Older blog entries for davidz (starting at number 3)


I've fixed a few bugs in the otherwise nice V4L patch from Matthew Mastracci, see here for a nice screenshot. Also got a patch from hadess for detecting read/write speed of optical drives - very neat. And I got around to integrating the firewire patch from Joe even though he says it's a bit cheesy still.

In other news I'm now using my fastest system (dual Athlon 1800, 512MB) as my main development box instead of my old Dell Inspiron 7500 (650 PIII, 128MB). Up until now, the fast box have just been collecting dust, more or less, the past 18 months, so about time. It's really a breeze, just sad it's not as portable :-)

elanthis : yeah, you need udev-012 (udev-013 isn't backwards compatible) and this file in /etc/dbus-1/system.d and then restart the bus. The string with an error in it, stems from the system bus instance, so it must be a bug in d-bus.

16 Jan 2004 (updated 16 Jan 2004 at 18:46 UTC) »
Thrown of the bus

Spent about six or seven hours finding out why the HAL daemon would suddenly exit with exit code 1. After building a new D-BUS RPM that respects the DBUS_VERBOSE environment variable, I found out that I was constructing garbage messages and the system bus daemon kicked me off which resulted in exit() being called somewhere in libdbus in my client (the HAL daemon). Hmm...

Hotplugging media

Yay, now HAL does the right thing when a user removes media without unmounting it - the code is quite ugly, polling and stuff, but it seems to work quite well at least for the limited set of devices (Compact Flash, CD and DVD discs) that I can test with.

Applications can now catch a D-BUS event saying that HAL unmounted, lazy-style even, the filesystems on the unplugged media.

elanthis : The HAL RPM that rml built was based on a 0.2.4 tarball from me. You should try the latest 0.2.5 here.


I'm currently at the curling rink between games in Danish Open - 802.11b is indeed pervasive. Had some coworkers from Jerusalem visiting this week; went out for beer wednesday; nice.

13 Jan 2004 (updated 13 Jan 2004 at 23:57 UTC) »

Yesterday I got myself a new card reader that supports removable storage (my old one simply powered down when removing the CF card), so now HAL supports media detection for both SCSI and IDE devices. Currently it's through polling (like magicdev), because of how Linux 2.6 currently works.

Today, rml put gnome-volume-manager into GNOME CVS, which I think is a pretty big deal for the HAL project - one implication is that I actually have to begin to care about API/ABI stability. Things are definitely coming into shape for hardware management on GNOME.

Been in meetings most of today at work; good to be home punching the keyboard. Started blogging through gnome-blog; neat.

Created an account for myself so I can post random notes about HAL and other stuff; neat, now I only need to find out how to mirror this to my existing weblog. Hmm..

Work on HAL is progressing nicely; work on adding I2C devices is progressing; just need to get my other system up to integrate the patch. Have exchanged a bunch of mails with rml about HAL and the soon-to-be-released gnome-volume-manager. Hope to finalize volume detection soon, at least for optical discs.. There was also an interesting thread on linux-hotplug-devel regarding media detection.

Still got two ISO's of FreeBSD 5.2-RC2 in my homedir waiting to be burned and installed.

New Advogato Features

New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.

Keep up with the latest Advogato features by reading the Advogato status blog.

If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!