I’ve set up a Bugzilla bug tracker at forknibbler.com, to use until a full-blown system is available at dragonflybsd.org. Be kind – it’s a woefully underpowered machine.
Leaf.dragonflybsd.org has been significantly upgraded; this is where shell accounts and the mail archive are located.
Joerg Sonnenberger has committed a large update, written by Andreas Hauser, to the port overrides for Xorg. It’s much more likely to work now, if you are looking to switch from XFree86.
“Micha³ ‘GiM’ Spadliñski” posted a link to a 80×15 DragonFly banner.
If you’re feeling religious: a petition for Flash on BSD.
The FreeBSD NVIDIA driver has been updated; Emiel Kollof has a test DragonFly version of this new update he’d like people to try out.
MAtthew Dillon warned that the ongoing VFS work will make the use of buildkernel/installkernel absolutely necessary for kernel installation, for anyone using filesystems other than those built into the kernel.
A discussion about the planned and ongoing VFS changes has led to some very informative posts (that’s 3 different links) from Matthew Dillon about what he plans to do.
The Center for Internet Security has released a “FreeBSD Benchmark and Scoring Tool“. This may also work on DragonFly.
Joerg Sonnenberger has committed the first half of Simon ‘corecode’ Schubert’s work on improving the IOCTL code, making support for (for instance) Linux calls much easier. Joerg also committed a change to module loading, suggested by Simon, which may allow modules such as AGP to load when otherwise they would not.
Matthew Dillon posted about the start of work on the new VFS subsystem.
Eirik Nygaard has added message-passed syscalls into the system. They aren’t running for all users by default, but they will be once proved stable.
dragonflybsd.org will be going though some upgrades soon, as Matthew Dillon noted in a recent post. Notably, leaf.dragonflybsd.org is going to be an AMD64 machine.
The qt33 override is gone; TrollTech has incorporated Joerg Sonnenberger’s patches.
dragonflybsd.org is now back on a T1, with a slightly better connection. So, now you don’t have to use the mirrors, though you should.
Matthew Dillon mentioned something of how well programming is going, and how close we are to a major step in the networking cleanup.
If your sound card previously did not work, a recent fix suggested by Barry Bouwsma and committed by Matthew Dillon may fix it. If you have a SF64-PCR sound card, a change in make.conf will be needed, as described in the link above.
There’s a review of DragonFly 1.0A over at Newsforge by Jem Matzan of The Jem Report. It has valid points (Using the ports system is a hack) and some invalid ones (The reviewer wouldn’t benchmark with SMP off.)
Joerg Sonnenberger has a x.org port that uses the same layout as the FreeBSD port, available for testing.
Adam Kirchoff noted his computer would have trouble when he hooked up a USB2 CDROM; Matthew Dillon followed up by pointing out EHCI (the subsystem that deals with USB2) is somewhat troublesome, though UHCI (USB1) should work fine. (I’m linking to the start and end of the thread.)