tech-pkg archive

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

Re: glibmm build issues with gcc 4.8



Havard Eidnes <he%NetBSD.org@localhost> writes:

> netbsd-6/macppc:

Thanks for the data.

> Well, it's a matter of whether we essentially disable support
> for lots of platforms by insisting on the use of a modern GCC (if
> insistence is what's being suggested).

It is not being suggested.  I ask everyone to read the actual design
instead of making assumptions based on comments from other people who
also have not read the design ;-) This will get improved some more, but
it's a good introduction to the issues.

https://wiki.netbsd.org/pkgsrc/gcc/

As for "disabling", nobody is talking about that or ever meant that.

What we are saying, (a summary to calm things down, but please discuss
relative to the wiki page contents) is:

  on a platform with an old gcc version, which causes many packages to
  have trouble, if we can successfully build something modern (like
  gcc5), then we'll set it up so that's used for all C++ building

the point is that this would be flipped on for a given OS/version/cpu
when we know that makes things better, and not enabled when it makes
things worse.

People still seem upset about this, and I don't understand why.

I did not realize earlier that newer gcc from pkgsrc is troubled on a
large number of CPUs (on which gcc in base is ok), so the cpu part above
was not explicit.


There is an issue about the pain of building a compiler imposed on
people who only want to build packages that would have built with the
base compiler.  But that's largely about build time, not if things work
or not.

Attachment: signature.asc
Description: PGP signature



Home | Main Index | Thread Index | Old Index