Merge pull request #9 from JuliaGPU/jps/buildkite-initial

Add Buildkite CI

.buildkite/pipeline.yml

@@ -0,0 +1,164 @@
+steps:
+
+  # CUDA
+  - label: "CUDA - Julia v1.10"
+    plugins:
+      - JuliaCI/julia#v1:
+          version: "1.10"
+    command: |
+      julia -e 'using Pkg
+
+                println("--- :julia: Instantiating environment")
+                Pkg.add("CUDA")
+                Pkg.develop(path=".")
+
+                println("+++ :julia: Running tests")
+                Pkg.test("AcceleratedKernels", test_args=["--CUDA"])'
+    agents:
+      queue: "juliagpu"
+      cuda: "*"
+    if: build.message !~ /\[skip tests\]/
+    timeout_in_minutes: 60
+
+  - label: "CUDA - Julia v1.11"
+    plugins:
+      - JuliaCI/julia#v1:
+          version: "1.11"
+    command: |
+      julia -e 'using Pkg
+
+                println("--- :julia: Instantiating environment")
+                Pkg.add("CUDA")
+                Pkg.develop(path=".")
+
+                println("+++ :julia: Running tests")
+                Pkg.test("AcceleratedKernels", test_args=["--CUDA"])'
+    agents:
+      queue: "juliagpu"
+      cuda: "*"
+    if: build.message !~ /\[skip tests\]/
+    timeout_in_minutes: 60
+
+
+  # AMDGPU
+  - label: "AMDGPU - Julia v1.10"
+    plugins:
+      - JuliaCI/julia#v1:
+          version: "1.10"
+    command: |
+      julia -e 'using Pkg
+
+                println("--- :julia: Instantiating environment")
+                Pkg.add("AMDGPU")
+                Pkg.develop(path=".")
+
+                println("+++ :julia: Running tests")
+                Pkg.test("AcceleratedKernels", test_args=["--AMDGPU"])'
+    agents:
+      queue: "juliagpu"
+      rocm: "*"
+      rocmgpu: "*"
+    if: build.message !~ /\[skip tests\]/
+    timeout_in_minutes: 60
+
+  - label: "AMDGPU - Julia v1.11"
+    plugins:
+      - JuliaCI/julia#v1:
+          version: "1.11"
+    command: |
+      julia -e 'using Pkg
+
+                println("--- :julia: Instantiating environment")
+                Pkg.add("AMDGPU")
+                Pkg.develop(path=".")
+
+                println("+++ :julia: Running tests")
+                Pkg.test("AcceleratedKernels", test_args=["--AMDGPU"])'
+    agents:
+      queue: "juliagpu"
+      rocm: "*"
+      rocmgpu: "*"
+    if: build.message !~ /\[skip tests\]/
+    timeout_in_minutes: 60
+
+
+  # oneAPI
+  - label: "oneAPI - Julia v1.10"
+    plugins:
+      - JuliaCI/julia#v1:
+          version: "1.10"
+    command: |
+      julia -e 'using Pkg
+
+                println("--- :julia: Instantiating environment")
+                Pkg.add("oneAPI")
+                Pkg.develop(path=".")
+
+                println("+++ :julia: Running tests")
+                Pkg.test("AcceleratedKernels", test_args=["--oneAPI"])'
+    agents:
+      queue: "juliagpu"
+      intel: "*"
+    if: build.message !~ /\[skip tests\]/
+    timeout_in_minutes: 60
+
+  - label: "oneAPI - Julia v1.11"
+    plugins:
+      - JuliaCI/julia#v1:
+          version: "1.11"
+    command: |
+      julia -e 'using Pkg
+
+                println("--- :julia: Instantiating environment")
+                Pkg.add("oneAPI")
+                Pkg.develop(path=".")
+
+                println("+++ :julia: Running tests")
+                Pkg.test("AcceleratedKernels", test_args=["--oneAPI"])'
+    agents:
+      queue: "juliagpu"
+      intel: "*"
+    if: build.message !~ /\[skip tests\]/
+    timeout_in_minutes: 60
+
+
+  # Metal
+  - label: "Metal - Julia v1.10"
+    plugins:
+      - JuliaCI/julia#v1:
+          version: "1.10"
+    command: |
+      julia -e 'using Pkg
+
+                println("--- :julia: Instantiating environment")
+                Pkg.add("Metal")
+                Pkg.develop(path=".")
+
+                println("+++ :julia: Running tests")
+                Pkg.test("AcceleratedKernels", test_args=["--Metal"])'
+    agents:
+      queue: "juliaecosystem"
+      os: "macos"
+      arch: "aarch64"
+    if: build.message !~ /\[skip tests\]/
+    timeout_in_minutes: 60
+
+  - label: "Metal - Julia v1.11"
+    plugins:
+      - JuliaCI/julia#v1:
+          version: "1.11"
+    command: |
+      julia -e 'using Pkg
+
+                println("--- :julia: Instantiating environment")
+                Pkg.add("Metal")
+                Pkg.develop(path=".")
+
+                println("+++ :julia: Running tests")
+                Pkg.test("AcceleratedKernels", test_args=["--Metal"])'
+    agents:
+      queue: "juliaecosystem"
+      os: "macos"
+      arch: "aarch64"
+    if: build.message !~ /\[skip tests\]/
+    timeout_in_minutes: 60

