CompatHelper: add new compat entry for CUDA at version 4, (keep existing compat)

.github/workflows/CI.yml

@@ -0,0 +1,36 @@
+name: CI
+on:
+  push:
+    branches:
+      - main
+    tags: ['*']
+  pull_request:
+concurrency:
+  # Skip intermediate builds: always.
+  # Cancel intermediate builds: only if it is a pull request build.
+  group: ${{ github.workflow }}-${{ github.ref }}
+  cancel-in-progress: ${{ startsWith(github.ref, 'refs/pull/') }}
+jobs:
+  test:
+    name: Julia ${{ matrix.version }} - ${{ matrix.os }} - ${{ matrix.arch }} - ${{ github.event_name }}
+    runs-on: ${{ matrix.os }}
+    strategy:
+      fail-fast: false
+      matrix:
+        version:
+          - '1.0'
+          - '1.8'
+          - 'nightly'
+        os:
+          - ubuntu-latest
+        arch:
+          - x64
+    steps:
+      - uses: actions/checkout@v2
+      - uses: julia-actions/setup-julia@v1
+        with:
+          version: ${{ matrix.version }}
+          arch: ${{ matrix.arch }}
+      - uses: julia-actions/cache@v1
+      - uses: julia-actions/julia-buildpkg@v1
+      - uses: julia-actions/julia-runtest@v1

.github/workflows/CompatHelper.yml

@@ -0,0 +1,16 @@
+name: CompatHelper
+on:
+  schedule:
+    - cron: 0 0 * * *
+  workflow_dispatch:
+jobs:
+  CompatHelper:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Pkg.add("CompatHelper")
+        run: julia -e 'using Pkg; Pkg.add("CompatHelper")'
+      - name: CompatHelper.main()
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+          COMPATHELPER_PRIV: ${{ secrets.DOCUMENTER_KEY }}
+        run: julia -e 'using CompatHelper; CompatHelper.main()'

.github/workflows/TagBot.yml

@@ -0,0 +1,15 @@
+name: TagBot
+on:
+  issue_comment:
+    types:
+      - created
+  workflow_dispatch:
+jobs:
+  TagBot:
+    if: github.event_name == 'workflow_dispatch' || github.actor == 'JuliaTagBot'
+    runs-on: ubuntu-latest
+    steps:
+      - uses: JuliaRegistries/TagBot@v1
+        with:
+          token: ${{ secrets.GITHUB_TOKEN }}
+          ssh: ${{ secrets.DOCUMENTER_KEY }}

.gitignore

@@ -0,0 +1,2 @@
+/Manifest.toml
+*.sif

LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2023 brooks
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

Project.toml

@@ -0,0 +1,19 @@
+name = "spectralGPU"
+uuid = "1f941c8d-9269-406b-be03-2a5a5e88bc1c"
+authors = ["brooks"]
+version = "1.0.0-DEV"
+
+[deps]
+CUDA = "052768ef-5323-5732-b1bb-66c8b64840ba"
+FFTW = "7a1cc6ca-52ef-59f5-83cd-3a7055c09341"
+LazyGrids = "7031d0ef-c40d-4431-b2f8-61a8d2f650db"
+
+[compat]
+CUDA = "4"
+julia = "1"
+
+[extras]
+Test = "8dfed614-e22c-5e08-85e1-65c5234f0b40"
+
+[targets]
+test = ["Test"]

README.md

@@ -0,0 +1,3 @@
+# spectralGPU
+
+[![Build Status](https://github.com/vanillabrooks/spectralGPU.jl/actions/workflows/CI.yml/badge.svg?branch=main)](https://github.com/vanillabrooks/spectralGPU.jl/actions/workflows/CI.yml?query=branch%3Amain)

base.apptainer

@@ -0,0 +1,20 @@
+Bootstrap: docker
+From: ubuntu:20.04
+
+%files
+	./ /spectralGPU
+
+%post
+	rm /bin/sh && ln -s /bin/bash /bin/sh
+
+	apt update -y && apt upgrade -y
+	apt install -y curl libssl-dev
+
+	curl -fsSL https://install.julialang.org | sh -s -- --yes
+
+	. /root/.profile
+
+	julia --version
+
+%apprun distribute
+	echo "hello!"

