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
75 lines
2.8 KiB
LLVM
75 lines
2.8 KiB
LLVM
; This test checks if two similar functions, f1 and f2, can be merged locally within a single module
|
|
; while parameterizing a difference in their global variables, g1 and g2.
|
|
; To achieve this, we create two instances of the global merging function, f1.Tgm and f2.Tgm,
|
|
; which are tail-called from thunks f1 and f2 respectively.
|
|
; These identical functions, f1.Tgm and f2.Tgm, will be folded by the linker via Identical Code Folding (ICF).
|
|
|
|
; RUN: opt -mtriple=arm64-apple-darwin -S --passes=global-merge-func %s | FileCheck %s
|
|
|
|
; A merging instance is created with additional parameter.
|
|
; CHECK: define internal i32 @f1.Tgm(i32 %0, ptr %1)
|
|
; CHECK-NEXT: entry:
|
|
; CHECK-NEXT: %idxprom = sext i32 %0 to i64
|
|
; CHECK-NEXT: %arrayidx = getelementptr inbounds [0 x i32], ptr @g, i64 0, i64 %idxprom
|
|
; CHECK-NEXT: %2 = load i32, ptr %arrayidx, align 4
|
|
; CHECK-NEXT: %3 = load volatile i32, ptr %1, align 4
|
|
; CHECK-NEXT: %mul = mul nsw i32 %3, %2
|
|
; CHECK-NEXT: %add = add nsw i32 %mul, 1
|
|
; CHECK-NEXT: ret i32 %add
|
|
|
|
; The original function becomes a thunk passing g1.
|
|
; CHECK: define i32 @f1(i32 %a)
|
|
; CHECK-NEXT: %1 = tail call i32 @f1.Tgm(i32 %a, ptr @g1)
|
|
; CHECK-NEXT: ret i32 %1
|
|
|
|
; A same sequence is produced for f2.Tgm.
|
|
; CHECK: define internal i32 @f2.Tgm(i32 %0, ptr %1)
|
|
; CHECK-NEXT: entry:
|
|
; CHECK-NEXT: %idxprom = sext i32 %0 to i64
|
|
; CHECK-NEXT: %arrayidx = getelementptr inbounds [0 x i32], ptr @g, i64 0, i64 %idxprom
|
|
; CHECK-NEXT: %2 = load i32, ptr %arrayidx, align 4
|
|
; CHECK-NEXT: %3 = load volatile i32, ptr %1, align 4
|
|
; CHECK-NEXT: %mul = mul nsw i32 %3, %2
|
|
; CHECK-NEXT: %add = add nsw i32 %mul, 1
|
|
; CHECK-NEXT: ret i32 %add
|
|
|
|
; The original function becomes a thunk passing g2.
|
|
; CHECK: define i32 @f2(i32 %a)
|
|
; CHECK-NEXT: %1 = tail call i32 @f2.Tgm(i32 %a, ptr @g2)
|
|
; CHECK-NEXT: ret i32 %1
|
|
|
|
; RUN: llc -mtriple=arm64-apple-darwin -enable-global-merge-func=true < %s | FileCheck %s --check-prefix=MERGE
|
|
; RUN: llc -mtriple=arm64-apple-darwin -enable-global-merge-func=false < %s | FileCheck %s --check-prefix=NOMERGE
|
|
|
|
; MERGE: _f1.Tgm
|
|
; MERGE: _f2.Tgm
|
|
|
|
; NOMERGE-NOT: _f1.Tgm
|
|
; NOMERGE-NOT: _f2.Tgm
|
|
|
|
@g = external local_unnamed_addr global [0 x i32], align 4
|
|
@g1 = external global i32, align 4
|
|
@g2 = external global i32, align 4
|
|
|
|
define i32 @f1(i32 %a) {
|
|
entry:
|
|
%idxprom = sext i32 %a to i64
|
|
%arrayidx = getelementptr inbounds [0 x i32], ptr @g, i64 0, i64 %idxprom
|
|
%0 = load i32, ptr %arrayidx, align 4
|
|
%1 = load volatile i32, ptr @g1, align 4
|
|
%mul = mul nsw i32 %1, %0
|
|
%add = add nsw i32 %mul, 1
|
|
ret i32 %add
|
|
}
|
|
|
|
define i32 @f2(i32 %a) {
|
|
entry:
|
|
%idxprom = sext i32 %a to i64
|
|
%arrayidx = getelementptr inbounds [0 x i32], ptr @g, i64 0, i64 %idxprom
|
|
%0 = load i32, ptr %arrayidx, align 4
|
|
%1 = load volatile i32, ptr @g2, align 4
|
|
%mul = mul nsw i32 %1, %0
|
|
%add = add nsw i32 %mul, 1
|
|
ret i32 %add
|
|
}
|