[llvm] r326747 - [X86] Add silvermont fp arithmetic cost model tests
Simon Pilgrim via llvm-commits
llvm-commits at lists.llvm.org
Mon Mar 5 14:13:23 PST 2018
Author: rksimon
Date: Mon Mar 5 14:13:22 2018
New Revision: 326747
URL: http://llvm.org/viewvc/llvm-project?rev=326747&view=rev
Log:
[X86] Add silvermont fp arithmetic cost model tests
Add silvermont to existing high coverage tests instead of repeating in slm-arith-costs.ll
Modified:
llvm/trunk/test/Analysis/CostModel/X86/arith-fp.ll
Modified: llvm/trunk/test/Analysis/CostModel/X86/arith-fp.ll
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/Analysis/CostModel/X86/arith-fp.ll?rev=326747&r1=326746&r2=326747&view=diff
==============================================================================
--- llvm/trunk/test/Analysis/CostModel/X86/arith-fp.ll (original)
+++ llvm/trunk/test/Analysis/CostModel/X86/arith-fp.ll Mon Mar 5 14:13:22 2018
@@ -4,6 +4,7 @@
; RUN: opt < %s -enable-no-nans-fp-math -cost-model -analyze -mtriple=x86_64-apple-macosx10.8.0 -mattr=+avx2,+fma | FileCheck %s --check-prefix=CHECK --check-prefix=AVX2
; RUN: opt < %s -enable-no-nans-fp-math -cost-model -analyze -mtriple=x86_64-apple-macosx10.8.0 -mattr=+avx512f | FileCheck %s --check-prefix=CHECK --check-prefix=AVX512 --check-prefix=AVX512F
; RUN: opt < %s -enable-no-nans-fp-math -cost-model -analyze -mtriple=x86_64-apple-macosx10.8.0 -mattr=+avx512f,+avx512bw | FileCheck %s --check-prefix=CHECK --check-prefix=AVX512 --check-prefix=AVX512BW
+; RUN: opt < %s -enable-no-nans-fp-math -cost-model -analyze -mtriple=x86_64-apple-macosx10.8.0 -mcpu=slm | FileCheck %s --check-prefix=CHECK --check-prefix=SLM
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"
@@ -15,24 +16,28 @@ define i32 @fadd(i32 %arg) {
; AVX: cost of 1 {{.*}} %F32 = fadd
; AVX2: cost of 1 {{.*}} %F32 = fadd
; AVX512: cost of 1 {{.*}} %F32 = fadd
+ ; SLM: cost of 1 {{.*}} %F32 = fadd
%F32 = fadd float undef, undef
; SSE2: cost of 2 {{.*}} %V4F32 = fadd
; SSE42: cost of 1 {{.*}} %V4F32 = fadd
; AVX: cost of 1 {{.*}} %V4F32 = fadd
; AVX2: cost of 1 {{.*}} %V4F32 = fadd
; AVX512: cost of 1 {{.*}} %V4F32 = fadd
+ ; SLM: cost of 1 {{.*}} %V4F32 = fadd
%V4F32 = fadd <4 x float> undef, undef
; SSE2: cost of 4 {{.*}} %V8F32 = fadd
; SSE42: cost of 2 {{.*}} %V8F32 = fadd
; AVX: cost of 2 {{.*}} %V8F32 = fadd
; AVX2: cost of 1 {{.*}} %V8F32 = fadd
; AVX512: cost of 1 {{.*}} %V8F32 = fadd
+ ; SLM: cost of 2 {{.*}} %V8F32 = fadd
%V8F32 = fadd <8 x float> undef, undef
; SSE2: cost of 8 {{.*}} %V16F32 = fadd
; SSE42: cost of 4 {{.*}} %V16F32 = fadd
; AVX: cost of 4 {{.*}} %V16F32 = fadd
; AVX2: cost of 2 {{.*}} %V16F32 = fadd
; AVX512: cost of 1 {{.*}} %V16F32 = fadd
+ ; SLM: cost of 4 {{.*}} %V16F32 = fadd
%V16F32 = fadd <16 x float> undef, undef
; SSE2: cost of 2 {{.*}} %F64 = fadd
@@ -40,24 +45,28 @@ define i32 @fadd(i32 %arg) {
; AVX: cost of 1 {{.*}} %F64 = fadd
; AVX2: cost of 1 {{.*}} %F64 = fadd
; AVX512: cost of 1 {{.*}} %F64 = fadd
+ ; SLM: cost of 1 {{.*}} %F64 = fadd
%F64 = fadd double undef, undef
; SSE2: cost of 2 {{.*}} %V2F64 = fadd
; SSE42: cost of 1 {{.*}} %V2F64 = fadd
; AVX: cost of 1 {{.*}} %V2F64 = fadd
; AVX2: cost of 1 {{.*}} %V2F64 = fadd
; AVX512: cost of 1 {{.*}} %V2F64 = fadd
+ ; SLM: cost of 2 {{.*}} %V2F64 = fadd
%V2F64 = fadd <2 x double> undef, undef
; SSE2: cost of 4 {{.*}} %V4F64 = fadd
; SSE42: cost of 2 {{.*}} %V4F64 = fadd
; AVX: cost of 2 {{.*}} %V4F64 = fadd
; AVX2: cost of 1 {{.*}} %V4F64 = fadd
