From e71d3acc44450618a44b6337e8daa352c680b9e8 Mon Sep 17 00:00:00 2001
From: Michael K <michael.keiblinger@gmail.com>
Date: Wed, 27 Nov 2024 18:39:48 +0100
Subject: [PATCH] use distinct kernels in many kernels launch
---
tests/many_kernels_launch/main.cpp | 36 +++++++++++-------------------
1 file changed, 13 insertions(+), 23 deletions(-)
diff --git a/tests/many_kernels_launch/main.cpp b/tests/many_kernels_launch/main.cpp
index e813e92..f64bc9f 100644
--- a/tests/many_kernels_launch/main.cpp
+++ b/tests/many_kernels_launch/main.cpp
@@ -32,7 +32,6 @@ int main() {
libreCuDeviceGetName(name_buffer, 256, device);
std::cout << "Device Name: " + std::string(name_buffer) << std::endl;
- LibreCUmodule module{};
// read cubin file
uint8_t *image;
@@ -46,39 +45,28 @@ int main() {
std::memcpy(image, bytes.data(), bytes.size());
n_bytes = bytes.size();
}
- CUDA_CHECK(libreCuModuleLoadData(&module, image, n_bytes));
- // read functions
- uint32_t num_funcs{};
- CUDA_CHECK(libreCuModuleGetFunctionCount(&num_funcs, module));
- std::cout << "Num functions: " << num_funcs << std::endl;
-
- auto *functions = new LibreCUFunction[num_funcs];
- CUDA_CHECK(libreCuModuleEnumerateFunctions(functions, num_funcs, module));
-
- for (size_t i = 0; i < num_funcs; i++) {
- LibreCUFunction func = functions[i];
- const char *func_name{};
- CUDA_CHECK(libreCuFuncGetName(&func_name, func));
- std::cout << " function \"" << func_name << "\"" << std::endl;
+ size_t num_kernels = 1025;
+ LibreCUmodule modules[num_kernels];
+ for (int i = 0; i < num_kernels; i++) {
+ CUDA_CHECK(libreCuModuleLoadData(modules + i, image, n_bytes));
}
- delete[] functions;
-
- // find function
- LibreCUFunction func{};
- CUDA_CHECK(libreCuModuleGetFunction(&func, module, "emtpy_kernel"));
+ // find functions
+ LibreCUFunction funcs[num_kernels];
+ for (int i = 0; i < num_kernels; i++) {
+ CUDA_CHECK(libreCuModuleGetFunction(funcs + i, modules[i], "emtpy_kernel"));
+ }
// create stream
LibreCUstream stream{};
CUDA_CHECK(libreCuStreamCreate(&stream, 0));
void *params[] = {};
- size_t num_kernels = 1025;
auto start = std::chrono::high_resolution_clock::now();
for (int i = 0; i < num_kernels; ++i) {
- CUDA_CHECK(libreCuLaunchKernel(func,
+ CUDA_CHECK(libreCuLaunchKernel(funcs[i],
1, 1, 1,
1, 1, 1,
0,
@@ -105,7 +93,9 @@ int main() {
CUDA_CHECK(libreCuStreamDestroy(stream));
// unload module
- CUDA_CHECK(libreCuModuleUnload(module));
+ for (int i = 0; i < num_kernels; ++i) {
+ CUDA_CHECK(libreCuModuleUnload(modules[i]));
+ }
// destroy ctx
CUDA_CHECK(libreCuCtxDestroy(ctx));