> Do you think D-Bus is better solution for async communication between process > of > different user levels? I'm not very familiar to dbus yet. But I can change > the IPC > interface easily.
Tying it completely into dbus/hal would be neat. When you plug in the N810 when in flash mode, hal would pick up the USB event from the kernel and automatically run an 0xffff daemon. That daemon would connect to the device (it'd have to be told which one since you can have multiple in theory, or it just is single-instance and handles all) and connects to dbus as well. Then, it broadcasts some sort of "device available" message that the GUI flasher could catch to display the device in the list. Of course, the flasher daemon needs to take queries too for when the gui has just started up. If you really want to make it independent of the device used (Nokia, Neo, HTC, ...) then you should probably write a d-bus/hal spec for this that flashers can use to define everything the gui needs... Overall though, it sounds like a lot of work. I'm quite content using the command line and, while I'm writing this anyway, thanks for that :) Oh and finally a shameless plug: I'm working on figuring out the wireless chip so we can run a mac80211-based driver on the device, I have already written an SPI frontend for the p54 driver but the p54 common code can't handle this chip yet. Help would be welcome, contact me privately. Links, so far, are: * http://cx3110x.sipsolutions.net/ * http://johannes.sipsolutions.net/N810 * http://article.gmane.org/gmane.linux.kernel.wireless.general/11997 johannes
signature.asc
Description: This is a digitally signed message part
_______________________________________________ 0xffff mailing list 0xffff@lists.nopcode.org https://lists.nopcode.org/mailman/listinfo/0xffff