; AVX512: cost of 1 {{.*}} %V4F64 = fadd
+ ; SLM: cost of 4 {{.*}} %V4F64 = fadd
%V4F64 = fadd <4 x double> undef, undef
; SSE2: cost of 8 {{.*}} %V8F64 = fadd
; SSE42: cost of 4 {{.*}} %V8F64 = fadd
; AVX: cost of 4 {{.*}} %V8F64 = fadd
; AVX2: cost of 2 {{.*}} %V8F64 = fadd
; AVX512: cost of 1 {{.*}} %V8F64 = fadd
+ ; SLM: cost of 8 {{.*}} %V8F64 = fadd
%V8F64 = fadd <8 x double> undef, undef
ret i32 undef
@@ -70,24 +79,28 @@ define i32 @fsub(i32 %arg) {
; AVX: cost of 1 {{.*}} %F32 = fsub
; AVX2: cost of 1 {{.*}} %F32 = fsub
; AVX512: cost of 1 {{.*}} %F32 = fsub
+ ; SLM: cost of 1 {{.*}} %F32 = fsub
%F32 = fsub float undef, undef
; SSE2: cost of 2 {{.*}} %V4F32 = fsub
; SSE42: cost of 1 {{.*}} %V4F32 = fsub
; AVX: cost of 1 {{.*}} %V4F32 = fsub
; AVX2: cost of 1 {{.*}} %V4F32 = fsub
; AVX512: cost of 1 {{.*}} %V4F32 = fsub
+ ; SLM: cost of 1 {{.*}} %V4F32 = fsub
%V4F32 = fsub <4 x float> undef, undef
; SSE2: cost of 4 {{.*}} %V8F32 = fsub
; SSE42: cost of 2 {{.*}} %V8F32 = fsub
; AVX: cost of 2 {{.*}} %V8F32 = fsub
; AVX2: cost of 1 {{.*}} %V8F32 = fsub
; AVX512: cost of 1 {{.*}} %V8F32 = fsub
+ ; SLM: cost of 2 {{.*}} %V8F32 = fsub
%V8F32 = fsub <8 x float> undef, undef
; SSE2: cost of 8 {{.*}} %V16F32 = fsub
; SSE42: cost of 4 {{.*}} %V16F32 = fsub
; AVX: cost of 4 {{.*}} %V16F32 = fsub
; AVX2: cost of 2 {{.*}} %V16F32 = fsub
; AVX512: cost of 1 {{.*}} %V16F32 = fsub
+ ; SLM: cost of 4 {{.*}} %V16F32 = fsub
%V16F32 = fsub <16 x float> undef, undef
; SSE2: cost of 2 {{.*}} %F64 = fsub
@@ -95,24 +108,28 @@ define i32 @fsub(i32 %arg) {
; AVX: cost of 1 {{.*}} %F64 = fsub
; AVX2: cost of 1 {{.*}} %F64 = fsub
; AVX512: cost of 1 {{.*}} %F64 = fsub
+ ; SLM: cost of 1 {{.*}} %F64 = fsub
%F64 = fsub double undef, undef
; SSE2: cost of 2 {{.*}} %V2F64 = fsub
; SSE42: cost of 1 {{.*}} %V2F64 = fsub
; AVX: cost of 1 {{.*}} %V2F64 = fsub
; AVX2: cost of 1 {{.*}} %V2F64 = fsub
; AVX512: cost of 1 {{.*}} %V2F64 = fsub
+ ; SLM: cost of 2 {{.*}} %V2F64 = fsub
%V2F64 = fsub <2 x double> undef, undef
; SSE2: cost of 4 {{.*}} %V4F64 = fsub
; SSE42: cost of 2 {{.*}} %V4F64 = fsub
; AVX: cost of 2 {{.*}} %V4F64 = fsub
; AVX2: cost of 1 {{.*}} %V4F64 = fsub
; AVX512: cost of 1 {{.*}} %V4F64 = fsub
+ ; SLM: cost of 4 {{.*}} %V4F64 = fsub
%V4F64 = fsub <4 x double> undef, undef
; SSE2: cost of 8 {{.*}} %V8F64 = fsub
; SSE42: cost of 4 {{.*}} %V8F64 = fsub
; AVX: cost of 4 {{.*}} %V8F64 = fsub
; AVX2: cost of 2 {{.*}} %V8F64 = fsub
; AVX512: cost of 1 {{.*}} %V8F64 = fsub
+ ; SLM: cost of 8 {{.*}} %V8F64 = fsub
%V8F64 = fsub <8 x double> undef, undef
ret i32 undef
@@ -125,24 +142,28 @@ define i32 @fmul(i32 %arg) {
; AVX: cost of 1 {{.*}} %F32 = fmul
; AVX2: cost of 1 {{.*}} %F32 = fmul
; AVX512: cost of 1 {{.*}} %F32 = fmul
+ ; SLM: cost of 1 {{.*}} %F32 = fmul
%F32 = fmul float undef, undef
; SSE2: cost of 2 {{.*}} %V4F32 = fmul
; SSE42: cost of 1 {{.*}} %V4F32 = fmul
; AVX: cost of 1 {{.*}} %V4F32 = fmul
; AVX2: cost of 1 {{.*}} %V4F32 = fmul
; AVX512: cost of 1 {{.*}} %V4F32 = fmul
+ ; SLM: cost of 2 {{.*}} %V4F32 = fmul
%V4F32 = fmul <4 x float> undef, undef
; SSE2: cost of 4 {{.*}} %V8F32 = fmul
; SSE42: cost of 2 {{.*}} %V8F32 = fmul
; AVX: cost of 2 {{.*}} %V8F32 = fmul
; AVX2: cost of 1 {{.*}} %V8F32 = fmul
; AVX512: cost of 1 {{.*}} %V8F32 = fmul
+ ; SLM: cost of 4 {{.*}} %V8F32 = fmul
%V8F32 = fmul <8 x float> undef, undef
; SSE2: cost of 8 {{.*}} %V16F32 = fmul
; SSE42: cost of 4 {{.*}} %V16F32 = fmul
; AVX: cost of 4 {{.*}} %V16F32 = fmul
