<64 x i8> %a, i8 %b) {
; AVX1-LABEL: 'splatvar_shift_v64i8'
; AVX1-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <64 x i8> poison, i8 %b, i32 0
; AVX1-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %splat = shufflevector <64 x i8> %insert, <64 x i8> poison, <64 x i32> zeroinitializer
-; AVX1-NEXT: Cost Model: Found an estimated cost of 100 for instruction: %shift = ashr <64 x i8> %a, %splat
+; AVX1-NEXT: Cost Model: Found an estimated cost of 88 for instruction: %shift = ashr <64 x i8> %a, %splat
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
; AVX2-LABEL: 'splatvar_shift_v64i8'
@@ -1153,7 +1153,7 @@ define <64 x i8> @splatvar_shift_v64i8(<64 x i8> %a, i8 %b) {
; BTVER2-LABEL: 'splatvar_shift_v64i8'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <64 x i8> poison, i8 %b, i32 0
; BTVER2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %splat = shufflevector <64 x i8> %insert, <64 x i8> poison, <64 x i32> zeroinitializer
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 100 for instruction: %shift = ashr <64 x i8> %a, %splat
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 88 for instruction: %shift = ashr <64 x i8> %a, %splat
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
%insert = insertelement <64 x i8> poison, i8 %b, i32 0
@@ -1172,7 +1172,7 @@ define <2 x i64> @constant_shift_v2i64(<2 x i64> %a) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
; AVX1-LABEL: 'constant_shift_v2i64'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = ashr <2 x i64> %a, <i64 1, i64 7>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %shift = ashr <2 x i64> %a, <i64 1, i64 7>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
; AVX2-LABEL: 'constant_shift_v2i64'
@@ -1188,7 +1188,7 @@ define <2 x i64> @constant_shift_v2i64(<2 x i64> %a) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
; BTVER2-LABEL: 'constant_shift_v2i64'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = ashr <2 x i64> %a, <i64 1, i64 7>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %shift = ashr <2 x i64> %a, <i64 1, i64 7>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
%shift = ashr <2 x i64> %a, <i64 1, i64 7>
@@ -1201,7 +1201,7 @@ define <4 x i64> @constant_shift_v4i64(<4 x i64> %a) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
; AVX1-LABEL: 'constant_shift_v4i64'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 26 for instruction: %shift = ashr <4 x i64> %a, <i64 1, i64 7, i64 15, i64 31>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = ashr <4 x i64> %a, <i64 1, i64 7, i64 15, i64 31>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
; AVX2-LABEL: 'constant_shift_v4i64'
@@ -1217,7 +1217,7 @@ define <4 x i64> @constant_shift_v4i64(<4 x i64> %a) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
; BTVER2-LABEL: 'constant_shift_v4i64'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 26 for instruction: %shift = ashr <4 x i64> %a, <i64 1, i64 7, i64 15, i64 31>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = ashr <4 x i64> %a, <i64 1, i64 7, i64 15, i64 31>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
%shift = ashr <4 x i64> %a, <i64 1, i64 7, i64 15, i64 31>
@@ -1230,7 +1230,7 @@ define <8 x i64> @constant_shift_v8i64(<8 x i64> %a) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
; AVX1-LABEL: 'constant_shift_v8i64'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 52 for instruction: %shift = ashr <8 x i64> %a, <i64 1, i64 7, i64 15, i64 31, i64 1, i64 7, i64 15, i64 31>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %shift = ashr <8 x i64> %a, <i64 1, i64 7, i64 15, i64 31, i64 1, i64 7, i64 15, i64 31>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
; AVX2-LABEL: 'constant_shift_v8i64'
@@ -1246,7 +1246,7 @@ define <8 x i64> @constant_shift_v8i64(<8 x i64> %a) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
; BTVER2-LABEL: 'constant_shift_v8i64'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 52 for instruction: %shift = ashr <8 x i64> %a, <i64 1, i64 7, i64 15, i64 31, i64 1, i64 7, i64 15, i64 31>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %shift = ashr <8 x i64> %a, <i64 1, i64 7, i64 15, i64 31, i64 1, i64 7, i64 15, i64 31>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
%shift = ashr <8 x i64> %a, <i64 1, i64 7, i64 15, i64 31, i64 1, i64 7, i64 15, i64 31>
@@ -1263,7 +1263,7 @@ define <4 x i32> @constant_shift_v4i32(<4 x i32> %a) {
; 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>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %shift = ashr <4 x i32> %a, <i32 4, i32 5, i32 6, i32 7>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
;
; AVX2-LABEL: 'constant_shift_v4i32'
@@ -1279,7 +1279,7 @@ define <4 x i32> @constant_shift_v4i32(<4 x i32> %a) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
;
; BTVER2-LABEL: 'constant_shift_v4i32'
-; BTVER2-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>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %shift = ashr <4 x i32> %a, <i32 4, i32 5, i32 6, i32 7>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
;
%shift = ashr <4 x i32> %a, <i32 4, i32 5, i32 6, i32 7>
@@ -1296,7 +1296,7 @@ define <8 x i32> @constant_shift_v8i32(<8 x i32> %a) {
; 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>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %shift = ashr <8 x i32> %a, <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
;
; AVX2-LABEL: 'constant_shift_v8i32'
@@ -1316,7 +1316,7 @@ define <8 x i32> @constant_shift_v8i32(<8 x i32> %a) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
;
; BTVER2-LABEL: 'constant_shift_v8i32'
-; BTVER2-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>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %shift = ashr <8 x i32> %a, <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
;
%shift = ashr <8 x i32> %a, <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>
@@ -1333,7 +1333,7 @@ define <16 x i32> @constant_shift_v16i32(<16 x i32> %a) {
; 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>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 28 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>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
;
; AVX2-LABEL: 'constant_shift_v16i32'
@@ -1353,7 +1353,7 @@ define <16 x i32> @constant_shift_v16i32(<16 x i32> %a) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
;
; BTVER2-LABEL: 'constant_shift_v16i32'
-; BTVER2-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>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 28 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>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
;
%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>
@@ -1370,7 +1370,7 @@ define <8 x i16> @constant_shift_v8i16(<8 x i16> %a) {
; SSE42-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
;
; AVX1-LABEL: 'constant_shift_v8i16'
-; AVX1-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>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %shift = ashr <8 x i16> %a, <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
;
; AVX2-LABEL: 'constant_shift_v8i16'
@@ -1398,7 +1398,7 @@ define <8 x i16> @constant_shift_v8i16(<8 x i16> %a) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
;
; BTVER2-LABEL: 'constant_shift_v8i16'
-; BTVER2-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>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %shift = ashr <8 x i16> %a, <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
;
%shift = ashr <8 x i16> %a, <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>
@@ -1415,7 +1415,7 @@ define <16 x i16> @constant_shift_v16i16(<16 x i16> %a) {
; 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>
+; AVX1-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>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
;
; AVX2-LABEL: 'constant_shift_v16i16'
@@ -1443,7 +1443,7 @@ define <16 x i16> @constant_shift_v16i16(<16 x i16> %a) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
;
; BTVER2-LABEL: 'constant_shift_v16i16'
-; BTVER2-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>
+; BTVER2-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>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
;
%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>
@@ -1460,7 +1460,7 @@ define <32 x i16> @constant_shift_v32i16(<32 x i16> %a) {
; 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>
+; AVX1-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>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
;
; AVX2-LABEL: 'constant_shift_v32i16'
@@ -1488,7 +1488,7 @@ define <32 x i16> @constant_shift_v32i16(<32 x i16> %a) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
;
; BTVER2-LABEL: 'constant_shift_v32i16'
-; BTVER2-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>
+; BTVER2-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>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
;
%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>
@@ -1505,7 +1505,7 @@ define <16 x i8> @constant_shift_v16i8(<16 x i8> %a) {
; SSE42-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; AVX1-LABEL: 'constant_shift_v16i8'
-; AVX1-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>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 21 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>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; AVX2-LABEL: 'constant_shift_v16i8'
@@ -1533,7 +1533,7 @@ define <16 x i8> @constant_shift_v16i8(<16 x i8> %a) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; BTVER2-LABEL: 'constant_shift_v16i8'
-; BTVER2-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>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 21 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>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
%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>
@@ -1550,7 +1550,7 @@ define <32 x i8> @constant_shift_v32i8(<32 x i8> %a) {
; 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>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 44 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>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
; AVX2-LABEL: 'constant_shift_v32i8'
@@ -1578,7 +1578,7 @@ define <32 x i8> @constant_shift_v32i8(<32 x i8> %a) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
; BTVER2-LABEL: 'constant_shift_v32i8'
-; BTVER2-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>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 44 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>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
%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>
@@ -1595,7 +1595,7 @@ define <64 x i8> @constant_shift_v64i8(<64 x i8> %a) {
; 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>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 88 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>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
; AVX2-LABEL: 'constant_shift_v64i8'
@@ -1623,7 +1623,7 @@ define <64 x i8> @constant_shift_v64i8(<64 x i8> %a) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
; BTVER2-LABEL: 'constant_shift_v64i8'
-; BTVER2-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>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 88 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>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
%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/llvm/test/Analysis/CostModel/X86/vshift-ashr-cost.ll b/llvm/test/Analysis/CostModel/X86/vshift-ashr-cost.ll
index 8b8cbfb2197f1..51433745aa0bc 100644
--- a/llvm/test/Analysis/CostModel/X86/vshift-ashr-cost.ll
+++ b/llvm/test/Analysis/CostModel/X86/vshift-ashr-cost.ll
@@ -28,7 +28,7 @@ define <2 x i64> @var_shift_v2i64(<2 x i64> %a, <2 x i64> %b) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
; AVX1-LABEL: 'var_shift_v2i64'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = ashr <2 x i64> %a, %b
+; AVX1-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %shift = ashr <2 x i64> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
; AVX2-LABEL: 'var_shift_v2i64'
@@ -44,7 +44,7 @@ define <2 x i64> @var_shift_v2i64(<2 x i64> %a, <2 x i64> %b) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
; BTVER2-LABEL: 'var_shift_v2i64'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = ashr <2 x i64> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %shift = ashr <2 x i64> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
%shift = ashr <2 x i64> %a, %b
@@ -57,7 +57,7 @@ define <4 x i64> @var_shift_v4i64(<4 x i64> %a, <4 x i64> %b) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
; AVX1-LABEL: 'var_shift_v4i64'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 26 for instruction: %shift = ashr <4 x i64> %a, %b
+; AVX1-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = ashr <4 x i64> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
; AVX2-LABEL: 'var_shift_v4i64'
@@ -73,7 +73,7 @@ define <4 x i64> @var_shift_v4i64(<4 x i64> %a, <4 x i64> %b) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
; BTVER2-LABEL: 'var_shift_v4i64'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 26 for instruction: %shift = ashr <4 x i64> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = ashr <4 x i64> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
%shift = ashr <4 x i64> %a, %b
@@ -86,7 +86,7 @@ define <8 x i64> @var_shift_v8i64(<8 x i64> %a, <8 x i64> %b) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
; AVX1-LABEL: 'var_shift_v8i64'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 52 for instruction: %shift = ashr <8 x i64> %a, %b
+; AVX1-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %shift = ashr <8 x i64> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
; AVX2-LABEL: 'var_shift_v8i64'
@@ -102,7 +102,7 @@ define <8 x i64> @var_shift_v8i64(<8 x i64> %a, <8 x i64> %b) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
; BTVER2-LABEL: 'var_shift_v8i64'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 52 for instruction: %shift = ashr <8 x i64> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %shift = ashr <8 x i64> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
%shift = ashr <8 x i64> %a, %b
@@ -119,7 +119,7 @@ define <4 x i32> @var_shift_v4i32(<4 x i32> %a, <4 x i32> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %shift = ashr <4 x i32> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
;
; AVX2-LABEL: 'var_shift_v4i32'
@@ -135,7 +135,7 @@ define <4 x i32> @var_shift_v4i32(<4 x i32> %a, <4 x i32> %b) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
;
; BTVER2-LABEL: 'var_shift_v4i32'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = ashr <4 x i32> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %shift = ashr <4 x i32> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
;
%shift = ashr <4 x i32> %a, %b
@@ -152,7 +152,7 @@ define <8 x i32> @var_shift_v8i32(<8 x i32> %a, <8 x i32> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %shift = ashr <8 x i32> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
;
; AVX2-LABEL: 'var_shift_v8i32'
@@ -172,7 +172,7 @@ define <8 x i32> @var_shift_v8i32(<8 x i32> %a, <8 x i32> %b) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
;
; BTVER2-LABEL: 'var_shift_v8i32'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 26 for instruction: %shift = ashr <8 x i32> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %shift = ashr <8 x i32> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
;
%shift = ashr <8 x i32> %a, %b
@@ -189,7 +189,7 @@ define <16 x i32> @var_shift_v16i32(<16 x i32> %a, <16 x i32> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 28 for instruction: %shift = ashr <16 x i32> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
;
; AVX2-LABEL: 'var_shift_v16i32'
@@ -209,7 +209,7 @@ define <16 x i32> @var_shift_v16i32(<16 x i32> %a, <16 x i32> %b) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
;
; BTVER2-LABEL: 'var_shift_v16i32'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 52 for instruction: %shift = ashr <16 x i32> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 28 for instruction: %shift = ashr <16 x i32> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
;
%shift = ashr <16 x i32> %a, %b
@@ -226,7 +226,7 @@ define <8 x i16> @var_shift_v8i16(<8 x i16> %a, <8 x i16> %b) {
; SSE42-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
;
; AVX1-LABEL: 'var_shift_v8i16'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %shift = ashr <8 x i16> %a, %b
+; AVX1-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %shift = ashr <8 x i16> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
;
; AVX2-LABEL: 'var_shift_v8i16'
@@ -254,7 +254,7 @@ define <8 x i16> @var_shift_v8i16(<8 x i16> %a, <8 x i16> %b) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
;
; BTVER2-LABEL: 'var_shift_v8i16'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %shift = ashr <8 x i16> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %shift = ashr <8 x i16> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
;
%shift = ashr <8 x i16> %a, %b
@@ -271,7 +271,7 @@ define <16 x i16> @var_shift_v16i16(<16 x i16> %a, <16 x i16> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 28 for instruction: %shift = ashr <16 x i16> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
;
; AVX2-LABEL: 'var_shift_v16i16'
@@ -299,7 +299,7 @@ define <16 x i16> @var_shift_v16i16(<16 x i16> %a, <16 x i16> %b) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
;
; BTVER2-LABEL: 'var_shift_v16i16'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 30 for instruction: %shift = ashr <16 x i16> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 28 for instruction: %shift = ashr <16 x i16> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
;
%shift = ashr <16 x i16> %a, %b
@@ -316,7 +316,7 @@ define <32 x i16> @var_shift_v32i16(<32 x i16> %a, <32 x i16> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 56 for instruction: %shift = ashr <32 x i16> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
;
; AVX2-LABEL: 'var_shift_v32i16'
@@ -344,7 +344,7 @@ define <32 x i16> @var_shift_v32i16(<32 x i16> %a, <32 x i16> %b) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
;
; BTVER2-LABEL: 'var_shift_v32i16'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 60 for instruction: %shift = ashr <32 x i16> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 56 for instruction: %shift = ashr <32 x i16> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
;
%shift = ashr <32 x i16> %a, %b
@@ -361,7 +361,7 @@ define <16 x i8> @var_shift_v16i8(<16 x i8> %a, <16 x i8> %b) {
; SSE42-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; AVX1-LABEL: 'var_shift_v16i8'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %shift = ashr <16 x i8> %a, %b
+; AVX1-NEXT: Cost Model: Found an estimated cost of 21 for instruction: %shift = ashr <16 x i8> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; AVX2-LABEL: 'var_shift_v16i8'
@@ -389,7 +389,7 @@ define <16 x i8> @var_shift_v16i8(<16 x i8> %a, <16 x i8> %b) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; BTVER2-LABEL: 'var_shift_v16i8'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %shift = ashr <16 x i8> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 21 for instruction: %shift = ashr <16 x i8> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
%shift = ashr <16 x i8> %a, %b
@@ -406,7 +406,7 @@ define <32 x i8> @var_shift_v32i8(<32 x i8> %a, <32 x i8> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 44 for instruction: %shift = ashr <32 x i8> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
