[bug#60847] [PATCH] Enable cross-compilation for the pyproject-build-system.

[Maxim Cournoyer]

Hi Ludovic,

Ludovic Courtès <ludo@gnu.org> writes:

> Hi Maxim,
>
> Maxim Cournoyer <maxim.cournoyer@gmail.com> skribis:
>
>> Ludovic Courtès <ludo@gnu.org> writes:
>>
>>> Hello,
>>>
>>> Maxim Cournoyer <maxim.cournoyer@gmail.com> skribis:
>>>
>>>> +++ b/guix/packages.scm
>>>> @@ -1864,28 +1864,30 @@ (define* (bag->derivation bag #:optional context)
>>>
>>> […]
>>>
>>>> +  (let ((builder-name (procedure-name (bag-build bag))))
>>>> +    (if (or (bag-target bag)
>>>> +            (eq? 'pyproject-build builder-name))
>>>> +        (bag->cross-derivation bag)
>>>
>>> This one part is a showstopper to me, for two reasons:
>>>
>>>   1. We cannot rely on ‘procedure-name’ (it’s a debugging aid and it’s
>>>      not guaranteed to return something useful).
>>>
>>>   2. Special-casing build systems here is not okay: the bag and build
>>>      system abstractions exist to maintain separation of concerns.
>>>
>>> I understand there’s an actual bug to fix and the desire to fix a more
>>> common issue, but I think this one approach is not the way forward.
>>>
>>> I hope that makes sense!
>>
>> I agree this is not "pretty", but it would be a "temporary" kludge until
>
> To make sure there’s no misunderstanding, I think that’s not okay even
> as a “temporary kludge” (I’m not against temporary kludges in general,
> I’ve done my share of that ;-), but this would be way too brittle.)

"way too brittle" is a strong warning :-).  I don't understand what is
so brittle about it?  It's not like we change the build system names
often.

>> all the build systems can be migrated (and the package adjusted for) the
>> "new" way, which is: native-inputs and inputs always co-exist, whether
>> the build is a native one or a cross one.
>
> Again, I’m not sure about the “new way”.  I’m sorry I lack the bandwidth
> to review this quickly, especially a wide-ranging change like
> inputs/native-inputs.
>
> (Just yesterday I was fixing cross-compilation issues on ‘core-updates’;
> that gives a lot of insight as to how native-inputs/inputs play out in
> practice.)
>
> [...]
>
>> Otherwise, could you offer a concrete suggestion as the way forward?  I
>> appreciate the "that's not the way", but stopping short of suggesting a
>> better alternative leaves me wanting more :-).
>
> Yup, I’m sorry I don’t have any suggestions!

Thanks, it's useful to state that up front.  Otherwise the other side of
the line goes like "... so?" :-).

> Perhaps one suggestion would be: start the native-inputs/inputs change
> on a new branch, for all the build systems (or at least the main
> ones), and then try to build the ‘core’ subset (which includes
> cross-compilation) to get an idea of the kind of code simplification
> opportunities as well as issues that arise.  I feel like it’s hard to
> anticipate the impact of such a change without actually trying.

I can do this; the exercise would be similar to what I've done for
pyproject-build-system, where I was able to build complex packages (both
natively and cross-compiled) with the new same-bag representation
change, but to a larger scale.  It's more work, which is why I would
have preferred to contain its scope to get started, but the changes were
easy, IIRC.

Basically, it forces us to review all the (assoc-ref inputs
"package-name") or (search-input-file inputs "file") procedure calls and
remove the assumption that native-inputs are also inputs for native
builds (that'll fix multiple cross-compilation bugs at the same time, so
it's a good thing to do anyway).

To be continued!

-- 
Thanks,
Maxim