Opened 5 years ago
Closed 5 years ago
#23209 closed defect (duplicate)
Large performance regression since MPIR upgrade
Reported by:  mmezzarobba  Owned by:  

Priority:  critical  Milestone:  sageduplicate/invalid/wontfix 
Component:  performance  Keywords:  
Cc:  vdelecroix, jpflori, jdemeyer, vbraun  Merged in:  
Authors:  Reviewers:  Marc Mezzarobba  
Report Upstream:  N/A  Work issues:  
Branch:  u/mmezzarobba/mpir_gnu_as (Commits, GitHub, GitLab)  Commit:  26420450be58a7a301e975d9671ff16017ebeff0 
Dependencies:  Stopgaps: 
Description
Due to a bug introduced in #22493, MPIR is built without assembly support on many systems, causing a significant performance regression for pretty much everything in sage.
Change History (10)
comment:1 Changed 5 years ago by
 Component changed from PLEASE CHANGE to performance
 Status changed from new to needs_review
comment:2 followup: ↓ 4 Changed 5 years ago by
comment:3 Changed 5 years ago by
 Commit changed from ee4536bb60f11b46aaa889995f0e2c271c032fe9 to 26420450be58a7a301e975d9671ff16017ebeff0
Branch pushed to git repo; I updated commit sha1. This was a forced push. New commits:
2642045  Fix detection of GNU as version in mpir spkginstall

comment:4 in reply to: ↑ 2 Changed 5 years ago by
Replying to tscrim:
You should also bump the version to make sure this gets recompiled (i.e., treat this like a changed patch).
Done, thanks. (I hate working with sage packages.)
comment:5 Changed 5 years ago by
 Cc jdemeyer vbruan added
Unfortunately I don't have the technical expertise to comment on the change itself.
comment:6 Changed 5 years ago by
 Cc vbraun added; vbruan removed
comment:7 Changed 5 years ago by
This conflicts with beta10.
comment:8 followup: ↓ 9 Changed 5 years ago by
 Milestone changed from sage8.0 to sageduplicate/invalid/wontfix
Actually, I think this was inadvertently fixed by #23146.
comment:9 in reply to: ↑ 8 Changed 5 years ago by
 Status changed from needs_review to positive_review
comment:10 Changed 5 years ago by
 Resolution set to duplicate
 Reviewers set to Marc Mezzarobba
 Status changed from positive_review to closed
You should also bump the version to make sure this gets recompiled (i.e., treat this like a changed patch).