Re: Fixing regressions and serious issues

[Jean Abou Samra]

Le 14/09/2022 à 22:16, Jonas Hahnfeld a écrit :
> On Wed, 2022-07-20 at 11:39 +0200, Jonas Hahnfeld via Discussions on
> LilyPond development wrote:
> What do we do about this one? Over the past couple of weeks, I tried
> quite a number of ideas, with no success so far.



Thanks a lot for working on this even if it didn't succeed so far.

Just in case for others: Jonas shared some details about what he tried
in https://github.com/ivmai/bdwgc/issues/454#issuecomment-1244375504


> Questions:
> a) Do we stick to the plan of branching next week, after the planned
> release of 2.23.13 this weekend?
> b) If we decide to branch and eventually arrive in December without a
> fix, do we block the release?
>
> At the current moment, branching without a "guaranteed" release date
> bears a certain risk that we will end up with something half-finished
> while blocking progress before resuming a new cycle of development
> releases. What do people think?



b) I would say yes. It would be sad, but for better or worse, our
significant part of our user base is on Windows, as far as I know.

a) I don't know.

One thing I can say is that finalizing !1510 (-dcompile-scheme-code)
is going to take me a day or two (not helped by being sick of working
on that problem), and it wouldn't be unreasonable to have it in the
unstable release before branching, as it changes the execution of Scheme
code in some significant respects (compiling is optional, but the new
error handling isn't). So I'd consider it reasonable to delay the release
to next week-end for now, and see what happens for the Windows crashes. I
might manage to spend some time on that next week, but I'm not sure,
and it's not like I'm particularly good at that kind of low-level work.

One thing we could try is biting the bullet and attempting to
cross-compile Guile 3 for MinGW. We might have a slight chance that
the problem goes away, or becomes different in a way that makes
it easier to understand. And this is something we want to do
on the long term anyway.

Jean