-
Notifications
You must be signed in to change notification settings - Fork 2k
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
tests/periph_gpio: reduce benchmark iterations #14640
tests/periph_gpio: reduce benchmark iterations #14640
Conversation
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.
If Murdock is happy, this sould be fine.
(But can Murdock even be truly happy with the current nrf52dk
situation?)
I'm wondering whether it will solve the benchmark problem on ATmega. When I was benchmarking the new GPIO API in PR #14607, I had either to reduce the number of tests to 1000 or to remove the |
that (results completely nonsense) probably was caused by #14639! |
ah, didn't read good, you had that figured out already... |
Unfortunatly, 100k iterations are is still too much for ATmega. With
It makes no sense that Just for information, real values are:
which makes much more sense. |
Yeah, a year ago when I was already working on the new GPIO API. But this should be solved with PR #14639. |
That's why, I also used an own version in RIOT/tests/periph_gpio_exp/main.c Lines 33 to 48 in 70db5f7
|
It seems that Murdock failed due to a reason that is not related to this PR. |
Our MCU's are predictable enough that 100k iterations should suffice.
f751684
to
07c70a9
Compare
too much shouldn't happen, just take longer. I think this pr didn't include the irq disable fix PR. I've rebased, so it should now. Can you double check? |
Cool, now it is correct 👍
|
Unrelated tests failed:
will re-trigger. |
thanks for confirming! |
Murdock is happy, failed tests before were not related to this PR. |
Contribution description
Our MCU's are predictable enough that 100k iterations should suffice.
No need to wait tens of seconds.
Testing procedure
Issues/PRs references
Found in #12457, was about to increase test time for z1.