Skip to content

Remove optimization flags from clang codegen tests #113714

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 3 commits into from
Oct 28, 2024
Merged

Remove optimization flags from clang codegen tests #113714

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
5c47d3e
Remove -O3 in clang codegen test
MatzeB Oct 25, 2024
2c2009a
add mem2reg, do not change avx-cast-builtins
MatzeB Oct 25, 2024
35f8c19
Address review feedback
MatzeB Oct 28, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
104 changes: 83 additions & 21 deletions clang/test/CodeGen/X86/avx-cmp-builtins.c
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
// RUN: %clang_cc1 -flax-vector-conversions=none -ffreestanding %s -O3 -triple=x86_64-apple-darwin -target-feature +avx -emit-llvm -o - | FileCheck %s
// RUN: %clang_cc1 -flax-vector-conversions=none -ffreestanding %s -O3 -triple=i386-apple-darwin -target-feature +avx -emit-llvm -o - | FileCheck %s
// FIXME: The shufflevector instructions in test_cmpgt_sd are relying on O3 here.
// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 5
// RUN: %clang_cc1 -flax-vector-conversions=none -ffreestanding %s -triple=x86_64-- -target-feature +avx -disable-O0-optnone -emit-llvm -o - | opt -S -passes=mem2reg | FileCheck %s
// RUN: %clang_cc1 -flax-vector-conversions=none -ffreestanding %s -triple=i386-- -target-feature +avx -disable-O0-optnone -emit-llvm -o - | opt -S -passes=mem2reg | FileCheck %s


#include <immintrin.h>
Expand All @@ -9,62 +9,124 @@
// Test LLVM IR codegen of cmpXY instructions
//

// CHECK-LABEL: define dso_local <2 x double> @test_cmp_sd(
// CHECK-SAME: <2 x double> noundef [[A:%.*]], <2 x double> noundef [[B:%.*]]) #[[ATTR0:[0-9]+]] {
// CHECK-NEXT: [[ENTRY:.*:]]
// CHECK-NEXT: [[TMP0:%.*]] = call <2 x double> @llvm.x86.sse2.cmp.sd(<2 x double> [[A]], <2 x double> [[B]], i8 13)
// CHECK-NEXT: ret <2 x double> [[TMP0]]
//
__m128d test_cmp_sd(__m128d a, __m128d b) {
// Expects that the third argument in LLVM IR is immediate expression
// CHECK: @llvm.x86.sse2.cmp.sd({{.*}}, i8 13)
return _mm_cmp_sd(a, b, _CMP_GE_OS);
}

// CHECK-LABEL: define dso_local <4 x float> @test_cmp_ss(
// CHECK-SAME: <4 x float> noundef [[A:%.*]], <4 x float> noundef [[B:%.*]]) #[[ATTR0]] {
// CHECK-NEXT: [[ENTRY:.*:]]
// CHECK-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.x86.sse.cmp.ss(<4 x float> [[A]], <4 x float> [[B]], i8 13)
// CHECK-NEXT: ret <4 x float> [[TMP0]]
//
__m128 test_cmp_ss(__m128 a, __m128 b) {
// Expects that the third argument in LLVM IR is immediate expression
// CHECK: @llvm.x86.sse.cmp.ss({{.*}}, i8 13)
return _mm_cmp_ss(a, b, _CMP_GE_OS);
}

// CHECK-LABEL: define dso_local <4 x float> @test_cmpgt_ss(
// CHECK-SAME: <4 x float> noundef [[A:%.*]], <4 x float> noundef [[B:%.*]]) #[[ATTR0]] {
// CHECK-NEXT: [[ENTRY:.*:]]
// CHECK-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.x86.sse.cmp.ss(<4 x float> [[B]], <4 x float> [[A]], i8 1)
// CHECK-NEXT: [[SHUFFLE_I:%.*]] = shufflevector <4 x float> [[A]], <4 x float> [[TMP0]], <4 x i32> <i32 4, i32 1, i32 2, i32 3>
// CHECK-NEXT: ret <4 x float> [[SHUFFLE_I]]
//
__m128 test_cmpgt_ss(__m128 a, __m128 b) {
// CHECK: @llvm.x86.sse.cmp.ss({{.*}}, i8 1)
// CHECK: shufflevector <{{.*}}, <4 x i32> <i32 0, i32 5, i32 6, i32 7>
return _mm_cmpgt_ss(a, b);
}

