BUG, SIMD: Fix rounding large numbers on SSE2#22750
Merged
seberg merged 1 commit intonumpy:mainfrom Dec 7, 2022
Merged
Conversation
Before SSE41, there were no native instructions for rounding operations on double precision. We usually emulate it by assuming that the `MXCR` register is set to rounding, adding a large number `2^52` to `X` and then subtracting it back to eliminate any excess precision as long as `|X|` is less than `2^52` otherwise returns `X.` The current emulated intrinics `npyv_[rint,floor, ceil, trunc]_f64` was not checking whether `|x|` equal or large `2^52` which leads to losing accuracy on large numbers.
Member
|
LGTM, thanks Sayed! |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
closes #22170
Before SSE41, there were no native instructions for rounding
operations on double precision. We usually emulate it by assuming
that the
MXCRregister is set to rounding, adding alarge number
2^52toXand then subtracting it back toeliminate any excess precision as long as
|X|is less than2^52otherwise returns
X.The current emulated intrinsics
npyv_[rint, floor, ceil, trunc]_f64was not checking whether
|x|equal or large2^52which leadsto losing accuracy on large numbers.