[llvm] 6ba1427 - [CostModel][X86] Add cost kinds test coverage for fp arithmetic operators
Simon Pilgrim via llvm-commits
llvm-commits at lists.llvm.org
Sat Aug 13 08:43:42 PDT 2022
Author: Simon Pilgrim
Date: 2022-08-13T16:42:43+01:00
New Revision: 6ba14272255e492d96250979804423e678f2de40
URL: https://github.com/llvm/llvm-project/commit/6ba14272255e492d96250979804423e678f2de40
DIFF: https://github.com/llvm/llvm-project/commit/6ba14272255e492d96250979804423e678f2de40.diff
LOG: [CostModel][X86] Add cost kinds test coverage for fp arithmetic operators
Added:
llvm/test/Analysis/CostModel/X86/arith-fp-costkinds.ll
Modified:
Removed:
################################################################################
diff --git a/llvm/test/Analysis/CostModel/X86/arith-fp-costkinds.ll b/llvm/test/Analysis/CostModel/X86/arith-fp-costkinds.ll
new file mode 100644
index 000000000000..e20dde0797fc
--- /dev/null
+++ b/llvm/test/Analysis/CostModel/X86/arith-fp-costkinds.ll
@@ -0,0 +1,660 @@
+; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py
+; RUN: opt -mtriple=x86_64-- -passes="print<cost-model>" 2>&1 -disable-output -mattr=+sse2 -cost-kind=latency < %s | FileCheck %s --check-prefixes=LATE
+; RUN: opt -mtriple=x86_64-- -passes="print<cost-model>" 2>&1 -disable-output -mattr=+sse2 -cost-kind=code-size < %s | FileCheck %s --check-prefixes=SIZE,SSE-SIZE
+; RUN: opt -mtriple=x86_64-- -passes="print<cost-model>" 2>&1 -disable-output -mattr=+sse2 -cost-kind=size-latency < %s | FileCheck %s --check-prefixes=SIZE_LATE,SSE-SIZE_LATE
+;
+; RUN: opt -mtriple=x86_64-- -passes="print<cost-model>" 2>&1 -disable-output -mattr=+avx -cost-kind=latency < %s | FileCheck %s --check-prefixes=LATE
+; RUN: opt -mtriple=x86_64-- -passes="print<cost-model>" 2>&1 -disable-output -mattr=+avx -cost-kind=code-size < %s | FileCheck %s --check-prefixes=SIZE,AVX-SIZE
+; RUN: opt -mtriple=x86_64-- -passes="print<cost-model>" 2>&1 -disable-output -mattr=+avx -cost-kind=size-latency < %s | FileCheck %s --check-prefixes=SIZE_LATE,AVX-SIZE_LATE,AVX1-SIZE_LATE
+;
+; RUN: opt -mtriple=x86_64-- -passes="print<cost-model>" 2>&1 -disable-output -mattr=+avx2 -cost-kind=latency < %s | FileCheck %s --check-prefixes=LATE
+; RUN: opt -mtriple=x86_64-- -passes="print<cost-model>" 2>&1 -disable-output -mattr=+avx2 -cost-kind=code-size < %s | FileCheck %s --check-prefixes=SIZE,AVX-SIZE
+; RUN: opt -mtriple=x86_64-- -passes="print<cost-model>" 2>&1 -disable-output -mattr=+avx2 -cost-kind=size-latency < %s | FileCheck %s --check-prefixes=SIZE_LATE,AVX-SIZE_LATE,AVX2-SIZE_LATE
+
+target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128"
+target triple = "x86_64-apple-macosx10.8.0"
+
+define i32 @fadd(i32 %arg) {
+; LATE-LABEL: 'fadd'
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %F32 = fadd float undef, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4F32 = fadd <4 x float> undef, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V8F32 = fadd <8 x float> undef, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V16F32 = fadd <16 x float> undef, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %F64 = fadd double undef, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2F64 = fadd <2 x double> undef, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4F64 = fadd <4 x double> undef, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V8F64 = fadd <8 x double> undef, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+; SIZE-LABEL: 'fadd'
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %F32 = fadd float undef, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V4F32 = fadd <4 x float> undef, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V8F32 = fadd <8 x float> undef, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V16F32 = fadd <16 x float> undef, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %F64 = fadd double undef, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V2F64 = fadd <2 x double> undef, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V4F64 = fadd <4 x double> undef, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V8F64 = fadd <8 x double> undef, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+; SIZE_LATE-LABEL: 'fadd'
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %F32 = fadd float undef, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V4F32 = fadd <4 x float> undef, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V8F32 = fadd <8 x float> undef, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V16F32 = fadd <16 x float> undef, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %F64 = fadd double undef, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V2F64 = fadd <2 x double> undef, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V4F64 = fadd <4 x double> undef, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V8F64 = fadd <8 x double> undef, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+ %F32 = fadd float undef, undef
+ %V4F32 = fadd <4 x float> undef, undef
+ %V8F32 = fadd <8 x float> undef, undef
+ %V16F32 = fadd <16 x float> undef, undef
+
+ %F64 = fadd double undef, undef
+ %V2F64 = fadd <2 x double> undef, undef
+ %V4F64 = fadd <4 x double> undef, undef
+ %V8F64 = fadd <8 x double> undef, undef
+
+ ret i32 undef
+}
+
+define i32 @fsub(i32 %arg) {
+; LATE-LABEL: 'fsub'
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %F32 = fsub float undef, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4F32 = fsub <4 x float> undef, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V8F32 = fsub <8 x float> undef, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V16F32 = fsub <16 x float> undef, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %F64 = fsub double undef, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2F64 = fsub <2 x double> undef, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4F64 = fsub <4 x double> undef, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V8F64 = fsub <8 x double> undef, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+; SIZE-LABEL: 'fsub'
