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
104 lines
3.4 KiB
LLVM
104 lines
3.4 KiB
LLVM
; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
|
|
;
|
|
; Test cases in this file are intended to be run with both reassociate and
|
|
; gvn passes enabled.
|
|
;
|
|
; Test numbering remains continuous across:
|
|
; - InstCombine/fast-basictest.ll
|
|
; - PhaseOrdering/fast-basictest.ll
|
|
; - PhaseOrdering/fast-reassociate-gvn.ll
|
|
; - Reassociate/fast-basictest.ll
|
|
;
|
|
; RUN: opt < %s -passes=reassociate,gvn -S | FileCheck %s --check-prefixes=CHECK,REASSOC_AND_GVN --allow-unused-prefixes
|
|
; RUN: opt < %s -O2 -S | FileCheck %s --check-prefixes=CHECK,O2 --allow-unused-prefixes
|
|
|
|
@fe = external global float
|
|
@fa = external global float
|
|
@fb = external global float
|
|
@fc = external global float
|
|
@ff = external global float
|
|
|
|
; If two sums of the same operands in different order are counted with 'fast'
|
|
; flag and then stored to global variables, we can reuse the same value twice.
|
|
; Sums:
|
|
; - test3: (a+b)+c and (a+c)+b
|
|
; - test4: c+(a+b) and (c+a)+b
|
|
; - test5: c+(b+a) and (c+a)+b
|
|
; TODO: check if 'reassoc' flag is technically enough for this optimization
|
|
; (currently the transformation is not done with 'reassoc' only).
|
|
|
|
define void @test3() {
|
|
; CHECK-LABEL: @test3(
|
|
; CHECK-NEXT: [[A:%.*]] = load float, ptr @fa, align 4
|
|
; CHECK-NEXT: [[B:%.*]] = load float, ptr @fb, align 4
|
|
; CHECK-NEXT: [[C:%.*]] = load float, ptr @fc, align 4
|
|
; CHECK-NEXT: [[T1:%.*]] = fadd fast float [[B]], [[A]]
|
|
; CHECK-NEXT: [[T2:%.*]] = fadd fast float [[T1]], [[C]]
|
|
; CHECK-NEXT: store float [[T2]], ptr @fe, align 4
|
|
; CHECK-NEXT: store float [[T2]], ptr @ff, align 4
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
%A = load float, ptr @fa
|
|
%B = load float, ptr @fb
|
|
%C = load float, ptr @fc
|
|
%t1 = fadd fast float %A, %B
|
|
%t2 = fadd fast float %t1, %C
|
|
%t3 = fadd fast float %A, %C
|
|
%t4 = fadd fast float %t3, %B
|
|
; e = (a+b)+c;
|
|
store float %t2, ptr @fe
|
|
; f = (a+c)+b
|
|
store float %t4, ptr @ff
|
|
ret void
|
|
}
|
|
|
|
define void @test4() {
|
|
; CHECK-LABEL: @test4(
|
|
; CHECK-NEXT: [[A:%.*]] = load float, ptr @fa, align 4
|
|
; CHECK-NEXT: [[B:%.*]] = load float, ptr @fb, align 4
|
|
; CHECK-NEXT: [[C:%.*]] = load float, ptr @fc, align 4
|
|
; CHECK-NEXT: [[T1:%.*]] = fadd fast float [[B]], [[A]]
|
|
; CHECK-NEXT: [[T2:%.*]] = fadd fast float [[T1]], [[C]]
|
|
; CHECK-NEXT: store float [[T2]], ptr @fe, align 4
|
|
; CHECK-NEXT: store float [[T2]], ptr @ff, align 4
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
%A = load float, ptr @fa
|
|
%B = load float, ptr @fb
|
|
%C = load float, ptr @fc
|
|
%t1 = fadd fast float %A, %B
|
|
%t2 = fadd fast float %C, %t1
|
|
%t3 = fadd fast float %C, %A
|
|
%t4 = fadd fast float %t3, %B
|
|
; e = c+(a+b)
|
|
store float %t2, ptr @fe
|
|
; f = (c+a)+b
|
|
store float %t4, ptr @ff
|
|
ret void
|
|
}
|
|
|
|
define void @test5() {
|
|
; CHECK-LABEL: @test5(
|
|
; CHECK-NEXT: [[A:%.*]] = load float, ptr @fa, align 4
|
|
; CHECK-NEXT: [[B:%.*]] = load float, ptr @fb, align 4
|
|
; CHECK-NEXT: [[C:%.*]] = load float, ptr @fc, align 4
|
|
; CHECK-NEXT: [[T1:%.*]] = fadd fast float [[B]], [[A]]
|
|
; CHECK-NEXT: [[T2:%.*]] = fadd fast float [[T1]], [[C]]
|
|
; CHECK-NEXT: store float [[T2]], ptr @fe, align 4
|
|
; CHECK-NEXT: store float [[T2]], ptr @ff, align 4
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
%A = load float, ptr @fa
|
|
%B = load float, ptr @fb
|
|
%C = load float, ptr @fc
|
|
%t1 = fadd fast float %B, %A
|
|
%t2 = fadd fast float %C, %t1
|
|
%t3 = fadd fast float %C, %A
|
|
%t4 = fadd fast float %t3, %B
|
|
; e = c+(b+a)
|
|
store float %t2, ptr @fe
|
|
; f = (c+a)+b
|
|
store float %t4, ptr @ff
|
|
ret void
|
|
}
|