NVIDIA / NVIDIA/nvbench

Revisit handling dependent DLLs on Windows

Open
#369 1 comment 1 reaction 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Cuda
Stars
927
Forks
123
Avg merge
2d 10h
Merged PRs (30d)
2

Description

I worry that this will not be sufficient for downstream consumers -- eg. a project using nvbench might fail to find these at runtime without the user reconfig'ing their path manually.

Here's how we handled this "back in the day" before we yanked msvc the first time. It copied the dependent dlls into same dir as nvbench's binaries, so that transitive linking should work without extra setup (IIRC).

https://github.com/NVIDIA/nvbench/blob/pre_msvc_drop/cmake/NVBenchDependentDlls.cmake

Originally posted by @alliepiper in https://github.com/NVIDIA/nvbench/pull/354#discussion_r3269049306

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by comparing the current Windows dependency handling with the historical cmake/NVBenchDependentDlls.cmake implementation linked in the issue. Trace how nvbench binaries are produced and where their dependent DLLs are handled. Done means downstream Windows consumers can run without manually reconfiguring their path, with dependent DLLs beside the nvbench binaries.

Written by the indexing model from the issue text.

Assessment

Tech stack
cmake
Domain
build-system, operating-systems
Issue type
Bug
Difficulty
3/5
Estimated time
1-2 days
Activity status
Quiet
Clarity
Mostly clear
Newbie friendliness
58/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.