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %F32 = fsub float undef, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V4F32 = fsub <4 x float> undef, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V8F32 = fsub <8 x float> undef, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V16F32 = fsub <16 x float> undef, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %F64 = fsub double undef, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V2F64 = fsub <2 x double> undef, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V4F64 = fsub <4 x double> undef, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V8F64 = fsub <8 x double> undef, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+; SIZE_LATE-LABEL: 'fsub'
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %F32 = fsub float undef, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V4F32 = fsub <4 x float> undef, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V8F32 = fsub <8 x float> undef, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V16F32 = fsub <16 x float> undef, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %F64 = fsub double undef, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V2F64 = fsub <2 x double> undef, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V4F64 = fsub <4 x double> undef, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V8F64 = fsub <8 x double> undef, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+ %F32 = fsub float undef, undef
+ %V4F32 = fsub <4 x float> undef, undef
+ %V8F32 = fsub <8 x float> undef, undef
+ %V16F32 = fsub <16 x float> undef, undef
+
+ %F64 = fsub double undef, undef
+ %V2F64 = fsub <2 x double> undef, undef
+ %V4F64 = fsub <4 x double> undef, undef
+ %V8F64 = fsub <8 x double> undef, undef
+
+ ret i32 undef
+}
+
+define i32 @fneg_idiom(i32 %arg) {
+; LATE-LABEL: 'fneg_idiom'
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %F32 = fsub float -0.000000e+00, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4F32 = fsub <4 x float> <float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00>, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V8F32 = fsub <8 x float> <float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00>, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V16F32 = fsub <16 x float> <float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00>, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %F64 = fsub double -0.000000e+00, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2F64 = fsub <2 x double> <double -0.000000e+00, double -0.000000e+00>, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4F64 = fsub <4 x double> <double -0.000000e+00, double -0.000000e+00, double -0.000000e+00, double -0.000000e+00>, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V8F64 = fsub <8 x double> <double -0.000000e+00, double -0.000000e+00, double -0.000000e+00, double -0.000000e+00, double -0.000000e+00, double -0.000000e+00, double -0.000000e+00, double -0.000000e+00>, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+; SIZE-LABEL: 'fneg_idiom'
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %F32 = fsub float -0.000000e+00, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V4F32 = fsub <4 x float> <float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00>, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V8F32 = fsub <8 x float> <float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00>, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V16F32 = fsub <16 x float> <float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00>, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %F64 = fsub double -0.000000e+00, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V2F64 = fsub <2 x double> <double -0.000000e+00, double -0.000000e+00>, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V4F64 = fsub <4 x double> <double -0.000000e+00, double -0.000000e+00, double -0.000000e+00, double -0.000000e+00>, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V8F64 = fsub <8 x double> <double -0.000000e+00, double -0.000000e+00, double -0.000000e+00, double -0.000000e+00, double -0.000000e+00, double -0.000000e+00, double -0.000000e+00, double -0.000000e+00>, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+; SIZE_LATE-LABEL: 'fneg_idiom'
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %F32 = fsub float -0.000000e+00, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V4F32 = fsub <4 x float> <float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00>, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V8F32 = fsub <8 x float> <float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00>, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V16F32 = fsub <16 x float> <float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00>, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %F64 = fsub double -0.000000e+00, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V2F64 = fsub <2 x double> <double -0.000000e+00, double -0.000000e+00>, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V4F64 = fsub <4 x double> <double -0.000000e+00, double -0.000000e+00, double -0.000000e+00, double -0.000000e+00>, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V8F64 = fsub <8 x double> <double -0.000000e+00, double -0.000000e+00, double -0.000000e+00, double -0.000000e+00, double -0.000000e+00, double -0.000000e+00, double -0.000000e+00, double -0.000000e+00>, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+ %F32 = fsub float -0.0, undef
+ %V4F32 = fsub <4 x float> <float -0.0, float -0.0, float -0.0, float -0.0>, undef
+ %V8F32 = fsub <8 x float> <float -0.0, float -0.0, float -0.0, float -0.0, float -0.0, float -0.0, float -0.0, float -0.0>, undef
+ %V16F32 = fsub <16 x float> <float -0.0, float -0.0, float -0.0, float -0.0, float -0.0, float -0.0, float -0.0, float -0.0, float -0.0, float -0.0, float -0.0, float -0.0, float -0.0, float -0.0, float -0.0, float -0.0>, undef
+
+ %F64 = fsub double -0.0, undef
+ %V2F64 = fsub <2 x double> <double -0.0, double -0.0>, undef
+ %V4F64 = fsub <4 x double> <double -0.0, double -0.0, double -0.0, double -0.0>, undef
+ %V8F64 = fsub <8 x double> <double -0.0, double -0.0, double -0.0, double -0.0, double -0.0, double -0.0, double -0.0, double -0.0>, undef
+
+ ret i32 undef
+}
+
+define i32 @fneg(i32 %arg) {
+; LATE-LABEL: 'fneg'