; AVX2-LABEL: 'var_shift_v32i8'
@@ -434,7 +434,7 @@ define <32 x i8> @var_shift_v32i8(<32 x i8> %a, <32 x i8> %b) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
; BTVER2-LABEL: 'var_shift_v32i8'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 50 for instruction: %shift = ashr <32 x i8> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 44 for instruction: %shift = ashr <32 x i8> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
%shift = ashr <32 x i8> %a, %b
@@ -451,7 +451,7 @@ define <64 x i8> @var_shift_v64i8(<64 x i8> %a, <64 x i8> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 88 for instruction: %shift = ashr <64 x i8> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
; AVX2-LABEL: 'var_shift_v64i8'
@@ -479,7 +479,7 @@ define <64 x i8> @var_shift_v64i8(<64 x i8> %a, <64 x i8> %b) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
; BTVER2-LABEL: 'var_shift_v64i8'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 100 for instruction: %shift = ashr <64 x i8> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 88 for instruction: %shift = ashr <64 x i8> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
%shift = ashr <64 x i8> %a, %b
@@ -965,7 +965,7 @@ define <16 x i8> @splatvar_shift_v16i8(<16 x i8> %a, i8 %b) {
; AVX1-LABEL: 'splatvar_shift_v16i8'
; AVX1-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <16 x i8> undef, i8 %b, i32 0
; AVX1-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
-; AVX1-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %shift = ashr <16 x i8> %a, %splat
+; AVX1-NEXT: Cost Model: Found an estimated cost of 21 for instruction: %shift = ashr <16 x i8> %a, %splat
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; AVX2-LABEL: 'splatvar_shift_v16i8'
@@ -1007,7 +1007,7 @@ define <16 x i8> @splatvar_shift_v16i8(<16 x i8> %a, i8 %b) {
; BTVER2-LABEL: 'splatvar_shift_v16i8'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <16 x i8> undef, i8 %b, i32 0
; BTVER2-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
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %shift = ashr <16 x i8> %a, %splat
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 21 for instruction: %shift = ashr <16 x i8> %a, %splat
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
%insert = insertelement <16 x i8> undef, i8 %b, i32 0
@@ -1032,7 +1032,7 @@ define <32 x i8> @splatvar_shift_v32i8(<32 x i8> %a, i8 %b) {
; 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
; AVX1-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %splat = shufflevector <32 x i8> %insert, <32 x i8> undef, <32 x i32> zeroinitializer
-; AVX1-NEXT: Cost Model: Found an estimated cost of 50 for instruction: %shift = ashr <32 x i8> %a, %splat
+; AVX1-NEXT: Cost Model: Found an estimated cost of 44 for instruction: %shift = ashr <32 x i8> %a, %splat
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
; AVX2-LABEL: 'splatvar_shift_v32i8'
@@ -1080,7 +1080,7 @@ define <32 x i8> @splatvar_shift_v32i8(<32 x i8> %a, i8 %b) {
; BTVER2-LABEL: 'splatvar_shift_v32i8'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <32 x i8> undef, i8 %b, i32 0
; BTVER2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %splat = shufflevector <32 x i8> %insert, <32 x i8> undef, <32 x i32> zeroinitializer
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 50 for instruction: %shift = ashr <32 x i8> %a, %splat
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 44 for instruction: %shift = ashr <32 x i8> %a, %splat
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
%insert = insertelement <32 x i8> undef, i8 %b, i32 0
@@ -1105,7 +1105,7 @@ define <64 x i8> @splatvar_shift_v64i8(<64 x i8> %a, i8 %b) {
; 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
; AVX1-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %splat = shufflevector <64 x i8> %insert, <64 x i8> undef, <64 x i32> zeroinitializer
-; AVX1-NEXT: Cost Model: Found an estimated cost of 100 for instruction: %shift = ashr <64 x i8> %a, %splat
+; AVX1-NEXT: Cost Model: Found an estimated cost of 88 for instruction: %shift = ashr <64 x i8> %a, %splat
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
; AVX2-LABEL: 'splatvar_shift_v64i8'
@@ -1153,7 +1153,7 @@ define <64 x i8> @splatvar_shift_v64i8(<64 x i8> %a, i8 %b) {
; BTVER2-LABEL: 'splatvar_shift_v64i8'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <64 x i8> undef, i8 %b, i32 0
; BTVER2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %splat = shufflevector <64 x i8> %insert, <64 x i8> undef, <64 x i32> zeroinitializer
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 100 for instruction: %shift = ashr <64 x i8> %a, %splat
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 88 for instruction: %shift = ashr <64 x i8> %a, %splat
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
%insert = insertelement <64 x i8> undef, i8 %b, i32 0
@@ -1172,7 +1172,7 @@ define <2 x i64> @constant_shift_v2i64(<2 x i64> %a) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
; AVX1-LABEL: 'constant_shift_v2i64'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = ashr <2 x i64> %a, <i64 1, i64 7>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %shift = ashr <2 x i64> %a, <i64 1, i64 7>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
; AVX2-LABEL: 'constant_shift_v2i64'
@@ -1188,7 +1188,7 @@ define <2 x i64> @constant_shift_v2i64(<2 x i64> %a) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
; BTVER2-LABEL: 'constant_shift_v2i64'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = ashr <2 x i64> %a, <i64 1, i64 7>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %shift = ashr <2 x i64> %a, <i64 1, i64 7>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
%shift = ashr <2 x i64> %a, <i64 1, i64 7>
@@ -1201,7 +1201,7 @@ define <4 x i64> @constant_shift_v4i64(<4 x i64> %a) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
; AVX1-LABEL: 'constant_shift_v4i64'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 26 for instruction: %shift = ashr <4 x i64> %a, <i64 1, i64 7, i64 15, i64 31>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = ashr <4 x i64> %a, <i64 1, i64 7, i64 15, i64 31>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
; AVX2-LABEL: 'constant_shift_v4i64'
@@ -1217,7 +1217,7 @@ define <4 x i64> @constant_shift_v4i64(<4 x i64> %a) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
; BTVER2-LABEL: 'constant_shift_v4i64'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 26 for instruction: %shift = ashr <4 x i64> %a, <i64 1, i64 7, i64 15, i64 31>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = ashr <4 x i64> %a, <i64 1, i64 7, i64 15, i64 31>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
%shift = ashr <4 x i64> %a, <i64 1, i64 7, i64 15, i64 31>
@@ -1230,7 +1230,7 @@ define <8 x i64> @constant_shift_v8i64(<8 x i64> %a) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
; AVX1-LABEL: 'constant_shift_v8i64'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 52 for instruction: %shift = ashr <8 x i64> %a, <i64 1, i64 7, i64 15, i64 31, i64 1, i64 7, i64 15, i64 31>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %shift = ashr <8 x i64> %a, <i64 1, i64 7, i64 15, i64 31, i64 1, i64 7, i64 15, i64 31>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
; AVX2-LABEL: 'constant_shift_v8i64'
@@ -1246,7 +1246,7 @@ define <8 x i64> @constant_shift_v8i64(<8 x i64> %a) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
; BTVER2-LABEL: 'constant_shift_v8i64'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 52 for instruction: %shift = ashr <8 x i64> %a, <i64 1, i64 7, i64 15, i64 31, i64 1, i64 7, i64 15, i64 31>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %shift = ashr <8 x i64> %a, <i64 1, i64 7, i64 15, i64 31, i64 1, i64 7, i64 15, i64 31>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
%shift = ashr <8 x i64> %a, <i64 1, i64 7, i64 15, i64 31, i64 1, i64 7, i64 15, i64 31>
@@ -1263,7 +1263,7 @@ define <4 x i32> @constant_shift_v4i32(<4 x i32> %a) {
; 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>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %shift = ashr <4 x i32> %a, <i32 4, i32 5, i32 6, i32 7>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
;
; AVX2-LABEL: 'constant_shift_v4i32'
@@ -1279,7 +1279,7 @@ define <4 x i32> @constant_shift_v4i32(<4 x i32> %a) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
;
; BTVER2-LABEL: 'constant_shift_v4i32'
-; BTVER2-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>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %shift = ashr <4 x i32> %a, <i32 4, i32 5, i32 6, i32 7>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
;
%shift = ashr <4 x i32> %a, <i32 4, i32 5, i32 6, i32 7>
@@ -1296,7 +1296,7 @@ define <8 x i32> @constant_shift_v8i32(<8 x i32> %a) {
; 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>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %shift = ashr <8 x i32> %a, <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
;
; AVX2-LABEL: 'constant_shift_v8i32'
@@ -1316,7 +1316,7 @@ define <8 x i32> @constant_shift_v8i32(<8 x i32> %a) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
;
; BTVER2-LABEL: 'constant_shift_v8i32'
-; BTVER2-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>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %shift = ashr <8 x i32> %a, <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
;
%shift = ashr <8 x i32> %a, <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>
@@ -1333,7 +1333,7 @@ define <16 x i32> @constant_shift_v16i32(<16 x i32> %a) {
; 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>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 28 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>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
;
; AVX2-LABEL: 'constant_shift_v16i32'
@@ -1353,7 +1353,7 @@ define <16 x i32> @constant_shift_v16i32(<16 x i32> %a) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
;
; BTVER2-LABEL: 'constant_shift_v16i32'
-; BTVER2-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>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 28 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>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
;
%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>
@@ -1370,7 +1370,7 @@ define <8 x i16> @constant_shift_v8i16(<8 x i16> %a) {
; SSE42-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
;
; AVX1-LABEL: 'constant_shift_v8i16'
-; AVX1-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>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %shift = ashr <8 x i16> %a, <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
;
; AVX2-LABEL: 'constant_shift_v8i16'
@@ -1398,7 +1398,7 @@ define <8 x i16> @constant_shift_v8i16(<8 x i16> %a) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
;
; BTVER2-LABEL: 'constant_shift_v8i16'
-; BTVER2-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>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %shift = ashr <8 x i16> %a, <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
;
%shift = ashr <8 x i16> %a, <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>
@@ -1415,7 +1415,7 @@ define <16 x i16> @constant_shift_v16i16(<16 x i16> %a) {
; 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>
+; AVX1-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>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
;
; AVX2-LABEL: 'constant_shift_v16i16'
@@ -1443,7 +1443,7 @@ define <16 x i16> @constant_shift_v16i16(<16 x i16> %a) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
;
; BTVER2-LABEL: 'constant_shift_v16i16'
-; BTVER2-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>
+; BTVER2-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>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
;
%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>
@@ -1460,7 +1460,7 @@ define <32 x i16> @constant_shift_v32i16(<32 x i16> %a) {
; 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>
+; AVX1-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>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
;
; AVX2-LABEL: 'constant_shift_v32i16'
@@ -1488,7 +1488,7 @@ define <32 x i16> @constant_shift_v32i16(<32 x i16> %a) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
;
; BTVER2-LABEL: 'constant_shift_v32i16'
-; BTVER2-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>
+; BTVER2-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>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
;
%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>
@@ -1505,7 +1505,7 @@ define <16 x i8> @constant_shift_v16i8(<16 x i8> %a) {
; SSE42-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; AVX1-LABEL: 'constant_shift_v16i8'
-; AVX1-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>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 21 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>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; AVX2-LABEL: 'constant_shift_v16i8'
@@ -1533,7 +1533,7 @@ define <16 x i8> @constant_shift_v16i8(<16 x i8> %a) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; BTVER2-LABEL: 'constant_shift_v16i8'
-; BTVER2-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>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 21 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>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
%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>
@@ -1550,7 +1550,7 @@ define <32 x i8> @constant_shift_v32i8(<32 x i8> %a) {
; 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>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 44 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>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
; AVX2-LABEL: 'constant_shift_v32i8'
@@ -1578,7 +1578,7 @@ define <32 x i8> @constant_shift_v32i8(<32 x i8> %a) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
; BTVER2-LABEL: 'constant_shift_v32i8'
-; BTVER2-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>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 44 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>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
%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>
@@ -1595,7 +1595,7 @@ define <64 x i8> @constant_shift_v64i8(<64 x i8> %a) {
; 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>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 88 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>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
; AVX2-LABEL: 'constant_shift_v64i8'
@@ -1623,7 +1623,7 @@ define <64 x i8> @constant_shift_v64i8(<64 x i8> %a) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
; BTVER2-LABEL: 'constant_shift_v64i8'
-; BTVER2-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>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 88 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>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
%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/llvm/test/Analysis/CostModel/X86/vshift-lshr-cost-inseltpoison.ll b/llvm/test/Analysis/CostModel/X86/vshift-lshr-cost-inseltpoison.ll
index 58e7a96abc2b3..df26a345c17ca 100644
--- a/llvm/test/Analysis/CostModel/X86/vshift-lshr-cost-inseltpoison.ll
+++ b/llvm/test/Analysis/CostModel/X86/vshift-lshr-cost-inseltpoison.ll
@@ -28,7 +28,7 @@ define <2 x i64> @var_shift_v2i64(<2 x i64> %a, <2 x i64> %b) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
; AVX1-LABEL: 'var_shift_v2i64'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %shift = lshr <2 x i64> %a, %b
+; AVX1-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %shift = lshr <2 x i64> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
; AVX2-LABEL: 'var_shift_v2i64'
@@ -48,7 +48,7 @@ define <2 x i64> @var_shift_v2i64(<2 x i64> %a, <2 x i64> %b) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
; BTVER2-LABEL: 'var_shift_v2i64'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %shift = lshr <2 x i64> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %shift = lshr <2 x i64> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
%shift = lshr <2 x i64> %a, %b
@@ -61,7 +61,7 @@ define <4 x i64> @var_shift_v4i64(<4 x i64> %a, <4 x i64> %b) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
; AVX1-LABEL: 'var_shift_v4i64'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %shift = lshr <4 x i64> %a, %b
+; AVX1-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %shift = lshr <4 x i64> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
; AVX2-LABEL: 'var_shift_v4i64'
@@ -81,7 +81,7 @@ define <4 x i64> @var_shift_v4i64(<4 x i64> %a, <4 x i64> %b) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
; BTVER2-LABEL: 'var_shift_v4i64'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %shift = lshr <4 x i64> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %shift = lshr <4 x i64> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
%shift = lshr <4 x i64> %a, %b
@@ -94,7 +94,7 @@ define <8 x i64> @var_shift_v8i64(<8 x i64> %a, <8 x i64> %b) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
; AVX1-LABEL: 'var_shift_v8i64'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %shift = lshr <8 x i64> %a, %b
+; AVX1-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = lshr <8 x i64> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
; AVX2-LABEL: 'var_shift_v8i64'
@@ -114,7 +114,7 @@ define <8 x i64> @var_shift_v8i64(<8 x i64> %a, <8 x i64> %b) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
; BTVER2-LABEL: 'var_shift_v8i64'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %shift = lshr <8 x i64> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = lshr <8 x i64> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
%shift = lshr <8 x i64> %a, %b
@@ -131,7 +131,7 @@ define <4 x i32> @var_shift_v4i32(<4 x i32> %a, <4 x i32> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %shift = lshr <4 x i32> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
;
; AVX2-LABEL: 'var_shift_v4i32'
@@ -147,7 +147,7 @@ define <4 x i32> @var_shift_v4i32(<4 x i32> %a, <4 x i32> %b) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
;
; BTVER2-LABEL: 'var_shift_v4i32'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %shift = lshr <4 x i32> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %shift = lshr <4 x i32> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
;
%shift = lshr <4 x i32> %a, %b
@@ -164,7 +164,7 @@ define <8 x i32> @var_shift_v8i32(<8 x i32> %a, <8 x i32> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %shift = lshr <8 x i32> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
;
; AVX2-LABEL: 'var_shift_v8i32'
@@ -184,7 +184,7 @@ define <8 x i32> @var_shift_v8i32(<8 x i32> %a, <8 x i32> %b) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
;
; BTVER2-LABEL: 'var_shift_v8i32'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %shift = lshr <8 x i32> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %shift = lshr <8 x i32> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
;
%shift = lshr <8 x i32> %a, %b
@@ -201,7 +201,7 @@ define <16 x i32> @var_shift_v16i32(<16 x i32> %a, <16 x i32> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 28 for instruction: %shift = lshr <16 x i32> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
;
; AVX2-LABEL: 'var_shift_v16i32'
@@ -221,7 +221,7 @@ define <16 x i32> @var_shift_v16i32(<16 x i32> %a, <16 x i32> %b) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
;