benches/cpu_gpu.jl

@@ -0,0 +1,67 @@
+include("../src/spectralGPU.jl");
+using .spectralGPU: mesh, fft, markers, initial_condition, state, config, solver, Integrate
+
+using CUDA
+using BenchmarkTools
+
+#########
+######### GPU benchmark
+#########
+
+N = 128
+begin
+    local parallel = markers.SingleThreadGPU()
+
+    local K = mesh.wavenumbers_gpu(N)
+    local cfg = config.taylor_green_validation()
+    local st = state.create_state_gpu(N, K, cfg)
+    local msh = mesh.new_mesh(N)
+    local ic = markers.TaylorGreen()
+
+    local U = CuArray(zeros(N, N, N, 3))
+    local U_hat = CuArray(ComplexF64.(zeros(K.kn, N, N, 3)))
+    initial_condition.setup_initial_condition(parallel, ic, msh, U, U_hat)
+
+    r = @benchmark Integrate.integrate(
+        $parallel,
+        $K,
+        $cfg,
+        $st,
+        $U,
+        $U_hat,
+    )
+
+    println("GPU full solver integration")
+    display(r)
+    println("\n");
+end
+
+#########
+######### CPU benchmark
+#########
+
+begin
+    local parallel = markers.SingleThreadCPU()
+
+    local K = mesh.wavenumbers(N)
+    local cfg = config.taylor_green_validation()
+    local st = state.create_state(N, K, cfg)
+    local msh = mesh.new_mesh(N)
+    local ic = markers.TaylorGreen()
+
+    local U = zeros(N, N, N, 3)
+    local U_hat = ComplexF64.(zeros(K.kn, N, N, 3))
+    initial_condition.setup_initial_condition(parallel, ic, msh, U, U_hat)
+
+    r = @benchmark Integrate.integrate(
+        $parallel,
+        $K,
+        $cfg,
+        $st,
+        $U,
+        $U_hat,
+    )
+
+    println("CPU full solver integration")
+    display(r)
+end

benches/cross.jl

@@ -0,0 +1,39 @@
+include("../src/spectralGPU.jl");
+using .spectralGPU: mesh, fft, markers, initial_condition, state, config, solver, Integrate
+
+using CUDA
+using BenchmarkTools
+
+N = 64
+
+begin
+    local parallel = markers.SingleThreadGPU()
+    local re = 40.
+    local K = mesh.wavenumbers_gpu(N)
+    local cfg = config.create_config(N, re, 0.00001)
+    local st = state.create_state_gpu(N, K, cfg)
+    local U = CuArray(zeros(N, N, N, 3))
+    local U_hat = CuArray(ComplexF64.(zeros(K.kn, N, N, 3)))
+
+    r = @benchmark solver.cross!($parallel, $U, $st.curl; out = $st.dU)
+
+    println("GPU cross")
+    display(r)
+    println("")
+end
+
+begin
+    local parallel = markers.SingleThreadCPU()
+    local re = 40.
+    local K = mesh.wavenumbers(N)
+    local cfg = config.create_config(N, re, 0.00001)
+    local st = state.create_state(N, K, cfg)
+    local U = zeros(N, N, N, 3)
+    local U_hat = ComplexF64.(zeros(K.kn, N, N, 3))
+
+    r = @benchmark solver.cross!($parallel, $U, $st.curl; out = $st.dU)
+
+    println("CPU cross")
+    display(r)
+    println("")
+end