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %F32 = fneg float undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4F32 = fneg <4 x float> undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V8F32 = fneg <8 x float> undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V16F32 = fneg <16 x float> undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %F64 = fneg double undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2F64 = fneg <2 x double> undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4F64 = fneg <4 x double> undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V8F64 = fneg <8 x double> undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+; SIZE-LABEL: 'fneg'
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %F32 = fneg float undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V4F32 = fneg <4 x float> undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V8F32 = fneg <8 x float> undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V16F32 = fneg <16 x float> undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %F64 = fneg double undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V2F64 = fneg <2 x double> undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V4F64 = fneg <4 x double> undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V8F64 = fneg <8 x double> undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+; SIZE_LATE-LABEL: 'fneg'
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %F32 = fneg float undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V4F32 = fneg <4 x float> undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V8F32 = fneg <8 x float> undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V16F32 = fneg <16 x float> undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %F64 = fneg double undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V2F64 = fneg <2 x double> undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V4F64 = fneg <4 x double> undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V8F64 = fneg <8 x double> undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+ %F32 = fneg float undef
+ %V4F32 = fneg <4 x float> undef
+ %V8F32 = fneg <8 x float> undef
+ %V16F32 = fneg <16 x float> undef
+
+ %F64 = fneg double undef
+ %V2F64 = fneg <2 x double> undef
+ %V4F64 = fneg <4 x double> undef
+ %V8F64 = fneg <8 x double> undef
+
+ ret i32 undef
+}
+
+define i32 @fmul(i32 %arg) {
+; LATE-LABEL: 'fmul'
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %F32 = fmul float undef, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4F32 = fmul <4 x float> undef, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V8F32 = fmul <8 x float> undef, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V16F32 = fmul <16 x float> undef, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %F64 = fmul double undef, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2F64 = fmul <2 x double> undef, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4F64 = fmul <4 x double> undef, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V8F64 = fmul <8 x double> undef, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+; SIZE-LABEL: 'fmul'
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %F32 = fmul float undef, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V4F32 = fmul <4 x float> undef, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V8F32 = fmul <8 x float> undef, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V16F32 = fmul <16 x float> undef, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %F64 = fmul double undef, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V2F64 = fmul <2 x double> undef, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V4F64 = fmul <4 x double> undef, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V8F64 = fmul <8 x double> undef, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+; SIZE_LATE-LABEL: 'fmul'
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %F32 = fmul float undef, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V4F32 = fmul <4 x float> undef, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V8F32 = fmul <8 x float> undef, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V16F32 = fmul <16 x float> undef, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %F64 = fmul double undef, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V2F64 = fmul <2 x double> undef, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V4F64 = fmul <4 x double> undef, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V8F64 = fmul <8 x double> undef, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+ %F32 = fmul float undef, undef
+ %V4F32 = fmul <4 x float> undef, undef
+ %V8F32 = fmul <8 x float> undef, undef
+ %V16F32 = fmul <16 x float> undef, undef
+
+ %F64 = fmul double undef, undef
+ %V2F64 = fmul <2 x double> undef, undef
+ %V4F64 = fmul <4 x double> undef, undef
+ %V8F64 = fmul <8 x double> undef, undef
+
+ ret i32 undef
+}
+
+define i32 @fdiv(i32 %arg) {
+; LATE-LABEL: 'fdiv'
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %F32 = fdiv float undef, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4F32 = fdiv <4 x float> undef, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V8F32 = fdiv <8 x float> undef, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V16F32 = fdiv <16 x float> undef, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %F64 = fdiv double undef, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2F64 = fdiv <2 x double> undef, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4F64 = fdiv <4 x double> undef, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V8F64 = fdiv <8 x double> undef, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+; SIZE-LABEL: 'fdiv'
+; SIZE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %F32 = fdiv float undef, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4F32 = fdiv <4 x float> undef, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8F32 = fdiv <8 x float> undef, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V16F32 = fdiv <16 x float> undef, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %F64 = fdiv double undef, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V2F64 = fdiv <2 x double> undef, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4F64 = fdiv <4 x double> undef, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8F64 = fdiv <8 x double> undef, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+; SIZE_LATE-LABEL: 'fdiv'
