-
Notifications
You must be signed in to change notification settings - Fork 292
use -j$(nproc) for ctest #3431
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
base: main
Are you sure you want to change the base?
use -j$(nproc) for ctest #3431
Conversation
Signed-off-by: Mitchell <[email protected]>
exit 1 | ||
fi | ||
ctest --output-on-failure --test-dir build -E "ctest-nvqpp|ctest-targettests" | ||
ctest -j$(nproc) --output-on-failure --test-dir build -E "ctest-nvqpp|ctest-targettests" |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Amazing idea! 😄
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I had some inspiration 🙂
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Hope this speeds things up without causing thrashing or OOM errors.
That's my concern. I wonder if this was never done for reasons of tests not setting down properly, but we should see soon. |
I know that it causes issues on my machine (despite having 128GB memory), but that will be machine specific. It is also not very reliable to do this on a machine with a GPU because at least some of our tests expect to have access to the GPU's full memory, so the random run-to-run variation of timing will cause problems in that situation. (Still worth testing in the CI, though.) |
These tests are CPU only tests, so we should be good with GPU interference. Memory is a valid concern though, especially on shared runners. |
Looks like it doesn't really improve the time that much 🙁 |
CUDA Quantum Docs Bot: A preview of the documentation can be found here. |
note that llvm-lit is not explicitly added as well since it by default uses all available cpus[0].
[0] - https://llvm.org/docs/CommandGuide/lit.html#cmdoption-lit-j