; AVX2: cost of 2 {{.*}} %V16F32 = fmul
; AVX512: cost of 1 {{.*}} %V16F32 = fmul
+ ; SLM: cost of 8 {{.*}} %V16F32 = fmul
%V16F32 = fmul <16 x float> undef, undef
; SSE2: cost of 2 {{.*}} %F64 = fmul
@@ -150,24 +171,28 @@ define i32 @fmul(i32 %arg) {
; AVX: cost of 1 {{.*}} %F64 = fmul
; AVX2: cost of 1 {{.*}} %F64 = fmul
; AVX512: cost of 1 {{.*}} %F64 = fmul
+ ; SLM: cost of 2 {{.*}} %F64 = fmul
%F64 = fmul double undef, undef
; SSE2: cost of 2 {{.*}} %V2F64 = fmul
; SSE42: cost of 1 {{.*}} %V2F64 = fmul
; AVX: cost of 1 {{.*}} %V2F64 = fmul
; AVX2: cost of 1 {{.*}} %V2F64 = fmul
; AVX512: cost of 1 {{.*}} %V2F64 = fmul
+ ; SLM: cost of 4 {{.*}} %V2F64 = fmul
%V2F64 = fmul <2 x double> undef, undef
; SSE2: cost of 4 {{.*}} %V4F64 = fmul
; SSE42: cost of 2 {{.*}} %V4F64 = fmul
; AVX: cost of 2 {{.*}} %V4F64 = fmul
; AVX2: cost of 1 {{.*}} %V4F64 = fmul
; AVX512: cost of 1 {{.*}} %V4F64 = fmul
+ ; SLM: cost of 8 {{.*}} %V4F64 = fmul
%V4F64 = fmul <4 x double> undef, undef
; SSE2: cost of 8 {{.*}} %V8F64 = fmul
; SSE42: cost of 4 {{.*}} %V8F64 = fmul
; AVX: cost of 4 {{.*}} %V8F64 = fmul
; AVX2: cost of 2 {{.*}} %V8F64 = fmul
; AVX512: cost of 1 {{.*}} %V8F64 = fmul
+ ; SLM: cost of 16 {{.*}} %V8F64 = fmul
%V8F64 = fmul <8 x double> undef, undef
ret i32 undef
@@ -180,24 +205,28 @@ define i32 @fdiv(i32 %arg) {
; AVX: cost of 14 {{.*}} %F32 = fdiv
; AVX2: cost of 7 {{.*}} %F32 = fdiv
; AVX512: cost of 7 {{.*}} %F32 = fdiv
+ ; SLM: cost of 17 {{.*}} %F32 = fdiv
%F32 = fdiv float undef, undef
; SSE2: cost of 39 {{.*}} %V4F32 = fdiv
; SSE42: cost of 14 {{.*}} %V4F32 = fdiv
; AVX: cost of 14 {{.*}} %V4F32 = fdiv
; AVX2: cost of 7 {{.*}} %V4F32 = fdiv
; AVX512: cost of 7 {{.*}} %V4F32 = fdiv
+ ; SLM: cost of 39 {{.*}} %V4F32 = fdiv
%V4F32 = fdiv <4 x float> undef, undef
; SSE2: cost of 78 {{.*}} %V8F32 = fdiv
; SSE42: cost of 28 {{.*}} %V8F32 = fdiv
; AVX: cost of 28 {{.*}} %V8F32 = fdiv
; AVX2: cost of 14 {{.*}} %V8F32 = fdiv
; AVX512: cost of 14 {{.*}} %V8F32 = fdiv
+ ; SLM: cost of 78 {{.*}} %V8F32 = fdiv
%V8F32 = fdiv <8 x float> undef, undef
; SSE2: cost of 156 {{.*}} %V16F32 = fdiv
; SSE42: cost of 56 {{.*}} %V16F32 = fdiv
; AVX: cost of 56 {{.*}} %V16F32 = fdiv
; AVX2: cost of 28 {{.*}} %V16F32 = fdiv
; AVX512: cost of 2 {{.*}} %V16F32 = fdiv
+ ; SLM: cost of 156 {{.*}} %V16F32 = fdiv
%V16F32 = fdiv <16 x float> undef, undef
; SSE2: cost of 38 {{.*}} %F64 = fdiv
@@ -205,24 +234,28 @@ define i32 @fdiv(i32 %arg) {
; AVX: cost of 22 {{.*}} %F64 = fdiv
; AVX2: cost of 14 {{.*}} %F64 = fdiv
; AVX512: cost of 14 {{.*}} %F64 = fdiv
+ ; SLM: cost of 32 {{.*}} %F64 = fdiv
%F64 = fdiv double undef, undef
; SSE2: cost of 69 {{.*}} %V2F64 = fdiv
; SSE42: cost of 22 {{.*}} %V2F64 = fdiv
; AVX: cost of 22 {{.*}} %V2F64 = fdiv
; AVX2: cost of 14 {{.*}} %V2F64 = fdiv
; AVX512: cost of 14 {{.*}} %V2F64 = fdiv
+ ; SLM: cost of 69 {{.*}} %V2F64 = fdiv
%V2F64 = fdiv <2 x double> undef, undef
; SSE2: cost of 138 {{.*}} %V4F64 = fdiv
; SSE42: cost of 44 {{.*}} %V4F64 = fdiv
; AVX: cost of 44 {{.*}} %V4F64 = fdiv
; AVX2: cost of 28 {{.*}} %V4F64 = fdiv
; AVX512: cost of 28 {{.*}} %V4F64 = fdiv
+ ; SLM: cost of 138 {{.*}} %V4F64 = fdiv
%V4F64 = fdiv <4 x double> undef, undef
; SSE2: cost of 276 {{.*}} %V8F64 = fdiv
; SSE42: cost of 88 {{.*}} %V8F64 = fdiv
; AVX: cost of 88 {{.*}} %V8F64 = fdiv
; AVX2: cost of 56 {{.*}} %V8F64 = fdiv
; AVX512: cost of 2 {{.*}} %V8F64 = fdiv
+ ; SLM: cost of 276 {{.*}} %V8F64 = fdiv
%V8F64 = fdiv <8 x double> undef, undef