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %F32 = fdiv float undef, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4F32 = fdiv <4 x float> undef, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8F32 = fdiv <8 x float> undef, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V16F32 = fdiv <16 x float> undef, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %F64 = fdiv double undef, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V2F64 = fdiv <2 x double> undef, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4F64 = fdiv <4 x double> undef, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8F64 = fdiv <8 x double> undef, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+ %F32 = fdiv float undef, undef
+ %V4F32 = fdiv <4 x float> undef, undef
+ %V8F32 = fdiv <8 x float> undef, undef
+ %V16F32 = fdiv <16 x float> undef, undef
+
+ %F64 = fdiv double undef, undef
+ %V2F64 = fdiv <2 x double> undef, undef
+ %V4F64 = fdiv <4 x double> undef, undef
+ %V8F64 = fdiv <8 x double> undef, undef
+
+ ret i32 undef
+}
+
+define i32 @frem(i32 %arg) {
+; LATE-LABEL: 'frem'
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %F32 = frem float undef, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4F32 = frem <4 x float> undef, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V8F32 = frem <8 x float> undef, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V16F32 = frem <16 x float> undef, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %F64 = frem double undef, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2F64 = frem <2 x double> undef, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4F64 = frem <4 x double> undef, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V8F64 = frem <8 x double> undef, undef
+; LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+; SIZE-LABEL: 'frem'
+; SIZE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %F32 = frem float undef, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4F32 = frem <4 x float> undef, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8F32 = frem <8 x float> undef, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V16F32 = frem <16 x float> undef, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %F64 = frem double undef, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V2F64 = frem <2 x double> undef, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4F64 = frem <4 x double> undef, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8F64 = frem <8 x double> undef, undef
+; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+; SIZE_LATE-LABEL: 'frem'
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %F32 = frem float undef, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4F32 = frem <4 x float> undef, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8F32 = frem <8 x float> undef, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V16F32 = frem <16 x float> undef, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %F64 = frem double undef, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V2F64 = frem <2 x double> undef, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4F64 = frem <4 x double> undef, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8F64 = frem <8 x double> undef, undef
+; SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+ %F32 = frem float undef, undef
+ %V4F32 = frem <4 x float> undef, undef
+ %V8F32 = frem <8 x float> undef, undef
+ %V16F32 = frem <16 x float> undef, undef
+
+ %F64 = frem double undef, undef
+ %V2F64 = frem <2 x double> undef, undef
+ %V4F64 = frem <4 x double> undef, undef
+ %V8F64 = frem <8 x double> undef, undef
+
+ ret i32 undef
+}
+
+define i32 @fsqrt(i32 %arg) {
+; LATE-LABEL: 'fsqrt'
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %F32 = call float @llvm.sqrt.f32(float undef)
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4F32 = call <4 x float> @llvm.sqrt.v4f32(<4 x float> undef)
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V8F32 = call <8 x float> @llvm.sqrt.v8f32(<8 x float> undef)
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V16F32 = call <16 x float> @llvm.sqrt.v16f32(<16 x float> undef)
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %F64 = call double @llvm.sqrt.f64(double undef)
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2F64 = call <2 x double> @llvm.sqrt.v2f64(<2 x double> undef)
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4F64 = call <4 x double> @llvm.sqrt.v4f64(<4 x double> undef)
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V8F64 = call <8 x double> @llvm.sqrt.v8f64(<8 x double> undef)
+; LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+; SSE-SIZE-LABEL: 'fsqrt'
+; SSE-SIZE-NEXT: Cost Model: Found an estimated cost of 28 for instruction: %F32 = call float @llvm.sqrt.f32(float undef)
+; SSE-SIZE-NEXT: Cost Model: Found an estimated cost of 56 for instruction: %V4F32 = call <4 x float> @llvm.sqrt.v4f32(<4 x float> undef)
+; SSE-SIZE-NEXT: Cost Model: Found an estimated cost of 112 for instruction: %V8F32 = call <8 x float> @llvm.sqrt.v8f32(<8 x float> undef)
+; SSE-SIZE-NEXT: Cost Model: Found an estimated cost of 224 for instruction: %V16F32 = call <16 x float> @llvm.sqrt.v16f32(<16 x float> undef)
+; SSE-SIZE-NEXT: Cost Model: Found an estimated cost of 32 for instruction: %F64 = call double @llvm.sqrt.f64(double undef)
+; SSE-SIZE-NEXT: Cost Model: Found an estimated cost of 32 for instruction: %V2F64 = call <2 x double> @llvm.sqrt.v2f64(<2 x double> undef)
+; SSE-SIZE-NEXT: Cost Model: Found an estimated cost of 64 for instruction: %V4F64 = call <4 x double> @llvm.sqrt.v4f64(<4 x double> undef)
+; SSE-SIZE-NEXT: Cost Model: Found an estimated cost of 128 for instruction: %V8F64 = call <8 x double> @llvm.sqrt.v8f64(<8 x double> undef)
+; SSE-SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+; SSE-SIZE_LATE-LABEL: 'fsqrt'
+; SSE-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 28 for instruction: %F32 = call float @llvm.sqrt.f32(float undef)
+; SSE-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 56 for instruction: %V4F32 = call <4 x float> @llvm.sqrt.v4f32(<4 x float> undef)
+; SSE-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 112 for instruction: %V8F32 = call <8 x float> @llvm.sqrt.v8f32(<8 x float> undef)
+; SSE-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 224 for instruction: %V16F32 = call <16 x float> @llvm.sqrt.v16f32(<16 x float> undef)
+; SSE-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 32 for instruction: %F64 = call double @llvm.sqrt.f64(double undef)
