Merge branch 'weak_damping' into 'main' (998523a9) · Commits · Cianciosa, Mark / graph_framework

CMakeLists.txt

+1 −0

Original line number	Diff line number	Diff line
		@@ -6,6 +6,7 @@ project (rays CXX)
		# Build Options
		#-------------------------------------------------------------------------------
		option (USE_PCH "Enable the use of precompiled headers" ON)
		option (USE_STATIC "Limits the dyamics for testing." OFF)

		#-------------------------------------------------------------------------------
		# Set the cmake module path.

graph_benchmark/xrays_bench.cpp

+4 −3

Original line number	Diff line number	Diff line
		@@ -41,8 +41,7 @@ void bench_runner() {
		timeing::measure_diagnostic_threaded timing;

		for (size_t i = 0, ie = threads.size(); i < ie; i++) {
		threads[i] = std::thread([&timing, batch, extra] (const size_t thread_number,
		const size_t num_threads) -> void {
		threads[i] = std::thread([&timing, batch, extra] (const size_t thread_number) -> void {

		const size_t local_num_rays = batch
		+ (extra > thread_number ? 1 : 0);
		@@ -91,7 +90,7 @@ void bench_runner() {
		}
		solve.sync_host();
		timing.end_time(thread_number);
		}, i, threads.size());
		}, i);
		}

		for (std::thread &t : threads) {
		@@ -111,6 +110,8 @@ void bench_runner() {
		//------------------------------------------------------------------------------
		int main(int argc, const char * argv[]) {
		START_GPU
		(void)argc;
		(void)argv;

		bench_runner<float, 1000, 10, 100000> ();
		bench_runner<double, 1000, 10, 100000> ();

graph_driver/xrays.cpp

+34 −26

Original line number	Diff line number	Diff line
		@@ -131,16 +131,17 @@ void trace_ray(const size_t num_times,
		const size_t extra = num_rays%threads.size();

		for (size_t i = 0, ie = threads.size(); i < ie; i++) {
		threads[i] = std::thread([num_times, sub_steps, num_rays, batch, extra] (const size_t thread_number,
		const size_t num_threads) -> void {
		threads[i] = std::thread([num_times, sub_steps, batch, extra] (const size_t thread_number) -> void {

		const size_t num_steps = num_times/sub_steps;
		const size_t local_num_rays = batch
		+ (extra > thread_number ? 1 : 0);

		const uint64_t seed = static_cast<uint64_t> (std::chrono::system_clock::to_time_t(std::chrono::system_clock::now()));
		//const uint64_t seed = 0;
		std::mt19937_64 engine((thread_number + 1)*seed);
		#ifndef STATIC
		std::mt19937_64 engine((thread_number + 1)*static_cast<uint64_t> (std::chrono::system_clock::to_time_t(std::chrono::system_clock::now())));
		#else
		std::mt19937_64 engine(thread_number + 1);
		#endif
		std::uniform_int_distribution<size_t> int_dist(0, local_num_rays - 1);

		auto omega = graph::variable<T, SAFE_MATH> (local_num_rays, "\\omega");
		@@ -156,7 +157,7 @@ void trace_ray(const size_t num_times,

		// Inital conditions.
		if constexpr (jit::is_float<T> ()) {
		#if 0
		#if 1
		init_efit<T, float, SAFE_MATH> (omega, x, y, z,
		ky, kz, engine,
		local_num_rays);
		@@ -166,7 +167,7 @@ void trace_ray(const size_t num_times,
		local_num_rays);
		#endif
		} else {
		#if 0
		#if 1
		init_efit<T, double, SAFE_MATH> (omega, x, y, z,
		ky, kz, engine,
		local_num_rays);
		@@ -176,18 +177,18 @@ void trace_ray(const size_t num_times,
		local_num_rays);
		#endif
		}
		#if 0
		#if 1
		kx->set(static_cast<T> (-700.0));
		#else
		kx->set(static_cast<T> (-30.0));
		#endif
		auto eq = equilibrium::make_vmec<T, SAFE_MATH> (VMEC_FILE);
		//auto eq = equilibrium::make_efit<T, SAFE_MATH> (EFIT_FILE);
		//auto eq = equilibrium::make_vmec<T, SAFE_MATH> (VMEC_FILE);
		auto eq = equilibrium::make_efit<T, SAFE_MATH> (EFIT_FILE);
		//auto eq = equilibrium::make_slab_density<T, SAFE_MATH> ();
		//auto eq = equilibrium::make_slab_field<T, SAFE_MATH> ();
		//auto eq = equilibrium::make_no_magnetic_field<T, SAFE_MATH> ();

