-
Notifications
You must be signed in to change notification settings - Fork 246
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
timing: Always split histogram with zero as a boundary #436
base: master
Are you sure you want to change the base?
Conversation
In principle, I think this is a good idea. Can you provide an example output?
I'm not sure about this though, as it breaks the understanding that the area of each bar (excepting perhaps the two ends) are proportional to their frequency. Perhaps it is better to have two historgrams instead? |
Example output : The And yeah, possibly two distinct histograms would make more sense this would also provide more granularity ...
|
If the timing histogram has both positive and negative numbers, then we ensure that zero is used as a boundary and that at least 25% of the bins are dedicated to both positive and negative side so you get a decent level of details on both sides. Obviously that means bin size is different for positive / negative sides. Signed-off-by: Sylvain Munaut <[email protected]>
Just added a slightly different version that also prints exactly how many EPs are in there (in addition to the bar). I find it useful to check when I actually except a certain number of EP to fail to make sure only that # and not more are failing.
|
@eddiehung feedback ? |
If the timing histogram has both positive and negative numbers,
then we ensure that zero is used as a boundary and that at least
25% of the bins are dedicated to both positive and negative side
so you get a decent level of details on both sides.
Obviously that means bin size is different for positive / negative
sides.
Signed-off-by: Sylvain Munaut [email protected]