; BTVER2-LABEL: 'var_shift_v16i32'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 48 for instruction: %shift = lshr <16 x i32> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 28 for instruction: %shift = lshr <16 x i32> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
;
%shift = lshr <16 x i32> %a, %b
@@ -238,7 +238,7 @@ define <8 x i16> @var_shift_v8i16(<8 x i16> %a, <8 x i16> %b) {
; SSE42-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
;
; AVX1-LABEL: 'var_shift_v8i16'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %shift = lshr <8 x i16> %a, %b
+; AVX1-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %shift = lshr <8 x i16> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
;
; AVX2-LABEL: 'var_shift_v8i16'
@@ -266,7 +266,7 @@ define <8 x i16> @var_shift_v8i16(<8 x i16> %a, <8 x i16> %b) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
;
; BTVER2-LABEL: 'var_shift_v8i16'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %shift = lshr <8 x i16> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %shift = lshr <8 x i16> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
;
%shift = lshr <8 x i16> %a, %b
@@ -283,7 +283,7 @@ define <16 x i16> @var_shift_v16i16(<16 x i16> %a, <16 x i16> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 28 for instruction: %shift = lshr <16 x i16> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
;
; AVX2-LABEL: 'var_shift_v16i16'
@@ -311,7 +311,7 @@ define <16 x i16> @var_shift_v16i16(<16 x i16> %a, <16 x i16> %b) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
;
; BTVER2-LABEL: 'var_shift_v16i16'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 30 for instruction: %shift = lshr <16 x i16> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 28 for instruction: %shift = lshr <16 x i16> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
;
%shift = lshr <16 x i16> %a, %b
@@ -328,7 +328,7 @@ define <32 x i16> @var_shift_v32i16(<32 x i16> %a, <32 x i16> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 56 for instruction: %shift = lshr <32 x i16> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
;
; AVX2-LABEL: 'var_shift_v32i16'
@@ -356,7 +356,7 @@ define <32 x i16> @var_shift_v32i16(<32 x i16> %a, <32 x i16> %b) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
;
; BTVER2-LABEL: 'var_shift_v32i16'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 60 for instruction: %shift = lshr <32 x i16> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 56 for instruction: %shift = lshr <32 x i16> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
;
%shift = lshr <32 x i16> %a, %b
@@ -373,7 +373,7 @@ define <16 x i8> @var_shift_v16i8(<16 x i8> %a, <16 x i8> %b) {
; SSE42-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; AVX1-LABEL: 'var_shift_v16i8'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = lshr <16 x i8> %a, %b
+; AVX1-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %shift = lshr <16 x i8> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; AVX2-LABEL: 'var_shift_v16i8'
@@ -401,7 +401,7 @@ define <16 x i8> @var_shift_v16i8(<16 x i8> %a, <16 x i8> %b) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; BTVER2-LABEL: 'var_shift_v16i8'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = lshr <16 x i8> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %shift = lshr <16 x i8> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
%shift = lshr <16 x i8> %a, %b
@@ -418,7 +418,7 @@ define <32 x i8> @var_shift_v32i8(<32 x i8> %a, <32 x i8> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 23 for instruction: %shift = lshr <32 x i8> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
; AVX2-LABEL: 'var_shift_v32i8'
@@ -446,7 +446,7 @@ define <32 x i8> @var_shift_v32i8(<32 x i8> %a, <32 x i8> %b) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
; BTVER2-LABEL: 'var_shift_v32i8'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 26 for instruction: %shift = lshr <32 x i8> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 23 for instruction: %shift = lshr <32 x i8> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
%shift = lshr <32 x i8> %a, %b
@@ -463,7 +463,7 @@ define <64 x i8> @var_shift_v64i8(<64 x i8> %a, <64 x i8> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 46 for instruction: %shift = lshr <64 x i8> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
; AVX2-LABEL: 'var_shift_v64i8'
@@ -491,7 +491,7 @@ define <64 x i8> @var_shift_v64i8(<64 x i8> %a, <64 x i8> %b) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
; BTVER2-LABEL: 'var_shift_v64i8'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 52 for instruction: %shift = lshr <64 x i8> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 46 for instruction: %shift = lshr <64 x i8> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
%shift = lshr <64 x i8> %a, %b
@@ -977,7 +977,7 @@ define <16 x i8> @splatvar_shift_v16i8(<16 x i8> %a, i8 %b) {
; AVX1-LABEL: 'splatvar_shift_v16i8'
; AVX1-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <16 x i8> poison, i8 %b, i32 0
; AVX1-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <16 x i8> %insert, <16 x i8> poison, <16 x i32> zeroinitializer
-; AVX1-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = lshr <16 x i8> %a, %splat
+; AVX1-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %shift = lshr <16 x i8> %a, %splat
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; AVX2-LABEL: 'splatvar_shift_v16i8'
@@ -1019,7 +1019,7 @@ define <16 x i8> @splatvar_shift_v16i8(<16 x i8> %a, i8 %b) {
; BTVER2-LABEL: 'splatvar_shift_v16i8'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <16 x i8> poison, i8 %b, i32 0
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <16 x i8> %insert, <16 x i8> poison, <16 x i32> zeroinitializer
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = lshr <16 x i8> %a, %splat
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %shift = lshr <16 x i8> %a, %splat
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
%insert = insertelement <16 x i8> poison, i8 %b, i32 0
@@ -1044,7 +1044,7 @@ define <32 x i8> @splatvar_shift_v32i8(<32 x i8> %a, i8 %b) {
; AVX1-LABEL: 'splatvar_shift_v32i8'
; AVX1-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <32 x i8> poison, i8 %b, i32 0
; AVX1-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %splat = shufflevector <32 x i8> %insert, <32 x i8> poison, <32 x i32> zeroinitializer
-; AVX1-NEXT: Cost Model: Found an estimated cost of 26 for instruction: %shift = lshr <32 x i8> %a, %splat
+; AVX1-NEXT: Cost Model: Found an estimated cost of 23 for instruction: %shift = lshr <32 x i8> %a, %splat
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
; AVX2-LABEL: 'splatvar_shift_v32i8'
@@ -1092,7 +1092,7 @@ define <32 x i8> @splatvar_shift_v32i8(<32 x i8> %a, i8 %b) {
; BTVER2-LABEL: 'splatvar_shift_v32i8'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <32 x i8> poison, i8 %b, i32 0
; BTVER2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %splat = shufflevector <32 x i8> %insert, <32 x i8> poison, <32 x i32> zeroinitializer
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 26 for instruction: %shift = lshr <32 x i8> %a, %splat
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 23 for instruction: %shift = lshr <32 x i8> %a, %splat
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
%insert = insertelement <32 x i8> poison, i8 %b, i32 0
@@ -1117,7 +1117,7 @@ define <64 x i8> @splatvar_shift_v64i8(<64 x i8> %a, i8 %b) {
; AVX1-LABEL: 'splatvar_shift_v64i8'
; AVX1-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <64 x i8> poison, i8 %b, i32 0
; AVX1-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %splat = shufflevector <64 x i8> %insert, <64 x i8> poison, <64 x i32> zeroinitializer
-; AVX1-NEXT: Cost Model: Found an estimated cost of 52 for instruction: %shift = lshr <64 x i8> %a, %splat
+; AVX1-NEXT: Cost Model: Found an estimated cost of 46 for instruction: %shift = lshr <64 x i8> %a, %splat
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
; AVX2-LABEL: 'splatvar_shift_v64i8'
@@ -1165,7 +1165,7 @@ define <64 x i8> @splatvar_shift_v64i8(<64 x i8> %a, i8 %b) {
; BTVER2-LABEL: 'splatvar_shift_v64i8'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <64 x i8> poison, i8 %b, i32 0
; BTVER2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %splat = shufflevector <64 x i8> %insert, <64 x i8> poison, <64 x i32> zeroinitializer
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 52 for instruction: %shift = lshr <64 x i8> %a, %splat
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 46 for instruction: %shift = lshr <64 x i8> %a, %splat
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
%insert = insertelement <64 x i8> poison, i8 %b, i32 0
@@ -1184,7 +1184,7 @@ define <2 x i64> @constant_shift_v2i64(<2 x i64> %a) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
; AVX1-LABEL: 'constant_shift_v2i64'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %shift = lshr <2 x i64> %a, <i64 1, i64 7>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %shift = lshr <2 x i64> %a, <i64 1, i64 7>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
; AVX2-LABEL: 'constant_shift_v2i64'
@@ -1200,7 +1200,7 @@ define <2 x i64> @constant_shift_v2i64(<2 x i64> %a) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
; BTVER2-LABEL: 'constant_shift_v2i64'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %shift = lshr <2 x i64> %a, <i64 1, i64 7>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %shift = lshr <2 x i64> %a, <i64 1, i64 7>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
%shift = lshr <2 x i64> %a, <i64 1, i64 7>
@@ -1213,7 +1213,7 @@ define <4 x i64> @constant_shift_v4i64(<4 x i64> %a) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
; AVX1-LABEL: 'constant_shift_v4i64'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %shift = lshr <4 x i64> %a, <i64 1, i64 7, i64 15, i64 31>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %shift = lshr <4 x i64> %a, <i64 1, i64 7, i64 15, i64 31>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
; AVX2-LABEL: 'constant_shift_v4i64'
@@ -1233,7 +1233,7 @@ define <4 x i64> @constant_shift_v4i64(<4 x i64> %a) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
; BTVER2-LABEL: 'constant_shift_v4i64'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %shift = lshr <4 x i64> %a, <i64 1, i64 7, i64 15, i64 31>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %shift = lshr <4 x i64> %a, <i64 1, i64 7, i64 15, i64 31>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
%shift = lshr <4 x i64> %a, <i64 1, i64 7, i64 15, i64 31>
@@ -1246,7 +1246,7 @@ define <8 x i64> @constant_shift_v8i64(<8 x i64> %a) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
; AVX1-LABEL: 'constant_shift_v8i64'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %shift = lshr <8 x i64> %a, <i64 1, i64 7, i64 15, i64 31, i64 1, i64 7, i64 15, i64 31>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = lshr <8 x i64> %a, <i64 1, i64 7, i64 15, i64 31, i64 1, i64 7, i64 15, i64 31>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
; AVX2-LABEL: 'constant_shift_v8i64'
@@ -1266,7 +1266,7 @@ define <8 x i64> @constant_shift_v8i64(<8 x i64> %a) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
; BTVER2-LABEL: 'constant_shift_v8i64'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %shift = lshr <8 x i64> %a, <i64 1, i64 7, i64 15, i64 31, i64 1, i64 7, i64 15, i64 31>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = lshr <8 x i64> %a, <i64 1, i64 7, i64 15, i64 31, i64 1, i64 7, i64 15, i64 31>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
%shift = lshr <8 x i64> %a, <i64 1, i64 7, i64 15, i64 31, i64 1, i64 7, i64 15, i64 31>
@@ -1283,7 +1283,7 @@ define <4 x i32> @constant_shift_v4i32(<4 x i32> %a) {
; 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>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %shift = lshr <4 x i32> %a, <i32 4, i32 5, i32 6, i32 7>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
;
; AVX2-LABEL: 'constant_shift_v4i32'
@@ -1299,7 +1299,7 @@ define <4 x i32> @constant_shift_v4i32(<4 x i32> %a) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
;
; BTVER2-LABEL: 'constant_shift_v4i32'
-; BTVER2-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>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %shift = lshr <4 x i32> %a, <i32 4, i32 5, i32 6, i32 7>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
;
%shift = lshr <4 x i32> %a, <i32 4, i32 5, i32 6, i32 7>
@@ -1316,7 +1316,7 @@ define <8 x i32> @constant_shift_v8i32(<8 x i32> %a) {
; 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>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %shift = lshr <8 x i32> %a, <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
;
; AVX2-LABEL: 'constant_shift_v8i32'
@@ -1336,7 +1336,7 @@ define <8 x i32> @constant_shift_v8i32(<8 x i32> %a) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
;
; BTVER2-LABEL: 'constant_shift_v8i32'
-; BTVER2-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>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %shift = lshr <8 x i32> %a, <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
;
%shift = lshr <8 x i32> %a, <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>
@@ -1353,7 +1353,7 @@ define <16 x i32> @constant_shift_v16i32(<16 x i32> %a) {
; 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>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 28 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>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
;
; AVX2-LABEL: 'constant_shift_v16i32'
@@ -1373,7 +1373,7 @@ define <16 x i32> @constant_shift_v16i32(<16 x i32> %a) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
;
; BTVER2-LABEL: 'constant_shift_v16i32'
-; BTVER2-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>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 28 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>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
;
%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>
@@ -1390,7 +1390,7 @@ define <8 x i16> @constant_shift_v8i16(<8 x i16> %a) {
; SSE42-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
;
; AVX1-LABEL: 'constant_shift_v8i16'
-; AVX1-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>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %shift = lshr <8 x i16> %a, <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
;
; AVX2-LABEL: 'constant_shift_v8i16'
@@ -1418,7 +1418,7 @@ define <8 x i16> @constant_shift_v8i16(<8 x i16> %a) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
;
; BTVER2-LABEL: 'constant_shift_v8i16'
-; BTVER2-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>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %shift = lshr <8 x i16> %a, <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
;
%shift = lshr <8 x i16> %a, <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>
@@ -1435,7 +1435,7 @@ define <16 x i16> @constant_shift_v16i16(<16 x i16> %a) {
; 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>
+; AVX1-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>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
;
; AVX2-LABEL: 'constant_shift_v16i16'
@@ -1463,7 +1463,7 @@ define <16 x i16> @constant_shift_v16i16(<16 x i16> %a) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
;
; BTVER2-LABEL: 'constant_shift_v16i16'
-; BTVER2-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>
+; BTVER2-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>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
;
%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>
@@ -1480,7 +1480,7 @@ define <32 x i16> @constant_shift_v32i16(<32 x i16> %a) {
; 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>
+; AVX1-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>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
;
; AVX2-LABEL: 'constant_shift_v32i16'
@@ -1508,7 +1508,7 @@ define <32 x i16> @constant_shift_v32i16(<32 x i16> %a) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
;
; BTVER2-LABEL: 'constant_shift_v32i16'
-; BTVER2-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>
+; BTVER2-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>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
;
%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>
@@ -1525,7 +1525,7 @@ define <16 x i8> @constant_shift_v16i8(<16 x i8> %a) {
; SSE42-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; AVX1-LABEL: 'constant_shift_v16i8'
-; AVX1-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>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 11 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>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; AVX2-LABEL: 'constant_shift_v16i8'
@@ -1553,7 +1553,7 @@ define <16 x i8> @constant_shift_v16i8(<16 x i8> %a) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; BTVER2-LABEL: 'constant_shift_v16i8'
-; BTVER2-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>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 11 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>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
%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>
@@ -1570,7 +1570,7 @@ define <32 x i8> @constant_shift_v32i8(<32 x i8> %a) {
; 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>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 23 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>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
; AVX2-LABEL: 'constant_shift_v32i8'
@@ -1598,7 +1598,7 @@ define <32 x i8> @constant_shift_v32i8(<32 x i8> %a) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
; BTVER2-LABEL: 'constant_shift_v32i8'
-; BTVER2-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>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 23 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>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
%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>
@@ -1615,7 +1615,7 @@ define <64 x i8> @constant_shift_v64i8(<64 x i8> %a) {
; 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>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 46 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>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
; AVX2-LABEL: 'constant_shift_v64i8'
@@ -1643,7 +1643,7 @@ define <64 x i8> @constant_shift_v64i8(<64 x i8> %a) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
; BTVER2-LABEL: 'constant_shift_v64i8'
-; BTVER2-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>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 46 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>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
%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/llvm/test/Analysis/CostModel/X86/vshift-lshr-cost.ll b/llvm/test/Analysis/CostModel/X86/vshift-lshr-cost.ll
index 0cf3f4b41aced..2dcf071a7ebb5 100644
--- a/llvm/test/Analysis/CostModel/X86/vshift-lshr-cost.ll
+++ b/llvm/test/Analysis/CostModel/X86/vshift-lshr-cost.ll
@@ -28,7 +28,7 @@ define <2 x i64> @var_shift_v2i64(<2 x i64> %a, <2 x i64> %b) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
; AVX1-LABEL: 'var_shift_v2i64'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %shift = lshr <2 x i64> %a, %b
+; AVX1-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %shift = lshr <2 x i64> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
; AVX2-LABEL: 'var_shift_v2i64'
@@ -48,7 +48,7 @@ define <2 x i64> @var_shift_v2i64(<2 x i64> %a, <2 x i64> %b) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