ret i32 undef
@@ -235,24 +268,28 @@ define i32 @frem(i32 %arg) {
; AVX: cost of 2 {{.*}} %F32 = frem
; AVX2: cost of 2 {{.*}} %F32 = frem
; AVX512: cost of 2 {{.*}} %F32 = frem
+ ; SLM: cost of 2 {{.*}} %F32 = frem
%F32 = frem float undef, undef
; SSE2: cost of 14 {{.*}} %V4F32 = frem
; SSE42: cost of 14 {{.*}} %V4F32 = frem
; AVX: cost of 14 {{.*}} %V4F32 = frem
; AVX2: cost of 14 {{.*}} %V4F32 = frem
; AVX512: cost of 14 {{.*}} %V4F32 = frem
+ ; SLM: cost of 14 {{.*}} %V4F32 = frem
%V4F32 = frem <4 x float> undef, undef
; SSE2: cost of 28 {{.*}} %V8F32 = frem
; SSE42: cost of 28 {{.*}} %V8F32 = frem
; AVX: cost of 30 {{.*}} %V8F32 = frem
; AVX2: cost of 30 {{.*}} %V8F32 = frem
; AVX512: cost of 30 {{.*}} %V8F32 = frem
+ ; SLM: cost of 28 {{.*}} %V8F32 = frem
%V8F32 = frem <8 x float> undef, undef
; SSE2: cost of 56 {{.*}} %V16F32 = frem
; SSE42: cost of 56 {{.*}} %V16F32 = frem
; AVX: cost of 60 {{.*}} %V16F32 = frem
; AVX2: cost of 60 {{.*}} %V16F32 = frem
; AVX512: cost of 62 {{.*}} %V16F32 = frem
+ ; SLM: cost of 56 {{.*}} %V16F32 = frem
%V16F32 = frem <16 x float> undef, undef
; SSE2: cost of 2 {{.*}} %F64 = frem
@@ -260,24 +297,28 @@ define i32 @frem(i32 %arg) {
; AVX: cost of 2 {{.*}} %F64 = frem
; AVX2: cost of 2 {{.*}} %F64 = frem
; AVX512: cost of 2 {{.*}} %F64 = frem
+ ; SLM: cost of 2 {{.*}} %F64 = frem
%F64 = frem double undef, undef
; SSE2: cost of 6 {{.*}} %V2F64 = frem
; SSE42: cost of 6 {{.*}} %V2F64 = frem
; AVX: cost of 6 {{.*}} %V2F64 = frem
; AVX2: cost of 6 {{.*}} %V2F64 = frem
; AVX512: cost of 6 {{.*}} %V2F64 = frem
+ ; SLM: cost of 6 {{.*}} %V2F64 = frem
%V2F64 = frem <2 x double> undef, undef
; SSE2: cost of 12 {{.*}} %V4F64 = frem
; SSE42: cost of 12 {{.*}} %V4F64 = frem
; AVX: cost of 14 {{.*}} %V4F64 = frem
; AVX2: cost of 14 {{.*}} %V4F64 = frem
; AVX512: cost of 14 {{.*}} %V4F64 = frem
+ ; SLM: cost of 12 {{.*}} %V4F64 = frem
%V4F64 = frem <4 x double> undef, undef
; SSE2: cost of 24 {{.*}} %V8F64 = frem
; SSE42: cost of 24 {{.*}} %V8F64 = frem
; AVX: cost of 28 {{.*}} %V8F64 = frem
; AVX2: cost of 28 {{.*}} %V8F64 = frem
; AVX512: cost of 30 {{.*}} %V8F64 = frem
+ ; SLM: cost of 24 {{.*}} %V8F64 = frem
%V8F64 = frem <8 x double> undef, undef
ret i32 undef
@@ -290,24 +331,28 @@ define i32 @fsqrt(i32 %arg) {
; AVX: cost of 14 {{.*}} %F32 = call float @llvm.sqrt.f32
; AVX2: cost of 7 {{.*}} %F32 = call float @llvm.sqrt.f32
; AVX512: cost of 7 {{.*}} %F32 = call float @llvm.sqrt.f32
+ ; SLM: cost of 18 {{.*}} %F32 = call float @llvm.sqrt.f32
%F32 = call float @llvm.sqrt.f32(float undef)
; SSE2: cost of 56 {{.*}} %V4F32 = call <4 x float> @llvm.sqrt.v4f32
; SSE42: cost of 18 {{.*}} %V4F32 = call <4 x float> @llvm.sqrt.v4f32
; AVX: cost of 14 {{.*}} %V4F32 = call <4 x float> @llvm.sqrt.v4f32
; AVX2: cost of 7 {{.*}} %V4F32 = call <4 x float> @llvm.sqrt.v4f32
; AVX512: cost of 7 {{.*}} %V4F32 = call <4 x float> @llvm.sqrt.v4f32
+ ; SLM: cost of 18 {{.*}} %V4F32 = call <4 x float> @llvm.sqrt.v4f32
%V4F32 = call <4 x float> @llvm.sqrt.v4f32(<4 x float> undef)
; SSE2: cost of 112 {{.*}} %V8F32 = call <8 x float> @llvm.sqrt.v8f32
; SSE42: cost of 36 {{.*}} %V8F32 = call <8 x float> @llvm.sqrt.v8f32
; AVX: cost of 28 {{.*}} %V8F32 = call <8 x float> @llvm.sqrt.v8f32
; AVX2: cost of 14 {{.*}} %V8F32 = call <8 x float> @llvm.sqrt.v8f32
; AVX512: cost of 14 {{.*}} %V8F32 = call <8 x float> @llvm.sqrt.v8f32
+ ; SLM: cost of 36 {{.*}} %V8F32 = call <8 x float> @llvm.sqrt.v8f32
%V8F32 = call <8 x float> @llvm.sqrt.v8f32(<8 x float> undef)
