Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

A better scheme for computing benchmark names in absence of progname #70

Open
rrnewton opened this issue Nov 20, 2014 · 1 comment
Open

Comments

@rrnewton
Copy link
Member

Currently if progname is set manually, that had better be a unique identifier for the benchmark.

Otherwise, it currently falls back to basename of the target path. But sometimes the target path just ends in "Makefile". It should really do one of the following:

  • Use the full target path
  • Use a strategy to report a subset of the target path but ensure that that subset is UNIQUE among the benchmark suite. Also, in A/B/Makefile, it might be good to take B, but this will probably fall out of a more general policy for finding positions in the path that yield unique names.

The problem with the latter strategy is that it's too complicated and can surprise people. Extending the benchmark suite could make non-local changes in the names of the existing suite. So for now, we'll stick with the first strategy.

@rrnewton
Copy link
Member Author

Update: these respective behaviors are captured by canonicalBenchName vs prettyBenchName, though the latter isn't implemented yet and is just aliased to the former presently.

This ticket can stay open until the latter policy is implemented.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

1 participant