Re: Flock CFP: Language SIGs discussion

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On 7/13/23 11:32, Fabio Valentini wrote:
> On Thu, Jul 13, 2023 at 5:25 PM Demi Marie Obenour
> <demiobenour@xxxxxxxxx> wrote:
>>
>> On 7/5/23 02:22, Jens-Ulrik Petersen wrote:
>>> I have submitted a Flock proposal to have a common discussion session for
>>> (modern) Language SIGs. I think for this to be successful we need
>>> representatives from various Language SIGs to be there (Rust, Haskell,
>>> OCaml, Golang and of course Python and older ecosystems like Perl, R, TeX
>>> come to mind immediately - surely there are more). Language packaging
>>> experts are also welcome of course independently or affiliated to one or
>>> more language SIGs. Of course I also hope there will be broad attendance by
>>> interested contributors.
>>>
>>> The idea is to talk about common and distinct problems faced, both to learn
>>> from each other and to come up with practical ideas and plans for generally
>>> easing Fedora's mass packaging efforts.
>>>
>>> If you plan to be at Flock and are willing and able to represent your
>>> Language SIG at this Flock session do please reply or reach out to me. I
>>> think each SIG could do a brief presentation there to kick off the dialogue.
>>>
>>> Thanks, Jens
>>
>> This could be made much easier if Fedora’s build system supported automated
>> cascading rebuilds of transitive dependencies.  Haskell and OCaml are currently
>> linked statically in Fedora, but could be linked dynamically if cascading
>> rebuilds were supported.  Rust is likely, IMO, gain improved support for dynamic
>> linking in the future.
>>
>> I am _not_ going to start a debate as to whether requiring cascading rebuilds
>> is a good idea.  That requirement comes from Haskell, OCaml, and Rust, not me,
>> and so any complaints should be directed there.  This subthread is strictly
>> about changes to Fedora’s build system that make it easier to implement
>> cascading rebuilds.
> 
> I'm not sure how often I have to repeat myself, but dynamic linking
> for Rust crates is NOT GOING TO HAPPEN. Even if the build system
> supported doing this, the average number of updates for Rust crates I
> push every week would keep the build system busy for a month, not even
> mentioning the fact that it would blow up package sizes exponentially
> (yes, *exponentially*, by about a factor of 2^n, with the number n of
> optional features they support). Please stop mentioning Rust in this
> context, it is misleading.
> 
> Fabio

Why is static linking any better?  I’m not suggesting building the
Cartesian product of all possible build configurations.  That would be
absurd.  From my perspective it looks like static linking actually
requires _more_ effort from the builders: they must build everything
for each application that uses it instead of just building it once.
Also if Rust winds up being a significant fraction of the overall system,
the storage requirements of statically linking everything that uses Rust
will be a serious problem.
-- 
Sincerely,
Demi Marie Obenour (she/her/hers)
_______________________________________________
devel mailing list -- devel@xxxxxxxxxxxxxxxxxxxxxxx
To unsubscribe send an email to devel-leave@xxxxxxxxxxxxxxxxxxxxxxx
Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: https://lists.fedoraproject.org/archives/list/devel@xxxxxxxxxxxxxxxxxxxxxxx
Do not reply to spam, report it: https://pagure.io/fedora-infrastructure/new_issue




[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Index of Archives]     [Fedora Announce]     [Fedora Users]     [Fedora Kernel]     [Fedora Testing]     [Fedora Formulas]     [Fedora PHP Devel]     [Kernel Development]     [Fedora Legacy]     [Fedora Maintainers]     [Fedora Desktop]     [PAM]     [Red Hat Development]     [Gimp]     [Yosemite News]

  Powered by Linux