derek.holloway
7502bf64a4
CodeQL / Analyze (actions) (push) Canceled after 0s
CodeQL / Analyze (python) (push) Canceled after 0s
pre-commit / pre-commit (push) Canceled after 0s
Multi-Arch CI / setup (push) Canceled after 0s
Publish build_manylinux_x86_64 images / publish_build_manylinux_x86_64 (push) Canceled after 0s
Setup (Multi-Arch) / setup (push) Canceled after 0s
Multi-Arch CI / Linux::${{ fromJSON(needs.setup.outputs.linux_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubi10 image / publish_no_rocm_image_ubi10 (push) Canceled after 0s
Multi-Arch CI / Windows::${{ fromJSON(needs.setup.outputs.windows_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Multi-Arch CI / Manifest Diff (push) Canceled after 0s
TheRock Manifest Diff Report / Generate Manifest Diff Report (push) Canceled after 0s
Multi-Arch CI / CI Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_ocl_rt images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_rocgdb images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Unit Tests / Unit Tests :: ubuntu-24.04 (push) Canceled after 0s
Unit Tests / Unit Tests :: windows-2022 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_media (push) Canceled after 0s
Unit Tests / Unit Tests Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_openmpi (push) Canceled after 0s
98 lines
3.5 KiB
LLVM
98 lines
3.5 KiB
LLVM
; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
|
|
; RUN: opt < %s -passes=instcombine -S | FileCheck %s
|
|
; RUN: opt < %s -passes=instcombine -S -data-layout=n32 | FileCheck %s
|
|
; RUN: opt < %s -passes=instcombine -S -data-layout=n32:64 | FileCheck %s
|
|
; RUN: opt < %s -passes=instcombine -S -data-layout=n32:64:128 | FileCheck %s
|
|
|
|
declare void @llvm.memcpy.p0.p0.i32(ptr nocapture, ptr nocapture, i32, i1) nounwind
|
|
|
|
; memcpy can be expanded inline with load/store (based on the datalayout?)
|
|
|
|
define void @copy_1_byte(ptr %d, ptr %s) {
|
|
; CHECK-LABEL: @copy_1_byte(
|
|
; CHECK-NEXT: [[TMP1:%.*]] = load i8, ptr [[S:%.*]], align 1
|
|
; CHECK-NEXT: store i8 [[TMP1]], ptr [[D:%.*]], align 1
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
call void @llvm.memcpy.p0.p0.i32(ptr %d, ptr %s, i32 1, i1 false)
|
|
ret void
|
|
}
|
|
|
|
define void @copy_2_bytes(ptr %d, ptr %s) {
|
|
; CHECK-LABEL: @copy_2_bytes(
|
|
; CHECK-NEXT: [[TMP1:%.*]] = load i16, ptr [[S:%.*]], align 1
|
|
; CHECK-NEXT: store i16 [[TMP1]], ptr [[D:%.*]], align 1
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
call void @llvm.memcpy.p0.p0.i32(ptr %d, ptr %s, i32 2, i1 false)
|
|
ret void
|
|
}
|
|
|
|
; We don't expand small non-power-of-2. Should we? Might be a target-dependent choice.
|
|
|
|
define void @copy_3_bytes(ptr %d, ptr %s) {
|
|
; CHECK-LABEL: @copy_3_bytes(
|
|
; CHECK-NEXT: call void @llvm.memcpy.p0.p0.i32(ptr noundef nonnull align 1 dereferenceable(3) [[D:%.*]], ptr noundef nonnull align 1 dereferenceable(3) [[S:%.*]], i32 3, i1 false)
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
call void @llvm.memcpy.p0.p0.i32(ptr %d, ptr %s, i32 3, i1 false)
|
|
ret void
|
|
}
|
|
|
|
define void @copy_4_bytes(ptr %d, ptr %s) {
|
|
; CHECK-LABEL: @copy_4_bytes(
|
|
; CHECK-NEXT: [[TMP1:%.*]] = load i32, ptr [[S:%.*]], align 1
|
|
; CHECK-NEXT: store i32 [[TMP1]], ptr [[D:%.*]], align 1
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
call void @llvm.memcpy.p0.p0.i32(ptr %d, ptr %s, i32 4, i1 false)
|
|
ret void
|
|
}
|
|
|
|
; We don't expand small non-power-of-2. Should we? Might be a target-dependent choice.
|
|
|
|
define void @copy_5_bytes(ptr %d, ptr %s) {
|
|
; CHECK-LABEL: @copy_5_bytes(
|
|
; CHECK-NEXT: call void @llvm.memcpy.p0.p0.i32(ptr noundef nonnull align 1 dereferenceable(5) [[D:%.*]], ptr noundef nonnull align 1 dereferenceable(5) [[S:%.*]], i32 5, i1 false)
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
call void @llvm.memcpy.p0.p0.i32(ptr %d, ptr %s, i32 5, i1 false)
|
|
ret void
|
|
}
|
|
|
|
define void @copy_8_bytes(ptr %d, ptr %s) {
|
|
; CHECK-LABEL: @copy_8_bytes(
|
|
; CHECK-NEXT: [[TMP1:%.*]] = load i64, ptr [[S:%.*]], align 1
|
|
; CHECK-NEXT: store i64 [[TMP1]], ptr [[D:%.*]], align 1
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
call void @llvm.memcpy.p0.p0.i32(ptr %d, ptr %s, i32 8, i1 false)
|
|
ret void
|
|
}
|
|
|
|
define void @copy_16_bytes(ptr %d, ptr %s) {
|
|
; CHECK-LABEL: @copy_16_bytes(
|
|
; CHECK-NEXT: call void @llvm.memcpy.p0.p0.i32(ptr noundef nonnull align 1 dereferenceable(16) [[D:%.*]], ptr noundef nonnull align 1 dereferenceable(16) [[S:%.*]], i32 16, i1 false)
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
call void @llvm.memcpy.p0.p0.i32(ptr %d, ptr %s, i32 16, i1 false)
|
|
ret void
|
|
}
|
|
|
|
define void @copy_8_bytes_noalias(ptr %d, ptr %s) {
|
|
; CHECK-LABEL: @copy_8_bytes_noalias(
|
|
; CHECK-NEXT: [[TMP1:%.*]] = load i64, ptr [[S:%.*]], align 1, !alias.scope [[META0:![0-9]+]], !noalias [[META3:![0-9]+]]
|
|
; CHECK-NEXT: store i64 [[TMP1]], ptr [[D:%.*]], align 1, !alias.scope [[META0]], !noalias [[META3]]
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
call void @llvm.memcpy.p0.p0.i32(ptr %d, ptr %s, i32 8, i1 false), !alias.scope !4, !noalias !5
|
|
ret void
|
|
}
|
|
|
|
!0 = distinct !{!0, !"The domain"}
|
|
!1 = distinct !{!1}
|
|
!2 = !{!2, !0}
|
|
!3 = !{!3, !1}
|
|
!4 = !{!2}
|
|
!5 = !{!3}
|