derek/TheRock
1
0 Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity
Files
main
TheRock/compiler/amd-llvm/llvm/test/Transforms/InstSimplify/ConstProp/sse.ll
T
derek.holloway 7502bf64a4
CodeQL / Analyze (actions) (push) Canceled after 0s
Details
CodeQL / Analyze (python) (push) Canceled after 0s
Details
pre-commit / pre-commit (push) Canceled after 0s
Details
Multi-Arch CI / setup (push) Canceled after 0s
Details
Publish build_manylinux_x86_64 images / publish_build_manylinux_x86_64 (push) Canceled after 0s
Details
Setup (Multi-Arch) / setup (push) Canceled after 0s
Details
Multi-Arch CI / Linux::${{ fromJSON(needs.setup.outputs.linux_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Details
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Details
Publish no_rocm_image_ubi10 image / publish_no_rocm_image_ubi10 (push) Canceled after 0s
Details
Multi-Arch CI / Windows::${{ fromJSON(needs.setup.outputs.windows_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Details
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Details
Multi-Arch CI / Manifest Diff (push) Canceled after 0s
Details
TheRock Manifest Diff Report / Generate Manifest Diff Report (push) Canceled after 0s
Details
Multi-Arch CI / CI Summary (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04 (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04_ocl_rt images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Details
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Details
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04_rocgdb images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Details
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Details
Unit Tests / Unit Tests :: ubuntu-24.04 (push) Canceled after 0s
Details
Unit Tests / Unit Tests :: windows-2022 (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_media (push) Canceled after 0s
Details
Unit Tests / Unit Tests Summary (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_openmpi (push) Canceled after 0s
Details
Flatten subprojects into a single base monorepo
2026-09-02 16:34:47 -07:00

209 lines
8.1 KiB
LLVM
Raw Permalink Blame History

; RUN: opt < %s -passes=instsimplify -S | FileCheck %s
; REQUIRES: x86-registered-target
define i1 @test_sse_cvts_exact() nounwind readnone {
; CHECK-LABEL: @test_sse_cvts_exact(
; CHECK-NOT: call
; CHECK: ret i1 true
entry:
%i0 = tail call i32 @llvm.x86.sse.cvtss2si(<4 x float> <float 3.0, float undef, float undef, float undef>) nounwind
%i1 = tail call i64 @llvm.x86.sse.cvtss2si64(<4 x float> <float 3.0, float undef, float undef, float undef>) nounwind
%i2 = call i32 @llvm.x86.sse2.cvtsd2si(<2 x double> <double 7.0, double undef>) nounwind
%i3 = call i64 @llvm.x86.sse2.cvtsd2si64(<2 x double> <double 7.0, double undef>) nounwind
%sum02 = add i32 %i0, %i2
%sum13 = add i64 %i1, %i3
%cmp02 = icmp eq i32 %sum02, 10
%cmp13 = icmp eq i64 %sum13, 10
%b = and i1 %cmp02, %cmp13
ret i1 %b
}
; Inexact values should not fold as they are dependent on rounding mode
define i1 @test_sse_cvts_inexact() nounwind readnone {
; CHECK-LABEL: @test_sse_cvts_inexact(
; CHECK: call
; CHECK: call
; CHECK: call
; CHECK: call
entry:
%i0 = tail call i32 @llvm.x86.sse.cvtss2si(<4 x float> <float 1.75, float undef, float undef, float undef>) nounwind
%i1 = tail call i64 @llvm.x86.sse.cvtss2si64(<4 x float> <float 1.75, float undef, float undef, float undef>) nounwind