Project.toml

@@ -1,7 +1,7 @@
 name = "AcceleratedKernels"
 uuid = "6a4ca0a5-0e36-4168-a932-d9be78d558f1"
 authors = ["Andrei-Leonard Nicusan <leonard@evophase.co.uk> and contributors"]
-version = "0.3.0-DEV"
+version = "0.2.1"
 
 [deps]
 ArgCheck = "dce04be8-c92d-5529-be00-80e4d2c0e197"
@@ -14,12 +14,12 @@ Polyester = "f517fe37-dbe3-4b94-8317-1923a5111588"
 Unrolled = "9602ed7d-8fef-5bc8-8597-8f21381861e8"
 
 [compat]
-ArgCheck = "2.1"
+ArgCheck = "2"
 DocStringExtensions = "0.9"
-GPUArraysCore = "0.2"
+GPUArraysCore = "0.1, 0.2"
 KernelAbstractions = "0.9"
-Markdown = "1.10"
-OhMyThreads = "0.7.0"
+Markdown = "1"
+OhMyThreads = "0.7"
 Polyester = "0.7"
-Unrolled = "0.1.5"
+Unrolled = "0.1"
 julia = "1.10"

README.md

@@ -4,12 +4,170 @@
 
 [![Stable](https://img.shields.io/badge/docs-stable-blue.svg)](https://juliagpu.github.io/AcceleratedKernels.jl/stable/)
 [![Dev](https://img.shields.io/badge/docs-dev-blue.svg)](https://juliagpu.github.io/AcceleratedKernels.jl/dev/)
-[![CI-CPU](https://github.com/juliagpu/AcceleratedKernels.jl/actions/workflows/CI-CPU.yml/badge.svg)](https://github.com/juliagpu/AcceleratedKernels.jl/actions/workflows/CI-CPU.yml)
 [![Aqua QA](https://raw.githubusercontent.com/JuliaTesting/Aqua.jl/master/badge.svg)](https://github.com/JuliaTesting/Aqua.jl)
 [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
 