benches/curl.jl

@@ -0,0 +1,37 @@
+include("../src/spectralGPU.jl");
+using .spectralGPU: mesh, fft, markers, initial_condition, state, config, solver, Integrate
+
+using CUDA
+using BenchmarkTools
+
+N = 64
+
+begin
+    local parallel = markers.SingleThreadGPU()
+    local re = 40.
+    local K = mesh.wavenumbers_gpu(N)
+    local cfg = config.create_config(N, re, 0.00001)
+    local st = state.create_state_gpu(N, K, cfg)
+    local U_hat = CuArray(ComplexF64.(zeros(K.kn, N, N, 3)))
+
+    r = @benchmark solver.curl!($parallel, $K, $U_hat; out = $st.curl)
+
+    println("GPU curl")
+    display(r)
+    println("")
+end
+
+begin
+    local parallel = markers.SingleThreadCPU()
+    local re = 40.
+    local K = mesh.wavenumbers(N)
+    local cfg = config.create_config(N, re, 0.00001)
+    local st = state.create_state(N, K, cfg)
+    local U_hat = ComplexF64.(zeros(K.kn, N, N, 3))
+
+    r = @benchmark solver.curl!($parallel, $K, $U_hat; out = $st.curl)
+
+    println("CPU curl")
+    display(r)
+    println("")
+end

benches/fft.jl

@@ -0,0 +1,99 @@
+include("../src/spectralGPU.jl");
+using .spectralGPU: mesh, fft, markers, initial_condition, state, config, solver, Integrate
+
+using CUDA
+using BenchmarkTools
+
+#########
+######### GPU (forward) benchmark
+#########
+
+N = 64
+
+begin
+    parallel = markers.SingleThreadGPU()
+
+    K = mesh.wavenumbers_gpu(N)
+
+    U = CuArray(zeros(N, N, N, 3))
+    U_hat = CuArray(ComplexF64.(zeros(K.kn, N, N, 3)))
+
+    msh = mesh.new_mesh(N)
+    ic = markers.TaylorGreen()
+    initial_condition.setup_initial_condition(parallel, ic, msh, U, U_hat)
+
+    r = @benchmark @views fft.fftn_mpi!($parallel, $U[:, :, :, 1], $U_hat[:, :, :, 1])
+
+    println("gpu forward FFT")
+    display(r)
+    println("")
+end
+
+#########
+######### CPU (forward) benchmark
+#########
+
+begin
+    parallel = markers.SingleThreadCPU()
+
+    K = mesh.wavenumbers(N)
+
+    U = zeros(N, N, N, 3)
+    U_hat = ComplexF64.(zeros(K.kn, N, N, 3))
+
+    msh = mesh.new_mesh(N)
+    ic = markers.TaylorGreen()
+    initial_condition.setup_initial_condition(parallel, ic, msh, U, U_hat)
+
+    r = @benchmark @views fft.fftn_mpi!($parallel, $U[:, :, :, 1], $U_hat[:, :, :, 1])
+
+    println("cpu forward FFT")
+    display(r)
+    println("")
+end
+
+#########
+######### GPU (inverse) benchmark
+#########
+
+begin
+    parallel = markers.SingleThreadGPU()
+
+    K = mesh.wavenumbers_gpu(N)
+
+    U = CuArray(zeros(N, N, N, 3))
+    U_hat = CuArray(ComplexF64.(zeros(K.kn, N, N, 3)))
+
+    msh = mesh.new_mesh(N)
+    ic = markers.TaylorGreen()
+    initial_condition.setup_initial_condition(parallel, ic, msh, U, U_hat)
+
+    r = @benchmark @views fft.ifftn_mpi!($parallel, $K, $U_hat[:, :, :, 1], $U[:, :, :, 1])
+
+    println("gpu inverse FFT")
+    display(r)
+    println("")
+end
+
+#########
+######### CPU (inverse) benchmark
+#########
+
+begin
+    parallel = markers.SingleThreadCPU()
+
+    K = mesh.wavenumbers(N)
+
+    U = zeros(N, N, N, 3)
+    U_hat = ComplexF64.(zeros(K.kn, N, N, 3))
+
+    msh = mesh.new_mesh(N)
+    ic = markers.TaylorGreen()
+    initial_condition.setup_initial_condition(parallel, ic, msh, U, U_hat)
+
+    r = @benchmark @views fft.ifftn_mpi!($parallel, $K, $U_hat[:, :, :, 1], $U[:, :, :, 1])
+
+    println("cpu inverse FFT")
+    display(r)
+    println("")
+end