; SSE2: cost of 224 {{.*}} %V16F32 = call <16 x float> @llvm.sqrt.v16f32
; SSE42: cost of 72 {{.*}} %V16F32 = call <16 x float> @llvm.sqrt.v16f32
; AVX: cost of 56 {{.*}} %V16F32 = call <16 x float> @llvm.sqrt.v16f32
; AVX2: cost of 28 {{.*}} %V16F32 = call <16 x float> @llvm.sqrt.v16f32
; AVX512: cost of 1 {{.*}} %V16F32 = call <16 x float> @llvm.sqrt.v16f32
+ ; SLM: cost of 72 {{.*}} %V16F32 = call <16 x float> @llvm.sqrt.v16f32
%V16F32 = call <16 x float> @llvm.sqrt.v16f32(<16 x float> undef)
; SSE2: cost of 32 {{.*}} %F64 = call double @llvm.sqrt.f64
@@ -315,24 +360,28 @@ define i32 @fsqrt(i32 %arg) {
; AVX: cost of 21 {{.*}} %F64 = call double @llvm.sqrt.f64
; AVX2: cost of 14 {{.*}} %F64 = call double @llvm.sqrt.f64
; AVX512: cost of 14 {{.*}} %F64 = call double @llvm.sqrt.f64
+ ; SLM: cost of 32 {{.*}} %F64 = call double @llvm.sqrt.f64
%F64 = call double @llvm.sqrt.f64(double undef)
; SSE2: cost of 32 {{.*}} %V2F64 = call <2 x double> @llvm.sqrt.v2f64
; SSE42: cost of 32 {{.*}} %V2F64 = call <2 x double> @llvm.sqrt.v2f64
; AVX: cost of 21 {{.*}} %V2F64 = call <2 x double> @llvm.sqrt.v2f64
; AVX2: cost of 14 {{.*}} %V2F64 = call <2 x double> @llvm.sqrt.v2f64
; AVX512: cost of 14 {{.*}} %V2F64 = call <2 x double> @llvm.sqrt.v2f64
+ ; SLM: cost of 32 {{.*}} %V2F64 = call <2 x double> @llvm.sqrt.v2f64
%V2F64 = call <2 x double> @llvm.sqrt.v2f64(<2 x double> undef)
; SSE2: cost of 64 {{.*}} %V4F64 = call <4 x double> @llvm.sqrt.v4f64
; SSE42: cost of 64 {{.*}} %V4F64 = call <4 x double> @llvm.sqrt.v4f64
; AVX: cost of 43 {{.*}} %V4F64 = call <4 x double> @llvm.sqrt.v4f64
; AVX2: cost of 28 {{.*}} %V4F64 = call <4 x double> @llvm.sqrt.v4f64
; AVX512: cost of 28 {{.*}} %V4F64 = call <4 x double> @llvm.sqrt.v4f64
+ ; SLM: cost of 64 {{.*}} %V4F64 = call <4 x double> @llvm.sqrt.v4f64
%V4F64 = call <4 x double> @llvm.sqrt.v4f64(<4 x double> undef)
; SSE2: cost of 128 {{.*}} %V8F64 = call <8 x double> @llvm.sqrt.v8f64
; SSE42: cost of 128 {{.*}} %V8F64 = call <8 x double> @llvm.sqrt.v8f64
; AVX: cost of 86 {{.*}} %V8F64 = call <8 x double> @llvm.sqrt.v8f64
; AVX2: cost of 56 {{.*}} %V8F64 = call <8 x double> @llvm.sqrt.v8f64
; AVX512: cost of 1 {{.*}} %V8F64 = call <8 x double> @llvm.sqrt.v8f64
+ ; SLM: cost of 128 {{.*}} %V8F64 = call <8 x double> @llvm.sqrt.v8f64
%V8F64 = call <8 x double> @llvm.sqrt.v8f64(<8 x double> undef)
ret i32 undef
@@ -345,24 +394,28 @@ define i32 @fabs(i32 %arg) {
; AVX: cost of 2 {{.*}} %F32 = call float @llvm.fabs.f32
; AVX2: cost of 2 {{.*}} %F32 = call float @llvm.fabs.f32
; AVX512: cost of 2 {{.*}} %F32 = call float @llvm.fabs.f32
+ ; SLM: cost of 2 {{.*}} %F32 = call float @llvm.fabs.f32
%F32 = call float @llvm.fabs.f32(float undef)
; SSE2: cost of 2 {{.*}} %V4F32 = call <4 x float> @llvm.fabs.v4f32
; SSE42: cost of 2 {{.*}} %V4F32 = call <4 x float> @llvm.fabs.v4f32
; AVX: cost of 2 {{.*}} %V4F32 = call <4 x float> @llvm.fabs.v4f32
; AVX2: cost of 2 {{.*}} %V4F32 = call <4 x float> @llvm.fabs.v4f32
; AVX512: cost of 2 {{.*}} %V4F32 = call <4 x float> @llvm.fabs.v4f32
+ ; SLM: cost of 2 {{.*}} %V4F32 = call <4 x float> @llvm.fabs.v4f32
%V4F32 = call <4 x float> @llvm.fabs.v4f32(<4 x float> undef)
; SSE2: cost of 4 {{.*}} %V8F32 = call <8 x float> @llvm.fabs.v8f32
; SSE42: cost of 4 {{.*}} %V8F32 = call <8 x float> @llvm.fabs.v8f32
; AVX: cost of 2 {{.*}} %V8F32 = call <8 x float> @llvm.fabs.v8f32
; AVX2: cost of 2 {{.*}} %V8F32 = call <8 x float> @llvm.fabs.v8f32
