-
Notifications
You must be signed in to change notification settings - Fork 7
Rework sqrt cbrt #1328
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: develop
Are you sure you want to change the base?
Rework sqrt cbrt #1328
Conversation
|
Oh hi @mborland I don't know if we are supposed to work in the Alliance repo or Boost. I thought this was in the Alliance, but I'm mistakenly in the boost repo |
|
See also #1311 |
|
Looking at the Boost calendar, we have until the 18th for new libs and another week for larger changes. If this one fits, it might be nice. Let's see if it seems stable. |
|
Hi Matt (@mborland) I also added some trial counts to the root tests. So if this unduly bogs down some of the slow runners, then we can walk those changes down. Do please keep an eye out for anything like excessively increased CI/CD time. I see recent total times arond 3 hours. So we need to see if I added to much to the sanitizers or something weird like that. |
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## develop #1328 +/- ##
=========================================
+ Coverage 98.8% 98.8% +0.1%
=========================================
Files 278 278
Lines 18013 18034 +21
Branches 1916 1918 +2
=========================================
+ Hits 17790 17813 +23
+ Misses 223 221 -2
... and 7 files with indirect coverage changes Continue to review full report in Codecov by Sentry.
🚀 New features to boost your workflow:
|
The alliance repo was moved into boost now that we've been accepted and release is coming up. Your access should be exactly the same. |
This PR reworks and improves the existing
sqrtandcbrtimpls.sqrtreduce the argument further tocbrtreduce the argument further tocbrtto expected value.sqrtandcbrt.This is one that I've had planned for a while. I think it's about as good as I'm going to get it with the current technology.
Note that these calculations are still slow and could DEFINITELY use efficiency improvements, for instance with a better initial guess.