On Thu, Apr 03, 2003 at 03:44:28PM +0200, Sven Luther wrote:
> On Thu, Apr 03, 2003 at 11:21:36PM +1000, Daniel Stone wrote:
> >
> > Porting - m68k, hppa, mips, mipsel, {free,net}bsd-*. Merging Michel
> > Daenzer's radeon-ddc patch - it doesn't apply at all cleanly.
>
> I think Mark (from upstream) will be looking at Michel's patches, and it
> will no more be needed at upstream. Why does Michel's patch not apply
> cleanly ? Because he diffed them against his DRI build or something such ?
>
> BTW, the *BSD ports should work relatively ok, should they not, many of
> the X upstream use *BSD as developpment plateform.
One would think so. One would be... er... not entirely correct.
More usefully, the *code* more or less builds just fine (mostly).
However, the assumptions made are based on a tightly-coupled-to-kernel,
specific-to-that-BSD userland, *not* on a Debian GNU-ish userland; changing
this involves quite a bit of hunting down arcane Imake information, and
a few minor patches to teach the system to know when to use which set of
values.
For NetBSD, 4.(er... 2? I think?) this has been done, and it more or less
works; I have running X stuff on my not-a-buildd box, inside the chroot.
I haven't even *tried* putting a console on it, and for all I know it may
blow up merrily at that point.
Unless someone else volunteers, I intend to work with Mr. Stone (and
whomever else may be necessary) to make sure that 4.3 doesn't regress on
this. Unfortunately, at the moment the port is having issues with some bits
of toolchain sanity that's filed upstream, and we can't do a whole lot
until that's fixed (or at least, doing it is relatively pointless, unless
it's a new patchset, since it won't produce useable results).
--
Joel Baker <fenton@debian.org>
Attachment:
pgp9L_S8e8o9I.pgp
Description: PGP signature