GCC withdraw
Sam Fourman Jr.
sfourman at gmail.com
Sat Aug 24 10:30:26 UTC 2013
> If the 150 ports that only work with gcc, all work with a ports
> > gcc and do not need the gcc from base, would the following be OK ?
> >
> > - 9.x gcc default and clang in base;
> > - 10.x clang default and gcc in ports;
>
> Well, we write rules and we brake them. ;-)
>
> Just say that we know we brake them but it's inevitable because...
> And go futher.
>
I am not a developer, just a user, so I am not versed in all of the
issues but I
would REALLY like to see gcc moved to ports for 10.x
In my opinion this just needs to happen, if ports break, we deal with that
on a case by case basis.
FreeBSD as a community made the decision to move to clang as a compiler, and
moving gcc to ports enforces that decision, I prefer the "rip the band aid
off" approach
because it brings issues to light faster, and now people have real reasons
to fix things.
Now, I am aware that other architectures like ARM etc. need gcc in base for
basic things
like building kernel/world, because clang cant do this yet.
Maybe this is over simplifying it a bit but can't we just modify scripts in
some way
to pull gcc from ports into base, for these platforms at build time? SVN
*is* in base now (svnlite)
>From an outside look at this, it seems to me that we're holding back the
amd64 platform
just because the developer activity is a little more sparse than we would
prefer on other platforms.
Other platforms are important and they are needed, but those platforms are
the ones that
need patched up, they are the ones that need the band-aids implemented so
that gcc still works
for them.
So I vote, let's not give ourselves the burden of "lugging" dead weight in
base
for another 5 years. (in 2017 do we still want to be worrying about gcc in
base?)
So in the name of progress, let's make a comfortable final resting place
for gcc in our ports tree
and look to clang for our future.
Thoughts,
Sam Fourman Jr.
More information about the freebsd-current
mailing list