; BTVER2-LABEL: 'var_shift_v2i64'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %shift = lshr <2 x i64> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %shift = lshr <2 x i64> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
%shift = lshr <2 x i64> %a, %b
@@ -61,7 +61,7 @@ define <4 x i64> @var_shift_v4i64(<4 x i64> %a, <4 x i64> %b) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
; AVX1-LABEL: 'var_shift_v4i64'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %shift = lshr <4 x i64> %a, %b
+; AVX1-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %shift = lshr <4 x i64> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
; AVX2-LABEL: 'var_shift_v4i64'
@@ -81,7 +81,7 @@ define <4 x i64> @var_shift_v4i64(<4 x i64> %a, <4 x i64> %b) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
; BTVER2-LABEL: 'var_shift_v4i64'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %shift = lshr <4 x i64> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %shift = lshr <4 x i64> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
%shift = lshr <4 x i64> %a, %b
@@ -94,7 +94,7 @@ define <8 x i64> @var_shift_v8i64(<8 x i64> %a, <8 x i64> %b) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
; AVX1-LABEL: 'var_shift_v8i64'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %shift = lshr <8 x i64> %a, %b
+; AVX1-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = lshr <8 x i64> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
; AVX2-LABEL: 'var_shift_v8i64'
@@ -114,7 +114,7 @@ define <8 x i64> @var_shift_v8i64(<8 x i64> %a, <8 x i64> %b) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
; BTVER2-LABEL: 'var_shift_v8i64'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %shift = lshr <8 x i64> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = lshr <8 x i64> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
%shift = lshr <8 x i64> %a, %b
@@ -131,7 +131,7 @@ define <4 x i32> @var_shift_v4i32(<4 x i32> %a, <4 x i32> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %shift = lshr <4 x i32> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
;
; AVX2-LABEL: 'var_shift_v4i32'
@@ -147,7 +147,7 @@ define <4 x i32> @var_shift_v4i32(<4 x i32> %a, <4 x i32> %b) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
;
; BTVER2-LABEL: 'var_shift_v4i32'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %shift = lshr <4 x i32> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %shift = lshr <4 x i32> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
;
%shift = lshr <4 x i32> %a, %b
@@ -164,7 +164,7 @@ define <8 x i32> @var_shift_v8i32(<8 x i32> %a, <8 x i32> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %shift = lshr <8 x i32> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
;
; AVX2-LABEL: 'var_shift_v8i32'
@@ -184,7 +184,7 @@ define <8 x i32> @var_shift_v8i32(<8 x i32> %a, <8 x i32> %b) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
;
; BTVER2-LABEL: 'var_shift_v8i32'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %shift = lshr <8 x i32> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %shift = lshr <8 x i32> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
;
%shift = lshr <8 x i32> %a, %b
@@ -201,7 +201,7 @@ define <16 x i32> @var_shift_v16i32(<16 x i32> %a, <16 x i32> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 28 for instruction: %shift = lshr <16 x i32> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
;
; AVX2-LABEL: 'var_shift_v16i32'
@@ -221,7 +221,7 @@ define <16 x i32> @var_shift_v16i32(<16 x i32> %a, <16 x i32> %b) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
;
; BTVER2-LABEL: 'var_shift_v16i32'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 48 for instruction: %shift = lshr <16 x i32> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 28 for instruction: %shift = lshr <16 x i32> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
;
%shift = lshr <16 x i32> %a, %b
@@ -238,7 +238,7 @@ define <8 x i16> @var_shift_v8i16(<8 x i16> %a, <8 x i16> %b) {
; SSE42-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
;
; AVX1-LABEL: 'var_shift_v8i16'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %shift = lshr <8 x i16> %a, %b
+; AVX1-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %shift = lshr <8 x i16> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
;
; AVX2-LABEL: 'var_shift_v8i16'
@@ -266,7 +266,7 @@ define <8 x i16> @var_shift_v8i16(<8 x i16> %a, <8 x i16> %b) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
;
; BTVER2-LABEL: 'var_shift_v8i16'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %shift = lshr <8 x i16> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %shift = lshr <8 x i16> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
;
%shift = lshr <8 x i16> %a, %b
@@ -283,7 +283,7 @@ define <16 x i16> @var_shift_v16i16(<16 x i16> %a, <16 x i16> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 28 for instruction: %shift = lshr <16 x i16> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
;
; AVX2-LABEL: 'var_shift_v16i16'
@@ -311,7 +311,7 @@ define <16 x i16> @var_shift_v16i16(<16 x i16> %a, <16 x i16> %b) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
;
; BTVER2-LABEL: 'var_shift_v16i16'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 30 for instruction: %shift = lshr <16 x i16> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 28 for instruction: %shift = lshr <16 x i16> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
;
%shift = lshr <16 x i16> %a, %b
@@ -328,7 +328,7 @@ define <32 x i16> @var_shift_v32i16(<32 x i16> %a, <32 x i16> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 56 for instruction: %shift = lshr <32 x i16> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
;
; AVX2-LABEL: 'var_shift_v32i16'
@@ -356,7 +356,7 @@ define <32 x i16> @var_shift_v32i16(<32 x i16> %a, <32 x i16> %b) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
;
; BTVER2-LABEL: 'var_shift_v32i16'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 60 for instruction: %shift = lshr <32 x i16> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 56 for instruction: %shift = lshr <32 x i16> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
;
%shift = lshr <32 x i16> %a, %b
@@ -373,7 +373,7 @@ define <16 x i8> @var_shift_v16i8(<16 x i8> %a, <16 x i8> %b) {
; SSE42-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; AVX1-LABEL: 'var_shift_v16i8'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = lshr <16 x i8> %a, %b
+; AVX1-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %shift = lshr <16 x i8> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; AVX2-LABEL: 'var_shift_v16i8'
@@ -401,7 +401,7 @@ define <16 x i8> @var_shift_v16i8(<16 x i8> %a, <16 x i8> %b) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; BTVER2-LABEL: 'var_shift_v16i8'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = lshr <16 x i8> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %shift = lshr <16 x i8> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
%shift = lshr <16 x i8> %a, %b
@@ -418,7 +418,7 @@ define <32 x i8> @var_shift_v32i8(<32 x i8> %a, <32 x i8> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 23 for instruction: %shift = lshr <32 x i8> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
; AVX2-LABEL: 'var_shift_v32i8'
@@ -446,7 +446,7 @@ define <32 x i8> @var_shift_v32i8(<32 x i8> %a, <32 x i8> %b) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
; BTVER2-LABEL: 'var_shift_v32i8'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 26 for instruction: %shift = lshr <32 x i8> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 23 for instruction: %shift = lshr <32 x i8> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
%shift = lshr <32 x i8> %a, %b
@@ -463,7 +463,7 @@ define <64 x i8> @var_shift_v64i8(<64 x i8> %a, <64 x i8> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 46 for instruction: %shift = lshr <64 x i8> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
; AVX2-LABEL: 'var_shift_v64i8'
@@ -491,7 +491,7 @@ define <64 x i8> @var_shift_v64i8(<64 x i8> %a, <64 x i8> %b) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
; BTVER2-LABEL: 'var_shift_v64i8'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 52 for instruction: %shift = lshr <64 x i8> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 46 for instruction: %shift = lshr <64 x i8> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
%shift = lshr <64 x i8> %a, %b
@@ -977,7 +977,7 @@ define <16 x i8> @splatvar_shift_v16i8(<16 x i8> %a, i8 %b) {
; AVX1-LABEL: 'splatvar_shift_v16i8'
; AVX1-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <16 x i8> undef, i8 %b, i32 0
; AVX1-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
-; AVX1-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = lshr <16 x i8> %a, %splat
+; AVX1-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %shift = lshr <16 x i8> %a, %splat
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; AVX2-LABEL: 'splatvar_shift_v16i8'
@@ -1019,7 +1019,7 @@ define <16 x i8> @splatvar_shift_v16i8(<16 x i8> %a, i8 %b) {
; BTVER2-LABEL: 'splatvar_shift_v16i8'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <16 x i8> undef, i8 %b, i32 0
; BTVER2-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
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = lshr <16 x i8> %a, %splat
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %shift = lshr <16 x i8> %a, %splat
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
%insert = insertelement <16 x i8> undef, i8 %b, i32 0
@@ -1044,7 +1044,7 @@ define <32 x i8> @splatvar_shift_v32i8(<32 x i8> %a, i8 %b) {
; 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
; AVX1-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %splat = shufflevector <32 x i8> %insert, <32 x i8> undef, <32 x i32> zeroinitializer
-; AVX1-NEXT: Cost Model: Found an estimated cost of 26 for instruction: %shift = lshr <32 x i8> %a, %splat
+; AVX1-NEXT: Cost Model: Found an estimated cost of 23 for instruction: %shift = lshr <32 x i8> %a, %splat
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
; AVX2-LABEL: 'splatvar_shift_v32i8'
@@ -1092,7 +1092,7 @@ define <32 x i8> @splatvar_shift_v32i8(<32 x i8> %a, i8 %b) {
; BTVER2-LABEL: 'splatvar_shift_v32i8'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <32 x i8> undef, i8 %b, i32 0
; BTVER2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %splat = shufflevector <32 x i8> %insert, <32 x i8> undef, <32 x i32> zeroinitializer
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 26 for instruction: %shift = lshr <32 x i8> %a, %splat
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 23 for instruction: %shift = lshr <32 x i8> %a, %splat
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
%insert = insertelement <32 x i8> undef, i8 %b, i32 0
@@ -1117,7 +1117,7 @@ define <64 x i8> @splatvar_shift_v64i8(<64 x i8> %a, i8 %b) {
; 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
; AVX1-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %splat = shufflevector <64 x i8> %insert, <64 x i8> undef, <64 x i32> zeroinitializer
-; AVX1-NEXT: Cost Model: Found an estimated cost of 52 for instruction: %shift = lshr <64 x i8> %a, %splat
+; AVX1-NEXT: Cost Model: Found an estimated cost of 46 for instruction: %shift = lshr <64 x i8> %a, %splat
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
; AVX2-LABEL: 'splatvar_shift_v64i8'
@@ -1165,7 +1165,7 @@ define <64 x i8> @splatvar_shift_v64i8(<64 x i8> %a, i8 %b) {
; BTVER2-LABEL: 'splatvar_shift_v64i8'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <64 x i8> undef, i8 %b, i32 0
; BTVER2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %splat = shufflevector <64 x i8> %insert, <64 x i8> undef, <64 x i32> zeroinitializer
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 52 for instruction: %shift = lshr <64 x i8> %a, %splat
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 46 for instruction: %shift = lshr <64 x i8> %a, %splat
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
%insert = insertelement <64 x i8> undef, i8 %b, i32 0
@@ -1184,7 +1184,7 @@ define <2 x i64> @constant_shift_v2i64(<2 x i64> %a) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
; AVX1-LABEL: 'constant_shift_v2i64'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %shift = lshr <2 x i64> %a, <i64 1, i64 7>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %shift = lshr <2 x i64> %a, <i64 1, i64 7>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
; AVX2-LABEL: 'constant_shift_v2i64'
@@ -1200,7 +1200,7 @@ define <2 x i64> @constant_shift_v2i64(<2 x i64> %a) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
; BTVER2-LABEL: 'constant_shift_v2i64'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %shift = lshr <2 x i64> %a, <i64 1, i64 7>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %shift = lshr <2 x i64> %a, <i64 1, i64 7>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
%shift = lshr <2 x i64> %a, <i64 1, i64 7>
@@ -1213,7 +1213,7 @@ define <4 x i64> @constant_shift_v4i64(<4 x i64> %a) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
; AVX1-LABEL: 'constant_shift_v4i64'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %shift = lshr <4 x i64> %a, <i64 1, i64 7, i64 15, i64 31>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %shift = lshr <4 x i64> %a, <i64 1, i64 7, i64 15, i64 31>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
; AVX2-LABEL: 'constant_shift_v4i64'
@@ -1233,7 +1233,7 @@ define <4 x i64> @constant_shift_v4i64(<4 x i64> %a) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
; BTVER2-LABEL: 'constant_shift_v4i64'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %shift = lshr <4 x i64> %a, <i64 1, i64 7, i64 15, i64 31>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %shift = lshr <4 x i64> %a, <i64 1, i64 7, i64 15, i64 31>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
%shift = lshr <4 x i64> %a, <i64 1, i64 7, i64 15, i64 31>
@@ -1246,7 +1246,7 @@ define <8 x i64> @constant_shift_v8i64(<8 x i64> %a) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
; AVX1-LABEL: 'constant_shift_v8i64'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %shift = lshr <8 x i64> %a, <i64 1, i64 7, i64 15, i64 31, i64 1, i64 7, i64 15, i64 31>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = lshr <8 x i64> %a, <i64 1, i64 7, i64 15, i64 31, i64 1, i64 7, i64 15, i64 31>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
; AVX2-LABEL: 'constant_shift_v8i64'
@@ -1266,7 +1266,7 @@ define <8 x i64> @constant_shift_v8i64(<8 x i64> %a) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
; BTVER2-LABEL: 'constant_shift_v8i64'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %shift = lshr <8 x i64> %a, <i64 1, i64 7, i64 15, i64 31, i64 1, i64 7, i64 15, i64 31>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = lshr <8 x i64> %a, <i64 1, i64 7, i64 15, i64 31, i64 1, i64 7, i64 15, i64 31>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
%shift = lshr <8 x i64> %a, <i64 1, i64 7, i64 15, i64 31, i64 1, i64 7, i64 15, i64 31>
@@ -1283,7 +1283,7 @@ define <4 x i32> @constant_shift_v4i32(<4 x i32> %a) {
; 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>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %shift = lshr <4 x i32> %a, <i32 4, i32 5, i32 6, i32 7>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
;
; AVX2-LABEL: 'constant_shift_v4i32'
@@ -1299,7 +1299,7 @@ define <4 x i32> @constant_shift_v4i32(<4 x i32> %a) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
;
; BTVER2-LABEL: 'constant_shift_v4i32'
-; BTVER2-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>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %shift = lshr <4 x i32> %a, <i32 4, i32 5, i32 6, i32 7>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
;
%shift = lshr <4 x i32> %a, <i32 4, i32 5, i32 6, i32 7>
@@ -1316,7 +1316,7 @@ define <8 x i32> @constant_shift_v8i32(<8 x i32> %a) {
; 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>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %shift = lshr <8 x i32> %a, <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
;
; AVX2-LABEL: 'constant_shift_v8i32'
@@ -1336,7 +1336,7 @@ define <8 x i32> @constant_shift_v8i32(<8 x i32> %a) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
;
; BTVER2-LABEL: 'constant_shift_v8i32'
-; BTVER2-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>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %shift = lshr <8 x i32> %a, <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
;
%shift = lshr <8 x i32> %a, <i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>
@@ -1353,7 +1353,7 @@ define <16 x i32> @constant_shift_v16i32(<16 x i32> %a) {
; 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>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 28 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>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
;
; AVX2-LABEL: 'constant_shift_v16i32'
@@ -1373,7 +1373,7 @@ define <16 x i32> @constant_shift_v16i32(<16 x i32> %a) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
;
; BTVER2-LABEL: 'constant_shift_v16i32'
-; BTVER2-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>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 28 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>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
;
%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>
@@ -1390,7 +1390,7 @@ define <8 x i16> @constant_shift_v8i16(<8 x i16> %a) {
; SSE42-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
;
; AVX1-LABEL: 'constant_shift_v8i16'
-; AVX1-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>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %shift = lshr <8 x i16> %a, <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
;
; AVX2-LABEL: 'constant_shift_v8i16'
@@ -1418,7 +1418,7 @@ define <8 x i16> @constant_shift_v8i16(<8 x i16> %a) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
;
; BTVER2-LABEL: 'constant_shift_v8i16'
-; BTVER2-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>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %shift = lshr <8 x i16> %a, <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
;
%shift = lshr <8 x i16> %a, <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>
@@ -1435,7 +1435,7 @@ define <16 x i16> @constant_shift_v16i16(<16 x i16> %a) {
; 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>
+; AVX1-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>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
;
; AVX2-LABEL: 'constant_shift_v16i16'
@@ -1463,7 +1463,7 @@ define <16 x i16> @constant_shift_v16i16(<16 x i16> %a) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
;
; BTVER2-LABEL: 'constant_shift_v16i16'
-; BTVER2-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>
+; BTVER2-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>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
;
%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>
@@ -1480,7 +1480,7 @@ define <32 x i16> @constant_shift_v32i16(<32 x i16> %a) {
; 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>
+; AVX1-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>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
;
; AVX2-LABEL: 'constant_shift_v32i16'
@@ -1508,7 +1508,7 @@ define <32 x i16> @constant_shift_v32i16(<32 x i16> %a) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
;
; BTVER2-LABEL: 'constant_shift_v32i16'
-; BTVER2-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>
+; BTVER2-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>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
;
%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>
