nvbench

nvbench copied to clipboard

Bazel build rules

7

We are planning on using nvbench in our bazel monorepo. This will involve writing bazel build rules. It would be awesome if nvbench had the build rules in the repository...

mmcloughlin

How can the output metrics be interpreted

4

I'm fairly new to arena of profiling CUDA kernels and would like to learn more about the basic output metrics of this library. Specifically, looking at the output of ```nvbench.example.throughput```:...

seanajohnston

Async benchmarks always deadlock

1

The recent switch to lazy loading by default in CTK 12.2 seems to have broken the async benchmarks. This can be reproduced by `nvbench.example.axes`. The deadlock can be fixed by...

gevtushenko

Add option for reporting median runtime

1

I stumbled across [this Slack thread](https://nvidia.slack.com/archives/C01Q5NC7NT0/p1658214334637459) recently when I was trying to measure small kernels with nvbench and got fluctuating results. As @senior-zero notes in the thread, the variance of...

sleeepyjack

Tracking `nvbench` benchmark results

5

Is there any interest in tracking the results from `nvbench`? I'm considering adding an adapter for `nvbench` to my continuous benchmarking tool, Bencher: https://github.com/bencherdev/bencher And I figured I should check...

epompeii

When using this project as a with `CUDAHOSTCXX=g++-11` and `CXX=clang++-17` as dependency, I'm getting the following compilation error. ``` FAILED: _deps/nvbench-build/nvbench/CMakeFiles/nvbench.dir/blocking_kernel.cu.o ccache /usr/local/cuda/bin/nvcc -forward-unknown-to-host-compiler -ccbin=/usr/lib/ccache/g++ -Dnvbench_EXPORTS -I/home/stephan/projects/proviz-app-samples/debug/_deps/nvbench-src -I/home/stephan/projects/proviz-app-samples/debug/_deps/nvbench-build -I/home/stephan/projects/proviz-app-samples/debug/_deps/fmt-src/include -I/home/stephan/projects/proviz-app-samples/debug/_deps/nvbench-build/nvbench/detail...

theHamsta

It would be great if there was a way to say a benchmark is cuda graph compatible and have nvbench automatically record and replay the graph as a column for...

luitjens

Fixes cudaErrorInvalidValue when running on nvbench-created cuda stream

3

This PR fixes a minor issue that may occur when `nvbench` is run on multiple GPUs without a user-provided cuda stream. ## The issue The error that I observed in...

elstehle

[QST] Is there a CLI command to record GPU time for all iterations?

1

There are tags for `"nv/cold/time/gpu/stdev/relative"` and `"nv/cold/time/gpu/mean"`, but is there a way to record all iterations in the json output? This is an [option ](https://pytest-benchmark.readthedocs.io/en/latest/usage.html) to record all data in...

GregoryKimball

alliepiper