derek/TheRock
1
0 Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity
Files
4849f1eea335bb3333f3fbf66b138f72eb85cf6c
TheRock/compiler/amd-llvm/llvm/test/Transforms/LoopVectorize/induction-ptrcasts.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

87 lines
4.6 KiB
LLVM
Raw Blame History

; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --check-globals none --filter-out-after "^scalar.ph" --version 6
; RUN: opt -passes=loop-vectorize -force-vector-width=1 -force-vector-interleave=2 -S %s | FileCheck --check-prefix=VF1 %s
; RUN: opt -passes=loop-vectorize -force-vector-width=2 -force-vector-interleave=1 -S %s | FileCheck --check-prefix=VF2 %s
@f = external global i32, align 4
define void @int_iv_based_on_pointer_iv(ptr %A) {
; VF1-LABEL: define void @int_iv_based_on_pointer_iv(
; VF1-SAME: ptr [[A:%.*]]) {
; VF1-NEXT: [[ENTRY:.*:]]
; VF1-NEXT: [[SMIN:%.*]] = call i64 @llvm.smin.i64(i64 add (i64 ptrtoint (ptr @f to i64), i64 -4), i64 0)
; VF1-NEXT: [[TMP0:%.*]] = sub i64 add (i64 ptrtoint (ptr @f to i64), i64 -1), [[SMIN]]
; VF1-NEXT: [[TMP1:%.*]] = lshr i64 [[TMP0]], 2
; VF1-NEXT: [[TMP2:%.*]] = add nuw nsw i64 [[TMP1]], 1
; VF1-NEXT: [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[TMP2]], 2
; VF1-NEXT: br i1 [[MIN_ITERS_CHECK]], label %[[SCALAR_PH:.*]], label %[[VECTOR_PH:.*]]
; VF1: [[VECTOR_PH]]:
; VF1-NEXT: [[N_MOD_VF:%.*]] = urem i64 [[TMP2]], 2
; VF1-NEXT: [[N_VEC:%.*]] = sub i64 [[TMP2]], [[N_MOD_VF]]
; VF1-NEXT: [[TMP4:%.*]] = shl i64 [[N_VEC]], 2
; VF1-NEXT: [[TMP5:%.*]] = getelementptr i8, ptr null, i64 [[TMP4]]
; VF1-NEXT: br label %[[VECTOR_BODY:.*]]
; VF1: [[VECTOR_BODY]]:
; VF1-NEXT: [[INDEX:%.*]] = phi i64 [ 0, %[[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], %[[VECTOR_BODY]] ]
; VF1-NEXT: [[OFFSET_IDX:%.*]] = shl i64 [[INDEX]], 2
; VF1-NEXT: [[INDUCTION3:%.*]] = add i64 [[OFFSET_IDX]], 4
; VF1-NEXT: [[TMP7:%.*]] = getelementptr inbounds i8, ptr [[A]], i64 [[OFFSET_IDX]]
; VF1-NEXT: [[TMP8:%.*]] = getelementptr inbounds i8, ptr [[A]], i64 [[INDUCTION3]]
; VF1-NEXT: store i8 0, ptr [[TMP7]], align 1
; VF1-NEXT: store i8 0, ptr [[TMP8]], align 1
; VF1-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
; VF1-NEXT: [[TMP9:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; VF1-NEXT: br i1 [[TMP9]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
; VF1: [[MIDDLE_BLOCK]]:
; VF1-NEXT: [[CMP_N:%.*]] = icmp eq i64 [[TMP2]], [[N_VEC]]
; VF1-NEXT: br i1 [[CMP_N]], [[EXIT:label %.*]], label %[[SCALAR_PH]]
; VF1: [[SCALAR_PH]]:
;
; VF2-LABEL: define void @int_iv_based_on_pointer_iv(
; VF2-SAME: ptr [[A:%.*]]) {
; VF2-NEXT: [[ENTRY:.*:]]
; VF2-NEXT: [[SMIN:%.*]] = call i64 @llvm.smin.i64(i64 add (i64 ptrtoint (ptr @f to i64), i64 -4), i64 0)
; VF2-NEXT: [[TMP0:%.*]] = sub i64 add (i64 ptrtoint (ptr @f to i64), i64 -1), [[SMIN]]
; VF2-NEXT: [[TMP1:%.*]] = lshr i64 [[TMP0]], 2
; VF2-NEXT: [[TMP2:%.*]] = add nuw nsw i64 [[TMP1]], 1
; VF2-NEXT: [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[TMP2]], 2
; VF2-NEXT: br i1 [[MIN_ITERS_CHECK]], label %[[SCALAR_PH:.*]], label %[[VECTOR_PH:.*]]
; VF2: [[VECTOR_PH]]:
; VF2-NEXT: [[N_MOD_VF:%.*]] = urem i64 [[TMP2]], 2
; VF2-NEXT: [[N_VEC:%.*]] = sub i64 [[TMP2]], [[N_MOD_VF]]
; VF2-NEXT: [[TMP6:%.*]] = shl i64 [[N_VEC]], 2
; VF2-NEXT: [[TMP5:%.*]] = getelementptr i8, ptr null, i64 [[TMP6]]
; VF2-NEXT: br label %[[VECTOR_BODY:.*]]
; VF2: [[VECTOR_BODY]]:
; VF2-NEXT: [[INDEX:%.*]] = phi i64 [ 0, %[[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], %[[VECTOR_BODY]] ]
; VF2-NEXT: [[OFFSET_IDX:%.*]] = shl i64 [[INDEX]], 2
; VF2-NEXT: [[TMP4:%.*]] = add i64 [[OFFSET_IDX]], 4
; VF2-NEXT: [[TMP9:%.*]] = getelementptr inbounds i8, ptr [[A]], i64 [[OFFSET_IDX]]
; VF2-NEXT: [[TMP10:%.*]] = getelementptr inbounds i8, ptr [[A]], i64 [[TMP4]]
; VF2-NEXT: store i8 0, ptr [[TMP9]], align 1
; VF2-NEXT: store i8 0, ptr [[TMP10]], align 1
; VF2-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
; VF2-NEXT: [[TMP11:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; VF2-NEXT: br i1 [[TMP11]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
; VF2: [[MIDDLE_BLOCK]]:
; VF2-NEXT: [[CMP_N:%.*]] = icmp eq i64 [[TMP2]], [[N_VEC]]
; VF2-NEXT: br i1 [[CMP_N]], [[EXIT:label %.*]], label %[[SCALAR_PH]]
; VF2: [[SCALAR_PH]]:
;
entry:
br label %loop
loop:
%iv.int = phi i64 [ 0, %entry ], [ %iv.int.next, %loop ]
%iv.ptr = phi ptr [ null, %entry ], [ %iv.ptr.next, %loop ]
%iv.ptr.next = getelementptr inbounds i32, ptr %iv.ptr, i64 1
%gep.A = getelementptr inbounds i8, ptr %A, i64 %iv.int
store i8 0, ptr %gep.A
%iv.int.next = ptrtoint ptr %iv.ptr.next to i64
%sub.ptr.sub = sub i64 ptrtoint (ptr @f to i64), %iv.int.next
%cmp = icmp sgt i64 %sub.ptr.sub, 0
br i1 %cmp, label %loop, label %exit
exit:
ret void
}
Reference in new issue View Git Blame Copy Permalink