From a5dd08cac2991bbfcb304c8d4fc485381ff5f751 Mon Sep 17 00:00:00 2001 From: Roland Froese Date: Thu, 18 Apr 2019 18:12:09 +0000 Subject: [PowerPC] Add some PPC vec cost tests to prep for D60160 NFC llvm-svn: 358699 --- .../Analysis/CostModel/PowerPC/insert_extract.ll | 173 ++++++++++++++++++++- llvm/test/Analysis/CostModel/PowerPC/p9.ll | 68 -------- .../test/Analysis/CostModel/PowerPC/vector_unit.ll | 53 +++++++ 3 files changed, 225 insertions(+), 69 deletions(-) delete mode 100644 llvm/test/Analysis/CostModel/PowerPC/p9.ll create mode 100644 llvm/test/Analysis/CostModel/PowerPC/vector_unit.ll diff --git a/llvm/test/Analysis/CostModel/PowerPC/insert_extract.ll b/llvm/test/Analysis/CostModel/PowerPC/insert_extract.ll index 8dc003153a2..030af67d66d 100644 --- a/llvm/test/Analysis/CostModel/PowerPC/insert_extract.ll +++ b/llvm/test/Analysis/CostModel/PowerPC/insert_extract.ll @@ -1,16 +1,187 @@ -; RUN: opt < %s -cost-model -analyze -mtriple=powerpc64-unknown-linux-gnu -mcpu=pwr7 | FileCheck %s +; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py +; RUN: opt < %s -cost-model -analyze -mtriple=powerpc64-unknown-linux-gnu -mcpu=pwr7 -mattr=+vsx | FileCheck --check-prefix=CHECK-P7 %s +; RUN: opt < %s -cost-model -analyze -mtriple=powerpc64le-unknown-linux-gnu -mcpu=pwr8 -mattr=+vsx | FileCheck --check-prefix=CHECK-P8LE %s +; RUN: opt < %s -cost-model -analyze -mtriple=powerpc64-unknown-linux-gnu -mcpu=pwr9 -mattr=+vsx | FileCheck --check-prefix=CHECK-P9BE %s +; RUN: opt < %s -cost-model -analyze -mtriple=powerpc64le-unknown-linux-gnu -mcpu=pwr9 -mattr=+vsx | FileCheck --check-prefix=CHECK-P9LE %s + 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-f128:128:128-v128:128:128-n32:64" target triple = "powerpc64-unknown-linux-gnu" define i32 @insert(i32 %arg) { ; CHECK: cost of 10 {{.*}} insertelement +; CHECK-P7-LABEL: 'insert' +; CHECK-P7-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %x = insertelement <4 x i32> undef, i32 %arg, i32 0 +; CHECK-P7-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef +; +; CHECK-P8LE-LABEL: 'insert' +; CHECK-P8LE-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %x = insertelement <4 x i32> undef, i32 %arg, i32 0 +; CHECK-P8LE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef +; +; CHECK-P9BE-LABEL: 'insert' +; CHECK-P9BE-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %x = insertelement <4 x i32> undef, i32 %arg, i32 0 +; CHECK-P9BE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef +; +; CHECK-P9LE-LABEL: 'insert' +; CHECK-P9LE-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %x = insertelement <4 x i32> undef, i32 %arg, i32 0 +; CHECK-P9LE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef +; %x = insertelement <4 x i32> undef, i32 %arg, i32 0 ret i32 undef } define i32 @extract(<4 x i32> %arg) { ; CHECK: cost of 3 {{.*}} extractelement +; CHECK-P7-LABEL: 'extract' +; CHECK-P7-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %x = extractelement <4 x i32> %arg, i32 0 +; CHECK-P7-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 %x +; +; CHECK-P8LE-LABEL: 'extract' +; CHECK-P8LE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %x = extractelement <4 x i32> %arg, i32 0 +; CHECK-P8LE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 %x +; +; CHECK-P9BE-LABEL: 'extract' +; CHECK-P9BE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %x = extractelement <4 x i32> %arg, i32 0 +; CHECK-P9BE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 %x +; +; CHECK-P9LE-LABEL: 'extract' +; CHECK-P9LE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %x = extractelement <4 x i32> %arg, i32 0 +; CHECK-P9LE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 %x +; %x = extractelement <4 x i32> %arg, i32 0 ret i32 %x } +define void @test2xdouble(<2 x double> %arg1) { +; CHECK-P7-LABEL: 'test2xdouble' +; CHECK-P7-NEXT: Cost Model: Found an estimated cost of 0 for instruction: %v1 = extractelement <2 x double> %arg1, i32 0 +; CHECK-P7-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v2 = extractelement <2 x double> %arg1, i32 1 +; CHECK-P7-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void +; +; CHECK-P8LE-LABEL: 'test2xdouble' +; CHECK-P8LE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v1 = extractelement <2 x double> %arg1, i32 0 +; CHECK-P8LE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: %v2 = extractelement <2 x double> %arg1, i32 1 +; CHECK-P8LE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void +; +; CHECK-P9BE-LABEL: 'test2xdouble' +; CHECK-P9BE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: %v1 = extractelement <2 x double> %arg1, i32 0 +; CHECK-P9BE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2 = extractelement <2 x double> %arg1, i32 1 +; CHECK-P9BE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void +; +; CHECK-P9LE-LABEL: 'test2xdouble' +; CHECK-P9LE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v1 = extractelement <2 x double> %arg1, i32 0 +; CHECK-P9LE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: %v2 = extractelement <2 x double> %arg1, i32 1 +; CHECK-P9LE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void +; + %v1 = extractelement <2 x double> %arg1, i32 0 + %v2 = extractelement <2 x double> %arg1, i32 1 + ret void +} + +define void @test4xi32(<4 x i32> %v1, i32 %x1) { +; CHECK-P7-LABEL: 'test4xi32' +; CHECK-P7-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %v2 = insertelement <4 x i32> %v1, i32 %x1, i32 2 +; CHECK-P7-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void +; +; CHECK-P8LE-LABEL: 'test4xi32' +; CHECK-P8LE-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %v2 = insertelement <4 x i32> %v1, i32 %x1, i32 2 +; CHECK-P8LE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void +; +; CHECK-P9BE-LABEL: 'test4xi32' +; CHECK-P9BE-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %v2 = insertelement <4 x i32> %v1, i32 %x1, i32 2 +; CHECK-P9BE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void +; +; CHECK-P9LE-LABEL: 'test4xi32' +; CHECK-P9LE-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %v2 = insertelement <4 x i32> %v1, i32 %x1, i32 2 +; CHECK-P9LE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void +; + %v2 = insertelement <4 x i32> %v1, i32 %x1, i32 2 + ret void +} + +define void @vexti32(<4 x i32> %p1) { +; CHECK-P7-LABEL: 'vexti32' +; CHECK-P7-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %i1 = extractelement <4 x i32> %p1, i32 0 +; CHECK-P7-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %i2 = extractelement <4 x i32> %p1, i32 1 +; CHECK-P7-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %i3 = extractelement <4 x i32> %p1, i32 2 +; CHECK-P7-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %i4 = extractelement <4 x i32> %p1, i32 3 +; CHECK-P7-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void +; +; CHECK-P8LE-LABEL: 'vexti32' +; CHECK-P8LE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %i1 = extractelement <4 x i32> %p1, i32 0 +; CHECK-P8LE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %i2 = extractelement <4 x i32> %p1, i32 1 +; CHECK-P8LE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %i3 = extractelement <4 x i32> %p1, i32 2 +; CHECK-P8LE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %i4 = extractelement <4 x i32> %p1, i32 3 +; CHECK-P8LE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void +; +; CHECK-P9BE-LABEL: 'vexti32' +; CHECK-P9BE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %i1 = extractelement <4 x i32> %p1, i32 0 +; CHECK-P9BE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %i2 = extractelement <4 x i32> %p1, i32 1 +; CHECK-P9BE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %i3 = extractelement <4 x i32> %p1, i32 2 +; CHECK-P9BE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %i4 = extractelement <4 x i32> %p1, i32 3 +; CHECK-P9BE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void +; +; CHECK-P9LE-LABEL: 'vexti32' +; CHECK-P9LE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %i1 = extractelement <4 x i32> %p1, i32 0 +; CHECK-P9LE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %i2 = extractelement <4 x i32> %p1, i32 1 +; CHECK-P9LE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %i3 = extractelement <4 x i32> %p1, i32 2 +; CHECK-P9LE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %i4 = extractelement <4 x i32> %p1, i32 3 +; CHECK-P9LE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void +; + %i1 = extractelement <4 x i32> %p1, i32 0 + %i2 = extractelement <4 x i32> %p1, i32 1 + %i3 = extractelement <4 x i32> %p1, i32 2 + %i4 = extractelement <4 x i32> %p1, i32 3 + ret void +} + +define void @vexti64(<2 x i64> %p1) { +; CHECK-P7-LABEL: 'vexti64' +; CHECK-P7-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %i1 = extractelement <2 x i64> %p1, i32 0 +; CHECK-P7-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %i2 = extractelement <2 x i64> %p1, i32 1 +; CHECK-P7-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void +; +; CHECK-P8LE-LABEL: 'vexti64' +; CHECK-P8LE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %i1 = extractelement <2 x i64> %p1, i32 0 +; CHECK-P8LE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %i2 = extractelement <2 x i64> %p1, i32 1 +; CHECK-P8LE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void +; +; CHECK-P9BE-LABEL: 'vexti64' +; CHECK-P9BE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %i1 = extractelement <2 x i64> %p1, i32 0 +; CHECK-P9BE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %i2 = extractelement <2 x i64> %p1, i32 1 +; CHECK-P9BE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void +; +; CHECK-P9LE-LABEL: 'vexti64' +; CHECK-P9LE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %i1 = extractelement <2 x i64> %p1, i32 0 +; CHECK-P9LE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %i2 = extractelement <2 x i64> %p1, i32 1 +; CHECK-P9LE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void +; + %i1 = extractelement <2 x i64> %p1, i32 0 + %i2 = extractelement <2 x i64> %p1, i32 1 + ret void +} + +define void @vext(<8 x i16> %p1, <16 x i8> %p2) { +; CHECK-P7-LABEL: 'vext' +; CHECK-P7-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %i1 = extractelement <8 x i16> %p1, i32 0 +; CHECK-P7-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %i2 = extractelement <16 x i8> %p2, i32 0 +; CHECK-P7-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void +; +; CHECK-P8LE-LABEL: 'vext' +; CHECK-P8LE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %i1 = extractelement <8 x i16> %p1, i32 0 +; CHECK-P8LE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %i2 = extractelement <16 x i8> %p2, i32 0 +; CHECK-P8LE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void +; +; CHECK-P9BE-LABEL: 'vext' +; CHECK-P9BE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %i1 = extractelement <8 x i16> %p1, i32 0 +; CHECK-P9BE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %i2 = extractelement <16 x i8> %p2, i32 0 +; CHECK-P9BE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void +; +; CHECK-P9LE-LABEL: 'vext' +; CHECK-P9LE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %i1 = extractelement <8 x i16> %p1, i32 0 +; CHECK-P9LE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %i2 = extractelement <16 x i8> %p2, i32 0 +; CHECK-P9LE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void +; + %i1 = extractelement <8 x i16> %p1, i32 0 + %i2 = extractelement <16 x i8> %p2, i32 0 + ret void +} diff --git a/llvm/test/Analysis/CostModel/PowerPC/p9.ll b/llvm/test/Analysis/CostModel/PowerPC/p9.ll deleted file mode 100644 index 38ffc077628..00000000000 --- a/llvm/test/Analysis/CostModel/PowerPC/p9.ll +++ /dev/null @@ -1,68 +0,0 @@ -; RUN: opt < %s -cost-model -analyze -mtriple=powerpc64-unknown-linux-gnu -mcpu=pwr7 -mattr=+vsx | FileCheck %s -; RUN: opt < %s -cost-model -analyze -mtriple=powerpc64-unknown-linux-gnu -mcpu=pwr9 -mattr=+vsx | FileCheck --check-prefix=CHECK-P9 %s -; RUN: opt < %s -cost-model -analyze -mtriple=powerpc64le-unknown-linux-gnu -mcpu=pwr9 -mattr=+vsx | FileCheck --check-prefix=CHECK-LE %s - -define void @testi16(i16 %arg1, i16 %arg2, i16* %arg3) { - - %s1 = add i16 %arg1, %arg2 - %s2 = zext i16 %arg1 to i32 - %s3 = load i16, i16* %arg3 - store i16 %arg2, i16* %arg3 - %c = icmp eq i16 %arg1, %arg2 - - ret void - ; CHECK: cost of 1 {{.