 
-Parallel algorithm building blocks for the Julia ecosystem, targeting multithreaded CPUs, and GPUs via Intel oneAPI, AMD ROCm, Apple Metal and Nvidia CUDA (and any future backends added to the [JuliaGPU](https://juliagpu.org/) organisation).
+Parallel algorithm building blocks for the Julia ecosystem, targeting multithreaded CPUs, and GPUs via Intel oneAPI, AMD ROCm, Apple Metal and Nvidia CUDA (and any future backends added to the [JuliaGPU](https://juliagpu.org/) organisation) from a unified KernelAbstractions.jl codebase.
+
+
+<table>
+
+<tr>
+<th>AK Backend</th>
+<th>Julia Version</th>
+<th>CI Status</th>
+</tr>
+
+<tr>
+<td>
+
+CPU Single- and Multi-Threaded
+
+</td>
+<td>
+
+Julia LTS, Stable, Pre-Release
+
+x86 and x64
+
+Windows, Ubuntu, MacOS
+
+</td>
+<td>
+
+[![CI-CPU](https://github.com/juliagpu/AcceleratedKernels.jl/actions/workflows/CI-CPU.yml/badge.svg)](https://github.com/juliagpu/AcceleratedKernels.jl/actions/workflows/CI-CPU.yml)
+
+</td>
+</tr>
+
+<tr>
+<td rowspan=2>
+
+[CUDA](https://github.com/JuliaGPU/CUDA.jl)
+
+</td>
+<td>
+
+Julia v1.10
+
+</td>
+<td>
+
+[![Build status](https://badge.buildkite.com/5b8c747451b382a6b1ad0a1b566d565bc851fc59515792c62e.svg?step=CUDA%20-%20Julia%20v1.10)](https://buildkite.com/julialang/acceleratedkernels-dot-jl)
+
+</td>
+</tr>
+
+<tr>
+<td>
+
+Julia v1.11
+
+</td>
+<td>
+
+[![Build status](https://badge.buildkite.com/5b8c747451b382a6b1ad0a1b566d565bc851fc59515792c62e.svg?step=CUDA%20-%20Julia%20v1.11)](https://buildkite.com/julialang/acceleratedkernels-dot-jl)
+
+</td>
+</tr>
+
+<tr>
+<td rowspan=2>
+
+[AMDGPU](https://github.com/JuliaGPU/AMDGPU.jl)
+
+</td>
+<td>
+
+Julia v1.10
+
+</td>
+<td>
+
+[![Build status](https://badge.buildkite.com/5b8c747451b382a6b1ad0a1b566d565bc851fc59515792c62e.svg?step=AMDGPU%20-%20Julia%20v1.10)](https://buildkite.com/julialang/acceleratedkernels-dot-jl)
+
+</td>
+</tr>
+
+<tr>
+<td>
+
+Julia v1.11
+
+</td>
+<td>
+
+[![Build status](https://badge.buildkite.com/5b8c747451b382a6b1ad0a1b566d565bc851fc59515792c62e.svg?step=AMDGPU%20-%20Julia%20v1.11)](https://buildkite.com/julialang/acceleratedkernels-dot-jl)
+
+</td>
+</tr>
+
+<tr>
+<td rowspan=2>
+
+[oneAPI](https://github.com/JuliaGPU/oneAPI.jl)
+
+</td>
+<td>
+
+Julia v1.10
+
+</td>
+<td>
+
+[![Build status](https://badge.buildkite.com/5b8c747451b382a6b1ad0a1b566d565bc851fc59515792c62e.svg?step=oneAPI%20-%20Julia%20v1.10)](https://buildkite.com/julialang/acceleratedkernels-dot-jl)
+
+</td>
+</tr>
+
+<tr>
+<td>
+
+Julia v1.11
+
+</td>
+<td>
+
+[![Build status](https://badge.buildkite.com/5b8c747451b382a6b1ad0a1b566d565bc851fc59515792c62e.svg?step=oneAPI%20-%20Julia%20v1.11)](https://buildkite.com/julialang/acceleratedkernels-dot-jl)
+
+</td>
+</tr>
+
+<tr>
+<td rowspan=2>
+
+[Metal](https://github.com/JuliaGPU/Metal.jl)
+
+[Known Issue](https://github.com/JuliaGPU/AcceleratedKernels.jl/issues/10) 
+
+</td>
+<td>
+
+Julia v1.10
+
+</td>
+<td>
+
+[![Build status](https://badge.buildkite.com/5b8c747451b382a6b1ad0a1b566d565bc851fc59515792c62e.svg?step=Metal%20-%20Julia%20v1.10)](https://buildkite.com/julialang/acceleratedkernels-dot-jl)
+
+</td>
+</tr>
+
+<tr>
+<td>
+
+Julia v1.11
+
+</td>
+<td>
+
+[![Build status](https://badge.buildkite.com/5b8c747451b382a6b1ad0a1b566d565bc851fc59515792c62e.svg?step=Metal%20-%20Julia%20v1.11)](https://buildkite.com/julialang/acceleratedkernels-dot-jl)
+
+</td>
+</tr>
+
+</table>
 