// CHECK-LABEL: define dso_local <4 x float> @test_cmpge_ss(
// CHECK-SAME: <4 x float> noundef [[A:%.*]], <4 x float> noundef [[B:%.*]]) #[[ATTR0]] {
// CHECK-NEXT: [[ENTRY:.*:]]
// CHECK-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.x86.sse.cmp.ss(<4 x float> [[B]], <4 x float> [[A]], i8 2)
// CHECK-NEXT: [[SHUFFLE_I:%.*]] = shufflevector <4 x float> [[A]], <4 x float> [[TMP0]], <4 x i32> <i32 4, i32 1, i32 2, i32 3>
// CHECK-NEXT: ret <4 x float> [[SHUFFLE_I]]
//
__m128 test_cmpge_ss(__m128 a, __m128 b) {
// CHECK: @llvm.x86.sse.cmp.ss({{.*}}, i8 2)
// CHECK: shufflevector <{{.*}}, <4 x i32> <i32 0, i32 5, i32 6, i32 7>
return _mm_cmpge_ss(a, b);
}

// CHECK-LABEL: define dso_local <4 x float> @test_cmpngt_ss(
// CHECK-SAME: <4 x float> noundef [[A:%.*]], <4 x float> noundef [[B:%.*]]) #[[ATTR0]] {
// CHECK-NEXT: [[ENTRY:.*:]]
// CHECK-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.x86.sse.cmp.ss(<4 x float> [[B]], <4 x float> [[A]], i8 5)
// CHECK-NEXT: [[SHUFFLE_I:%.*]] = shufflevector <4 x float> [[A]], <4 x float> [[TMP0]], <4 x i32> <i32 4, i32 1, i32 2, i32 3>
// CHECK-NEXT: ret <4 x float> [[SHUFFLE_I]]
//
__m128 test_cmpngt_ss(__m128 a, __m128 b) {
// CHECK: @llvm.x86.sse.cmp.ss({{.*}}, i8 5)
// CHECK: shufflevector <{{.*}}, <4 x i32> <i32 0, i32 5, i32 6, i32 7>
return _mm_cmpngt_ss(a, b);
}

// CHECK-LABEL: define dso_local <4 x float> @test_cmpnge_ss(
// CHECK-SAME: <4 x float> noundef [[A:%.*]], <4 x float> noundef [[B:%.*]]) #[[ATTR0]] {
// CHECK-NEXT: [[ENTRY:.*:]]
// CHECK-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.x86.sse.cmp.ss(<4 x float> [[B]], <4 x float> [[A]], i8 6)
// CHECK-NEXT: [[SHUFFLE_I:%.*]] = shufflevector <4 x float> [[A]], <4 x float> [[TMP0]], <4 x i32> <i32 4, i32 1, i32 2, i32 3>
// CHECK-NEXT: ret <4 x float> [[SHUFFLE_I]]
//
__m128 test_cmpnge_ss(__m128 a, __m128 b) {
// CHECK: @llvm.x86.sse.cmp.ss({{.*}}, i8 6)
// CHECK: shufflevector <{{.*}}, <4 x i32> <i32 0, i32 5, i32 6, i32 7>
return _mm_cmpnge_ss(a, b);
}

// CHECK-LABEL: define dso_local <2 x double> @test_cmpgt_sd(
// CHECK-SAME: <2 x double> noundef [[A:%.*]], <2 x double> noundef [[B:%.*]]) #[[ATTR0]] {
// CHECK-NEXT: [[ENTRY:.*:]]
// CHECK-NEXT: [[TMP0:%.*]] = call <2 x double> @llvm.x86.sse2.cmp.sd(<2 x double> [[B]], <2 x double> [[A]], i8 1)
// CHECK-NEXT: [[VECEXT_I:%.*]] = extractelement <2 x double> [[TMP0]], i32 0
// CHECK-NEXT: [[VECINIT_I:%.*]] = insertelement <2 x double> poison, double [[VECEXT_I]], i32 0
// CHECK-NEXT: [[VECEXT1_I:%.*]] = extractelement <2 x double> [[A]], i32 1
// CHECK-NEXT: [[VECINIT2_I:%.*]] = insertelement <2 x double> [[VECINIT_I]], double [[VECEXT1_I]], i32 1
// CHECK-NEXT: ret <2 x double> [[VECINIT2_I]]
//
__m128d test_cmpgt_sd(__m128d a, __m128d b) {
// CHECK: @llvm.x86.sse2.cmp.sd({{.*}}, i8 1)
// CHECK: shufflevector <{{.*}}, <2 x i32> <i32 0, i32 3>
return _mm_cmpgt_sd(a, b);
}

