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
77 lines
2.6 KiB
LLVM
77 lines
2.6 KiB
LLVM
; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
|
|
; RUN: opt < %s -passes=reassociate -S | FileCheck %s
|
|
|
|
define i32 @shl_add(i8 %x) {
|
|
; CHECK-LABEL: @shl_add(
|
|
; CHECK-NEXT: [[CONV:%.*]] = zext i8 [[X:%.*]] to i32
|
|
; CHECK-NEXT: [[SHL:%.*]] = shl nuw nsw i32 [[CONV]], 8
|
|
; CHECK-NEXT: [[SHL2:%.*]] = shl nuw nsw i32 [[CONV]], 16
|
|
; CHECK-NEXT: [[ADD:%.*]] = add nuw nsw i32 [[SHL]], [[SHL2]]
|
|
; CHECK-NEXT: ret i32 [[ADD]]
|
|
;
|
|
%conv = zext i8 %x to i32
|
|
%shl = shl nuw nsw i32 %conv, 8
|
|
%shl2 = shl nuw nsw i32 %conv, 16
|
|
%add = or i32 %shl, %shl2
|
|
ret i32 %add
|
|
}
|
|
|
|
; If we don't know that operands have no common bits set,
|
|
; we can't convert the `or` into an `add`.
|
|
define i32 @test1(i32 %a, i32 %b) {
|
|
; CHECK-LABEL: @test1(
|
|
; CHECK-NEXT: [[C:%.*]] = or i32 [[B:%.*]], [[A:%.*]]
|
|
; CHECK-NEXT: [[C_PLUS_ONE:%.*]] = add i32 [[C]], 1
|
|
; CHECK-NEXT: ret i32 [[C_PLUS_ONE]]
|
|
;
|
|
%c = or i32 %a, %b
|
|
%c.plus.one = add i32 %c, 1
|
|
ret i32 %c.plus.one
|
|
}
|
|
|
|
; But if we *do* know that operands have no common bits set,
|
|
; we *can* convert the `or` into an `add`.
|
|
define i32 @test2(i32 %x, i32 %y) {
|
|
; CHECK-LABEL: @test2(
|
|
; CHECK-NEXT: [[X_NUMLZ:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[X:%.*]], i1 true), [[RNG0:!range !.*]]
|
|
; CHECK-NEXT: [[RES:%.*]] = add nuw nsw i32 [[X_NUMLZ]], -32
|
|
; CHECK-NEXT: [[RES_PLUS_ONE:%.*]] = add i32 [[RES]], [[Y:%.*]]
|
|
; CHECK-NEXT: ret i32 [[RES_PLUS_ONE]]
|
|
;
|
|
%x.numlz = tail call i32 @llvm.ctlz.i32(i32 %x, i1 true), !range !0
|
|
%res = or i32 %x.numlz, -32
|
|
%res.plus.one = add i32 %res, %y
|
|
ret i32 %res.plus.one
|
|
}
|
|
|
|
; And that allows reassociation in general.
|
|
define i32 @test3(i32 %x, i32 %bit) {
|
|
; CHECK-LABEL: @test3(
|
|
; CHECK-NEXT: [[X_NUMLZ:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[X:%.*]], i1 true), [[RNG0]]
|
|
; CHECK-NEXT: [[BIT_PLUS_ONE:%.*]] = add i32 [[BIT:%.*]], -31
|
|
; CHECK-NEXT: [[RES:%.*]] = add i32 [[BIT_PLUS_ONE]], [[X_NUMLZ]]
|
|
; CHECK-NEXT: ret i32 [[RES]]
|
|
;
|
|
%x.numlz = tail call i32 @llvm.ctlz.i32(i32 %x, i1 true), !range !0
|
|
%zero.minus.x.numactivebits = or i32 %x.numlz, -32
|
|
%bit.plus.one = add i32 %bit, 1
|
|
%res = add i32 %bit.plus.one, %zero.minus.x.numactivebits
|
|
ret i32 %res
|
|
}
|
|
|
|
; Test that disjoint allow reassociation.
|
|
define i32 @test4(i32 %a, i32 %b) {
|
|
; CHECK-LABEL: @test4(
|
|
; CHECK-NEXT: [[C:%.*]] = add i32 [[A:%.*]], 1
|
|
; CHECK-NEXT: [[C_PLUS_ONE:%.*]] = add i32 [[C]], [[B:%.*]]
|
|
; CHECK-NEXT: ret i32 [[C_PLUS_ONE]]
|
|
;
|
|
%c = or disjoint i32 %a, %b
|
|
%c.plus.one = add i32 %c, 1
|
|
ret i32 %c.plus.one
|
|
}
|
|
|
|
declare i32 @llvm.ctlz.i32(i32, i1 immarg) #2
|
|
|
|
!0 = !{i32 0, i32 33}
|