-
Notifications
You must be signed in to change notification settings - Fork 24
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
Fast abort of benchmarks when same executable is missing #224
Conversation
d6c9d8e
to
8c493e2
Compare
2458acc
to
3879bae
Compare
298d90a
to
14b3ea1
Compare
fix pylint issues return true false logic of completed_task create method to mark binary as missing give executor a binary missing attribute change logic of batch scheduler to consume first item in list add get_executor and is_binary_missing to run id
test all executors
- also mark it so that it isn’t considered a test
Using the executor object to identify run ids that won't work is catching only a part of them. Thus, instead of checking the executor object, I am now checking the executable path/filename as part of the cmdline. This also makes sure we only match the same executables. The problem here is that we can use the various variables in the command line expansions also in the path or executor names. Thus, looking at the cmdline after expansion is better. |
…executable This avoids issues with the command line being created early for runId identity Need to sort the runs so that we have a deterministic order.
… path We support interpolation of variables in executer path and excutable name, this, need to use the rendered version. Executor objects may also be instantiated multiple times, which means we can’t really use identity. Change used vocabulary to remove binary, since we use executable in the config file.
No description provided.