		#if 0
		#if 1
		const T endtime = static_cast<T> (2.0);
		#else
		const T endtime = static_cast<T> (0.2);
		@@ -270,7 +271,7 @@ void trace_ray(const size_t num_times,
		solve.sync_host();
		}

		}, i, threads.size());
		}, i);
		}

		for (std::thread &t : threads) {
		@@ -304,8 +305,7 @@ void calculate_power(const size_t num_times,
		const size_t extra = num_rays%threads.size();

		for (size_t i = 0, ie = threads.size(); i < ie; i++) {
		threads[i] = std::thread([num_times, sub_steps, num_rays, batch, extra] (const size_t thread_number,
		const size_t num_threads) -> void {
		threads[i] = std::thread([num_times, sub_steps, batch, extra] (const size_t thread_number) -> void {
		std::ostringstream stream;
		stream << "result" << thread_number << ".nc";

		@@ -323,21 +323,22 @@ void calculate_power(const size_t num_times,
		auto t = graph::variable<T, SAFE_MATH> (local_num_rays, "t");
		auto kamp = graph::variable<T, SAFE_MATH> (local_num_rays, "kamp");

		auto eq = equilibrium::make_vmec<T, SAFE_MATH> (VMEC_FILE);
		//auto eq = equilibrium::make_efit<T, SAFE_MATH> (EFIT_FILE);
		//auto eq = equilibrium::make_vmec<T, SAFE_MATH> (VMEC_FILE);
		auto eq = equilibrium::make_efit<T, SAFE_MATH> (EFIT_FILE);
		//auto eq = equilibrium::make_slab_density<T, SAFE_MATH> ();
		//auto eq = equilibrium::make_slab_field<T, SAFE_MATH> ();
		//auto eq = equilibrium::make_no_magnetic_field<T, SAFE_MATH> ();

		absorption::root_finder<dispersion::hot_plasma<T, dispersion::z_erfi<T, SAFE_MATH>, SAFE_MATH>>
		root(kamp, omega, kx, ky, kz, x, y, z, t, eq,
		stream.str(), local_num_rays, thread_number);
		root.compile();
		//absorption::root_finder<dispersion::hot_plasma<T, dispersion::z_erfi<T, SAFE_MATH>, SAFE_MATH>>
		absorption::weak_damping<T, SAFE_MATH>
		power(kamp, omega, kx, ky, kz, x, y, z, t, eq,
		stream.str(), thread_number);
		power.compile();

		for (size_t j = 0, je = num_steps + 1; j < je; j++) {
		root.run(j);
		power.run(j);
		}
		}, i, threads.size());
		}, i);
		}

		for (std::thread &t : threads) {
		@@ -371,8 +372,7 @@ void bin_power(const size_t num_times,
		const size_t extra = num_rays%threads.size();

		for (size_t i = 0, ie = threads.size(); i < ie; i++) {
		threads[i] = std::thread([num_times, sub_steps, num_rays, batch, extra] (const size_t thread_number,
		const size_t num_threads) -> void {
		threads[i] = std::thread([num_times, sub_steps, batch, extra] (const size_t thread_number) -> void {
		std::ostringstream stream;
		stream << "result" << thread_number << ".nc";

		@@ -390,7 +390,9 @@ void bin_power(const size_t num_times,
		auto power = graph::variable<T, SAFE_MATH> (local_num_rays, static_cast<T> (1.0), "power");
		auto k_sum = graph::variable<T, SAFE_MATH> (local_num_rays, static_cast<T> (0.0), "k_sum");

		auto eq = equilibrium::make_vmec<T, SAFE_MATH> (VMEC_FILE);
		//auto eq = equilibrium::make_vmec<T, SAFE_MATH> (VMEC_FILE);
		auto eq = equilibrium::make_efit<T, SAFE_MATH> (EFIT_FILE);

		auto x_real = eq->get_x(x, y, z);
		auto y_real = eq->get_y(x, y, z);
		auto z_real = eq->get_z(x, y, z);
		@@ -467,7 +469,7 @@ void bin_power(const size_t num_times,
		}

		sync.join();
		}, i, threads.size());
		}, i);
		}

		for (std::thread &t : threads) {
		@@ -485,13 +487,19 @@ void bin_power(const size_t num_times,
		//------------------------------------------------------------------------------
		int main(int argc, const char * argv[]) {
		START_GPU
		(void)argc;
		(void)argv;
		const timeing::measure_diagnostic total("Total Time");

		jit::verbose = verbose;

		const size_t num_times = 100000;
		const size_t sub_steps = 100;
		#ifndef STATIC
		const size_t num_rays = 100000;
		#else
		const size_t num_rays = 1;
		#endif

		const bool use_safe_math = true;