; AVX512: cost of 2 {{.*}} %V8F32 = call <8 x float> @llvm.fabs.v8f32
+ ; SLM: cost of 4 {{.*}} %V8F32 = call <8 x float> @llvm.fabs.v8f32
%V8F32 = call <8 x float> @llvm.fabs.v8f32(<8 x float> undef)
; SSE2: cost of 8 {{.*}} %V16F32 = call <16 x float> @llvm.fabs.v16f32
; SSE42: cost of 8 {{.*}} %V16F32 = call <16 x float> @llvm.fabs.v16f32
; AVX: cost of 4 {{.*}} %V16F32 = call <16 x float> @llvm.fabs.v16f32
; AVX2: cost of 4 {{.*}} %V16F32 = call <16 x float> @llvm.fabs.v16f32
; AVX512: cost of 2 {{.*}} %V16F32 = call <16 x float> @llvm.fabs.v16f32
+ ; SLM: cost of 8 {{.*}} %V16F32 = call <16 x float> @llvm.fabs.v16f32
%V16F32 = call <16 x float> @llvm.fabs.v16f32(<16 x float> undef)
; SSE2: cost of 2 {{.*}} %F64 = call double @llvm.fabs.f64
@@ -370,24 +423,28 @@ define i32 @fabs(i32 %arg) {
; AVX: cost of 2 {{.*}} %F64 = call double @llvm.fabs.f64
; AVX2: cost of 2 {{.*}} %F64 = call double @llvm.fabs.f64
; AVX512: cost of 2 {{.*}} %F64 = call double @llvm.fabs.f64
+ ; SLM: cost of 2 {{.*}} %F64 = call double @llvm.fabs.f64
%F64 = call double @llvm.fabs.f64(double undef)
; SSE2: cost of 2 {{.*}} %V2F64 = call <2 x double> @llvm.fabs.v2f64
; SSE42: cost of 2 {{.*}} %V2F64 = call <2 x double> @llvm.fabs.v2f64
; AVX: cost of 2 {{.*}} %V2F64 = call <2 x double> @llvm.fabs.v2f64
; AVX2: cost of 2 {{.*}} %V2F64 = call <2 x double> @llvm.fabs.v2f64
; AVX512: cost of 2 {{.*}} %V2F64 = call <2 x double> @llvm.fabs.v2f64
+ ; SLM: cost of 2 {{.*}} %V2F64 = call <2 x double> @llvm.fabs.v2f64
%V2F64 = call <2 x double> @llvm.fabs.v2f64(<2 x double> undef)
; SSE2: cost of 4 {{.*}} %V4F64 = call <4 x double> @llvm.fabs.v4f64
; SSE42: cost of 4 {{.*}} %V4F64 = call <4 x double> @llvm.fabs.v4f64
; AVX: cost of 2 {{.*}} %V4F64 = call <4 x double> @llvm.fabs.v4f64
; AVX2: cost of 2 {{.*}} %V4F64 = call <4 x double> @llvm.fabs.v4f64
; AVX512: cost of 2 {{.*}} %V4F64 = call <4 x double> @llvm.fabs.v4f64
+ ; SLM: cost of 4 {{.*}} %V4F64 = call <4 x double> @llvm.fabs.v4f64
%V4F64 = call <4 x double> @llvm.fabs.v4f64(<4 x double> undef)
; SSE2: cost of 8 {{.*}} %V8F64 = call <8 x double> @llvm.fabs.v8f64
; SSE42: cost of 8 {{.*}} %V8F64 = call <8 x double> @llvm.fabs.v8f64
; AVX: cost of 4 {{.*}} %V8F64 = call <8 x double> @llvm.fabs.v8f64
; AVX2: cost of 4 {{.*}} %V8F64 = call <8 x double> @llvm.fabs.v8f64
; AVX512: cost of 2 {{.*}} %V8F64 = call <8 x double> @llvm.fabs.v8f64
+ ; SLM: cost of 8 {{.*}} %V8F64 = call <8 x double> @llvm.fabs.v8f64
%V8F64 = call <8 x double> @llvm.fabs.v8f64(<8 x double> undef)
ret i32 undef
@@ -400,24 +457,28 @@ define i32 @fcopysign(i32 %arg) {
; AVX: cost of 2 {{.*}} %F32 = call float @llvm.copysign.f32
; AVX2: cost of 2 {{.*}} %F32 = call float @llvm.copysign.f32
; AVX512: cost of 2 {{.*}} %F32 = call float @llvm.copysign.f32
+ ; SLM: cost of 2 {{.*}} %F32 = call float @llvm.copysign.f32
%F32 = call float @llvm.copysign.f32(float undef, float undef)
; SSE2: cost of 2 {{.*}} %V4F32 = call <4 x float> @llvm.copysign.v4f32
; SSE42: cost of 2 {{.*}} %V4F32 = call <4 x float> @llvm.copysign.v4f32
; AVX: cost of 2 {{.*}} %V4F32 = call <4 x float> @llvm.copysign.v4f32
; AVX2: cost of 2 {{.*}} %V4F32 = call <4 x float> @llvm.copysign.v4f32
; AVX512: cost of 2 {{.*}} %V4F32 = call <4 x float> @llvm.copysign.v4f32
+ ; SLM: cost of 2 {{.*}} %V4F32 = call <4 x float> @llvm.copysign.v4f32
%V4F32 = call <4 x float> @llvm.copysign.v4f32(<4 x float> undef, <4 x float> undef)
; SSE2: cost of 4 {{.*}} %V8F32 = call <8 x float> @llvm.copysign.v8f32
