> The people on clang architectures need to know that the gcc systems
>are
> different, that different decisions have been made. Education is
>way
> more important than consistancy.
I'm all for being educated about differences between architectures. I
think the current manual pages don't achieve it in this particular
regard. They don't, as far as I see, mention anywhere that there even
*are* gcc systems as opposed to clang systems. I've only learned about
it from this mailing list and from blog entries by Frederic Cambus.
> This manual page is not hurting you.
It does cause confusion (e.g. I see no way to find out from the manual
pages on which platforms the GNU assembler is part of the system and
on which platforms it is not; removing the man pages for certain
platforms might not be the solution, but the problem is real).
Regards
Stanislav
On Fr, 03 Mär 2023 09:43:16 -0700
Theo de Raadt <deraadt@openbsd.org> wrote:
> And I think you are INCORRECT.
>
> The #1 reason to make a manual page visible is for learning.
>
> The people on clang architectures need to know that the gcc systems
>are
> different, that different decisions have been made. Education is
>way
> more important than consistancy.
>
> "Stanislav Syekirin" wrote:
>
>> I agree. I would expect man pages for as(1), gcc(1), gcc-local(1)
>> etc. to be present if as and gcc are present, and absent if they are
>> absent. Or, alternatively, gcc-local(1) should document which
>> platforms use gcc and which don't.
>>
>> Regards
>> Stanislav
>>
>> On Do, 2 Mär 2023 22:47:08 +0000
>> Jason McIntyre <jmc@kerhand.co.uk> wrote:
>>
>> > i don;t think we should be installing gcc-local(1) on any archs
>> > where
>> > gcc isnt happening:
>> > $ uname -a
>> > OpenBSD manila.kerhand.co.uk 7.2 GENERIC.MP#22 amd64
>> > $ man gcc
>> > man: No entry for gcc in the manual.
>> > jmc
>> >
>>
>
No comments:
Post a Comment