%i2 = call i32 @llvm.x86.sse2.cvtsd2si(<2 x double> <double 1.75, double undef>) nounwind
%i3 = call i64 @llvm.x86.sse2.cvtsd2si64(<2 x double> <double 1.75, double undef>) nounwind
%sum02 = add i32 %i0, %i2
%sum13 = add i64 %i1, %i3
%cmp02 = icmp eq i32 %sum02, 4
%cmp13 = icmp eq i64 %sum13, 4
%b = and i1 %cmp02, %cmp13
ret i1 %b
}
; FLT_MAX/DBL_MAX should not fold
define i1 @test_sse_cvts_max() nounwind readnone {
; CHECK-LABEL: @test_sse_cvts_max(
; CHECK: call
; CHECK: call
; CHECK: call
; CHECK: call
entry:
%fm = bitcast <4 x i32> <i32 2139095039, i32 undef, i32 undef, i32 undef> to <4 x float>
%dm = bitcast <2 x i64> <i64 9218868437227405311, i64 undef> to <2 x double>
%i0 = tail call i32 @llvm.x86.sse.cvtss2si(<4 x float> %fm) nounwind
%i1 = tail call i64 @llvm.x86.sse.cvtss2si64(<4 x float> %fm) nounwind
%i2 = call i32 @llvm.x86.sse2.cvtsd2si(<2 x double> %dm) nounwind
%i3 = call i64 @llvm.x86.sse2.cvtsd2si64(<2 x double> %dm) nounwind
%sum02 = add i32 %i0, %i2
%sum13 = add i64 %i1, %i3
%sum02.sext = sext i32 %sum02 to i64
%b = icmp eq i64 %sum02.sext, %sum13
ret i1 %b
}
; INF should not fold
define i1 @test_sse_cvts_inf() nounwind readnone {
; CHECK-LABEL: @test_sse_cvts_inf(
; CHECK: call
; CHECK: call
; CHECK: call
; CHECK: call
entry:
%fm = bitcast <4 x i32> <i32 2139095040, i32 undef, i32 undef, i32 undef> to <4 x float>
%dm = bitcast <2 x i64> <i64 9218868437227405312, i64 undef> to <2 x double>
%i0 = tail call i32 @llvm.x86.sse.cvtss2si(<4 x float> %fm) nounwind
%i1 = tail call i64 @llvm.x86.sse.cvtss2si64(<4 x float> %fm) nounwind
%i2 = call i32 @llvm.x86.sse2.cvtsd2si(<2 x double> %dm) nounwind
%i3 = call i64 @llvm.x86.sse2.cvtsd2si64(<2 x double> %dm) nounwind
%sum02 = add i32 %i0, %i2
%sum13 = add i64 %i1, %i3
%sum02.sext = sext i32 %sum02 to i64
%b = icmp eq i64 %sum02.sext, %sum13
ret i1 %b
}
; NAN should not fold
define i1 @test_sse_cvts_nan() nounwind readnone {
; CHECK-LABEL: @test_sse_cvts_nan(
; CHECK: call
; CHECK: call
; CHECK: call
; CHECK: call
entry:
%fm = bitcast <4 x i32> <i32 2143289344, i32 undef, i32 undef, i32 undef> to <4 x float>
%dm = bitcast <2 x i64> <i64 9221120237041090560, i64 undef> to <2 x double>
%i0 = tail call i32 @llvm.x86.sse.cvtss2si(<4 x float> %fm) nounwind
%i1 = tail call i64 @llvm.x86.sse.cvtss2si64(<4 x float> %fm) nounwind
%i2 = call i32 @llvm.x86.sse2.cvtsd2si(<2 x double> %dm) nounwind
%i3 = call i64 @llvm.x86.sse2.cvtsd2si64(<2 x double> %dm) nounwind
%sum02 = add i32 %i0, %i2
%sum13 = add i64 %i1, %i3
%sum02.sext = sext i32 %sum02 to i64
%b = icmp eq i64 %sum02.sext, %sum13
ret i1 %b
}
define i1 @test_sse_cvtts_exact() nounwind readnone {
; CHECK-LABEL: @test_sse_cvtts_exact(
; CHECK-NOT: call
; CHECK: ret i1 true
entry:
%i0 = tail call i32 @llvm.x86.sse.cvttss2si(<4 x float> <float 3.0, float undef, float undef, float undef>) nounwind
%i1 = tail call i64 @llvm.x86.sse.cvttss2si64(<4 x float> <float 3.0, float undef, float undef, float undef>) nounwind
%i2 = call i32 @llvm.x86.sse2.cvttsd2si(<2 x double> <double 7.0, double undef>) nounwind
%i3 = call i64 @llvm.x86.sse2.cvttsd2si64(<2 x double> <double 7.0, double undef>) nounwind
%sum02 = add i32 %i0, %i2
%sum13 = add i64 %i1, %i3
%cmp02 = icmp eq i32 %sum02, 10
%cmp13 = icmp eq i64 %sum13, 10
%b = and i1 %cmp02, %cmp13
ret i1 %b
}
define i1 @test_sse_cvtts_inexact() nounwind readnone {
; CHECK-LABEL: @test_sse_cvtts_inexact(
; CHECK-NOT: call
; CHECK: ret i1 true
entry:
%i0 = tail call i32 @llvm.x86.sse.cvttss2si(<4 x float> <float 1.75, float undef, float undef, float undef>) nounwind
%i1 = tail call i64 @llvm.x86.sse.cvttss2si64(<4 x float> <float 1.75, float undef, float undef, float undef>) nounwind
%i2 = call i32 @llvm.x86.sse2.cvttsd2si(<2 x double> <double 1.75, double undef>) nounwind
%i3 = call i64 @llvm.x86.sse2.cvttsd2si64(<2 x double> <double 1.75, double undef>) nounwind
%sum02 = add i32 %i0, %i2
%sum13 = add i64 %i1, %i3
%cmp02 = icmp eq i32 %sum02, 2
%cmp13 = icmp eq i64 %sum13, 2
%b = and i1 %cmp02, %cmp13
ret i1 %b
}
; FLT_MAX/DBL_MAX should not fold
define i1 @test_sse_cvtts_max() nounwind readnone {
; CHECK-LABEL: @test_sse_cvtts_max(
; CHECK: call
; CHECK: call
; CHECK: call
; CHECK: call
entry:
%fm = bitcast <4 x i32> <i32 2139095039, i32 undef, i32 undef, i32 undef> to <4 x float>
%dm = bitcast <2 x i64> <i64 9218868437227405311, i64 undef> to <2 x double>
%i0 = tail call i32 @llvm.x86.sse.cvttss2si(<4 x float> %fm) nounwind
%i1 = tail call i64 @llvm.x86.sse.cvttss2si64(<4 x float> %fm) nounwind
%i2 = call i32 @llvm.x86.sse2.cvttsd2si(<2 x double> %dm) nounwind
%i3 = call i64 @llvm.x86.sse2.cvttsd2si64(<2 x double> %dm) nounwind
%sum02 = add i32 %i0, %i2
%sum13 = add i64 %i1, %i3
%sum02.sext = sext i32 %sum02 to i64
%b = icmp eq i64 %sum02.sext, %sum13
ret i1 %b
}
; INF should not fold
define i1 @test_sse_cvtts_inf() nounwind readnone {
; CHECK-LABEL: @test_sse_cvtts_inf(
; CHECK: call
; CHECK: call
; CHECK: call
; CHECK: call
entry:
%fm = bitcast <4 x i32> <i32 2139095040, i32 undef, i32 undef, i32 undef> to <4 x float>
%dm = bitcast <2 x i64> <i64 9218868437227405312, i64 undef> to <2 x double>
%i0 = tail call i32 @llvm.x86.sse.cvttss2si(<4 x float> %fm) nounwind
%i1 = tail call i64 @llvm.x86.sse.cvttss2si64(<4 x float> %fm) nounwind
%i2 = call i32 @llvm.x86.sse2.cvttsd2si(<2 x double> %dm) nounwind
%i3 = call i64 @llvm.x86.sse2.cvttsd2si64(<2 x double> %dm) nounwind
%sum02 = add i32 %i0, %i2
%sum13 = add i64 %i1, %i3
%sum02.sext = sext i32 %sum02 to i64
%b = icmp eq i64 %sum02.sext, %sum13
ret i1 %b
}
; NAN should not fold
define i1 @test_sse_cvtts_nan() nounwind readnone {
; CHECK-LABEL: @test_sse_cvtts_nan(
; CHECK: call
; CHECK: call
; CHECK: call
; CHECK: call
entry:
%fm = bitcast <4 x i32> <i32 2143289344, i32 undef, i32 undef, i32 undef> to <4 x float>
%dm = bitcast <2 x i64> <i64 9221120237041090560, i64 undef> to <2 x double>
%i0 = tail call i32 @llvm.x86.sse.cvttss2si(<4 x float> %fm) nounwind
%i1 = tail call i64 @llvm.x86.sse.cvttss2si64(<4 x float> %fm) nounwind
%i2 = call i32 @llvm.x86.sse2.cvttsd2si(<2 x double> %dm) nounwind
%i3 = call i64 @llvm.x86.sse2.cvttsd2si64(<2 x double> %dm) nounwind
%sum02 = add i32 %i0, %i2
%sum13 = add i64 %i1, %i3
%sum02.sext = sext i32 %sum02 to i64
%b = icmp eq i64 %sum02.sext, %sum13
ret i1 %b
}
declare i32 @llvm.x86.sse.cvtss2si(<4 x float>) nounwind readnone
declare i32 @llvm.x86.sse.cvttss2si(<4 x float>) nounwind readnone
declare i64 @llvm.x86.sse.cvtss2si64(<4 x float>) nounwind readnone
declare i64 @llvm.x86.sse.cvttss2si64(<4 x float>) nounwind readnone
declare i32 @llvm.x86.sse2.cvtsd2si(<2 x double>) nounwind readnone
declare i32 @llvm.x86.sse2.cvttsd2si(<2 x double>) nounwind readnone
declare i64 @llvm.x86.sse2.cvtsd2si64(<2 x double>) nounwind readnone
declare i64 @llvm.x86.sse2.cvttsd2si64(<2 x double>) nounwind readnone
Reference in new issue View Git Blame Copy Permalink