Hi Jose,

On Sun, Nov 23, 2025 at 01:11:29AM +0100, Jose E. Marchesi via Gcc wrote:
> We appreciate the trust in us and our work, and we can only assure you
> that the GCC community, hackers and users both, will not be disappointed
> and will not regret having accepted this front-end.  We will do our best
> to not be a burden to anyone, and on the contrary contribute with as
> much value as possible to the rest of the compiler.  We are very happy
> with the opportunity; it is a privilege to be in GCC, and we REALLY
> appreciate it for what it's worth.  So thank you and the community.

Very nice! It really was time to get this frontend and new vibrant
community added to GCC. Thanks for all your work making GCC a little
more diverse.

> - We interpret the first stipulation as not building the algol68
>   front-end when --enable-language=all is specified.  Is this correct?
>   In that case, we will adapt the patches to make it so.

The "stipulation" might be worded somewhat ambigiously. But please
make sure it is enabled for --enable-languages=all so it will picked
up by testers like the gcc-fullest buildbot. As others have pointed
out it is just about not enabling the frontend by default if not
explicitly added with --enable-languages.

> - Once the previous point is clarified, should we push the patch series
>   as submitted to gcc-patches, even if they are not bisectable, or is it
>   preferred to squash the patches into a single big commit?

If the tree doesn't build with some patches you might consider moving
the patch that enables the building of the new frontend last (assuming
everything else just builds when the frontend/runtime isn't enabled).

> As for suggestions on how to make it easier to maintain GCC front-ends
> outside of trunk, we don't have any specific suggestion right now, but
> it is an interesting topic and will follow up in a different thread
> sharing our experience, in the hope something useful can be extracted
> from it.

I would be interested in whether using a fork in the forge was more or
less helpful than having it as branch. Did it provide you more
flexibility or didn't it really matter? But first get the patches
merged, we can reflect on the experiences later.

Thanks,

Mark

Reply via email to