Re: non-Linux OSes

[Ryan Lortie <desrt desrt ca>]

hi

On Tue, Oct 22, 2013, at 6:11, Alexander Larsson wrote:
> By Darwin you mean OSX in general I assume?

yes.  I guess we need Cocoa here too in order to really be useful.

> I think the above list is a good start. But it is not good enough. We
> also need to specify which versions of the above OSes, and which
> versions of the compilers (especially important for e.g. msvc).

For MSVC I agree.  For the others, I don't think it's that big of a
problem (although it could become a nice opportunity).

As a result of our backcompat with world+dog, we've been pretty
conservative about what we require.  If we add a feature that appears in
new versions of Linux, for example, we always add the back-compat case
for both older Linux and other non-Linux systems.  I expect we'd pretty
much continue this way.

What would be nice is in the case of things that have long been in
POSIX, we could "end of life" old OSes and drop the backcompat hacks,
relying on the new feature.

> What about android? It runs "linux", but its sufficiently different than
> desktop linux to require its own testing.
>
> Also, what about iOS? As per e.g.
> http://stackoverflow.com/questions/5287960/compiling-glib-for-iphone:
>
>         Yes it can be built, (see:
>         http://stackoverflow.com/a/17733328/1856278) and there are
>         several Apps in the Appstore that uses it.

These were obvious omissions.  Thanks.

> llvm?

We've been mostly treating llvm as GCC for now and that's been working
mostly OK for us.  We occasionally see issues but they're almost always
clang bugs (of course, doesn't mean that we don't have to work around
these issues and that we shouldn't test...).

> It seems to me that it should be possible to have virtual instances of
> at least Linux (of various kinds), *Bsd, OpenSolaris, and Hurd running
> in some VM farm setup with "just" some work. But ideally we should be
> running copies of the non-free OSes too. 

This is more work than you think.  Not every free software operating
system has come the same distance that Linux distros have in terms of
ease of installation and none of them are as familiar to us as any Linux
system would be.  I'm not the only one who has tried to do this
piecemeal in the past and failed -- to even get the system up to the
"build deps installed" point is not trivial.  After that, keeping it
up-to-date would be an ongoing challenge, multiplied by 10 (for the
number of VMs we have, each totally different in its upgrade progress
and schedule).

I think we need outside help.

> Which these are virtualizable on x86? I know OSX is virtualizable, but
> there are licensing issues on non-apple hardware.

New copyright law in Canada creates explicit exceptions for this sort of
thing, so this is legal, in Canada at least. :)

> Solaris is availible
> on X86, maybe we can ask oracle for a few licenses for this (of course,
> ideally we should be testing on sparc hw too).

I think there are a couple of old (but beefy) sparc machines floating
around that some GNOME hackers received as gifts from Sun back in the
day...

Note: I just discovered that OpenSolaris lives on... Their homepage even
talks about how they're GNOME-based: http://openindiana.org/

Probably there will be a growing delta here with time, though.

> HP-UX is itanium and
> pa-risc only i believe, AIX is PPC/s390 only. Maybe we could score a
> power machine that can run VMs from somewhere, but s390/pa-risc/itanium
> hardware sounds unlikely, maybe we can soft-emulate them?.

Or we could exercise the other angle in this case and try to find some
people who would give us access to their existing systems (or set up VMs
for us, and maintain them).


Cheers