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
68 lines
2.4 KiB
LLVM
68 lines
2.4 KiB
LLVM
; RUN: opt -mtriple=amdgcn--amdhsa -S -passes=inline -inline-threshold=0 < %s | FileCheck %s
|
|
|
|
define void @use_flat_ptr_arg(ptr nocapture %p) {
|
|
entry:
|
|
%tmp1 = load float, ptr %p, align 4
|
|
%div = fdiv float 1.000000e+00, %tmp1
|
|
%add0 = fadd float %div, 1.0
|
|
%add1 = fadd float %add0, 1.0
|
|
%add2 = fadd float %add1, 1.0
|
|
%add3 = fadd float %add2, 1.0
|
|
%add4 = fadd float %add3, 1.0
|
|
%add5 = fadd float %add4, 1.0
|
|
%add6 = fadd float %add5, 1.0
|
|
%add7 = fadd float %add6, 1.0
|
|
%add8 = fadd float %add7, 1.0
|
|
%add9 = fadd float %add8, 1.0
|
|
%add10 = fadd float %add9, 1.0
|
|
store float %add10, ptr %p, align 4
|
|
ret void
|
|
}
|
|
|
|
define void @use_private_ptr_arg(ptr addrspace(5) nocapture %p) {
|
|
entry:
|
|
%tmp1 = load float, ptr addrspace(5) %p, align 4
|
|
%div = fdiv float 1.000000e+00, %tmp1
|
|
%add0 = fadd float %div, 1.0
|
|
%add1 = fadd float %add0, 1.0
|
|
%add2 = fadd float %add1, 1.0
|
|
%add3 = fadd float %add2, 1.0
|
|
%add4 = fadd float %add3, 1.0
|
|
%add5 = fadd float %add4, 1.0
|
|
%add6 = fadd float %add5, 1.0
|
|
%add7 = fadd float %add6, 1.0
|
|
%add8 = fadd float %add7, 1.0
|
|
%add9 = fadd float %add8, 1.0
|
|
%add10 = fadd float %add9, 1.0
|
|
store float %add10, ptr addrspace(5) %p, align 4
|
|
ret void
|
|
}
|
|
|
|
; Test that the inline threshold is boosted if called with an
|
|
; addrspacecasted' alloca.
|
|
; CHECK-LABEL: @test_inliner_flat_ptr(
|
|
; CHECK: call i32 @llvm.amdgcn.workitem.id.x()
|
|
; CHECK-NOT: call {{[.*]}}@
|
|
define amdgpu_kernel void @test_inliner_flat_ptr(ptr addrspace(1) nocapture %a, i32 %n) {
|
|
entry:
|
|
%pvt_arr = alloca [64 x float], align 4, addrspace(5)
|
|
%tid = tail call i32 @llvm.amdgcn.workitem.id.x()
|
|
%arrayidx = getelementptr inbounds float, ptr addrspace(1) %a, i32 %tid
|
|
%tmp2 = load float, ptr addrspace(1) %arrayidx, align 4
|
|
%add = add i32 %tid, 1
|
|
%arrayidx2 = getelementptr inbounds float, ptr addrspace(1) %a, i32 %add
|
|
%tmp5 = load float, ptr addrspace(1) %arrayidx2, align 4
|
|
%or = or i32 %tid, %n
|
|
%arrayidx5 = getelementptr inbounds [64 x float], ptr addrspace(5) %pvt_arr, i32 0, i32 %or
|
|
%arrayidx7 = getelementptr inbounds [64 x float], ptr addrspace(5) %pvt_arr, i32 0, i32 %or
|
|
%to.flat = addrspacecast ptr addrspace(5) %arrayidx7 to ptr
|
|
call void @use_private_ptr_arg(ptr addrspace(5) %arrayidx7)
|
|
call void @use_flat_ptr_arg(ptr %to.flat)
|
|
ret void
|
|
}
|
|
|
|
declare i32 @llvm.amdgcn.workitem.id.x() #1
|
|
|
|
attributes #0 = { noinline }
|
|
attributes #1 = { nounwind readnone }
|