@@ -1525,7 +1525,7 @@ define <16 x i8> @constant_shift_v16i8(<16 x i8> %a) {
; SSE42-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; AVX1-LABEL: 'constant_shift_v16i8'
-; AVX1-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>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 11 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>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; AVX2-LABEL: 'constant_shift_v16i8'
@@ -1553,7 +1553,7 @@ define <16 x i8> @constant_shift_v16i8(<16 x i8> %a) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; BTVER2-LABEL: 'constant_shift_v16i8'
-; BTVER2-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>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 11 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>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
%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>
@@ -1570,7 +1570,7 @@ define <32 x i8> @constant_shift_v32i8(<32 x i8> %a) {
; 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>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 23 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>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
; AVX2-LABEL: 'constant_shift_v32i8'
@@ -1598,7 +1598,7 @@ define <32 x i8> @constant_shift_v32i8(<32 x i8> %a) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
; BTVER2-LABEL: 'constant_shift_v32i8'
-; BTVER2-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>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 23 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>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
%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>
@@ -1615,7 +1615,7 @@ define <64 x i8> @constant_shift_v64i8(<64 x i8> %a) {
; 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>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 46 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>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
; AVX2-LABEL: 'constant_shift_v64i8'
@@ -1643,7 +1643,7 @@ define <64 x i8> @constant_shift_v64i8(<64 x i8> %a) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
; BTVER2-LABEL: 'constant_shift_v64i8'
-; BTVER2-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>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 46 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>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
%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/llvm/test/Analysis/CostModel/X86/vshift-shl-cost-inseltpoison.ll b/llvm/test/Analysis/CostModel/X86/vshift-shl-cost-inseltpoison.ll
index 914b725e55067..570b2e4a59ea9 100644
--- a/llvm/test/Analysis/CostModel/X86/vshift-shl-cost-inseltpoison.ll
+++ b/llvm/test/Analysis/CostModel/X86/vshift-shl-cost-inseltpoison.ll
@@ -28,7 +28,7 @@ define <2 x i64> @var_shift_v2i64(<2 x i64> %a, <2 x i64> %b) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
; AVX1-LABEL: 'var_shift_v2i64'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %shift = shl <2 x i64> %a, %b
+; AVX1-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %shift = shl <2 x i64> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
; AVX2-LABEL: 'var_shift_v2i64'
@@ -44,7 +44,7 @@ define <2 x i64> @var_shift_v2i64(<2 x i64> %a, <2 x i64> %b) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
; BTVER2-LABEL: 'var_shift_v2i64'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %shift = shl <2 x i64> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %shift = shl <2 x i64> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
%shift = shl <2 x i64> %a, %b
@@ -57,7 +57,7 @@ define <4 x i64> @var_shift_v4i64(<4 x i64> %a, <4 x i64> %b) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
; AVX1-LABEL: 'var_shift_v4i64'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %shift = shl <4 x i64> %a, %b
+; AVX1-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %shift = shl <4 x i64> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
; AVX2-LABEL: 'var_shift_v4i64'
@@ -77,7 +77,7 @@ define <4 x i64> @var_shift_v4i64(<4 x i64> %a, <4 x i64> %b) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
; BTVER2-LABEL: 'var_shift_v4i64'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %shift = shl <4 x i64> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %shift = shl <4 x i64> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
%shift = shl <4 x i64> %a, %b
@@ -90,7 +90,7 @@ define <8 x i64> @var_shift_v8i64(<8 x i64> %a, <8 x i64> %b) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
; AVX1-LABEL: 'var_shift_v8i64'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %shift = shl <8 x i64> %a, %b
+; AVX1-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = shl <8 x i64> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
; AVX2-LABEL: 'var_shift_v8i64'
@@ -110,7 +110,7 @@ define <8 x i64> @var_shift_v8i64(<8 x i64> %a, <8 x i64> %b) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
; BTVER2-LABEL: 'var_shift_v8i64'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %shift = shl <8 x i64> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = shl <8 x i64> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
%shift = shl <8 x i64> %a, %b
@@ -127,7 +127,7 @@ define <4 x i32> @var_shift_v4i32(<4 x i32> %a, <4 x i32> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %shift = shl <4 x i32> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
;
; AVX2-LABEL: 'var_shift_v4i32'
@@ -143,7 +143,7 @@ define <4 x i32> @var_shift_v4i32(<4 x i32> %a, <4 x i32> %b) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
;
; BTVER2-LABEL: 'var_shift_v4i32'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %shift = shl <4 x i32> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %shift = shl <4 x i32> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
;
%shift = shl <4 x i32> %a, %b
@@ -160,7 +160,7 @@ define <8 x i32> @var_shift_v8i32(<8 x i32> %a, <8 x i32> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %shift = shl <8 x i32> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
;
; AVX2-LABEL: 'var_shift_v8i32'
@@ -180,7 +180,7 @@ define <8 x i32> @var_shift_v8i32(<8 x i32> %a, <8 x i32> %b) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
;
; BTVER2-LABEL: 'var_shift_v8i32'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %shift = shl <8 x i32> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %shift = shl <8 x i32> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
;
%shift = shl <8 x i32> %a, %b
@@ -197,7 +197,7 @@ define <16 x i32> @var_shift_v16i32(<16 x i32> %a, <16 x i32> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %shift = shl <16 x i32> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
;
; AVX2-LABEL: 'var_shift_v16i32'
@@ -217,7 +217,7 @@ define <16 x i32> @var_shift_v16i32(<16 x i32> %a, <16 x i32> %b) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
;
; BTVER2-LABEL: 'var_shift_v16i32'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %shift = shl <16 x i32> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %shift = shl <16 x i32> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
;
%shift = shl <16 x i32> %a, %b
@@ -234,7 +234,7 @@ define <8 x i16> @var_shift_v8i16(<8 x i16> %a, <8 x i16> %b) {
; SSE42-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
;
; AVX1-LABEL: 'var_shift_v8i16'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %shift = shl <8 x i16> %a, %b
+; AVX1-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %shift = shl <8 x i16> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
;
; AVX2-LABEL: 'var_shift_v8i16'
@@ -262,7 +262,7 @@ define <8 x i16> @var_shift_v8i16(<8 x i16> %a, <8 x i16> %b) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
;
; BTVER2-LABEL: 'var_shift_v8i16'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %shift = shl <8 x i16> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %shift = shl <8 x i16> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
;
%shift = shl <8 x i16> %a, %b
@@ -279,7 +279,7 @@ define <16 x i16> @var_shift_v16i16(<16 x i16> %a, <16 x i16> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %shift = shl <16 x i16> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
;
; AVX2-LABEL: 'var_shift_v16i16'
@@ -307,7 +307,7 @@ define <16 x i16> @var_shift_v16i16(<16 x i16> %a, <16 x i16> %b) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
;
; BTVER2-LABEL: 'var_shift_v16i16'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 30 for instruction: %shift = shl <16 x i16> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %shift = shl <16 x i16> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
;
%shift = shl <16 x i16> %a, %b
@@ -324,7 +324,7 @@ define <32 x i16> @var_shift_v32i16(<32 x i16> %a, <32 x i16> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 26 for instruction: %shift = shl <32 x i16> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
;
; AVX2-LABEL: 'var_shift_v32i16'
@@ -352,7 +352,7 @@ define <32 x i16> @var_shift_v32i16(<32 x i16> %a, <32 x i16> %b) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
;
; BTVER2-LABEL: 'var_shift_v32i16'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 60 for instruction: %shift = shl <32 x i16> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 26 for instruction: %shift = shl <32 x i16> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
;
%shift = shl <32 x i16> %a, %b
@@ -369,7 +369,7 @@ define <16 x i8> @var_shift_v16i8(<16 x i8> %a, <16 x i8> %b) {
; SSE42-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; AVX1-LABEL: 'var_shift_v16i8'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %shift = shl <16 x i8> %a, %b
+; AVX1-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %shift = shl <16 x i8> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; AVX2-LABEL: 'var_shift_v16i8'
@@ -397,7 +397,7 @@ define <16 x i8> @var_shift_v16i8(<16 x i8> %a, <16 x i8> %b) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; BTVER2-LABEL: 'var_shift_v16i8'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %shift = shl <16 x i8> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %shift = shl <16 x i8> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
%shift = shl <16 x i8> %a, %b
@@ -414,7 +414,7 @@ define <32 x i8> @var_shift_v32i8(<32 x i8> %a, <32 x i8> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 22 for instruction: %shift = shl <32 x i8> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
; AVX2-LABEL: 'var_shift_v32i8'
@@ -442,7 +442,7 @@ define <32 x i8> @var_shift_v32i8(<32 x i8> %a, <32 x i8> %b) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
; BTVER2-LABEL: 'var_shift_v32i8'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %shift = shl <32 x i8> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 22 for instruction: %shift = shl <32 x i8> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
%shift = shl <32 x i8> %a, %b
@@ -459,7 +459,7 @@ define <64 x i8> @var_shift_v64i8(<64 x i8> %a, <64 x i8> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 44 for instruction: %shift = shl <64 x i8> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
; AVX2-LABEL: 'var_shift_v64i8'
@@ -487,7 +487,7 @@ define <64 x i8> @var_shift_v64i8(<64 x i8> %a, <64 x i8> %b) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
; BTVER2-LABEL: 'var_shift_v64i8'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 48 for instruction: %shift = shl <64 x i8> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 44 for instruction: %shift = shl <64 x i8> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
%shift = shl <64 x i8> %a, %b
@@ -973,7 +973,7 @@ define <16 x i8> @splatvar_shift_v16i8(<16 x i8> %a, i8 %b) {
; AVX1-LABEL: 'splatvar_shift_v16i8'
; AVX1-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <16 x i8> poison, i8 %b, i32 0
; AVX1-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <16 x i8> %insert, <16 x i8> poison, <16 x i32> zeroinitializer
-; AVX1-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %shift = shl <16 x i8> %a, %splat
+; AVX1-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %shift = shl <16 x i8> %a, %splat
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; AVX2-LABEL: 'splatvar_shift_v16i8'
@@ -1015,7 +1015,7 @@ define <16 x i8> @splatvar_shift_v16i8(<16 x i8> %a, i8 %b) {
; BTVER2-LABEL: 'splatvar_shift_v16i8'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <16 x i8> poison, i8 %b, i32 0
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %splat = shufflevector <16 x i8> %insert, <16 x i8> poison, <16 x i32> zeroinitializer
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %shift = shl <16 x i8> %a, %splat
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %shift = shl <16 x i8> %a, %splat
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
%insert = insertelement <16 x i8> poison, i8 %b, i32 0
@@ -1040,7 +1040,7 @@ define <32 x i8> @splatvar_shift_v32i8(<32 x i8> %a, i8 %b) {
; AVX1-LABEL: 'splatvar_shift_v32i8'
; AVX1-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <32 x i8> poison, i8 %b, i32 0
; AVX1-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %splat = shufflevector <32 x i8> %insert, <32 x i8> poison, <32 x i32> zeroinitializer
-; AVX1-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %shift = shl <32 x i8> %a, %splat
+; AVX1-NEXT: Cost Model: Found an estimated cost of 22 for instruction: %shift = shl <32 x i8> %a, %splat
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
; AVX2-LABEL: 'splatvar_shift_v32i8'
@@ -1088,7 +1088,7 @@ define <32 x i8> @splatvar_shift_v32i8(<32 x i8> %a, i8 %b) {
; BTVER2-LABEL: 'splatvar_shift_v32i8'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <32 x i8> poison, i8 %b, i32 0
; BTVER2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %splat = shufflevector <32 x i8> %insert, <32 x i8> poison, <32 x i32> zeroinitializer
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %shift = shl <32 x i8> %a, %splat
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 22 for instruction: %shift = shl <32 x i8> %a, %splat
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
%insert = insertelement <32 x i8> poison, i8 %b, i32 0
@@ -1113,7 +1113,7 @@ define <64 x i8> @splatvar_shift_v64i8(<64 x i8> %a, i8 %b) {
; AVX1-LABEL: 'splatvar_shift_v64i8'
; AVX1-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <64 x i8> poison, i8 %b, i32 0
; AVX1-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %splat = shufflevector <64 x i8> %insert, <64 x i8> poison, <64 x i32> zeroinitializer
-; AVX1-NEXT: Cost Model: Found an estimated cost of 48 for instruction: %shift = shl <64 x i8> %a, %splat
+; AVX1-NEXT: Cost Model: Found an estimated cost of 44 for instruction: %shift = shl <64 x i8> %a, %splat
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
; AVX2-LABEL: 'splatvar_shift_v64i8'
@@ -1161,7 +1161,7 @@ define <64 x i8> @splatvar_shift_v64i8(<64 x i8> %a, i8 %b) {
; BTVER2-LABEL: 'splatvar_shift_v64i8'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <64 x i8> poison, i8 %b, i32 0
; BTVER2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %splat = shufflevector <64 x i8> %insert, <64 x i8> poison, <64 x i32> zeroinitializer
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 48 for instruction: %shift = shl <64 x i8> %a, %splat
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 44 for instruction: %shift = shl <64 x i8> %a, %splat
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
%insert = insertelement <64 x i8> poison, i8 %b, i32 0
@@ -1180,7 +1180,7 @@ define <2 x i64> @constant_shift_v2i64(<2 x i64> %a) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
; AVX1-LABEL: 'constant_shift_v2i64'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %shift = shl <2 x i64> %a, <i64 1, i64 7>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %shift = shl <2 x i64> %a, <i64 1, i64 7>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
; AVX2-LABEL: 'constant_shift_v2i64'
@@ -1196,7 +1196,7 @@ define <2 x i64> @constant_shift_v2i64(<2 x i64> %a) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
; BTVER2-LABEL: 'constant_shift_v2i64'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %shift = shl <2 x i64> %a, <i64 1, i64 7>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %shift = shl <2 x i64> %a, <i64 1, i64 7>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
%shift = shl <2 x i64> %a, <i64 1, i64 7>
@@ -1209,7 +1209,7 @@ define <4 x i64> @constant_shift_v4i64(<4 x i64> %a) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
; AVX1-LABEL: 'constant_shift_v4i64'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %shift = shl <4 x i64> %a, <i64 1, i64 7, i64 15, i64 31>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %shift = shl <4 x i64> %a, <i64 1, i64 7, i64 15, i64 31>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
; AVX2-LABEL: 'constant_shift_v4i64'
@@ -1229,7 +1229,7 @@ define <4 x i64> @constant_shift_v4i64(<4 x i64> %a) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
; BTVER2-LABEL: 'constant_shift_v4i64'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %shift = shl <4 x i64> %a, <i64 1, i64 7, i64 15, i64 31>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %shift = shl <4 x i64> %a, <i64 1, i64 7, i64 15, i64 31>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
%shift = shl <4 x i64> %a, <i64 1, i64 7, i64 15, i64 31>
@@ -1242,7 +1242,7 @@ define <8 x i64> @constant_shift_v8i64(<8 x i64> %a) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
; AVX1-LABEL: 'constant_shift_v8i64'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %shift = shl <8 x i64> %a, <i64 1, i64 7, i64 15, i64 31, i64 1, i64 7, i64 15, i64 31>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = shl <8 x i64> %a, <i64 1, i64 7, i64 15, i64 31, i64 1, i64 7, i64 15, i64 31>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
; AVX2-LABEL: 'constant_shift_v8i64'
@@ -1262,7 +1262,7 @@ define <8 x i64> @constant_shift_v8i64(<8 x i64> %a) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
; BTVER2-LABEL: 'constant_shift_v8i64'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %shift = shl <8 x i64> %a, <i64 1, i64 7, i64 15, i64 31, i64 1, i64 7, i64 15, i64 31>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = shl <8 x i64> %a, <i64 1, i64 7, i64 15, i64 31, i64 1, i64 7, i64 15, i64 31>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
%shift = shl <8 x i64> %a, <i64 1, i64 7, i64 15, i64 31, i64 1, i64 7, i64 15, i64 31>
@@ -1465,7 +1465,7 @@ define <16 x i8> @constant_shift_v16i8(<16 x i8> %a) {
; SSE42-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; AVX1-LABEL: 'constant_shift_v16i8'
-; AVX1-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>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 10 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>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; AVX2-LABEL: 'constant_shift_v16i8'
@@ -1493,7 +1493,7 @@ define <16 x i8> @constant_shift_v16i8(<16 x i8> %a) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; BTVER2-LABEL: 'constant_shift_v16i8'
-; BTVER2-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>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 10 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>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
%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>
@@ -1510,7 +1510,7 @@ define <32 x i8> @constant_shift_v32i8(<32 x i8> %a) {
; 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>
+; AVX1-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>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