*}} add - ; CHECK: cost of 1 {{.*}} zext - ; CHECK: cost of 1 {{.*}} load - ; CHECK: cost of 1 {{.*}} store - ; CHECK: cost of 1 {{.*}} icmp - ; CHECK-P9: cost of 1 {{.*}} add - ; CHECK-P9: cost of 1 {{.*}} zext - ; CHECK-P9: cost of 1 {{.*}} load - ; CHECK-P9: cost of 1 {{.*}} store - ; CHECK-P9: cost of 1 {{.*}} icmp -} - -define void @test4xi16(<4 x i16> %arg1, <4 x i16> %arg2) { - - %v1 = add <4 x i16> %arg1, %arg2 - %v2 = zext <4 x i16> %arg1 to <4 x i32> - %v3 = shufflevector <4 x i16> %arg1, <4 x i16> undef, <4 x i32> zeroinitializer - %c = icmp eq <4 x i16> %arg1, %arg2 - - ret void - ; CHECK: cost of 1 {{.*}} add - ; CHECK: cost of 1 {{.*}} zext - ; CHECK: cost of 1 {{.*}} shufflevector - ; CHECK: cost of 1 {{.*}} icmp - ; CHECK-P9: cost of 2 {{.*}} add - ; CHECK-P9: cost of 2 {{.*}} zext - ; CHECK-P9: cost of 2 {{.*}} shufflevector - ; CHECK-P9: cost of 2 {{.*}} icmp -} - -define void @test4xi32(<4 x i32> %arg1, <4 x i32> %arg2, <4 x i32>* %arg3) { - - %v1 = load <4 x i32>, <4 x i32>* %arg3 - store <4 x i32> %arg2, <4 x i32>* %arg3 - - ret void - ; CHECK: cost of 1 {{.*}} load - ; CHECK: cost of 1 {{.*}} store - ; CHECK-P9: cost of 2 {{.*}} load - ; CHECK-P9: cost of 2 {{.*}} store -} - -define void @test2xdouble(<2 x double> %arg1) { - %v1 = extractelement <2 x double> %arg1, i32 0 - %v2 = extractelement <2 x double> %arg1, i32 1 - - ret void - ; CHECK: cost of 0 {{.*}} extractelement - ; CHECK: cost of 1 {{.*}} extractelement - ; CHECK-P9: cost of 0 {{.*}} extractelement - ; CHECK-P9: cost of 2 {{.*}} extractelement - ; CHECK-LE-LABEL: test2xdouble - ; CHECK-LE: cost of 2 {{.*}} extractelement - ; CHECK-LE: cost of 0 {{.*}} extractelement -} diff --git a/llvm/test/Analysis/CostModel/PowerPC/vector_unit.ll b/llvm/test/Analysis/CostModel/PowerPC/vector_unit.ll new file mode 100644 index 00000000000..d90e987b1fc --- /dev/null +++ b/llvm/test/Analysis/CostModel/PowerPC/vector_unit.ll @@ -0,0 +1,53 @@ +; RUN: opt < %s -cost-model -analyze -mtriple=powerpc64-unknown-linux-gnu -mcpu=pwr7 -mattr=+vsx | FileCheck %s +; RUN: opt < %s -cost-model -analyze -mtriple=powerpc64-unknown-linux-gnu -mcpu=pwr9 -mattr=+vsx | FileCheck --check-prefix=CHECK-P9 %s + +define void @testi16(i16 %arg1, i16 %arg2, i16* %arg3) { + + %s1 = add i16 %arg1, %arg2 + %s2 = zext i16 %arg1 to i32 + %s3 = load i16, i16* %arg3 + store i16 %arg2, i16* %arg3 + %c = icmp eq i16 %arg1, %arg2 + + ret void + ; CHECK: cost of 1 {{.*}} add + ; CHECK: cost of 1 {{.*}} zext + ; CHECK: cost of 1 {{.*}} load + ; CHECK: cost of 1 {{.*}} store + ; CHECK: cost of 1 {{.*}} icmp + ; CHECK-P9: cost of 1 {{.*}} add + ; CHECK-P9: cost of 1 {{.*}} zext + ; CHECK-P9: cost of 1 {{.*}} load + ; CHECK-P9: cost of 1 {{.*}} store + ; CHECK-P9: cost of 1 {{.*}} icmp +} + +define void @test4xi16(<4 x i16> %arg1, <4 x i16> %arg2) { + + %v1 = add <4 x i16> %arg1, %arg2 + %v2 = zext <4 x i16> %arg1 to <4 x i32> + %v3 = shufflevector <4 x i16> %arg1, <4 x i16> undef, <4 x i32> zeroinitializer + %c = icmp eq <4 x i16> %arg1, %arg2 + + ret void + ; CHECK: cost of 1 {{.*}} add + ; CHECK: cost of 1 {{.*}} zext + ; CHECK: cost of 1 {{.*}} shufflevector + ; CHECK: cost of 1 {{.*}} icmp + ; CHECK-P9: cost of 2 {{.*}} add + ; CHECK-P9: cost of 2 {{.*}} zext + ; CHECK-P9: cost of 2 {{.*}} shufflevector + ; CHECK-P9: cost of 2 {{.*}} icmp +} + +define void @test4xi32(<4 x i32> %arg1, <4 x i32> %arg2, <4 x i32>* %arg3) { + + %v1 = load <4 x i32>, <4 x i32>* %arg3 + store <4 x i32> %arg2, <4 x i32>* %arg3 + + ret void + ; CHECK: cost of 1 {{.*}} load + ; CHECK: cost of 1 {{.*}} store + ; CHECK-P9: cost of 2 {{.*}} load + ; CHECK-P9: cost of 2 {{.*}} store +} -- cgit v1.2.3