+; SSE-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 32 for instruction: %V2F64 = call <2 x double> @llvm.sqrt.v2f64(<2 x double> undef)
+; SSE-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 64 for instruction: %V4F64 = call <4 x double> @llvm.sqrt.v4f64(<4 x double> undef)
+; SSE-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 128 for instruction: %V8F64 = call <8 x double> @llvm.sqrt.v8f64(<8 x double> undef)
+; SSE-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+; AVX1-SIZE_LATE-LABEL: 'fsqrt'
+; AVX1-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %F32 = call float @llvm.sqrt.f32(float undef)
+; AVX1-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %V4F32 = call <4 x float> @llvm.sqrt.v4f32(<4 x float> undef)
+; AVX1-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 28 for instruction: %V8F32 = call <8 x float> @llvm.sqrt.v8f32(<8 x float> undef)
+; AVX1-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 56 for instruction: %V16F32 = call <16 x float> @llvm.sqrt.v16f32(<16 x float> undef)
+; AVX1-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 21 for instruction: %F64 = call double @llvm.sqrt.f64(double undef)
+; AVX1-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 21 for instruction: %V2F64 = call <2 x double> @llvm.sqrt.v2f64(<2 x double> undef)
+; AVX1-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 43 for instruction: %V4F64 = call <4 x double> @llvm.sqrt.v4f64(<4 x double> undef)
+; AVX1-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 86 for instruction: %V8F64 = call <8 x double> @llvm.sqrt.v8f64(<8 x double> undef)
+; AVX1-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+; AVX2-SIZE_LATE-LABEL: 'fsqrt'
+; AVX2-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %F32 = call float @llvm.sqrt.f32(float undef)
+; AVX2-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V4F32 = call <4 x float> @llvm.sqrt.v4f32(<4 x float> undef)
+; AVX2-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %V8F32 = call <8 x float> @llvm.sqrt.v8f32(<8 x float> undef)
+; AVX2-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 28 for instruction: %V16F32 = call <16 x float> @llvm.sqrt.v16f32(<16 x float> undef)
+; AVX2-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %F64 = call double @llvm.sqrt.f64(double undef)
+; AVX2-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %V2F64 = call <2 x double> @llvm.sqrt.v2f64(<2 x double> undef)
+; AVX2-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 28 for instruction: %V4F64 = call <4 x double> @llvm.sqrt.v4f64(<4 x double> undef)
+; AVX2-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 56 for instruction: %V8F64 = call <8 x double> @llvm.sqrt.v8f64(<8 x double> undef)
+; AVX2-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+ %F32 = call float @llvm.sqrt.f32(float undef)
+ %V4F32 = call <4 x float> @llvm.sqrt.v4f32(<4 x float> undef)
+ %V8F32 = call <8 x float> @llvm.sqrt.v8f32(<8 x float> undef)
+ %V16F32 = call <16 x float> @llvm.sqrt.v16f32(<16 x float> undef)
+
+ %F64 = call double @llvm.sqrt.f64(double undef)
+ %V2F64 = call <2 x double> @llvm.sqrt.v2f64(<2 x double> undef)
+ %V4F64 = call <4 x double> @llvm.sqrt.v4f64(<4 x double> undef)
+ %V8F64 = call <8 x double> @llvm.sqrt.v8f64(<8 x double> undef)
+
+ ret i32 undef
+}
+
+define i32 @fabs(i32 %arg) {
+; LATE-LABEL: 'fabs'
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %F32 = call float @llvm.fabs.f32(float undef)
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4F32 = call <4 x float> @llvm.fabs.v4f32(<4 x float> undef)
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V8F32 = call <8 x float> @llvm.fabs.v8f32(<8 x float> undef)
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V16F32 = call <16 x float> @llvm.fabs.v16f32(<16 x float> undef)
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %F64 = call double @llvm.fabs.f64(double undef)
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2F64 = call <2 x double> @llvm.fabs.v2f64(<2 x double> undef)
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4F64 = call <4 x double> @llvm.fabs.v4f64(<4 x double> undef)
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V8F64 = call <8 x double> @llvm.fabs.v8f64(<8 x double> undef)
+; LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+; SSE-SIZE-LABEL: 'fabs'
+; SSE-SIZE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %F32 = call float @llvm.fabs.f32(float undef)
+; SSE-SIZE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V4F32 = call <4 x float> @llvm.fabs.v4f32(<4 x float> undef)
+; SSE-SIZE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8F32 = call <8 x float> @llvm.fabs.v8f32(<8 x float> undef)
+; SSE-SIZE-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V16F32 = call <16 x float> @llvm.fabs.v16f32(<16 x float> undef)
+; SSE-SIZE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %F64 = call double @llvm.fabs.f64(double undef)
+; SSE-SIZE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2F64 = call <2 x double> @llvm.fabs.v2f64(<2 x double> undef)
+; SSE-SIZE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4F64 = call <4 x double> @llvm.fabs.v4f64(<4 x double> undef)
+; SSE-SIZE-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V8F64 = call <8 x double> @llvm.fabs.v8f64(<8 x double> undef)
+; SSE-SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+; SSE-SIZE_LATE-LABEL: 'fabs'
+; SSE-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %F32 = call float @llvm.fabs.f32(float undef)
+; SSE-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V4F32 = call <4 x float> @llvm.fabs.v4f32(<4 x float> undef)
+; SSE-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8F32 = call <8 x float> @llvm.fabs.v8f32(<8 x float> undef)
+; SSE-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V16F32 = call <16 x float> @llvm.fabs.v16f32(<16 x float> undef)
+; SSE-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %F64 = call double @llvm.fabs.f64(double undef)
+; SSE-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2F64 = call <2 x double> @llvm.fabs.v2f64(<2 x double> undef)
+; SSE-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4F64 = call <4 x double> @llvm.fabs.v4f64(<4 x double> undef)
+; SSE-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V8F64 = call <8 x double> @llvm.fabs.v8f64(<8 x double> undef)