; AVX2-LABEL: 'constant_shift_v32i8'
@@ -1538,7 +1538,7 @@ define <32 x i8> @constant_shift_v32i8(<32 x i8> %a) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
; BTVER2-LABEL: 'constant_shift_v32i8'
-; BTVER2-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>
+; BTVER2-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>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
%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>
@@ -1555,7 +1555,7 @@ define <64 x i8> @constant_shift_v64i8(<64 x i8> %a) {
; 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>
+; AVX1-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>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
; AVX2-LABEL: 'constant_shift_v64i8'
@@ -1583,7 +1583,7 @@ define <64 x i8> @constant_shift_v64i8(<64 x i8> %a) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
; BTVER2-LABEL: 'constant_shift_v64i8'
-; BTVER2-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>
+; BTVER2-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>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
%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>
@@ -2021,7 +2021,7 @@ define <2 x i64> @test5(<2 x i64> %a) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shl
;
; AVX1-LABEL: 'test5'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %shl = shl <2 x i64> %a, <i64 2, i64 3>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %shl = shl <2 x i64> %a, <i64 2, i64 3>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shl
;
; AVX2-LABEL: 'test5'
@@ -2037,7 +2037,7 @@ define <2 x i64> @test5(<2 x i64> %a) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shl
;
; BTVER2-LABEL: 'test5'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %shl = shl <2 x i64> %a, <i64 2, i64 3>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %shl = shl <2 x i64> %a, <i64 2, i64 3>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shl
;
%shl = shl <2 x i64> %a, <i64 2, i64 3>
@@ -2141,7 +2141,7 @@ define <4 x i64> @test8(<4 x i64> %a) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shl
;
; AVX1-LABEL: 'test8'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %shl = shl <4 x i64> %a, <i64 1, i64 2, i64 3, i64 4>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %shl = shl <4 x i64> %a, <i64 1, i64 2, i64 3, i64 4>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shl
;
; AVX2-LABEL: 'test8'
@@ -2161,7 +2161,7 @@ define <4 x i64> @test8(<4 x i64> %a) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shl
;
; BTVER2-LABEL: 'test8'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %shl = shl <4 x i64> %a, <i64 1, i64 2, i64 3, i64 4>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %shl = shl <4 x i64> %a, <i64 1, i64 2, i64 3, i64 4>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shl
;
%shl = shl <4 x i64> %a, <i64 1, i64 2, i64 3, i64 4>
@@ -2265,7 +2265,7 @@ define <8 x i64> @test11(<8 x i64> %a) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shl
;
; AVX1-LABEL: 'test11'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %shl = shl <8 x i64> %a, <i64 1, i64 1, i64 2, i64 3, i64 1, i64 1, i64 2, i64 3>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shl = shl <8 x i64> %a, <i64 1, i64 1, i64 2, i64 3, i64 1, i64 1, i64 2, i64 3>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shl
;
; AVX2-LABEL: 'test11'
@@ -2285,7 +2285,7 @@ define <8 x i64> @test11(<8 x i64> %a) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shl
;
; BTVER2-LABEL: 'test11'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %shl = shl <8 x i64> %a, <i64 1, i64 1, i64 2, i64 3, i64 1, i64 1, i64 2, i64 3>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shl = shl <8 x i64> %a, <i64 1, i64 1, i64 2, i64 3, i64 1, i64 1, i64 2, i64 3>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shl
;
%shl = shl <8 x i64> %a, <i64 1, i64 1, i64 2, i64 3, i64 1, i64 1, i64 2, i64 3>
diff --git a/llvm/test/Analysis/CostModel/X86/vshift-shl-cost.ll b/llvm/test/Analysis/CostModel/X86/vshift-shl-cost.ll
index fed023b6ca795..0eaa51b3b7653 100644
--- a/llvm/test/Analysis/CostModel/X86/vshift-shl-cost.ll
+++ b/llvm/test/Analysis/CostModel/X86/vshift-shl-cost.ll
@@ -28,7 +28,7 @@ define <2 x i64> @var_shift_v2i64(<2 x i64> %a, <2 x i64> %b) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
; AVX1-LABEL: 'var_shift_v2i64'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %shift = shl <2 x i64> %a, %b
+; AVX1-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %shift = shl <2 x i64> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
; AVX2-LABEL: 'var_shift_v2i64'
@@ -44,7 +44,7 @@ define <2 x i64> @var_shift_v2i64(<2 x i64> %a, <2 x i64> %b) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
; BTVER2-LABEL: 'var_shift_v2i64'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %shift = shl <2 x i64> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %shift = shl <2 x i64> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
%shift = shl <2 x i64> %a, %b
@@ -57,7 +57,7 @@ define <4 x i64> @var_shift_v4i64(<4 x i64> %a, <4 x i64> %b) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
; AVX1-LABEL: 'var_shift_v4i64'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %shift = shl <4 x i64> %a, %b
+; AVX1-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %shift = shl <4 x i64> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
; AVX2-LABEL: 'var_shift_v4i64'
@@ -77,7 +77,7 @@ define <4 x i64> @var_shift_v4i64(<4 x i64> %a, <4 x i64> %b) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
; BTVER2-LABEL: 'var_shift_v4i64'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %shift = shl <4 x i64> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %shift = shl <4 x i64> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
%shift = shl <4 x i64> %a, %b
@@ -90,7 +90,7 @@ define <8 x i64> @var_shift_v8i64(<8 x i64> %a, <8 x i64> %b) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
; AVX1-LABEL: 'var_shift_v8i64'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %shift = shl <8 x i64> %a, %b
+; AVX1-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = shl <8 x i64> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
; AVX2-LABEL: 'var_shift_v8i64'
@@ -110,7 +110,7 @@ define <8 x i64> @var_shift_v8i64(<8 x i64> %a, <8 x i64> %b) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
; BTVER2-LABEL: 'var_shift_v8i64'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %shift = shl <8 x i64> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = shl <8 x i64> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
%shift = shl <8 x i64> %a, %b
@@ -127,7 +127,7 @@ define <4 x i32> @var_shift_v4i32(<4 x i32> %a, <4 x i32> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %shift = shl <4 x i32> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
;
; AVX2-LABEL: 'var_shift_v4i32'
@@ -143,7 +143,7 @@ define <4 x i32> @var_shift_v4i32(<4 x i32> %a, <4 x i32> %b) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
;
; BTVER2-LABEL: 'var_shift_v4i32'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %shift = shl <4 x i32> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %shift = shl <4 x i32> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %shift
;
%shift = shl <4 x i32> %a, %b
@@ -160,7 +160,7 @@ define <8 x i32> @var_shift_v8i32(<8 x i32> %a, <8 x i32> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %shift = shl <8 x i32> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
;
; AVX2-LABEL: 'var_shift_v8i32'
@@ -180,7 +180,7 @@ define <8 x i32> @var_shift_v8i32(<8 x i32> %a, <8 x i32> %b) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
;
; BTVER2-LABEL: 'var_shift_v8i32'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %shift = shl <8 x i32> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %shift = shl <8 x i32> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %shift
;
%shift = shl <8 x i32> %a, %b
@@ -197,7 +197,7 @@ define <16 x i32> @var_shift_v16i32(<16 x i32> %a, <16 x i32> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %shift = shl <16 x i32> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
;
; AVX2-LABEL: 'var_shift_v16i32'
@@ -217,7 +217,7 @@ define <16 x i32> @var_shift_v16i32(<16 x i32> %a, <16 x i32> %b) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
;
; BTVER2-LABEL: 'var_shift_v16i32'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %shift = shl <16 x i32> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %shift = shl <16 x i32> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %shift
;
%shift = shl <16 x i32> %a, %b
@@ -234,7 +234,7 @@ define <8 x i16> @var_shift_v8i16(<8 x i16> %a, <8 x i16> %b) {
; SSE42-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
;
; AVX1-LABEL: 'var_shift_v8i16'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %shift = shl <8 x i16> %a, %b
+; AVX1-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %shift = shl <8 x i16> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
;
; AVX2-LABEL: 'var_shift_v8i16'
@@ -262,7 +262,7 @@ define <8 x i16> @var_shift_v8i16(<8 x i16> %a, <8 x i16> %b) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
;
; BTVER2-LABEL: 'var_shift_v8i16'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %shift = shl <8 x i16> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %shift = shl <8 x i16> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %shift
;
%shift = shl <8 x i16> %a, %b
@@ -279,7 +279,7 @@ define <16 x i16> @var_shift_v16i16(<16 x i16> %a, <16 x i16> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %shift = shl <16 x i16> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
;
; AVX2-LABEL: 'var_shift_v16i16'
@@ -307,7 +307,7 @@ define <16 x i16> @var_shift_v16i16(<16 x i16> %a, <16 x i16> %b) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
;
; BTVER2-LABEL: 'var_shift_v16i16'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 30 for instruction: %shift = shl <16 x i16> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %shift = shl <16 x i16> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %shift
;
%shift = shl <16 x i16> %a, %b
@@ -324,7 +324,7 @@ define <32 x i16> @var_shift_v32i16(<32 x i16> %a, <32 x i16> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 26 for instruction: %shift = shl <32 x i16> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
;
; AVX2-LABEL: 'var_shift_v32i16'
@@ -352,7 +352,7 @@ define <32 x i16> @var_shift_v32i16(<32 x i16> %a, <32 x i16> %b) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
;
; BTVER2-LABEL: 'var_shift_v32i16'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 60 for instruction: %shift = shl <32 x i16> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 26 for instruction: %shift = shl <32 x i16> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %shift
;
%shift = shl <32 x i16> %a, %b
@@ -369,7 +369,7 @@ define <16 x i8> @var_shift_v16i8(<16 x i8> %a, <16 x i8> %b) {
; SSE42-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; AVX1-LABEL: 'var_shift_v16i8'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %shift = shl <16 x i8> %a, %b
+; AVX1-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %shift = shl <16 x i8> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; AVX2-LABEL: 'var_shift_v16i8'
@@ -397,7 +397,7 @@ define <16 x i8> @var_shift_v16i8(<16 x i8> %a, <16 x i8> %b) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; BTVER2-LABEL: 'var_shift_v16i8'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %shift = shl <16 x i8> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %shift = shl <16 x i8> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
%shift = shl <16 x i8> %a, %b
@@ -414,7 +414,7 @@ define <32 x i8> @var_shift_v32i8(<32 x i8> %a, <32 x i8> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 22 for instruction: %shift = shl <32 x i8> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
; AVX2-LABEL: 'var_shift_v32i8'
@@ -442,7 +442,7 @@ define <32 x i8> @var_shift_v32i8(<32 x i8> %a, <32 x i8> %b) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
; BTVER2-LABEL: 'var_shift_v32i8'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %shift = shl <32 x i8> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 22 for instruction: %shift = shl <32 x i8> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
%shift = shl <32 x i8> %a, %b
@@ -459,7 +459,7 @@ define <64 x i8> @var_shift_v64i8(<64 x i8> %a, <64 x i8> %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
+; AVX1-NEXT: Cost Model: Found an estimated cost of 44 for instruction: %shift = shl <64 x i8> %a, %b
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
; AVX2-LABEL: 'var_shift_v64i8'
@@ -487,7 +487,7 @@ define <64 x i8> @var_shift_v64i8(<64 x i8> %a, <64 x i8> %b) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
; BTVER2-LABEL: 'var_shift_v64i8'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 48 for instruction: %shift = shl <64 x i8> %a, %b
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 44 for instruction: %shift = shl <64 x i8> %a, %b
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
%shift = shl <64 x i8> %a, %b
@@ -973,7 +973,7 @@ define <16 x i8> @splatvar_shift_v16i8(<16 x i8> %a, i8 %b) {
; AVX1-LABEL: 'splatvar_shift_v16i8'
; AVX1-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <16 x i8> undef, i8 %b, i32 0
; AVX1-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
-; AVX1-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %shift = shl <16 x i8> %a, %splat
+; AVX1-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %shift = shl <16 x i8> %a, %splat
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; AVX2-LABEL: 'splatvar_shift_v16i8'
@@ -1015,7 +1015,7 @@ define <16 x i8> @splatvar_shift_v16i8(<16 x i8> %a, i8 %b) {
; BTVER2-LABEL: 'splatvar_shift_v16i8'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <16 x i8> undef, i8 %b, i32 0
; BTVER2-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
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %shift = shl <16 x i8> %a, %splat
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %shift = shl <16 x i8> %a, %splat
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
%insert = insertelement <16 x i8> undef, i8 %b, i32 0
@@ -1040,7 +1040,7 @@ define <32 x i8> @splatvar_shift_v32i8(<32 x i8> %a, i8 %b) {
; 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
; AVX1-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %splat = shufflevector <32 x i8> %insert, <32 x i8> undef, <32 x i32> zeroinitializer
-; AVX1-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %shift = shl <32 x i8> %a, %splat
+; AVX1-NEXT: Cost Model: Found an estimated cost of 22 for instruction: %shift = shl <32 x i8> %a, %splat
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
; AVX2-LABEL: 'splatvar_shift_v32i8'
@@ -1088,7 +1088,7 @@ define <32 x i8> @splatvar_shift_v32i8(<32 x i8> %a, i8 %b) {
; BTVER2-LABEL: 'splatvar_shift_v32i8'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <32 x i8> undef, i8 %b, i32 0
; BTVER2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %splat = shufflevector <32 x i8> %insert, <32 x i8> undef, <32 x i32> zeroinitializer
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %shift = shl <32 x i8> %a, %splat
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 22 for instruction: %shift = shl <32 x i8> %a, %splat
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
%insert = insertelement <32 x i8> undef, i8 %b, i32 0
@@ -1113,7 +1113,7 @@ define <64 x i8> @splatvar_shift_v64i8(<64 x i8> %a, i8 %b) {
; 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
; AVX1-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %splat = shufflevector <64 x i8> %insert, <64 x i8> undef, <64 x i32> zeroinitializer
-; AVX1-NEXT: Cost Model: Found an estimated cost of 48 for instruction: %shift = shl <64 x i8> %a, %splat
+; AVX1-NEXT: Cost Model: Found an estimated cost of 44 for instruction: %shift = shl <64 x i8> %a, %splat
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
; AVX2-LABEL: 'splatvar_shift_v64i8'
@@ -1161,7 +1161,7 @@ define <64 x i8> @splatvar_shift_v64i8(<64 x i8> %a, i8 %b) {
; BTVER2-LABEL: 'splatvar_shift_v64i8'
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %insert = insertelement <64 x i8> undef, i8 %b, i32 0
; BTVER2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %splat = shufflevector <64 x i8> %insert, <64 x i8> undef, <64 x i32> zeroinitializer
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 48 for instruction: %shift = shl <64 x i8> %a, %splat
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 44 for instruction: %shift = shl <64 x i8> %a, %splat
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
%insert = insertelement <64 x i8> undef, i8 %b, i32 0
@@ -1180,7 +1180,7 @@ define <2 x i64> @constant_shift_v2i64(<2 x i64> %a) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
; AVX1-LABEL: 'constant_shift_v2i64'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %shift = shl <2 x i64> %a, <i64 1, i64 7>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %shift = shl <2 x i64> %a, <i64 1, i64 7>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
; AVX2-LABEL: 'constant_shift_v2i64'
@@ -1196,7 +1196,7 @@ define <2 x i64> @constant_shift_v2i64(<2 x i64> %a) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
; BTVER2-LABEL: 'constant_shift_v2i64'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %shift = shl <2 x i64> %a, <i64 1, i64 7>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %shift = shl <2 x i64> %a, <i64 1, i64 7>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shift
;
%shift = shl <2 x i64> %a, <i64 1, i64 7>
@@ -1209,7 +1209,7 @@ define <4 x i64> @constant_shift_v4i64(<4 x i64> %a) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
; AVX1-LABEL: 'constant_shift_v4i64'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %shift = shl <4 x i64> %a, <i64 1, i64 7, i64 15, i64 31>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %shift = shl <4 x i64> %a, <i64 1, i64 7, i64 15, i64 31>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
; AVX2-LABEL: 'constant_shift_v4i64'
@@ -1229,7 +1229,7 @@ define <4 x i64> @constant_shift_v4i64(<4 x i64> %a) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
; BTVER2-LABEL: 'constant_shift_v4i64'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %shift = shl <4 x i64> %a, <i64 1, i64 7, i64 15, i64 31>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %shift = shl <4 x i64> %a, <i64 1, i64 7, i64 15, i64 31>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shift
;
%shift = shl <4 x i64> %a, <i64 1, i64 7, i64 15, i64 31>
@@ -1242,7 +1242,7 @@ define <8 x i64> @constant_shift_v8i64(<8 x i64> %a) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
; AVX1-LABEL: 'constant_shift_v8i64'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %shift = shl <8 x i64> %a, <i64 1, i64 7, i64 15, i64 31, i64 1, i64 7, i64 15, i64 31>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = shl <8 x i64> %a, <i64 1, i64 7, i64 15, i64 31, i64 1, i64 7, i64 15, i64 31>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
; AVX2-LABEL: 'constant_shift_v8i64'
@@ -1262,7 +1262,7 @@ define <8 x i64> @constant_shift_v8i64(<8 x i64> %a) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
; BTVER2-LABEL: 'constant_shift_v8i64'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %shift = shl <8 x i64> %a, <i64 1, i64 7, i64 15, i64 31, i64 1, i64 7, i64 15, i64 31>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shift = shl <8 x i64> %a, <i64 1, i64 7, i64 15, i64 31, i64 1, i64 7, i64 15, i64 31>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shift
