Re: Future of jemalloc on FreeBSD after archive

On Jun 10, 2025, at 9:24 AM, Brooks Davis <brooks@freebsd.org> wrote:
> 
> On Tue, Jun 10, 2025 at 09:02:37AM -0700, Bakul Shah wrote:
>> On Jun 10, 2025, at 8:40???AM, Gleb Popov <arrowd@freebsd.org> wrote:
>>> 
>>> On Mon, Jun 9, 2025 at 11:08???PM Minsoo Choo <minsoochoo0122@proton.me> wrote:
>>>> 
>>>> As of June 2, 2025, jemalloc repository on Github has been archived. [1]
>>> 
>>> Is there any explanation why it got archived? It feels surprising as I
>>> recently heard news that Qt added some code to take advantage of
>>> jemalloc-specific functions.
>> 
>> From https://x.com/nateberkopec/status/1930010446410723533 & followups:
>> 
>> jemalloc mystery resolved: jemalloc has moved to the official Facebook github organization
>> 
>> context: for the last 2+ years the primary maintainers of jemalloc have been FB employees. this just formalizes Evans' stepping off the project.
>> 
>> jemalloc???s not dead ! The Meta fork is the new home, ???still actively developing here??? but issues now have to go in discussions https://github.com/facebook/jemalloc/discussions/7#discussioncomment-13394532
>> 
>> Contributor commit info shows pretty abrupt handoff from Jason Evans to Qi Wang (interwq) and David Goldblatt in 2017, so I presume this ???small team??? was primarily these 2. Note Qi Wang???s contributions tapering off, David stopped post-pandemic: LinkedIn says he left Meta in 2024
> 
> It's been maintained by Facebook forever so that's not a change.

The change is it becomes the primary repo.

> Why not move the repo or break the fork so issues work?  This feels like
> flailing around without understanding how github works and does not
> inspire confidence.

My guess is, the "move" might not be possible if they already
had a fork.  AIUI, issues don't get transferred to a fork.
But no idea really why they did what they did. I was just
answering Gleb's question!