Merge branch 'main' into ad

Project.toml

@@ -13,19 +13,22 @@ StaticArrays = "90137ffa-7385-5640-81b9-e52037218182"
 AMDGPU = "21141c5a-9bdb-4563-92ae-f87d6854732e"
 CUDA = "052768ef-5323-5732-b1bb-66c8b64840ba"
 Enzyme = "7da242da-08ed-463a-9acd-ee780be4f1d9"
+Metal = "dde4c033-4e86-420c-a63e-0dd931031962"
 Polyester = "f517fe37-dbe3-4b94-8317-1923a5111588"
 
 [extensions]
 ParallelStencil_AMDGPUExt = "AMDGPU"
 ParallelStencil_CUDAExt = "CUDA"
 ParallelStencil_EnzymeExt = "Enzyme"
+ParallelStencil_MetalExt = "Metal"
 
 [compat]
 AMDGPU = "0.6, 0.7, 0.8, 0.9, 1"
 CUDA = "3.12, 4, 5"
 CellArrays = "0.3"
 Enzyme = "0.12, 0.13"
 MacroTools = "0.5"
+Metal = "1.2"
 Polyester = "0.7"
 StaticArrays = "1"
 julia = "1.10" # Minimum version supporting Data module creation
@@ -35,4 +38,4 @@ TOML = "fa267f1f-6049-4f14-aa54-33bafae1ed76"
 Test = "8dfed614-e22c-5e08-85e1-65c5234f0b40"
 
 [targets]
-test = ["Test", "TOML", "AMDGPU", "CUDA", "Enzyme", "Polyester"]
+test = ["Test", "TOML", "AMDGPU", "CUDA", "Metal", "Enzyme", "Polyester"]

README.md

@@ -7,7 +7,7 @@ ParallelStencil empowers domain scientists to write architecture-agnostic high-l
 
 <a id="fig_teff">![Performance ParallelStencil Teff](docs/images/perf_ps2.png)</a>
 
-ParallelStencil relies on the native kernel programming capabilities of [CUDA.jl] and [AMDGPU.jl] and on [Base.Threads] for high-performance computations on GPUs and CPUs, respectively. It is seamlessly interoperable with [ImplicitGlobalGrid.jl], which renders the distributed parallelization of stencil-based GPU and CPU applications on a regular staggered grid almost trivial and enables close to ideal weak scaling of real-world applications on thousands of GPUs \[[1][JuliaCon20a], [2][JuliaCon20b], [3][JuliaCon19], [4][PASC19]\]. Moreover, ParallelStencil enables hiding communication behind computation with a simple macro call and without any particular restrictions on the package used for communication. ParallelStencil has been designed in conjunction with [ImplicitGlobalGrid.jl] for simplest possible usage by domain-scientists, rendering fast and interactive development of massively scalable high performance multi-GPU applications readily accessible to them. Furthermore, we have developed a self-contained approach for "Solving Nonlinear Multi-Physics on GPU Supercomputers with Julia" relying on ParallelStencil and [ImplicitGlobalGrid.jl] \[[1][JuliaCon20a]\]. ParallelStencil's feature to hide communication behind computation was showcased when a close to ideal weak scaling was demonstrated for a 3-D poro-hydro-mechanical real-world application on up to 1024 GPUs on the Piz Daint Supercomputer \[[1][JuliaCon20a]\]:
+ParallelStencil relies on the native kernel programming capabilities of [CUDA.jl], [AMDGPU.jl], [Metal.jl] and on [Base.Threads] for high-performance computations on GPUs and CPUs, respectively. It is seamlessly interoperable with [ImplicitGlobalGrid.jl], which renders the distributed parallelization of stencil-based GPU and CPU applications on a regular staggered grid almost trivial and enables close to ideal weak scaling of real-world applications on thousands of GPUs \[[1][JuliaCon20a], [2][JuliaCon20b], [3][JuliaCon19], [4][PASC19]\]. Moreover, ParallelStencil enables hiding communication behind computation with a simple macro call and without any particular restrictions on the package used for communication. ParallelStencil has been designed in conjunction with [ImplicitGlobalGrid.jl] for simplest possible usage by domain-scientists, rendering fast and interactive development of massively scalable high performance multi-GPU applications readily accessible to them. Furthermore, we have developed a self-contained approach for "Solving Nonlinear Multi-Physics on GPU Supercomputers with Julia" relying on ParallelStencil and [ImplicitGlobalGrid.jl] \[[1][JuliaCon20a]\]. ParallelStencil's feature to hide communication behind computation was showcased when a close to ideal weak scaling was demonstrated for a 3-D poro-hydro-mechanical real-world application on up to 1024 GPUs on the Piz Daint Supercomputer \[[1][JuliaCon20a]\]:
 
 ![Parallel efficiency of ParallelStencil with CUDA C backend](docs/images/par_eff_c_julia2.png)
 
@@ -33,7 +33,7 @@ Beyond traditional high-performance computing, ParallelStencil supports automati
 * [References](#references)
 