; SSE42: cost of 4 {{.*}} %V8F32 = call <8 x float> @llvm.copysign.v8f32
; AVX: cost of 2 {{.*}} %V8F32 = call <8 x float> @llvm.copysign.v8f32
; AVX2: cost of 2 {{.*}} %V8F32 = call <8 x float> @llvm.copysign.v8f32
; AVX512: cost of 2 {{.*}} %V8F32 = call <8 x float> @llvm.copysign.v8f32
+ ; SLM: cost of 4 {{.*}} %V8F32 = call <8 x float> @llvm.copysign.v8f32
%V8F32 = call <8 x float> @llvm.copysign.v8f32(<8 x float> undef, <8 x float> undef)
; SSE2: cost of 8 {{.*}} %V16F32 = call <16 x float> @llvm.copysign.v16f32
; SSE42: cost of 8 {{.*}} %V16F32 = call <16 x float> @llvm.copysign.v16f32
; AVX: cost of 4 {{.*}} %V16F32 = call <16 x float> @llvm.copysign.v16f32
; AVX2: cost of 4 {{.*}} %V16F32 = call <16 x float> @llvm.copysign.v16f32
; AVX512: cost of 2 {{.*}} %V16F32 = call <16 x float> @llvm.copysign.v16f32
+ ; SLM: cost of 8 {{.*}} %V16F32 = call <16 x float> @llvm.copysign.v16f32
%V16F32 = call <16 x float> @llvm.copysign.v16f32(<16 x float> undef, <16 x float> undef)
; SSE2: cost of 2 {{.*}} %F64 = call double @llvm.copysign.f64
@@ -425,24 +486,28 @@ define i32 @fcopysign(i32 %arg) {
; AVX: cost of 2 {{.*}} %F64 = call double @llvm.copysign.f64
; AVX2: cost of 2 {{.*}} %F64 = call double @llvm.copysign.f64
; AVX512: cost of 2 {{.*}} %F64 = call double @llvm.copysign.f64
+ ; SLM: cost of 2 {{.*}} %F64 = call double @llvm.copysign.f64
%F64 = call double @llvm.copysign.f64(double undef, double undef)
; SSE2: cost of 2 {{.*}} %V2F64 = call <2 x double> @llvm.copysign.v2f64
; SSE42: cost of 2 {{.*}} %V2F64 = call <2 x double> @llvm.copysign.v2f64
; AVX: cost of 2 {{.*}} %V2F64 = call <2 x double> @llvm.copysign.v2f64
; AVX2: cost of 2 {{.*}} %V2F64 = call <2 x double> @llvm.copysign.v2f64
; AVX512: cost of 2 {{.*}} %V2F64 = call <2 x double> @llvm.copysign.v2f64
+ ; SLM: cost of 2 {{.*}} %V2F64 = call <2 x double> @llvm.copysign.v2f64
%V2F64 = call <2 x double> @llvm.copysign.v2f64(<2 x double> undef, <2 x double> undef)
; SSE2: cost of 4 {{.*}} %V4F64 = call <4 x double> @llvm.copysign.v4f64
; SSE42: cost of 4 {{.*}} %V4F64 = call <4 x double> @llvm.copysign.v4f64
; AVX: cost of 2 {{.*}} %V4F64 = call <4 x double> @llvm.copysign.v4f64
; AVX2: cost of 2 {{.*}} %V4F64 = call <4 x double> @llvm.copysign.v4f64
; AVX512: cost of 2 {{.*}} %V4F64 = call <4 x double> @llvm.copysign.v4f64
+ ; SLM: cost of 4 {{.*}} %V4F64 = call <4 x double> @llvm.copysign.v4f64
%V4F64 = call <4 x double> @llvm.copysign.v4f64(<4 x double> undef, <4 x double> undef)
; SSE2: cost of 8 {{.*}} %V8F64 = call <8 x double> @llvm.copysign.v8f64
; SSE42: cost of 8 {{.*}} %V8F64 = call <8 x double> @llvm.copysign.v8f64
; AVX: cost of 4 {{.*}} %V8F64 = call <8 x double> @llvm.copysign.v8f64
; AVX2: cost of 4 {{.*}} %V8F64 = call <8 x double> @llvm.copysign.v8f64
; AVX512: cost of 2 {{.*}} %V8F64 = call <8 x double> @llvm.copysign.v8f64
+ ; SLM: cost of 8 {{.*}} %V8F64 = call <8 x double> @llvm.copysign.v8f64
%V8F64 = call <8 x double> @llvm.copysign.v8f64(<8 x double> undef, <8 x double> undef)
ret i32 undef
@@ -455,24 +520,28 @@ define i32 @fma(i32 %arg) {
; AVX: cost of 1 {{.*}} %F32 = call float @llvm.fma.f32
; AVX2: cost of 1 {{.*}} %F32 = call float @llvm.fma.f32
; AVX512: cost of 1 {{.*}} %F32 = call float @llvm.fma.f32
+ ; SLM: cost of 10 {{.*}} %F32 = call float @llvm.fma.f32
%F32 = call float @llvm.fma.f32(float undef, float undef, float undef)
; SSE2: cost of 43 {{.*}} %V4F32 = call <4 x float> @llvm.fma.v4f32
