Antonio Huete updated psm(4) using code from FreeBSD; I don’t think it’s been committed yet but the patch is available. This will be especially valuable to you if you have a synaptics touchpad; it enables many of the functions.
The dragonflybsd.org website is a wiki (ikiwiki, to be exact) and any page on it can be edited. You have to create an account, first, which only requires picking a username and password.
However! If you have an account on leaf.dragonflybsd.org where it is hosted, you can check it out via git, edit any number of pages using your favorite editor, and commit it back and the pages will automatically rebuild. The commit will even show up like any other change.
Pratyush Kshirsagar has put together a website describing his proportional RSS work for DragonFly. He may have more content in the future.
DragonFly 2.8 has been branched, with the release anticipated for the end of this month. I’m building packages right now for it, so we should have binary packages ready at release. A convenient git-summarized changelog exists to track differences since 2.7.3, at least.
Venkatesh Srinivas has a large writeup describing just how the memory allocator in DragonFly changed from 2.6 to the upcoming 2.8 version. In all that text, you may notice the cheering statistic that it gave a 20% improvement in sysbench results.
Something for everyone this week.
- Via sjg/IRC: The next platforms for DragonFly: Dragon 32 and Dragon 64.
- chmod -x chmod: a slideshow of possible solutions. (via) As ‘blinkkin’ pointed out on IRC: “hammer history /bin/chmod” and “cp /bin/chmod@@0xtransaction_id” would also fix it.
- 328 slides of git-wrangling tips (also via)
- How to pretend to be busy. I wish I had time for this.
- The first MUD, as a solution for class conflict, and not the fighter-vs-mage-vs-cleric type. (via)
Siju George found no equivalent of OpenBSD’s ‘afterboot‘ quick-start page in DragonFly, so he went and created it himself. Go, read.
YONETANI Tomokazu wrote out a nice explanation of acpi(4) and the myriad ACPI subsystems which can be enabled or disabled at boot time. If you do have booting problems, it’s usually ACPI, and it’s usually only one small part. Finding that small part is easier with this list.
Rumko came up with some changes for vkernel(7) that, among other things, made it possible to run them diskless; almost like booting a thin workstation.
Found via a random Google search: SSHGuard. It’s not available in pkgsrc, but it’s in other BSD packaging systems, and it lists DragonFly on the site as a possible host. It monitors various services and blocks access to overly aggressive connections using (on DragonFly) pf. This is similar to scripts discussed here in the past. It also may be useful in light of the recent FTPd problem.
OpenSSL on DragonFly can now dynamically load engines (cryptographic support modules) at runtime, thanks to Peter Avalos.
Something that always got with with Linux binary support was that I couldn’t get the Linux /proc filesystem to automatically mount on boot. I’d end up doing it by hand later, right after I tried to start a Linux binary and had all sorts of issues. Pierre Abbat had this same problem, and Sascha Wildner has the answer: “linux_load=yes” in /boot/loader.conf.
If you were thinking of buying a Western Digital Passport USB drive, it’s supported on DragonFly, thanks to Dylan Reinhold and Alex Hornung.
For those using the release version of DragonFly, the new C-based loader in 2.8 will look like this. Well, not exactly. This is from a proposal from Alex Hornung that removes some extra lines, but I expect this is what you’ll see.
The newest issue of BSD Magazine is all about VPNs and BSD. It’s free to read in PDF form.
The 200th (yay!) episode of BSDTalk has 14 minutes of conversation with Kjell Wooding, talking about mg, a sort of teeny emacs included with OpenBSD.
Matt Dillon and Venkatesh Srinivas conspired to fix another nmalloc issue, which should resolve any remaining problems people were having with Firefox, and possibly other applications as well. Due to an oversight of sorts, all locking operations on nmalloc’s depot were ineffective, as if there were no locking at all. Curiously, it worked remarkably well considering such a large race condition was present.
Here’s something: Pratyush Kshirsagar came along, saw the proportional RSS project idea, and did it. It’s nice to have a completed project just sort of fall out of the sky.