graph_framework.xcodeproj/project.pbxproj

+9 −15

Original line number	Diff line number	Diff line
		@@ -1336,8 +1336,9 @@
		"EFIT_FILE=\\\"/Users/m4c/Projects/graph_framework/graph_tests/efit.nc\\\"",
		"VMEC_FILE=\\\"/Users/m4c/Projects/graph_framework/graph_tests/vmec.nc\\\"",
		USE_METAL,
		"CXX_FLAGS=\\\"-g\\ -fsanitize=undefined\\ -fsanitize=float-divide-by-zero\\ -fsanitize-trap=all\\\"",
		"CXX_FLAGS=\\\"-g\\\"",
		"CXX=\\\"c++\\ -I/Users/m4c/Projects/graph_framework/graph_framework\\ -std=gnu++2a\\\"",
		STATIC,
		"DEBUG=1",
		"$(inherited)",
		);
		@@ -1349,16 +1350,12 @@
		GCC_WARN_UNUSED_VARIABLE = YES;
		HEADER_SEARCH_PATHS = /usr/local/include;
		LIBRARY_SEARCH_PATHS = /usr/local/lib;
		MACOSX_DEPLOYMENT_TARGET = 13.5;
		MACOSX_DEPLOYMENT_TARGET = 14.0;
		MTL_ENABLE_DEBUG_INFO = INCLUDE_SOURCE;
		MTL_FAST_MATH = YES;
		ONLY_ACTIVE_ARCH = YES;
		OTHER_CPLUSPLUSFLAGS = (
		"$(OTHER_CFLAGS)",
		"-fsanitize=float-divide-by-zero",
		);
		OTHER_CPLUSPLUSFLAGS = "$(OTHER_CFLAGS)";
		OTHER_LDFLAGS = (
		"-fsanitize=float-divide-by-zero",
		"-lnetcdf",
		"-ld_classic",
		"-rpath",
		@@ -1425,16 +1422,12 @@
		GCC_WARN_UNUSED_VARIABLE = YES;
		HEADER_SEARCH_PATHS = /usr/local/include;
		LIBRARY_SEARCH_PATHS = /usr/local/lib;
		MACOSX_DEPLOYMENT_TARGET = 13.5;
		MACOSX_DEPLOYMENT_TARGET = 14.0;
		MTL_ENABLE_DEBUG_INFO = NO;
		MTL_FAST_MATH = YES;
		ONLY_ACTIVE_ARCH = YES;
		OTHER_CPLUSPLUSFLAGS = (
		"$(OTHER_CFLAGS)",
		"-fsanitize=float-divide-by-zero",
		);
		OTHER_CPLUSPLUSFLAGS = "$(OTHER_CFLAGS)";
		OTHER_LDFLAGS = (
		"-fsanitize=float-divide-by-zero",
		"-lnetcdf",
		"-ld_classic",
		"-rpath",
		@@ -1453,8 +1446,9 @@
		EXECUTABLE_PREFIX = lib;
		GCC_PREPROCESSOR_DEFINITIONS = (
		USE_METAL,
		"CXX_FLAGS=\\\"-g\\ -fsanitize=undefined\\ -fsanitize=float-divide-by-zero\\ -fsanitize-trap=all\\\"",
		"CXX_FLAGS=\\\"-gl\\\"",
		"CXX=\\\"c++\\ -I/Users/m4c/Projects/graph_framework/graph_framework\\ -std=gnu++2a\\\"",
		STATIC,
		"DEBUG=1",
		"$(inherited)",
		);
		@@ -1497,7 +1491,7 @@
		DEVELOPMENT_TEAM = "";
		GCC_PREPROCESSOR_DEFINITIONS = "$(inherited)";
		MACOSX_DEPLOYMENT_TARGET = 13.5;
		OTHER_CPLUSPLUSFLAGS = "-fsanitize=float-divide-by-zero";
		OTHER_CPLUSPLUSFLAGS = "";
		PRODUCT_NAME = "$(TARGET_NAME)";
		SDKROOT = macosx;
		};

graph_framework/CMakeLists.txt

+1 −0

Original line number	Diff line number	Diff line
		@@ -18,6 +18,7 @@ target_compile_definitions (rays
		EFIT_FILE="${CMAKE_CURRENT_SOURCE_DIR}/../graph_tests/efit.nc"
		VMEC_FILE="${CMAKE_CURRENT_SOURCE_DIR}/../graph_tests/vmec.nc"
		$<$<BOOL:${USE_CUDA}>:HEADER_DIR="$<BUILD_INTERFACE:${CMAKE_CURRENT_SOURCE_DIR}>">
		$<$<BOOL:${USE_STATIC}>:STATIC>
		)
		target_include_directories (rays
		INTERFACE