+; SSE-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+; AVX-SIZE-LABEL: 'fabs'
+; AVX-SIZE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %F32 = call float @llvm.fabs.f32(float undef)
+; AVX-SIZE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V4F32 = call <4 x float> @llvm.fabs.v4f32(<4 x float> undef)
+; AVX-SIZE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V8F32 = call <8 x float> @llvm.fabs.v8f32(<8 x float> undef)
+; AVX-SIZE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V16F32 = call <16 x float> @llvm.fabs.v16f32(<16 x float> undef)
+; AVX-SIZE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %F64 = call double @llvm.fabs.f64(double undef)
+; AVX-SIZE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2F64 = call <2 x double> @llvm.fabs.v2f64(<2 x double> undef)
+; AVX-SIZE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V4F64 = call <4 x double> @llvm.fabs.v4f64(<4 x double> undef)
+; AVX-SIZE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8F64 = call <8 x double> @llvm.fabs.v8f64(<8 x double> undef)
+; AVX-SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+; AVX-SIZE_LATE-LABEL: 'fabs'
+; AVX-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %F32 = call float @llvm.fabs.f32(float undef)
+; AVX-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V4F32 = call <4 x float> @llvm.fabs.v4f32(<4 x float> undef)
+; AVX-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V8F32 = call <8 x float> @llvm.fabs.v8f32(<8 x float> undef)
+; AVX-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V16F32 = call <16 x float> @llvm.fabs.v16f32(<16 x float> undef)
+; AVX-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %F64 = call double @llvm.fabs.f64(double undef)
+; AVX-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2F64 = call <2 x double> @llvm.fabs.v2f64(<2 x double> undef)
+; AVX-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V4F64 = call <4 x double> @llvm.fabs.v4f64(<4 x double> undef)
+; AVX-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8F64 = call <8 x double> @llvm.fabs.v8f64(<8 x double> undef)
+; AVX-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+ %F32 = call float @llvm.fabs.f32(float undef)
+ %V4F32 = call <4 x float> @llvm.fabs.v4f32(<4 x float> undef)
+ %V8F32 = call <8 x float> @llvm.fabs.v8f32(<8 x float> undef)
+ %V16F32 = call <16 x float> @llvm.fabs.v16f32(<16 x float> undef)
+
+ %F64 = call double @llvm.fabs.f64(double undef)
+ %V2F64 = call <2 x double> @llvm.fabs.v2f64(<2 x double> undef)
+ %V4F64 = call <4 x double> @llvm.fabs.v4f64(<4 x double> undef)
+ %V8F64 = call <8 x double> @llvm.fabs.v8f64(<8 x double> undef)
+
+ ret i32 undef
+}
+
+define i32 @fcopysign(i32 %arg) {
+; LATE-LABEL: 'fcopysign'
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %F32 = call float @llvm.copysign.f32(float undef, float undef)
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4F32 = call <4 x float> @llvm.copysign.v4f32(<4 x float> undef, <4 x float> undef)
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V8F32 = call <8 x float> @llvm.copysign.v8f32(<8 x float> undef, <8 x float> undef)
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V16F32 = call <16 x float> @llvm.copysign.v16f32(<16 x float> undef, <16 x float> undef)
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %F64 = call double @llvm.copysign.f64(double undef, double undef)
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2F64 = call <2 x double> @llvm.copysign.v2f64(<2 x double> undef, <2 x double> undef)
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4F64 = call <4 x double> @llvm.copysign.v4f64(<4 x double> undef, <4 x double> undef)
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V8F64 = call <8 x double> @llvm.copysign.v8f64(<8 x double> undef, <8 x double> undef)
+; LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+; SSE-SIZE-LABEL: 'fcopysign'
+; SSE-SIZE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %F32 = call float @llvm.copysign.f32(float undef, float undef)
+; SSE-SIZE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V4F32 = call <4 x float> @llvm.copysign.v4f32(<4 x float> undef, <4 x float> undef)
+; SSE-SIZE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8F32 = call <8 x float> @llvm.copysign.v8f32(<8 x float> undef, <8 x float> undef)
+; SSE-SIZE-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V16F32 = call <16 x float> @llvm.copysign.v16f32(<16 x float> undef, <16 x float> undef)
+; SSE-SIZE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %F64 = call double @llvm.copysign.f64(double undef, double undef)
+; SSE-SIZE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2F64 = call <2 x double> @llvm.copysign.v2f64(<2 x double> undef, <2 x double> undef)
+; SSE-SIZE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4F64 = call <4 x double> @llvm.copysign.v4f64(<4 x double> undef, <4 x double> undef)
+; SSE-SIZE-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V8F64 = call <8 x double> @llvm.copysign.v8f64(<8 x double> undef, <8 x double> undef)
+; SSE-SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+; SSE-SIZE_LATE-LABEL: 'fcopysign'
+; SSE-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %F32 = call float @llvm.copysign.f32(float undef, float undef)
+; SSE-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V4F32 = call <4 x float> @llvm.copysign.v4f32(<4 x float> undef, <4 x float> undef)
+; SSE-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8F32 = call <8 x float> @llvm.copysign.v8f32(<8 x float> undef, <8 x float> undef)
+; SSE-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V16F32 = call <16 x float> @llvm.copysign.v16f32(<16 x float> undef, <16 x float> undef)
+; SSE-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %F64 = call double @llvm.copysign.f64(double undef, double undef)
+; SSE-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2F64 = call <2 x double> @llvm.copysign.v2f64(<2 x double> undef, <2 x double> undef)
+; SSE-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4F64 = call <4 x double> @llvm.copysign.v4f64(<4 x double> undef, <4 x double> undef)
+; SSE-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V8F64 = call <8 x double> @llvm.copysign.v8f64(<8 x double> undef, <8 x double> undef)
+; SSE-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+; AVX-SIZE-LABEL: 'fcopysign'
