finalizeTest(Test) | o2::benchmark::GPUbenchmark< chunk_t > | |
globalFinalize() | o2::benchmark::GPUbenchmark< chunk_t > | |
globalInit() | o2::benchmark::GPUbenchmark< chunk_t > | |
GPUbenchmark()=delete | o2::benchmark::GPUbenchmark< chunk_t > | |
GPUbenchmark(benchmarkOpts &opts) | o2::benchmark::GPUbenchmark< chunk_t > | inline |
initTest(Test) | o2::benchmark::GPUbenchmark< chunk_t > | |
measure(void(GPUbenchmark::*)(T...), const char *, T &&... args) | o2::benchmark::GPUbenchmark< chunk_t > | |
printDevices() | o2::benchmark::GPUbenchmark< chunk_t > | |
run() | o2::benchmark::GPUbenchmark< chunk_t > | |
runConcurrent(void(*kernel)(chunk_t *, size_t, T...), std::vector< std::pair< float, float > > &chunkRanges, int32_t nLaunches, int32_t dimStreams, int32_t nBlocks, int32_t nThreads, T &... args) | o2::benchmark::GPUbenchmark< chunk_t > | |
runDistributed(void(*kernel)(chunk_t **, size_t *, T...), std::vector< std::pair< float, float > > &chunkRanges, int32_t nLaunches, size_t nBlocks, int32_t nThreads, T &... args) | o2::benchmark::GPUbenchmark< chunk_t > | |
runSequential(void(*kernel)(chunk_t *, size_t, T...), std::pair< float, float > &chunkRanges, int32_t nLaunches, int32_t dimGrid, int32_t dimBlock, T &... args) | o2::benchmark::GPUbenchmark< chunk_t > | |
runTest(Test, Mode, KernelConfig) | o2::benchmark::GPUbenchmark< chunk_t > | |
~GPUbenchmark()=default | o2::benchmark::GPUbenchmark< chunk_t > | virtual |