 ## Parallelization and optimization with one macro call
-A simple call to `@parallel` is enough to parallelize and optimize a function and to launch it. The package used underneath for parallelization is defined in a call to `@init_parallel_stencil` beforehand. Supported are [CUDA.jl] and [AMDGPU.jl] for running on GPU and [Base.Threads] for CPU. The following example outlines how to run parallel computations on a GPU using the native kernel programming capabilities of [CUDA.jl] underneath (omitted lines are represented with `#(...)`, omitted arguments with `...`):
+A simple call to `@parallel` is enough to parallelize and optimize a function and to launch it. The package used underneath for parallelization is defined in a call to `@init_parallel_stencil` beforehand. Supported are [CUDA.jl], [AMDGPU.jl] and [Metal.jl] for running on GPU and [Base.Threads] for CPU. The following example outlines how to run parallel computations on a GPU using the native kernel programming capabilities of [CUDA.jl] underneath (omitted lines are represented with `#(...)`, omitted arguments with `...`):
 ```julia
 #(...)
 @init_parallel_stencil(CUDA,...)
@@ -554,6 +554,7 @@ Please open an issue to discuss your idea for a contribution beforehand. Further
 [CellArrays.jl]: https://github.com/omlins/CellArrays.jl
 [CUDA.jl]: https://github.com/JuliaGPU/CUDA.jl
 [AMDGPU.jl]: https://github.com/JuliaGPU/AMDGPU.jl
+[Metal.jl]: https://github.com/JuliaGPU/Metal.jl
 [Enzyme.jl]: https://github.com/EnzymeAD/Enzyme.jl
 [MacroTools.jl]: https://github.com/FluxML/MacroTools.jl
 [StaticArrays.jl]: https://github.com/JuliaArrays/StaticArrays.jl

ext/ParallelStencil_MetalExt.jl

@@ -0,0 +1,4 @@
+module ParallelStencil_MetalExt
+    include(joinpath(@__DIR__, "..", "src", "ParallelKernel", "MetalExt", "shared.jl"))
+    include(joinpath(@__DIR__, "..", "src", "ParallelKernel", "MetalExt", "allocators.jl"))
+end

src/FieldAllocators.jl

@@ -28,6 +28,7 @@ To see a description of a macro type `?<macroname>` (including the `@`).
 """
 module FieldAllocators
     import ..ParallelKernel
+    import ..ParallelStencil: check_initialized
     @doc replace(ParallelKernel.FieldAllocators.ALLOCATE_DOC,          "@init_parallel_kernel" => "@init_parallel_stencil") macro allocate(args...)     check_initialized(__module__); esc(:(ParallelStencil.ParallelKernel.FieldAllocators.@allocate($(args...)))); end
     @doc replace(ParallelKernel.FieldAllocators.FIELD_DOC,             "@init_parallel_kernel" => "@init_parallel_stencil") macro Field(args...)        check_initialized(__module__); esc(:(ParallelStencil.ParallelKernel.FieldAllocators.@Field($(args...)))); end
     @doc replace(ParallelKernel.FieldAllocators.VECTORFIELD_DOC,       "@init_parallel_kernel" => "@init_parallel_stencil") macro VectorField(args...)  check_initialized(__module__); esc(:(ParallelStencil.ParallelKernel.FieldAllocators.@VectorField($(args...)))); end
@@ -46,5 +47,19 @@ module FieldAllocators
     @doc replace(ParallelKernel.FieldAllocators.TENSORFIELD_COMP_DOC,  "@init_parallel_kernel" => "@init_parallel_stencil") macro XZField(args...)      check_initialized(__module__); esc(:(ParallelStencil.ParallelKernel.FieldAllocators.@XZField($(args...)))); end
     @doc replace(ParallelKernel.FieldAllocators.TENSORFIELD_COMP_DOC,  "@init_parallel_kernel" => "@init_parallel_stencil") macro YZField(args...)      check_initialized(__module__); esc(:(ParallelStencil.ParallelKernel.FieldAllocators.@YZField($(args...)))); end
 
+    macro IField(args...)        check_initialized(__module__); esc(:(ParallelStencil.ParallelKernel.FieldAllocators.@IField($(args...)))); end
+    macro XXYField(args...)      check_initialized(__module__); esc(:(ParallelStencil.ParallelKernel.FieldAllocators.@XXYField($(args...)))); end
+    macro XYYField(args...)      check_initialized(__module__); esc(:(ParallelStencil.ParallelKernel.FieldAllocators.@XYYField($(args...)))); end
+    macro XYZField(args...)      check_initialized(__module__); esc(:(ParallelStencil.ParallelKernel.FieldAllocators.@XYZField($(args...)))); end
+    macro XXYZField(args...)     check_initialized(__module__); esc(:(ParallelStencil.ParallelKernel.FieldAllocators.@XXYZField($(args...)))); end
+    macro XYYZField(args...)     check_initialized(__module__); esc(:(ParallelStencil.ParallelKernel.FieldAllocators.@XYYZField($(args...)))); end
+    macro XYZZField(args...)     check_initialized(__module__); esc(:(ParallelStencil.ParallelKernel.FieldAllocators.@XYZZField($(args...)))); end
+    macro XXYYField(args...)     check_initialized(__module__); esc(:(ParallelStencil.ParallelKernel.FieldAllocators.@XXYYField($(args...)))); end
+    macro XXZZField(args...)     check_initialized(__module__); esc(:(ParallelStencil.ParallelKernel.FieldAllocators.@XXZZField($(args...)))); end
+    macro YYZZField(args...)     check_initialized(__module__); esc(:(ParallelStencil.ParallelKernel.FieldAllocators.@YYZZField($(args...)))); end
+    macro XXYYZField(args...)    check_initialized(__module__); esc(:(ParallelStencil.ParallelKernel.FieldAllocators.@XXYYZField($(args...)))); end
+    macro XYYZZField(args...)    check_initialized(__module__); esc(:(ParallelStencil.ParallelKernel.FieldAllocators.@XYYZZField($(args...)))); end
+    macro XXYZZField(args...)    check_initialized(__module__); esc(:(ParallelStencil.ParallelKernel.FieldAllocators.@XXYZZField($(args...)))); end
+
     export @allocate, @Field, @VectorField, @BVectorField, @TensorField, @XField, @BXField, @YField, @BYField, @ZField, @BZField, @XXField, @YYField, @ZZField, @XYField, @XZField, @YZField
 end