+; AVX-SIZE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %F32 = call float @llvm.copysign.f32(float undef, float undef)
+; AVX-SIZE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V4F32 = call <4 x float> @llvm.copysign.v4f32(<4 x float> undef, <4 x float> undef)
+; AVX-SIZE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V8F32 = call <8 x float> @llvm.copysign.v8f32(<8 x float> undef, <8 x float> undef)
+; AVX-SIZE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V16F32 = call <16 x float> @llvm.copysign.v16f32(<16 x float> undef, <16 x float> undef)
+; AVX-SIZE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %F64 = call double @llvm.copysign.f64(double undef, double undef)
+; AVX-SIZE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2F64 = call <2 x double> @llvm.copysign.v2f64(<2 x double> undef, <2 x double> undef)
+; AVX-SIZE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V4F64 = call <4 x double> @llvm.copysign.v4f64(<4 x double> undef, <4 x double> undef)
+; AVX-SIZE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8F64 = call <8 x double> @llvm.copysign.v8f64(<8 x double> undef, <8 x double> undef)
+; AVX-SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+; AVX-SIZE_LATE-LABEL: 'fcopysign'
+; AVX-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %F32 = call float @llvm.copysign.f32(float undef, float undef)
+; AVX-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V4F32 = call <4 x float> @llvm.copysign.v4f32(<4 x float> undef, <4 x float> undef)
+; AVX-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V8F32 = call <8 x float> @llvm.copysign.v8f32(<8 x float> undef, <8 x float> undef)
+; AVX-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V16F32 = call <16 x float> @llvm.copysign.v16f32(<16 x float> undef, <16 x float> undef)
+; AVX-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %F64 = call double @llvm.copysign.f64(double undef, double undef)
+; AVX-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2F64 = call <2 x double> @llvm.copysign.v2f64(<2 x double> undef, <2 x double> undef)
+; AVX-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V4F64 = call <4 x double> @llvm.copysign.v4f64(<4 x double> undef, <4 x double> undef)
+; AVX-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8F64 = call <8 x double> @llvm.copysign.v8f64(<8 x double> undef, <8 x double> undef)
+; AVX-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+ %F32 = call float @llvm.copysign.f32(float undef, float undef)
+ %V4F32 = call <4 x float> @llvm.copysign.v4f32(<4 x float> undef, <4 x float> undef)
+ %V8F32 = call <8 x float> @llvm.copysign.v8f32(<8 x float> undef, <8 x float> undef)
+ %V16F32 = call <16 x float> @llvm.copysign.v16f32(<16 x float> undef, <16 x float> undef)
+
+ %F64 = call double @llvm.copysign.f64(double undef, double undef)
+ %V2F64 = call <2 x double> @llvm.copysign.v2f64(<2 x double> undef, <2 x double> undef)
+ %V4F64 = call <4 x double> @llvm.copysign.v4f64(<4 x double> undef, <4 x double> undef)
+ %V8F64 = call <8 x double> @llvm.copysign.v8f64(<8 x double> undef, <8 x double> undef)
+
+ ret i32 undef
+}
+
+define i32 @fma(i32 %arg) {
+; LATE-LABEL: 'fma'
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %F32 = call float @llvm.fma.f32(float undef, float undef, float undef)
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4F32 = call <4 x float> @llvm.fma.v4f32(<4 x float> undef, <4 x float> undef, <4 x float> undef)
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V8F32 = call <8 x float> @llvm.fma.v8f32(<8 x float> undef, <8 x float> undef, <8 x float> undef)
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V16F32 = call <16 x float> @llvm.fma.v16f32(<16 x float> undef, <16 x float> undef, <16 x float> undef)
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %F64 = call double @llvm.fma.f64(double undef, double undef, double undef)
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2F64 = call <2 x double> @llvm.fma.v2f64(<2 x double> undef, <2 x double> undef, <2 x double> undef)
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4F64 = call <4 x double> @llvm.fma.v4f64(<4 x double> undef, <4 x double> undef, <4 x double> undef)
+; LATE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V8F64 = call <8 x double> @llvm.fma.v8f64(<8 x double> undef, <8 x double> undef, <8 x double> undef)
+; LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+; SSE-SIZE-LABEL: 'fma'
+; SSE-SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %F32 = call float @llvm.fma.f32(float undef, float undef, float undef)
+; SSE-SIZE-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V4F32 = call <4 x float> @llvm.fma.v4f32(<4 x float> undef, <4 x float> undef, <4 x float> undef)
+; SSE-SIZE-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %V8F32 = call <8 x float> @llvm.fma.v8f32(<8 x float> undef, <8 x float> undef, <8 x float> undef)
+; SSE-SIZE-NEXT: Cost Model: Found an estimated cost of 28 for instruction: %V16F32 = call <16 x float> @llvm.fma.v16f32(<16 x float> undef, <16 x float> undef, <16 x float> undef)
+; SSE-SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %F64 = call double @llvm.fma.f64(double undef, double undef, double undef)
+; SSE-SIZE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2F64 = call <2 x double> @llvm.fma.v2f64(<2 x double> undef, <2 x double> undef, <2 x double> undef)
+; SSE-SIZE-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V4F64 = call <4 x double> @llvm.fma.v4f64(<4 x double> undef, <4 x double> undef, <4 x double> undef)
+; SSE-SIZE-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %V8F64 = call <8 x double> @llvm.fma.v8f64(<8 x double> undef, <8 x double> undef, <8 x double> undef)
+; SSE-SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+; SSE-SIZE_LATE-LABEL: 'fma'
+; SSE-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %F32 = call float @llvm.fma.f32(float undef, float undef, float undef)
+; SSE-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 43 for instruction: %V4F32 = call <4 x float> @llvm.fma.v4f32(<4 x float> undef, <4 x float> undef, <4 x float> undef)
+; SSE-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 86 for instruction: %V8F32 = call <8 x float> @llvm.fma.v8f32(<8 x float> undef, <8 x float> undef, <8 x float> undef)
+; SSE-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 172 for instruction: %V16F32 = call <16 x float> @llvm.fma.v16f32(<16 x float> undef, <16 x float> undef, <16 x float> undef)
