Re: Bug#598528: unstable on GNU/kFreeBSD
- To: email@example.com
- Subject: Re: Bug#598528: unstable on GNU/kFreeBSD
- From: Robert Millan <firstname.lastname@example.org>
- Date: Sat, 2 Oct 2010 13:37:10 +0200
- Message-id: <AANLkTikomAOnyTB+=T-aewtXy7DZy=_W+kHCwPs0SvYE@mail.gmail.com>
- In-reply-to: <20100930144319.GA21120@gaara.hadrons.org>
- References: <20100929182004.1429.77659.reportbug@thorin> <20100929185513.GH23663@sym.noone.org> <AANLkTikbdBjrbSV1=NWr6CURYhXCKoMUstgGQXunFqeW@mail.gmail.com> <20100930144319.GA21120@gaara.hadrons.org>
2010/9/30, Guillem Jover <email@example.com>:
> The real solution long term is to get KMS support in kFreeBSD, forking
> the UMS drivers (because at some point others will start dropping UMS
> support too) is just going to be painful.
> Robert Noland seems to have been working on adding GEM/TTM/KMS to
> FreeBSD. There's also been a GSoC this year to add support for that to
> DragonFly BSD (http://www.dragonflybsd.org/docs/developer/GEMdrmKMS/).
This name rings a bell. First hit in google "freebsd kms" is a message
from him saying (in reference to KMS) that he started a new job which
has cut his time for FreeBSD almost entirely.
As for the GSoC project for DragonFly, which if successful would presumably
be ported to FreeBSD after some time, it's unclear whether the "correct"
solution will arrive soon. My gut feeling is that at the very least
it'll take a
In any case, painful or not, in the meantime FreeBSD folks (including Noland)
are maintaining a forked version of xf86-video-intel:
I gave it a quick look and it doesn't seem to host any actual driver
development, but at least they keep it in sync with newer X server API.
I severely lack time myself, and actually I no longer even have this intel
board, but if someone has time and interest my recommendation would
be to approach FreeBSD folks (specially miwi and rnoland) and try to
setup a common upstream for maintaining their xf86-video-intel,
preferably at freedesktop if they would accept it.