// CHECK-LABEL: define dso_local <2 x double> @test_cmpge_sd(
// CHECK-SAME: <2 x double> noundef [[A:%.*]], <2 x double> noundef [[B:%.*]]) #[[ATTR0]] {
// CHECK-NEXT: [[ENTRY:.*:]]
// CHECK-NEXT: [[TMP0:%.*]] = call <2 x double> @llvm.x86.sse2.cmp.sd(<2 x double> [[B]], <2 x double> [[A]], i8 2)
// CHECK-NEXT: [[VECEXT_I:%.*]] = extractelement <2 x double> [[TMP0]], i32 0
// CHECK-NEXT: [[VECINIT_I:%.*]] = insertelement <2 x double> poison, double [[VECEXT_I]], i32 0
// CHECK-NEXT: [[VECEXT1_I:%.*]] = extractelement <2 x double> [[A]], i32 1
// CHECK-NEXT: [[VECINIT2_I:%.*]] = insertelement <2 x double> [[VECINIT_I]], double [[VECEXT1_I]], i32 1
// CHECK-NEXT: ret <2 x double> [[VECINIT2_I]]
//
__m128d test_cmpge_sd(__m128d a, __m128d b) {
// CHECK: @llvm.x86.sse2.cmp.sd({{.*}}, i8 2)
// CHECK: shufflevector <{{.*}}, <2 x i32> <i32 0, i32 3>
return _mm_cmpge_sd(a, b);
}

// CHECK-LABEL: define dso_local <2 x double> @test_cmpngt_sd(
// CHECK-SAME: <2 x double> noundef [[A:%.*]], <2 x double> noundef [[B:%.*]]) #[[ATTR0]] {
// CHECK-NEXT: [[ENTRY:.*:]]
// CHECK-NEXT: [[TMP0:%.*]] = call <2 x double> @llvm.x86.sse2.cmp.sd(<2 x double> [[B]], <2 x double> [[A]], i8 5)
// CHECK-NEXT: [[VECEXT_I:%.*]] = extractelement <2 x double> [[TMP0]], i32 0
// CHECK-NEXT: [[VECINIT_I:%.*]] = insertelement <2 x double> poison, double [[VECEXT_I]], i32 0
// CHECK-NEXT: [[VECEXT1_I:%.*]] = extractelement <2 x double> [[A]], i32 1
// CHECK-NEXT: [[VECINIT2_I:%.*]] = insertelement <2 x double> [[VECINIT_I]], double [[VECEXT1_I]], i32 1
// CHECK-NEXT: ret <2 x double> [[VECINIT2_I]]
//
__m128d test_cmpngt_sd(__m128d a, __m128d b) {
// CHECK: @llvm.x86.sse2.cmp.sd({{.*}}, i8 5)
// CHECK: shufflevector <{{.*}}, <2 x i32> <i32 0, i32 3>
return _mm_cmpngt_sd(a, b);
}

// CHECK-LABEL: define dso_local <2 x double> @test_cmpnge_sd(
// CHECK-SAME: <2 x double> noundef [[A:%.*]], <2 x double> noundef [[B:%.*]]) #[[ATTR0]] {
// CHECK-NEXT: [[ENTRY:.*:]]
// CHECK-NEXT: [[TMP0:%.*]] = call <2 x double> @llvm.x86.sse2.cmp.sd(<2 x double> [[B]], <2 x double> [[A]], i8 6)
// CHECK-NEXT: [[VECEXT_I:%.*]] = extractelement <2 x double> [[TMP0]], i32 0
// CHECK-NEXT: [[VECINIT_I:%.*]] = insertelement <2 x double> poison, double [[VECEXT_I]], i32 0
// CHECK-NEXT: [[VECEXT1_I:%.*]] = extractelement <2 x double> [[A]], i32 1
// CHECK-NEXT: [[VECINIT2_I:%.*]] = insertelement <2 x double> [[VECINIT_I]], double [[VECEXT1_I]], i32 1
// CHECK-NEXT: ret <2 x double> [[VECINIT2_I]]
//
__m128d test_cmpnge_sd(__m128d a, __m128d b) {
// CHECK: @llvm.x86.sse2.cmp.sd({{.*}}, i8 6)
// CHECK: shufflevector <{{.*}}, <2 x i32> <i32 0, i32 3>
return _mm_cmpnge_sd(a, b);
}
Loading
Loading