; SSE42: cost of 43 {{.*}} %V4F32 = call <4 x float> @llvm.fma.v4f32
; AVX: cost of 1 {{.*}} %V4F32 = call <4 x float> @llvm.fma.v4f32
; AVX2: cost of 1 {{.*}} %V4F32 = call <4 x float> @llvm.fma.v4f32
; AVX512: cost of 1 {{.*}} %V4F32 = call <4 x float> @llvm.fma.v4f32
+ ; SLM: cost of 43 {{.*}} %V4F32 = call <4 x float> @llvm.fma.v4f32
%V4F32 = call <4 x float> @llvm.fma.v4f32(<4 x float> undef, <4 x float> undef, <4 x float> undef)
; SSE2: cost of 86 {{.*}} %V8F32 = call <8 x float> @llvm.fma.v8f32
; SSE42: cost of 86 {{.*}} %V8F32 = call <8 x float> @llvm.fma.v8f32
; AVX: cost of 1 {{.*}} %V8F32 = call <8 x float> @llvm.fma.v8f32
; AVX2: cost of 1 {{.*}} %V8F32 = call <8 x float> @llvm.fma.v8f32
; AVX512: cost of 1 {{.*}} %V8F32 = call <8 x float> @llvm.fma.v8f32
+ ; SLM: cost of 86 {{.*}} %V8F32 = call <8 x float> @llvm.fma.v8f32
%V8F32 = call <8 x float> @llvm.fma.v8f32(<8 x float> undef, <8 x float> undef, <8 x float> undef)
; SSE2: cost of 172 {{.*}} %V16F32 = call <16 x float> @llvm.fma.v16f32
; SSE42: cost of 172 {{.*}} %V16F32 = call <16 x float> @llvm.fma.v16f32
; AVX: cost of 4 {{.*}} %V16F32 = call <16 x float> @llvm.fma.v16f32
; AVX2: cost of 4 {{.*}} %V16F32 = call <16 x float> @llvm.fma.v16f32
; AVX512: cost of 1 {{.*}} %V16F32 = call <16 x float> @llvm.fma.v16f32
+ ; SLM: cost of 172 {{.*}} %V16F32 = call <16 x float> @llvm.fma.v16f32
%V16F32 = call <16 x float> @llvm.fma.v16f32(<16 x float> undef, <16 x float> undef, <16 x float> undef)
; SSE2: cost of 10 {{.*}} %F64 = call double @llvm.fma.f64
@@ -480,24 +549,28 @@ define i32 @fma(i32 %arg) {
; AVX: cost of 1 {{.*}} %F64 = call double @llvm.fma.f64
; AVX2: cost of 1 {{.*}} %F64 = call double @llvm.fma.f64
; AVX512: cost of 1 {{.*}} %F64 = call double @llvm.fma.f64
+ ; SLM: cost of 10 {{.*}} %F64 = call double @llvm.fma.f64
%F64 = call double @llvm.fma.f64(double undef, double undef, double undef)
; SSE2: cost of 21 {{.*}} %V2F64 = call <2 x double> @llvm.fma.v2f64
; SSE42: cost of 21 {{.*}} %V2F64 = call <2 x double> @llvm.fma.v2f64
; AVX: cost of 1 {{.*}} %V2F64 = call <2 x double> @llvm.fma.v2f64
; AVX2: cost of 1 {{.*}} %V2F64 = call <2 x double> @llvm.fma.v2f64
; AVX512: cost of 1 {{.*}} %V2F64 = call <2 x double> @llvm.fma.v2f64
+ ; SLM: cost of 21 {{.*}} %V2F64 = call <2 x double> @llvm.fma.v2f64
%V2F64 = call <2 x double> @llvm.fma.v2f64(<2 x double> undef, <2 x double> undef, <2 x double> undef)
; SSE2: cost of 42 {{.*}} %V4F64 = call <4 x double> @llvm.fma.v4f64
; SSE42: cost of 42 {{.*}} %V4F64 = call <4 x double> @llvm.fma.v4f64
; AVX: cost of 1 {{.*}} %V4F64 = call <4 x double> @llvm.fma.v4f64
; AVX2: cost of 1 {{.*}} %V4F64 = call <4 x double> @llvm.fma.v4f64
; AVX512: cost of 1 {{.*}} %V4F64 = call <4 x double> @llvm.fma.v4f64
+ ; SLM: cost of 42 {{.*}} %V4F64 = call <4 x double> @llvm.fma.v4f64
%V4F64 = call <4 x double> @llvm.fma.v4f64(<4 x double> undef, <4 x double> undef, <4 x double> undef)
; SSE2: cost of 84 {{.*}} %V8F64 = call <8 x double> @llvm.fma.v8f64
; SSE42: cost of 84 {{.*}} %V8F64 = call <8 x double> @llvm.fma.v8f64
; AVX: cost of 4 {{.*}} %V8F64 = call <8 x double> @llvm.fma.v8f64
; AVX2: cost of 4 {{.*}} %V8F64 = call <8 x double> @llvm.fma.v8f64
; AVX512: cost of 1 {{.*}} %V8F64 = call <8 x double> @llvm.fma.v8f64
+ ; SLM: cost of 84 {{.*}} %V8F64 = call <8 x double> @llvm.fma.v8f64
%V8F64 = call <8 x double> @llvm.fma.v8f64(<8 x double> undef, <8 x double> undef, <8 x double> undef)
ret i32 undef
More information about the llvm-commits
mailing list