+; SSE-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %F64 = call double @llvm.fma.f64(double undef, double undef, double undef)
+; SSE-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 21 for instruction: %V2F64 = call <2 x double> @llvm.fma.v2f64(<2 x double> undef, <2 x double> undef, <2 x double> undef)
+; SSE-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 42 for instruction: %V4F64 = call <4 x double> @llvm.fma.v4f64(<4 x double> undef, <4 x double> undef, <4 x double> undef)
+; SSE-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 84 for instruction: %V8F64 = call <8 x double> @llvm.fma.v8f64(<8 x double> undef, <8 x double> undef, <8 x double> undef)
+; SSE-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+; AVX-SIZE-LABEL: 'fma'
+; AVX-SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %F32 = call float @llvm.fma.f32(float undef, float undef, float undef)
+; AVX-SIZE-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V4F32 = call <4 x float> @llvm.fma.v4f32(<4 x float> undef, <4 x float> undef, <4 x float> undef)
+; AVX-SIZE-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %V8F32 = call <8 x float> @llvm.fma.v8f32(<8 x float> undef, <8 x float> undef, <8 x float> undef)
+; AVX-SIZE-NEXT: Cost Model: Found an estimated cost of 32 for instruction: %V16F32 = call <16 x float> @llvm.fma.v16f32(<16 x float> undef, <16 x float> undef, <16 x float> undef)
+; AVX-SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %F64 = call double @llvm.fma.f64(double undef, double undef, double undef)
+; AVX-SIZE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2F64 = call <2 x double> @llvm.fma.v2f64(<2 x double> undef, <2 x double> undef, <2 x double> undef)
+; AVX-SIZE-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V4F64 = call <4 x double> @llvm.fma.v4f64(<4 x double> undef, <4 x double> undef, <4 x double> undef)
+; AVX-SIZE-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %V8F64 = call <8 x double> @llvm.fma.v8f64(<8 x double> undef, <8 x double> undef, <8 x double> undef)
+; AVX-SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+; AVX-SIZE_LATE-LABEL: 'fma'
+; AVX-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %F32 = call float @llvm.fma.f32(float undef, float undef, float undef)
+; AVX-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 43 for instruction: %V4F32 = call <4 x float> @llvm.fma.v4f32(<4 x float> undef, <4 x float> undef, <4 x float> undef)
+; AVX-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 88 for instruction: %V8F32 = call <8 x float> @llvm.fma.v8f32(<8 x float> undef, <8 x float> undef, <8 x float> undef)
+; AVX-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 176 for instruction: %V16F32 = call <16 x float> @llvm.fma.v16f32(<16 x float> undef, <16 x float> undef, <16 x float> undef)
+; AVX-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %F64 = call double @llvm.fma.f64(double undef, double undef, double undef)
+; AVX-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 21 for instruction: %V2F64 = call <2 x double> @llvm.fma.v2f64(<2 x double> undef, <2 x double> undef, <2 x double> undef)
+; AVX-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 43 for instruction: %V4F64 = call <4 x double> @llvm.fma.v4f64(<4 x double> undef, <4 x double> undef, <4 x double> undef)
+; AVX-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 86 for instruction: %V8F64 = call <8 x double> @llvm.fma.v8f64(<8 x double> undef, <8 x double> undef, <8 x double> undef)
+; AVX-SIZE_LATE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
+;
+ %F32 = call float @llvm.fma.f32(float undef, float undef, float undef)
+ %V4F32 = call <4 x float> @llvm.fma.v4f32(<4 x float> undef, <4 x float> undef, <4 x float> undef)
+ %V8F32 = call <8 x float> @llvm.fma.v8f32(<8 x float> undef, <8 x float> undef, <8 x float> undef)
+ %V16F32 = call <16 x float> @llvm.fma.v16f32(<16 x float> undef, <16 x float> undef, <16 x float> undef)
+
+ %F64 = call double @llvm.fma.f64(double undef, double undef, double undef)
+ %V2F64 = call <2 x double> @llvm.fma.v2f64(<2 x double> undef, <2 x double> undef, <2 x double> undef)
+ %V4F64 = call <4 x double> @llvm.fma.v4f64(<4 x double> undef, <4 x double> undef, <4 x double> undef)
+ %V8F64 = call <8 x double> @llvm.fma.v8f64(<8 x double> undef, <8 x double> undef, <8 x double> undef)
+
+ ret i32 undef
+}
+
+declare float @llvm.sqrt.f32(float)
+declare <4 x float> @llvm.sqrt.v4f32(<4 x float>)
+declare <8 x float> @llvm.sqrt.v8f32(<8 x float>)
+declare <16 x float> @llvm.sqrt.v16f32(<16 x float>)
+
+declare double @llvm.sqrt.f64(double)
+declare <2 x double> @llvm.sqrt.v2f64(<2 x double>)
+declare <4 x double> @llvm.sqrt.v4f64(<4 x double>)
+declare <8 x double> @llvm.sqrt.v8f64(<8 x double>)
+
+declare float @llvm.fabs.f32(float)
+declare <4 x float> @llvm.fabs.v4f32(<4 x float>)
+declare <8 x float> @llvm.fabs.v8f32(<8 x float>)
+declare <16 x float> @llvm.fabs.v16f32(<16 x float>)
+
+declare double @llvm.fabs.f64(double)
+declare <2 x double> @llvm.fabs.v2f64(<2 x double>)
+declare <4 x double> @llvm.fabs.v4f64(<4 x double>)
+declare <8 x double> @llvm.fabs.v8f64(<8 x double>)
+
+declare float @llvm.copysign.f32(float, float)
+declare <4 x float> @llvm.copysign.v4f32(<4 x float>, <4 x float>)
+declare <8 x float> @llvm.copysign.v8f32(<8 x float>, <8 x float>)
+declare <16 x float> @llvm.copysign.v16f32(<16 x float>, <16 x float>)
+
+declare double @llvm.copysign.f64(double, double)
+declare <2 x double> @llvm.copysign.v2f64(<2 x double>, <2 x double>)
+declare <4 x double> @llvm.copysign.v4f64(<4 x double>, <4 x double>)
+declare <8 x double> @llvm.copysign.v8f64(<8 x double>, <8 x double>)
+
+declare float @llvm.fma.f32(float, float, float)
+declare <4 x float> @llvm.fma.v4f32(<4 x float>, <4 x float>, <4 x float>)
+declare <8 x float> @llvm.fma.v8f32(<8 x float>, <8 x float>, <8 x float>)
+declare <16 x float> @llvm.fma.v16f32(<16 x float>, <16 x float>, <16 x float>)
+
+declare double @llvm.fma.f64(double, double, double)
+declare <2 x double> @llvm.fma.v2f64(<2 x double>, <2 x double>, <2 x double>)
+declare <4 x double> @llvm.fma.v4f64(<4 x double>, <4 x double>, <4 x double>)
+declare <8 x double> @llvm.fma.v8f64(<8 x double>, <8 x double>, <8 x double>)
More information about the llvm-commits
mailing list