 
 - [1. What's Different?](#1-whats-different)
@@ -43,11 +201,11 @@ Again, this is only possible because of the unique Julia compilation model, the
 
 
 ## 2. Status
-The AcceleratedKernels.jl sorters were adopted as the official [AMDGPU algorithms](https://github.com/JuliaGPU/AMDGPU.jl/pull/688)! The API is starting to stabilise; it follows the Julia standard library fairly closely - additionally exposing all temporary arrays for memory reuse. For any new ideas / requests, please join the conversation on [Julia Discourse](https://discourse.julialang.org/t/ann-acceleratedkernels-jl-cross-architecture-parallel-algorithms-for-julias-gpu-backends/119698/16) or post [an issue](https://github.com/juliagpu/AcceleratedKernels.jl/issues).
+The AcceleratedKernels.jl sorters were adopted as the official [AMDGPU algorithms](https://github.com/JuliaGPU/AMDGPU.jl/pull/688)! The API is starting to stabilise; it follows the Julia standard library fairly closely - and additionally exposing all temporary arrays for memory reuse. For any new ideas / requests, please join the conversation on [Julia Discourse](https://discourse.julialang.org/t/ann-acceleratedkernels-jl-cross-architecture-parallel-algorithms-for-julias-gpu-backends/119698/16) or post [an issue](https://github.com/juliagpu/AcceleratedKernels.jl/issues).
 
-We have an extensive test suite; however, I only ran them locally on the oneAPI (laptop Intel UHD Graphics 620), CUDA (laptop with Nvidia Quadro RTX 4000 and data centre Nvidia A100-40), Metal (Mac M2 and M3), and AMD (data centre AMD MI210) backends. Some kinks might still exist for some platform / OS permutations before a CI is set up.
+We have an extensive randomised test suite that we run on the CPU (single- and multi-threaded) backend on Windows, Ubuntu and MacOS for Julia LTS, Stable, and Pre-Release, plus the CUDA, AMDGPU, oneAPI and Metal backends on the [JuliaGPU buildkite](https://github.com/JuliaGPU/buildkite).
 
-AcceleratedKernels.jl will also be a fundamental building block of applications developed at [EvoPhase](https://evophase.co.uk/), so it will see continuous heavy use with industry backing. Long-term stability, performance improvements and support are priorities for us.
+AcceleratedKernels.jl is also be a fundamental building block of applications developed at [EvoPhase](https://evophase.co.uk/), so it will see continuous heavy use with industry backing. Long-term stability, performance improvements and support are priorities for us.
 
 
 ## 3. Benchmarks
@@ -560,8 +718,6 @@ Leave out to test the CPU backend:
 $> julia -e 'import Pkg; Pkg.test("AcceleratedKernels.jl")
 ```
 
-**TODO**: talk with the JuliaGPU team to add library to their [BuildKite agents](https://github.com/JuliaGPU/buildkite) CI.
-
 
 ## 8. Issues and Debugging
 As the compilation pipeline of GPU kernels is different to that of base Julia, error messages also look different - for example, where Julia would insert an exception when a variable name was not defined (e.g. we had a typo), a GPU kernel throwing exceptions cannot be compiled and instead you'll see some cascading errors like `"[...] compiling [...] resulted in invalid LLVM IR"` caused by `"Reason: unsupported use of an undefined name"` resulting in `"Reason: unsupported dynamic function invocation"`, etc.

prototype/Project.toml

@@ -4,4 +4,5 @@ Aqua = "4c88cf16-eb10-579e-8560-4a9242c79595"
 BenchmarkTools = "6e4b80f9-dd63-53aa-95a3-0cdb28fa8baf"
 Cthulhu = "f68482b8-f384-11e8-15f7-abe071a5a75f"
 KernelAbstractions = "63c18a36-062a-441e-b654-da1e3ab1ce7c"
+Metal = "dde4c033-4e86-420c-a63e-0dd931031962"
 PProf = "e4faabce-9ead-11e9-39d9-4379958e3056"