nvbench
nvbench copied to clipboard
NVIDIA
CUDA Kernel Benchmarking Library
The `test_export` tests added in #42 do not work with MSVC when run from a VS2019 dev prompt. The test executables fail to run:  This could be fixed by...
## Problem Today, I can provide a meaningful name for a type axes using the `set_type_axes_names` function. For example, ``` using ctcs_types = nvbench::type_list; NVBENCH_BENCH_TYPES(copy_type_conversion_sweep, NVBENCH_TYPE_AXES(ctcs_types, ctcs_types)) .set_type_axes_names({"In", "Out"}); ```...
## Update The information below is relevant, but outdated. #42 is adding install rules for NVBench. It's still extremely important to make sure that either: 1. All projects involved are...
When using nvbench::main in a target, nvbench adds an include path for nvbench using `-I` instead of `-isystem`. This causes tools like clang-tidy to pick up nvbench as outside of...
It would be nice to have a utility similar to Google Benchmarks ["DoNotOptimize" ](https://github.com/google/benchmark/blob/713b9177183375c8b1b25595e33daf2a1625df5b/include/benchmark/benchmark.h#L339-L377) that attempts to prevent compiler optimizations from discarding results. GBench's version uses inline assembly tricks that...
Recent results show that noise could be increased up to 50% due to X-Server running on the device. To warn users about the noisy environment, we could check if GPU...
One of my favorite things about [Catch2](https://github.com/catchorg/Catch2/blob/devel/docs/tutorial.md#test-cases-and-sections) is that you can provide a plain text, human readable description with each test. ``` TEST_CASE( "Factorials are computed", "[factorial]" ) { REQUIRE(...
Hi there, I would like to integrate nvbench on my C++ apps. The method that runs the GPU kernel is a template method as follows. ``` template int gemm_cutlass_launch_int(nvbench::state& state)...
The `--profile` flag executes a benchmark a single time to enable profiling a benchmark with tools like Nsight Systems and Nsight Compute. These tools are[ incompatible with concurrent use of...