;
%shift = shl <8 x i64> %a, <i64 1, i64 7, i64 15, i64 31, i64 1, i64 7, i64 15, i64 31>
@@ -1465,7 +1465,7 @@ define <16 x i8> @constant_shift_v16i8(<16 x i8> %a) {
; SSE42-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; AVX1-LABEL: 'constant_shift_v16i8'
-; AVX1-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>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 10 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>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; AVX2-LABEL: 'constant_shift_v16i8'
@@ -1493,7 +1493,7 @@ define <16 x i8> @constant_shift_v16i8(<16 x i8> %a) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
; BTVER2-LABEL: 'constant_shift_v16i8'
-; BTVER2-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>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 10 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>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %shift
;
%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>
@@ -1510,7 +1510,7 @@ define <32 x i8> @constant_shift_v32i8(<32 x i8> %a) {
; 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>
+; AVX1-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>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
; AVX2-LABEL: 'constant_shift_v32i8'
@@ -1538,7 +1538,7 @@ define <32 x i8> @constant_shift_v32i8(<32 x i8> %a) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
; BTVER2-LABEL: 'constant_shift_v32i8'
-; BTVER2-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>
+; BTVER2-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>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %shift
;
%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>
@@ -1555,7 +1555,7 @@ define <64 x i8> @constant_shift_v64i8(<64 x i8> %a) {
; 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>
+; AVX1-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>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
; AVX2-LABEL: 'constant_shift_v64i8'
@@ -1583,7 +1583,7 @@ define <64 x i8> @constant_shift_v64i8(<64 x i8> %a) {
; AVX512BWVL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
; BTVER2-LABEL: 'constant_shift_v64i8'
-; BTVER2-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>
+; BTVER2-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>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %shift
;
%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>
@@ -2021,7 +2021,7 @@ define <2 x i64> @test5(<2 x i64> %a) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shl
;
; AVX1-LABEL: 'test5'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %shl = shl <2 x i64> %a, <i64 2, i64 3>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %shl = shl <2 x i64> %a, <i64 2, i64 3>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shl
;
; AVX2-LABEL: 'test5'
@@ -2037,7 +2037,7 @@ define <2 x i64> @test5(<2 x i64> %a) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shl
;
; BTVER2-LABEL: 'test5'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %shl = shl <2 x i64> %a, <i64 2, i64 3>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %shl = shl <2 x i64> %a, <i64 2, i64 3>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %shl
;
%shl = shl <2 x i64> %a, <i64 2, i64 3>
@@ -2141,7 +2141,7 @@ define <4 x i64> @test8(<4 x i64> %a) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shl
;
; AVX1-LABEL: 'test8'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %shl = shl <4 x i64> %a, <i64 1, i64 2, i64 3, i64 4>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %shl = shl <4 x i64> %a, <i64 1, i64 2, i64 3, i64 4>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shl
;
; AVX2-LABEL: 'test8'
@@ -2161,7 +2161,7 @@ define <4 x i64> @test8(<4 x i64> %a) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shl
;
; BTVER2-LABEL: 'test8'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %shl = shl <4 x i64> %a, <i64 1, i64 2, i64 3, i64 4>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %shl = shl <4 x i64> %a, <i64 1, i64 2, i64 3, i64 4>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %shl
;
%shl = shl <4 x i64> %a, <i64 1, i64 2, i64 3, i64 4>
@@ -2265,7 +2265,7 @@ define <8 x i64> @test11(<8 x i64> %a) {
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shl
;
; AVX1-LABEL: 'test11'
-; AVX1-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %shl = shl <8 x i64> %a, <i64 1, i64 1, i64 2, i64 3, i64 1, i64 1, i64 2, i64 3>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shl = shl <8 x i64> %a, <i64 1, i64 1, i64 2, i64 3, i64 1, i64 1, i64 2, i64 3>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shl
;
; AVX2-LABEL: 'test11'
@@ -2285,7 +2285,7 @@ define <8 x i64> @test11(<8 x i64> %a) {
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shl
;
; BTVER2-LABEL: 'test11'
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %shl = shl <8 x i64> %a, <i64 1, i64 1, i64 2, i64 3, i64 1, i64 1, i64 2, i64 3>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %shl = shl <8 x i64> %a, <i64 1, i64 1, i64 2, i64 3, i64 1, i64 1, i64 2, i64 3>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %shl
;
%shl = shl <8 x i64> %a, <i64 1, i64 1, i64 2, i64 3, i64 1, i64 1, i64 2, i64 3>
diff --git a/llvm/test/Transforms/SLPVectorizer/X86/alternate-int-inseltpoison.ll b/llvm/test/Transforms/SLPVectorizer/X86/alternate-int-inseltpoison.ll
index 84c13464aa17b..a6867fbff0193 100644
--- a/llvm/test/Transforms/SLPVectorizer/X86/alternate-int-inseltpoison.ll
+++ b/llvm/test/Transforms/SLPVectorizer/X86/alternate-int-inseltpoison.ll
@@ -184,12 +184,10 @@ define <8 x i32> @ashr_shl_v8i32_const(<8 x i32> %a) {
; SLM-NEXT: ret <8 x i32> [[TMP3]]
;
; AVX1-LABEL: @ashr_shl_v8i32_const(
-; AVX1-NEXT: [[TMP1:%.*]] = shufflevector <8 x i32> [[A:%.*]], <8 x i32> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
-; AVX1-NEXT: [[TMP2:%.*]] = ashr <4 x i32> [[TMP1]], <i32 2, i32 2, i32 2, i32 2>
-; AVX1-NEXT: [[TMP3:%.*]] = shufflevector <8 x i32> [[A]], <8 x i32> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
-; AVX1-NEXT: [[TMP4:%.*]] = shl <4 x i32> [[TMP3]], <i32 3, i32 3, i32 3, i32 3>
-; AVX1-NEXT: [[R72:%.*]] = shufflevector <4 x i32> [[TMP2]], <4 x i32> [[TMP4]], <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
-; AVX1-NEXT: ret <8 x i32> [[R72]]
+; AVX1-NEXT: [[TMP1:%.*]] = ashr <8 x i32> [[A:%.*]], <i32 2, i32 2, i32 2, i32 2, i32 3, i32 3, i32 3, i32 3>
+; AVX1-NEXT: [[TMP2:%.*]] = shl <8 x i32> [[A]], <i32 2, i32 2, i32 2, i32 2, i32 3, i32 3, i32 3, i32 3>
+; AVX1-NEXT: [[TMP3:%.*]] = shufflevector <8 x i32> [[TMP1]], <8 x i32> [[TMP2]], <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 12, i32 13, i32 14, i32 15>
+; AVX1-NEXT: ret <8 x i32> [[TMP3]]
;
; AVX2-LABEL: @ashr_shl_v8i32_const(
; AVX2-NEXT: [[TMP1:%.*]] = ashr <8 x i32> [[A:%.*]], <i32 2, i32 2, i32 2, i32 2, i32 3, i32 3, i32 3, i32 3>
@@ -281,30 +279,18 @@ define <8 x i32> @ashr_lshr_shl_v8i32(<8 x i32> %a, <8 x i32> %b) {
; SLM-NEXT: ret <8 x i32> [[R72]]
;
; AVX1-LABEL: @ashr_lshr_shl_v8i32(
-; AVX1-NEXT: [[A0:%.*]] = extractelement <8 x i32> [[A:%.*]], i32 0
-; AVX1-NEXT: [[A1:%.*]] = extractelement <8 x i32> [[A]], i32 1
-; AVX1-NEXT: [[A2:%.*]] = extractelement <8 x i32> [[A]], i32 2
-; AVX1-NEXT: [[A3:%.*]] = extractelement <8 x i32> [[A]], i32 3
-; AVX1-NEXT: [[B0:%.*]] = extractelement <8 x i32> [[B:%.*]], i32 0
-; AVX1-NEXT: [[B1:%.*]] = extractelement <8 x i32> [[B]], i32 1
-; AVX1-NEXT: [[B2:%.*]] = extractelement <8 x i32> [[B]], i32 2
-; AVX1-NEXT: [[B3:%.*]] = extractelement <8 x i32> [[B]], i32 3
-; AVX1-NEXT: [[AB0:%.*]] = ashr i32 [[A0]], [[B0]]
-; AVX1-NEXT: [[AB1:%.*]] = ashr i32 [[A1]], [[B1]]
-; AVX1-NEXT: [[AB2:%.*]] = lshr i32 [[A2]], [[B2]]
-; AVX1-NEXT: [[AB3:%.*]] = lshr i32 [[A3]], [[B3]]
-; AVX1-NEXT: [[TMP1:%.*]] = shufflevector <8 x i32> [[A]], <8 x i32> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
-; AVX1-NEXT: [[TMP2:%.*]] = shufflevector <8 x i32> [[B]], <8 x i32> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
-; AVX1-NEXT: [[TMP3:%.*]] = lshr <4 x i32> [[TMP1]], [[TMP2]]
-; AVX1-NEXT: [[TMP4:%.*]] = shl <4 x i32> [[TMP1]], [[TMP2]]
+; AVX1-NEXT: [[TMP1:%.*]] = shufflevector <8 x i32> [[A:%.*]], <8 x i32> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+; AVX1-NEXT: [[TMP2:%.*]] = shufflevector <8 x i32> [[B:%.*]], <8 x i32> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+; AVX1-NEXT: [[TMP3:%.*]] = ashr <4 x i32> [[TMP1]], [[TMP2]]
+; AVX1-NEXT: [[TMP4:%.*]] = lshr <4 x i32> [[TMP1]], [[TMP2]]
; AVX1-NEXT: [[TMP5:%.*]] = shufflevector <4 x i32> [[TMP3]], <4 x i32> [[TMP4]], <4 x i32> <i32 0, i32 1, i32 6, i32 7>
-; AVX1-NEXT: [[R0:%.*]] = insertelement <8 x i32> poison, i32 [[AB0]], i32 0
-; AVX1-NEXT: [[R1:%.*]] = insertelement <8 x i32> [[R0]], i32 [[AB1]], i32 1
-; AVX1-NEXT: [[R2:%.*]] = insertelement <8 x i32> [[R1]], i32 [[AB2]], i32 2
-; AVX1-NEXT: [[R3:%.*]] = insertelement <8 x i32> [[R2]], i32 [[AB3]], i32 3
-; AVX1-NEXT: [[TMP6:%.*]] = shufflevector <4 x i32> [[TMP5]], <4 x i32> undef, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef>
-; AVX1-NEXT: [[R71:%.*]] = shufflevector <8 x i32> [[R3]], <8 x i32> [[TMP6]], <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 8, i32 9, i32 10, i32 11>
-; AVX1-NEXT: ret <8 x i32> [[R71]]
+; AVX1-NEXT: [[TMP6:%.*]] = shufflevector <8 x i32> [[A]], <8 x i32> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
+; AVX1-NEXT: [[TMP7:%.*]] = shufflevector <8 x i32> [[B]], <8 x i32> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
+; AVX1-NEXT: [[TMP8:%.*]] = lshr <4 x i32> [[TMP6]], [[TMP7]]
+; AVX1-NEXT: [[TMP9:%.*]] = shl <4 x i32> [[TMP6]], [[TMP7]]
+; AVX1-NEXT: [[TMP10:%.*]] = shufflevector <4 x i32> [[TMP8]], <4 x i32> [[TMP9]], <4 x i32> <i32 0, i32 1, i32 6, i32 7>
+; AVX1-NEXT: [[R72:%.*]] = shufflevector <4 x i32> [[TMP5]], <4 x i32> [[TMP10]], <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+; AVX1-NEXT: ret <8 x i32> [[R72]]
;
; AVX2-LABEL: @ashr_lshr_shl_v8i32(
; AVX2-NEXT: [[TMP1:%.*]] = shufflevector <8 x i32> [[A:%.*]], <8 x i32> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
diff --git a/llvm/test/Transforms/SLPVectorizer/X86/alternate-int.ll b/llvm/test/Transforms/SLPVectorizer/X86/alternate-int.ll
index 9a3973b4bba59..78a4c5b71429f 100644
--- a/llvm/test/Transforms/SLPVectorizer/X86/alternate-int.ll
+++ b/llvm/test/Transforms/SLPVectorizer/X86/alternate-int.ll
@@ -184,12 +184,10 @@ define <8 x i32> @ashr_shl_v8i32_const(<8 x i32> %a) {
; SLM-NEXT: ret <8 x i32> [[TMP3]]
;
; AVX1-LABEL: @ashr_shl_v8i32_const(
-; AVX1-NEXT: [[TMP1:%.*]] = shufflevector <8 x i32> [[A:%.*]], <8 x i32> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
-; AVX1-NEXT: [[TMP2:%.*]] = ashr <4 x i32> [[TMP1]], <i32 2, i32 2, i32 2, i32 2>
-; AVX1-NEXT: [[TMP3:%.*]] = shufflevector <8 x i32> [[A]], <8 x i32> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
-; AVX1-NEXT: [[TMP4:%.*]] = shl <4 x i32> [[TMP3]], <i32 3, i32 3, i32 3, i32 3>
-; AVX1-NEXT: [[R72:%.*]] = shufflevector <4 x i32> [[TMP2]], <4 x i32> [[TMP4]], <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
-; AVX1-NEXT: ret <8 x i32> [[R72]]
+; AVX1-NEXT: [[TMP1:%.*]] = ashr <8 x i32> [[A:%.*]], <i32 2, i32 2, i32 2, i32 2, i32 3, i32 3, i32 3, i32 3>
+; AVX1-NEXT: [[TMP2:%.*]] = shl <8 x i32> [[A]], <i32 2, i32 2, i32 2, i32 2, i32 3, i32 3, i32 3, i32 3>
+; AVX1-NEXT: [[TMP3:%.*]] = shufflevector <8 x i32> [[TMP1]], <8 x i32> [[TMP2]], <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 12, i32 13, i32 14, i32 15>
+; AVX1-NEXT: ret <8 x i32> [[TMP3]]
;
; AVX2-LABEL: @ashr_shl_v8i32_const(
; AVX2-NEXT: [[TMP1:%.*]] = ashr <8 x i32> [[A:%.*]], <i32 2, i32 2, i32 2, i32 2, i32 3, i32 3, i32 3, i32 3>
@@ -281,30 +279,18 @@ define <8 x i32> @ashr_lshr_shl_v8i32(<8 x i32> %a, <8 x i32> %b) {
; SLM-NEXT: ret <8 x i32> [[R72]]
;
; AVX1-LABEL: @ashr_lshr_shl_v8i32(
-; AVX1-NEXT: [[A0:%.*]] = extractelement <8 x i32> [[A:%.*]], i32 0
-; AVX1-NEXT: [[A1:%.*]] = extractelement <8 x i32> [[A]], i32 1
-; AVX1-NEXT: [[A2:%.*]] = extractelement <8 x i32> [[A]], i32 2
-; AVX1-NEXT: [[A3:%.*]] = extractelement <8 x i32> [[A]], i32 3
-; AVX1-NEXT: [[B0:%.*]] = extractelement <8 x i32> [[B:%.*]], i32 0
-; AVX1-NEXT: [[B1:%.*]] = extractelement <8 x i32> [[B]], i32 1
-; AVX1-NEXT: [[B2:%.*]] = extractelement <8 x i32> [[B]], i32 2
-; AVX1-NEXT: [[B3:%.*]] = extractelement <8 x i32> [[B]], i32 3
-; AVX1-NEXT: [[AB0:%.*]] = ashr i32 [[A0]], [[B0]]
-; AVX1-NEXT: [[AB1:%.*]] = ashr i32 [[A1]], [[B1]]
-; AVX1-NEXT: [[AB2:%.*]] = lshr i32 [[A2]], [[B2]]
-; AVX1-NEXT: [[AB3:%.*]] = lshr i32 [[A3]], [[B3]]
-; AVX1-NEXT: [[TMP1:%.*]] = shufflevector <8 x i32> [[A]], <8 x i32> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
-; AVX1-NEXT: [[TMP2:%.*]] = shufflevector <8 x i32> [[B]], <8 x i32> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
-; AVX1-NEXT: [[TMP3:%.*]] = lshr <4 x i32> [[TMP1]], [[TMP2]]
-; AVX1-NEXT: [[TMP4:%.*]] = shl <4 x i32> [[TMP1]], [[TMP2]]
+; AVX1-NEXT: [[TMP1:%.*]] = shufflevector <8 x i32> [[A:%.*]], <8 x i32> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+; AVX1-NEXT: [[TMP2:%.*]] = shufflevector <8 x i32> [[B:%.*]], <8 x i32> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+; AVX1-NEXT: [[TMP3:%.*]] = ashr <4 x i32> [[TMP1]], [[TMP2]]
+; AVX1-NEXT: [[TMP4:%.*]] = lshr <4 x i32> [[TMP1]], [[TMP2]]
; AVX1-NEXT: [[TMP5:%.*]] = shufflevector <4 x i32> [[TMP3]], <4 x i32> [[TMP4]], <4 x i32> <i32 0, i32 1, i32 6, i32 7>
-; AVX1-NEXT: [[R0:%.*]] = insertelement <8 x i32> undef, i32 [[AB0]], i32 0
-; AVX1-NEXT: [[R1:%.*]] = insertelement <8 x i32> [[R0]], i32 [[AB1]], i32 1
-; AVX1-NEXT: [[R2:%.*]] = insertelement <8 x i32> [[R1]], i32 [[AB2]], i32 2
-; AVX1-NEXT: [[R3:%.*]] = insertelement <8 x i32> [[R2]], i32 [[AB3]], i32 3
-; AVX1-NEXT: [[TMP6:%.*]] = shufflevector <4 x i32> [[TMP5]], <4 x i32> undef, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef>
-; AVX1-NEXT: [[R71:%.*]] = shufflevector <8 x i32> [[R3]], <8 x i32> [[TMP6]], <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 8, i32 9, i32 10, i32 11>
-; AVX1-NEXT: ret <8 x i32> [[R71]]
+; AVX1-NEXT: [[TMP6:%.*]] = shufflevector <8 x i32> [[A]], <8 x i32> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
+; AVX1-NEXT: [[TMP7:%.*]] = shufflevector <8 x i32> [[B]], <8 x i32> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
+; AVX1-NEXT: [[TMP8:%.*]] = lshr <4 x i32> [[TMP6]], [[TMP7]]
+; AVX1-NEXT: [[TMP9:%.*]] = shl <4 x i32> [[TMP6]], [[TMP7]]
+; AVX1-NEXT: [[TMP10:%.*]] = shufflevector <4 x i32> [[TMP8]], <4 x i32> [[TMP9]], <4 x i32> <i32 0, i32 1, i32 6, i32 7>
+; AVX1-NEXT: [[R72:%.*]] = shufflevector <4 x i32> [[TMP5]], <4 x i32> [[TMP10]], <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+; AVX1-NEXT: ret <8 x i32> [[R72]]
;
; AVX2-LABEL: @ashr_lshr_shl_v8i32(
; AVX2-NEXT: [[TMP1:%.*]] = shufflevector <8 x i32> [[A:%.*]], <8 x i32> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
diff --git a/llvm/test/Transforms/SLPVectorizer/X86/powof2div.ll b/llvm/test/Transforms/SLPVectorizer/X86/powof2div.ll
index e6fb9cae52ecf..431f8b14e1f00 100644
--- a/llvm/test/Transforms/SLPVectorizer/X86/powof2div.ll
+++ b/llvm/test/Transforms/SLPVectorizer/X86/powof2div.ll
@@ -1,6 +1,6 @@
; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
-; RUN: opt < %s -basic-aa -slp-vectorizer -S -mtriple=x86_64-unknown-linux-gnu -mattr=+avx | FileCheck %s --check-prefixes=CHECK,AVX1
-; RUN: opt < %s -basic-aa -slp-vectorizer -S -mtriple=x86_64-unknown-linux-gnu -mattr=+avx2 | FileCheck %s --check-prefixes=CHECK,AVX2
+; RUN: opt < %s -basic-aa -slp-vectorizer -S -mtriple=x86_64-unknown-linux-gnu -mattr=+avx | FileCheck %s
+; RUN: opt < %s -basic-aa -slp-vectorizer -S -mtriple=x86_64-unknown-linux-gnu -mattr=+avx2 | FileCheck %s
define void @powof2div_uniform(i32* noalias nocapture %a, i32* noalias nocapture readonly %b, i32* noalias nocapture readonly %c){
; CHECK-LABEL: @powof2div_uniform(
@@ -58,59 +58,26 @@ entry:
}
define void @powof2div_nonuniform(i32* noalias nocapture %a, i32* noalias nocapture readonly %b, i32* noalias nocapture readonly %c){
-; AVX1-LABEL: @powof2div_nonuniform(
-; AVX1-NEXT: entry:
-; AVX1-NEXT: [[TMP0:%.*]] = load i32, i32* [[B:%.*]], align 4
-; AVX1-NEXT: [[TMP1:%.*]] = load i32, i32* [[C:%.*]], align 4
-; AVX1-NEXT: [[ADD:%.*]] = add nsw i32 [[TMP1]], [[TMP0]]
-; AVX1-NEXT: [[DIV:%.*]] = sdiv i32 [[ADD]], 2
-; AVX1-NEXT: store i32 [[DIV]], i32* [[A:%.*]], align 4
-; AVX1-NEXT: [[ARRAYIDX3:%.*]] = getelementptr inbounds i32, i32* [[B]], i64 1
-; AVX1-NEXT: [[TMP2:%.*]] = load i32, i32* [[ARRAYIDX3]], align 4
-; AVX1-NEXT: [[ARRAYIDX4:%.*]] = getelementptr inbounds i32, i32* [[C]], i64 1
-; AVX1-NEXT: [[TMP3:%.*]] = load i32, i32* [[ARRAYIDX4]], align 4
-; AVX1-NEXT: [[ADD5:%.*]] = add nsw i32 [[TMP3]], [[TMP2]]
-; AVX1-NEXT: [[DIV6:%.*]] = sdiv i32 [[ADD5]], 4
-; AVX1-NEXT: [[ARRAYIDX7:%.*]] = getelementptr inbounds i32, i32* [[A]], i64 1
-; AVX1-NEXT: store i32 [[DIV6]], i32* [[ARRAYIDX7]], align 4
-; AVX1-NEXT: [[ARRAYIDX8:%.*]] = getelementptr inbounds i32, i32* [[B]], i64 2
-; AVX1-NEXT: [[TMP4:%.*]] = load i32, i32* [[ARRAYIDX8]], align 4
-; AVX1-NEXT: [[ARRAYIDX9:%.*]] = getelementptr inbounds i32, i32* [[C]], i64 2
-; AVX1-NEXT: [[TMP5:%.*]] = load i32, i32* [[ARRAYIDX9]], align 4
-; AVX1-NEXT: [[ADD10:%.*]] = add nsw i32 [[TMP5]], [[TMP4]]
-; AVX1-NEXT: [[DIV11:%.*]] = sdiv i32 [[ADD10]], 8
-; AVX1-NEXT: [[ARRAYIDX12:%.*]] = getelementptr inbounds i32, i32* [[A]], i64 2
-; AVX1-NEXT: store i32 [[DIV11]], i32* [[ARRAYIDX12]], align 4
-; AVX1-NEXT: [[ARRAYIDX13:%.*]] = getelementptr inbounds i32, i32* [[B]], i64 3
-; AVX1-NEXT: [[TMP6:%.*]] = load i32, i32* [[ARRAYIDX13]], align 4
-; AVX1-NEXT: [[ARRAYIDX14:%.*]] = getelementptr inbounds i32, i32* [[C]], i64 3
-; AVX1-NEXT: [[TMP7:%.*]] = load i32, i32* [[ARRAYIDX14]], align 4
-; AVX1-NEXT: [[ADD15:%.*]] = add nsw i32 [[TMP7]], [[TMP6]]
-; AVX1-NEXT: [[DIV16:%.*]] = sdiv i32 [[ADD15]], 16
-; AVX1-NEXT: [[ARRAYIDX17:%.*]] = getelementptr inbounds i32, i32* [[A]], i64 3
-; AVX1-NEXT: store i32 [[DIV16]], i32* [[ARRAYIDX17]], align 4
-; AVX1-NEXT: ret void
-;
-; AVX2-LABEL: @powof2div_nonuniform(
-; AVX2-NEXT: entry:
-; AVX2-NEXT: [[ARRAYIDX3:%.*]] = getelementptr inbounds i32, i32* [[B:%.*]], i64 1
-; AVX2-NEXT: [[ARRAYIDX4:%.*]] = getelementptr inbounds i32, i32* [[C:%.*]], i64 1
-; AVX2-NEXT: [[ARRAYIDX7:%.*]] = getelementptr inbounds i32, i32* [[A:%.*]], i64 1
-; AVX2-NEXT: [[ARRAYIDX8:%.*]] = getelementptr inbounds i32, i32* [[B]], i64 2
-; AVX2-NEXT: [[ARRAYIDX9:%.*]] = getelementptr inbounds i32, i32* [[C]], i64 2
-; AVX2-NEXT: [[ARRAYIDX12:%.*]] = getelementptr inbounds i32, i32* [[A]], i64 2
-; AVX2-NEXT: [[ARRAYIDX13:%.*]] = getelementptr inbounds i32, i32* [[B]], i64 3
-; AVX2-NEXT: [[TMP0:%.*]] = bitcast i32* [[B]] to <4 x i32>*
-; AVX2-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* [[TMP0]], align 4
-; AVX2-NEXT: [[ARRAYIDX14:%.*]] = getelementptr inbounds i32, i32* [[C]], i64 3
-; AVX2-NEXT: [[TMP2:%.*]] = bitcast i32* [[C]] to <4 x i32>*
-; AVX2-NEXT: [[TMP3:%.*]] = load <4 x i32>, <4 x i32>* [[TMP2]], align 4
-; AVX2-NEXT: [[TMP4:%.*]] = add nsw <4 x i32> [[TMP3]], [[TMP1]]
-; AVX2-NEXT: [[TMP5:%.*]] = sdiv <4 x i32> [[TMP4]], <i32 2, i32 4, i32 8, i32 16>
-; AVX2-NEXT: [[ARRAYIDX17:%.*]] = getelementptr inbounds i32, i32* [[A]], i64 3
-; AVX2-NEXT: [[TMP6:%.*]] = bitcast i32* [[A]] to <4 x i32>*
-; AVX2-NEXT: store <4 x i32> [[TMP5]], <4 x i32>* [[TMP6]], align 4
-; AVX2-NEXT: ret void
+; CHECK-LABEL: @powof2div_nonuniform(
+; CHECK-NEXT: entry:
+; CHECK-NEXT: [[ARRAYIDX3:%.*]] = getelementptr inbounds i32, i32* [[B:%.*]], i64 1
+; CHECK-NEXT: [[ARRAYIDX4:%.*]] = getelementptr inbounds i32, i32* [[C:%.*]], i64 1
+; CHECK-NEXT: [[ARRAYIDX7:%.*]] = getelementptr inbounds i32, i32* [[A:%.*]], i64 1
+; CHECK-NEXT: [[ARRAYIDX8:%.*]] = getelementptr inbounds i32, i32* [[B]], i64 2
+; CHECK-NEXT: [[ARRAYIDX9:%.*]] = getelementptr inbounds i32, i32* [[C]], i64 2
+; CHECK-NEXT: [[ARRAYIDX12:%.*]] = getelementptr inbounds i32, i32* [[A]], i64 2
+; CHECK-NEXT: [[ARRAYIDX13:%.*]] = getelementptr inbounds i32, i32* [[B]], i64 3
+; CHECK-NEXT: [[TMP0:%.*]] = bitcast i32* [[B]] to <4 x i32>*
+; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* [[TMP0]], align 4
+; CHECK-NEXT: [[ARRAYIDX14:%.*]] = getelementptr inbounds i32, i32* [[C]], i64 3
+; CHECK-NEXT: [[TMP2:%.*]] = bitcast i32* [[C]] to <4 x i32>*
+; CHECK-NEXT: [[TMP3:%.*]] = load <4 x i32>, <4 x i32>* [[TMP2]], align 4
+; CHECK-NEXT: [[TMP4:%.*]] = add nsw <4 x i32> [[TMP3]], [[TMP1]]
+; CHECK-NEXT: [[TMP5:%.*]] = sdiv <4 x i32> [[TMP4]], <i32 2, i32 4, i32 8, i32 16>
+; CHECK-NEXT: [[ARRAYIDX17:%.*]] = getelementptr inbounds i32, i32* [[A]], i64 3
+; CHECK-NEXT: [[TMP6:%.*]] = bitcast i32* [[A]] to <4 x i32>*
+; CHECK-NEXT: store <4 x i32> [[TMP5]], <4 x i32>* [[TMP6]], align 4
+; CHECK-NEXT: ret void
;
entry:
%0 = load i32, i32* %b, align 4
diff --git a/llvm/test/Transforms/SLPVectorizer/X86/shift-ashr.ll b/llvm/test/Transforms/SLPVectorizer/X86/shift-ashr.ll
index f1c8fae162434..927da3fac7b8a 100644
--- a/llvm/test/Transforms/SLPVectorizer/X86/shift-ashr.ll
+++ b/llvm/test/Transforms/SLPVectorizer/X86/shift-ashr.ll
@@ -1,10 +1,10 @@
; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt < %s -mtriple=x86_64-unknown -basic-aa -slp-vectorizer -S | FileCheck %s --check-prefix=SSE
-; RUN: opt < %s -mtriple=x86_64-unknown -mcpu=corei7-avx -basic-aa -slp-vectorizer -S | FileCheck %s --check-prefix=AVX1
-; RUN: opt < %s -mtriple=x86_64-unknown -mcpu=core-avx2 -basic-aa -slp-vectorizer -S | FileCheck %s --check-prefix=AVX2
+; RUN: opt < %s -mtriple=x86_64-unknown -mcpu=corei7-avx -basic-aa -slp-vectorizer -S | FileCheck %s --check-prefixes=AVX,AVX1
+; RUN: opt < %s -mtriple=x86_64-unknown -mcpu=core-avx2 -basic-aa -slp-vectorizer -S | FileCheck %s --check-prefixes=AVX,AVX2
; RUN: opt < %s -mtriple=x86_64-unknown -mcpu=knl -basic-aa -slp-vectorizer -S | FileCheck %s --check-prefix=AVX512
; RUN: opt < %s -mtriple=x86_64-unknown -mcpu=skx -mattr=-prefer-256-bit -basic-aa -slp-vectorizer -S | FileCheck %s --check-prefix=AVX512
-; RUN: opt < %s -mtriple=x86_64-unknown -mcpu=skx -mattr=+prefer-256-bit -basic-aa -slp-vectorizer -S | FileCheck %s --check-prefix=AVX2
+; RUN: opt < %s -mtriple=x86_64-unknown -mcpu=skx -mattr=+prefer-256-bit -basic-aa -slp-vectorizer -S | FileCheck %s --check-prefixes=AVX,AVX2
; RUN: opt < %s -mtriple=x86_64-unknown -mcpu=bdver4 -basic-aa -slp-vectorizer -S | FileCheck %s --check-prefix=XOP
@a64 = common global [8 x i64] zeroinitializer, align 64
@@ -223,35 +223,16 @@ define void @ashr_v16i32() {
; SSE-NEXT: store i32 [[R15]], i32* getelementptr inbounds ([16 x i32], [16 x i32]* @c32, i32 0, i64 15), align 4
; SSE-NEXT: ret void
;
-; AVX1-LABEL: @ashr_v16i32(
-; AVX1-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([16 x i32]* @a32 to <4 x i32>*), align 4
-; AVX1-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* bitcast (i32* getelementptr inbounds ([16 x i32], [16 x i32]* @a32, i32 0, i64 4) to <4 x i32>*), align 4
-; AVX1-NEXT: [[TMP3:%.*]] = load <4 x i32>, <4 x i32>* bitcast (i32* getelementptr inbounds ([16 x i32], [16 x i32]* @a32, i32 0, i64 8) to <4 x i32>*), align 4
-; AVX1-NEXT: [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* bitcast (i32* getelementptr inbounds ([16 x i32], [16 x i32]* @a32, i32 0, i64 12) to <4 x i32>*), align 4
-; AVX1-NEXT: [[TMP5:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([16 x i32]* @b32 to <4 x i32>*), align 4
-; AVX1-NEXT: [[TMP6:%.*]] = load <4 x i32>, <4 x i32>* bitcast (i32* getelementptr inbounds ([16 x i32], [16 x i32]* @b32, i32 0, i64 4) to <4 x i32>*), align 4
-; AVX1-NEXT: [[TMP7:%.*]] = load <4 x i32>, <4 x i32>* bitcast (i32* getelementptr inbounds ([16 x i32], [16 x i32]* @b32, i32 0, i64 8) to <4 x i32>*), align 4
-; AVX1-NEXT: [[TMP8:%.*]] = load <4 x i32>, <4 x i32>* bitcast (i32* getelementptr inbounds ([16 x i32], [16 x i32]* @b32, i32 0, i64 12) to <4 x i32>*), align 4
-; AVX1-NEXT: [[TMP9:%.*]] = ashr <4 x i32> [[TMP1]], [[TMP5]]
-; AVX1-NEXT: [[TMP10:%.*]] = ashr <4 x i32> [[TMP2]], [[TMP6]]
-; AVX1-NEXT: [[TMP11:%.*]] = ashr <4 x i32> [[TMP3]], [[TMP7]]
-; AVX1-NEXT: [[TMP12:%.*]] = ashr <4 x i32> [[TMP4]], [[TMP8]]
-; AVX1-NEXT: store <4 x i32> [[TMP9]], <4 x i32>* bitcast ([16 x i32]* @c32 to <4 x i32>*), align 4
-; AVX1-NEXT: store <4 x i32> [[TMP10]], <4 x i32>* bitcast (i32* getelementptr inbounds ([16 x i32], [16 x i32]* @c32, i32 0, i64 4) to <4 x i32>*), align 4
-; AVX1-NEXT: store <4 x i32> [[TMP11]], <4 x i32>* bitcast (i32* getelementptr inbounds ([16 x i32], [16 x i32]* @c32, i32 0, i64 8) to <4 x i32>*), align 4
-; AVX1-NEXT: store <4 x i32> [[TMP12]], <4 x i32>* bitcast (i32* getelementptr inbounds ([16 x i32], [16 x i32]* @c32, i32 0, i64 12) to <4 x i32>*), align 4
-; AVX1-NEXT: ret void
-;
-; AVX2-LABEL: @ashr_v16i32(
-; AVX2-NEXT: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([16 x i32]* @a32 to <8 x i32>*), align 4
-; AVX2-NEXT: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* bitcast (i32* getelementptr inbounds ([16 x i32], [16 x i32]* @a32, i32 0, i64 8) to <8 x i32>*), align 4
-; AVX2-NEXT: [[TMP3:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([16 x i32]* @b32 to <8 x i32>*), align 4
-; AVX2-NEXT: [[TMP4:%.*]] = load <8 x i32>, <8 x i32>* bitcast (i32* getelementptr inbounds ([16 x i32], [16 x i32]* @b32, i32 0, i64 8) to <8 x i32>*), align 4
-; AVX2-NEXT: [[TMP5:%.*]] = ashr <8 x i32> [[TMP1]], [[TMP3]]
-; AVX2-NEXT: [[TMP6:%.*]] = ashr <8 x i32> [[TMP2]], [[TMP4]]
-; AVX2-NEXT: store <8 x i32> [[TMP5]], <8 x i32>* bitcast ([16 x i32]* @c32 to <8 x i32>*), align 4
-; AVX2-NEXT: store <8 x i32> [[TMP6]], <8 x i32>* bitcast (i32* getelementptr inbounds ([16 x i32], [16 x i32]* @c32, i32 0, i64 8) to <8 x i32>*), align 4
-; AVX2-NEXT: ret void
+; AVX-LABEL: @ashr_v16i32(
+; AVX-NEXT: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([16 x i32]* @a32 to <8 x i32>*), align 4
+; AVX-NEXT: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* bitcast (i32* getelementptr inbounds ([16 x i32], [16 x i32]* @a32, i32 0, i64 8) to <8 x i32>*), align 4
+; AVX-NEXT: [[TMP3:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([16 x i32]* @b32 to <8 x i32>*), align 4
+; AVX-NEXT: [[TMP4:%.*]] = load <8 x i32>, <8 x i32>* bitcast (i32* getelementptr inbounds ([16 x i32], [16 x i32]* @b32, i32 0, i64 8) to <8 x i32>*), align 4
+; AVX-NEXT: [[TMP5:%.*]] = ashr <8 x i32> [[TMP1]], [[TMP3]]
+; AVX-NEXT: [[TMP6:%.*]] = ashr <8 x i32> [[TMP2]], [[TMP4]]
+; AVX-NEXT: store <8 x i32> [[TMP5]], <8 x i32>* bitcast ([16 x i32]* @c32 to <8 x i32>*), align 4
+; AVX-NEXT: store <8 x i32> [[TMP6]], <8 x i32>* bitcast (i32* getelementptr inbounds ([16 x i32], [16 x i32]* @c32, i32 0, i64 8) to <8 x i32>*), align 4
+; AVX-NEXT: ret void
;
; AVX512-LABEL: @ashr_v16i32(
; AVX512-NEXT: [[TMP1:%.*]] = load <16 x i32>, <16 x i32>* bitcast ([16 x i32]* @a32 to <16 x i32>*), align 4
diff --git a/llvm/test/Transforms/SLPVectorizer/X86/shift-lshr.ll b/llvm/test/Transforms/SLPVectorizer/X86/shift-lshr.ll
index 858489cbf5b93..95ee1229a98a7 100644
--- a/llvm/test/Transforms/SLPVectorizer/X86/shift-lshr.ll
+++ b/llvm/test/Transforms/SLPVectorizer/X86/shift-lshr.ll
@@ -1,10 +1,10 @@
; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt < %s -mtriple=x86_64-unknown -basic-aa -slp-vectorizer -S | FileCheck %s --check-prefix=SSE
-; RUN: opt < %s -mtriple=x86_64-unknown -mcpu=corei7-avx -basic-aa -slp-vectorizer -S | FileCheck %s --check-prefix=AVX --check-prefix=AVX1
-; RUN: opt < %s -mtriple=x86_64-unknown -mcpu=core-avx2 -basic-aa -slp-vectorizer -S | FileCheck %s --check-prefix=AVX --check-prefix=AVX2
+; RUN: opt < %s -mtriple=x86_64-unknown -mcpu=corei7-avx -basic-aa -slp-vectorizer -S | FileCheck %s --check-prefixes=AVX
+; RUN: opt < %s -mtriple=x86_64-unknown -mcpu=core-avx2 -basic-aa -slp-vectorizer -S | FileCheck %s --check-prefixes=AVX
; RUN: opt < %s -mtriple=x86_64-unknown -mcpu=knl -basic-aa -slp-vectorizer -S | FileCheck %s --check-prefix=AVX512
; RUN: opt < %s -mtriple=x86_64-unknown -mcpu=skx -mattr=-prefer-256-bit -basic-aa -slp-vectorizer -S | FileCheck %s --check-prefix=AVX512
-; RUN: opt < %s -mtriple=x86_64-unknown -mcpu=skx -mattr=+prefer-256-bit -basic-aa -slp-vectorizer -S | FileCheck %s --check-prefix=AVX --check-prefix=AVX2
+; RUN: opt < %s -mtriple=x86_64-unknown -mcpu=skx -mattr=+prefer-256-bit -basic-aa -slp-vectorizer -S | FileCheck %s --check-prefixes=AVX
; RUN: opt < %s -mtriple=x86_64-unknown -mcpu=bdver4 -basic-aa -slp-vectorizer -S | FileCheck %s --check-prefix=XOP
@a64 = common global [8 x i64] zeroinitializer, align 64
@@ -40,35 +40,16 @@ define void @lshr_v8i64() {
; SSE-NEXT: store <2 x i64> [[TMP12]], <2 x i64>* bitcast (i64* getelementptr inbounds ([8 x i64], [8 x i64]* @c64, i32 0, i64 6) to <2 x i64>*), align 8
; SSE-NEXT: ret void
;
-; AVX1-LABEL: @lshr_v8i64(
-; AVX1-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([8 x i64]* @a64 to <2 x i64>*), align 8
-; AVX1-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* bitcast (i64* getelementptr inbounds ([8 x i64], [8 x i64]* @a64, i32 0, i64 2) to <2 x i64>*), align 8
-; AVX1-NEXT: [[TMP3:%.*]] = load <2 x i64>, <2 x i64>* bitcast (i64* getelementptr inbounds ([8 x i64], [8 x i64]* @a64, i32 0, i64 4) to <2 x i64>*), align 8
-; AVX1-NEXT: [[TMP4:%.*]] = load <2 x i64>, <2 x i64>* bitcast (i64* getelementptr inbounds ([8 x i64], [8 x i64]* @a64, i32 0, i64 6) to <2 x i64>*), align 8
-; AVX1-NEXT: [[TMP5:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([8 x i64]* @b64 to <2 x i64>*), align 8
-; AVX1-NEXT: [[TMP6:%.*]] = load <2 x i64>, <2 x i64>* bitcast (i64* getelementptr inbounds ([8 x i64], [8 x i64]* @b64, i32 0, i64 2) to <2 x i64>*), align 8
-; AVX1-NEXT: [[TMP7:%.*]] = load <2 x i64>, <2 x i64>* bitcast (i64* getelementptr inbounds ([8 x i64], [8 x i64]* @b64, i32 0, i64 4) to <2 x i64>*), align 8
-; AVX1-NEXT: [[TMP8:%.*]] = load <2 x i64>, <2 x i64>* bitcast (i64* getelementptr inbounds ([8 x i64], [8 x i64]* @b64, i32 0, i64 6) to <2 x i64>*), align 8
-; AVX1-NEXT: [[TMP9:%.*]] = lshr <2 x i64> [[TMP1]], [[TMP5]]
-; AVX1-NEXT: [[TMP10:%.*]] = lshr <2 x i64> [[TMP2]], [[TMP6]]
-; AVX1-NEXT: [[TMP11:%.*]] = lshr <2 x i64> [[TMP3]], [[TMP7]]
-; AVX1-NEXT: [[TMP12:%.*]] = lshr <2 x i64> [[TMP4]], [[TMP8]]
-; AVX1-NEXT: store <2 x i64> [[TMP9]], <2 x i64>* bitcast ([8 x i64]* @c64 to <2 x i64>*), align 8
-; AVX1-NEXT: store <2 x i64> [[TMP10]], <2 x i64>* bitcast (i64* getelementptr inbounds ([8 x i64], [8 x i64]* @c64, i32 0, i64 2) to <2 x i64>*), align 8
-; AVX1-NEXT: store <2 x i64> [[TMP11]], <2 x i64>* bitcast (i64* getelementptr inbounds ([8 x i64], [8 x i64]* @c64, i32 0, i64 4) to <2 x i64>*), align 8
-; AVX1-NEXT: store <2 x i64> [[TMP12]], <2 x i64>* bitcast (i64* getelementptr inbounds ([8 x i64], [8 x i64]* @c64, i32 0, i64 6) to <2 x i64>*), align 8
-; AVX1-NEXT: ret void
-;
-; AVX2-LABEL: @lshr_v8i64(
-; AVX2-NEXT: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([8 x i64]* @a64 to <4 x i64>*), align 8
-; AVX2-NEXT: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* bitcast (i64* getelementptr inbounds ([8 x i64], [8 x i64]* @a64, i32 0, i64 4) to <4 x i64>*), align 8
-; AVX2-NEXT: [[TMP3:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([8 x i64]* @b64 to <4 x i64>*), align 8
-; AVX2-NEXT: [[TMP4:%.*]] = load <4 x i64>, <4 x i64>* bitcast (i64* getelementptr inbounds ([8 x i64], [8 x i64]* @b64, i32 0, i64 4) to <4 x i64>*), align 8
-; AVX2-NEXT: [[TMP5:%.*]] = lshr <4 x i64> [[TMP1]], [[TMP3]]
-; AVX2-NEXT: [[TMP6:%.*]] = lshr <4 x i64> [[TMP2]], [[TMP4]]
-; AVX2-NEXT: store <4 x i64> [[TMP5]], <4 x i64>* bitcast ([8 x i64]* @c64 to <4 x i64>*), align 8
-; AVX2-NEXT: store <4 x i64> [[TMP6]], <4 x i64>* bitcast (i64* getelementptr inbounds ([8 x i64], [8 x i64]* @c64, i32 0, i64 4) to <4 x i64>*), align 8
-; AVX2-NEXT: ret void
+; AVX-LABEL: @lshr_v8i64(
+; AVX-NEXT: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([8 x i64]* @a64 to <4 x i64>*), align 8
+; AVX-NEXT: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* bitcast (i64* getelementptr inbounds ([8 x i64], [8 x i64]* @a64, i32 0, i64 4) to <4 x i64>*), align 8
+; AVX-NEXT: [[TMP3:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([8 x i64]* @b64 to <4 x i64>*), align 8
+; AVX-NEXT: [[TMP4:%.*]] = load <4 x i64>, <4 x i64>* bitcast (i64* getelementptr inbounds ([8 x i64], [8 x i64]* @b64, i32 0, i64 4) to <4 x i64>*), align 8
+; AVX-NEXT: [[TMP5:%.*]] = lshr <4 x i64> [[TMP1]], [[TMP3]]
+; AVX-NEXT: [[TMP6:%.*]] = lshr <4 x i64> [[TMP2]], [[TMP4]]
+; AVX-NEXT: store <4 x i64> [[TMP5]], <4 x i64>* bitcast ([8 x i64]* @c64 to <4 x i64>*), align 8
+; AVX-NEXT: store <4 x i64> [[TMP6]], <4 x i64>* bitcast (i64* getelementptr inbounds ([8 x i64], [8 x i64]* @c64, i32 0, i64 4) to <4 x i64>*), align 8
+; AVX-NEXT: ret void
;
; AVX512-LABEL: @lshr_v8i64(
; AVX512-NEXT: [[TMP1:%.*]] = load <8 x i64>, <8 x i64>* bitcast ([8 x i64]* @a64 to <8 x i64>*), align 8
diff --git a/llvm/test/Transforms/SLPVectorizer/X86/shift-shl.ll b/llvm/test/Transforms/SLPVectorizer/X86/shift-shl.ll
index 371593d8c626b..5fa5659805c16 100644
--- a/llvm/test/Transforms/SLPVectorizer/X86/shift-shl.ll
+++ b/llvm/test/Transforms/SLPVectorizer/X86/shift-shl.ll
@@ -1,10 +1,10 @@
; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt < %s -mtriple=x86_64-unknown -basic-aa -slp-vectorizer -S | FileCheck %s --check-prefix=SSE
-; RUN: opt < %s -mtriple=x86_64-unknown -mcpu=corei7-avx -basic-aa -slp-vectorizer -S | FileCheck %s --check-prefix=AVX --check-prefix=AVX1
-; RUN: opt < %s -mtriple=x86_64-unknown -mcpu=core-avx2 -basic-aa -slp-vectorizer -S | FileCheck %s --check-prefix=AVX --check-prefix=AVX2
+; RUN: opt < %s -mtriple=x86_64-unknown -mcpu=corei7-avx -basic-aa -slp-vectorizer -S | FileCheck %s --check-prefixes=AVX
+; RUN: opt < %s -mtriple=x86_64-unknown -mcpu=core-avx2 -basic-aa -slp-vectorizer -S | FileCheck %s --check-prefixes=AVX
; RUN: opt < %s -mtriple=x86_64-unknown -mcpu=knl -basic-aa -slp-vectorizer -S | FileCheck %s --check-prefix=AVX512
; RUN: opt < %s -mtriple=x86_64-unknown -mcpu=skx -mattr=-prefer-256-bit -basic-aa -slp-vectorizer -S | FileCheck %s --check-prefix=AVX512
-; RUN: opt < %s -mtriple=x86_64-unknown -mcpu=skx -mattr=+prefer-256-bit -basic-aa -slp-vectorizer -S | FileCheck %s --check-prefix=AVX --check-prefix=AVX2
+; RUN: opt < %s -mtriple=x86_64-unknown -mcpu=skx -mattr=+prefer-256-bit -basic-aa -slp-vectorizer -S | FileCheck %s --check-prefixes=AVX
; RUN: opt < %s -mtriple=x86_64-unknown -mcpu=bdver4 -basic-aa -slp-vectorizer -S | FileCheck %s --check-prefix=XOP
@a64 = common global [8 x i64] zeroinitializer, align 64
@@ -40,35 +40,16 @@ define void @shl_v8i64() {
; SSE-NEXT: store <2 x i64> [[TMP12]], <2 x i64>* bitcast (i64* getelementptr inbounds ([8 x i64], [8 x i64]* @c64, i32 0, i64 6) to <2 x i64>*), align 8
; SSE-NEXT: ret void
;
-; AVX1-LABEL: @shl_v8i64(
-; AVX1-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([8 x i64]* @a64 to <2 x i64>*), align 8
-; AVX1-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* bitcast (i64* getelementptr inbounds ([8 x i64], [8 x i64]* @a64, i32 0, i64 2) to <2 x i64>*), align 8
-; AVX1-NEXT: [[TMP3:%.*]] = load <2 x i64>, <2 x i64>* bitcast (i64* getelementptr inbounds ([8 x i64], [8 x i64]* @a64, i32 0, i64 4) to <2 x i64>*), align 8
-; AVX1-NEXT: [[TMP4:%.*]] = load <2 x i64>, <2 x i64>* bitcast (i64* getelementptr inbounds ([8 x i64], [8 x i64]* @a64, i32 0, i64 6) to <2 x i64>*), align 8
-; AVX1-NEXT: [[TMP5:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([8 x i64]* @b64 to <2 x i64>*), align 8
-; AVX1-NEXT: [[TMP6:%.*]] = load <2 x i64>, <2 x i64>* bitcast (i64* getelementptr inbounds ([8 x i64], [8 x i64]* @b64, i32 0, i64 2) to <2 x i64>*), align 8
-; AVX1-NEXT: [[TMP7:%.*]] = load <2 x i64>, <2 x i64>* bitcast (i64* getelementptr inbounds ([8 x i64], [8 x i64]* @b64, i32 0, i64 4) to <2 x i64>*), align 8
-; AVX1-NEXT: [[TMP8:%.*]] = load <2 x i64>, <2 x i64>* bitcast (i64* getelementptr inbounds ([8 x i64], [8 x i64]* @b64, i32 0, i64 6) to <2 x i64>*), align 8
-; AVX1-NEXT: [[TMP9:%.*]] = shl <2 x i64> [[TMP1]], [[TMP5]]
-; AVX1-NEXT: [[TMP10:%.*]] = shl <2 x i64> [[TMP2]], [[TMP6]]
-; AVX1-NEXT: [[TMP11:%.*]] = shl <2 x i64> [[TMP3]], [[TMP7]]
-; AVX1-NEXT: [[TMP12:%.*]] = shl <2 x i64> [[TMP4]], [[TMP8]]
-; AVX1-NEXT: store <2 x i64> [[TMP9]], <2 x i64>* bitcast ([8 x i64]* @c64 to <2 x i64>*), align 8
-; AVX1-NEXT: store <2 x i64> [[TMP10]], <2 x i64>* bitcast (i64* getelementptr inbounds ([8 x i64], [8 x i64]* @c64, i32 0, i64 2) to <2 x i64>*), align 8
-; AVX1-NEXT: store <2 x i64> [[TMP11]], <2 x i64>* bitcast (i64* getelementptr inbounds ([8 x i64], [8 x i64]* @c64, i32 0, i64 4) to <2 x i64>*), align 8
-; AVX1-NEXT: store <2 x i64> [[TMP12]], <2 x i64>* bitcast (i64* getelementptr inbounds ([8 x i64], [8 x i64]* @c64, i32 0, i64 6) to <2 x i64>*), align 8
-; AVX1-NEXT: ret void
-;
-; AVX2-LABEL: @shl_v8i64(
-; AVX2-NEXT: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([8 x i64]* @a64 to <4 x i64>*), align 8
-; AVX2-NEXT: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* bitcast (i64* getelementptr inbounds ([8 x i64], [8 x i64]* @a64, i32 0, i64 4) to <4 x i64>*), align 8
-; AVX2-NEXT: [[TMP3:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([8 x i64]* @b64 to <4 x i64>*), align 8
-; AVX2-NEXT: [[TMP4:%.*]] = load <4 x i64>, <4 x i64>* bitcast (i64* getelementptr inbounds ([8 x i64], [8 x i64]* @b64, i32 0, i64 4) to <4 x i64>*), align 8
-; AVX2-NEXT: [[TMP5:%.*]] = shl <4 x i64> [[TMP1]], [[TMP3]]
-; AVX2-NEXT: [[TMP6:%.*]] = shl <4 x i64> [[TMP2]], [[TMP4]]
-; AVX2-NEXT: store <4 x i64> [[TMP5]], <4 x i64>* bitcast ([8 x i64]* @c64 to <4 x i64>*), align 8
-; AVX2-NEXT: store <4 x i64> [[TMP6]], <4 x i64>* bitcast (i64* getelementptr inbounds ([8 x i64], [8 x i64]* @c64, i32 0, i64 4) to <4 x i64>*), align 8
-; AVX2-NEXT: ret void
+; AVX-LABEL: @shl_v8i64(
+; AVX-NEXT: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([8 x i64]* @a64 to <4 x i64>*), align 8
+; AVX-NEXT: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* bitcast (i64* getelementptr inbounds ([8 x i64], [8 x i64]* @a64, i32 0, i64 4) to <4 x i64>*), align 8
+; AVX-NEXT: [[TMP3:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([8 x i64]* @b64 to <4 x i64>*), align 8
+; AVX-NEXT: [[TMP4:%.*]] = load <4 x i64>, <4 x i64>* bitcast (i64* getelementptr inbounds ([8 x i64], [8 x i64]* @b64, i32 0, i64 4) to <4 x i64>*), align 8
+; AVX-NEXT: [[TMP5:%.*]] = shl <4 x i64> [[TMP1]], [[TMP3]]
+; AVX-NEXT: [[TMP6:%.*]] = shl <4 x i64> [[TMP2]], [[TMP4]]
+; AVX-NEXT: store <4 x i64> [[TMP5]], <4 x i64>* bitcast ([8 x i64]* @c64 to <4 x i64>*), align 8
+; AVX-NEXT: store <4 x i64> [[TMP6]], <4 x i64>* bitcast (i64* getelementptr inbounds ([8 x i64], [8 x i64]* @c64, i32 0, i64 4) to <4 x i64>*), align 8
+; AVX-NEXT: ret void
;
; AVX512-LABEL: @shl_v8i64(
; AVX512-NEXT: [[TMP1:%.*]] = load <8 x i64>, <8 x i64>* bitcast ([8 x i64]* @a64 to <8 x i64>*), align 8
More information about the llvm-commits
mailing list