| Sarvatt | http://paste.ubuntu.com/429194/ | 00:00 |
|---|---|---|
| Sarvatt | something like that as your xorg.conf | 00:00 |
| Sarvatt | palm detect does some screwy things on a few of my laptops though, might need to mess with PalmMinWidth and PalmMinZ values | 00:06 |
| Sarvatt | does it not stick when you enable palm detect in the touchpad properties in gnome? | 00:06 |
| Sarvatt | or was that a gpointing-device-settings option.. | 00:06 |
| Sarvatt | ah yeah it is, bdmurray try just installing gpointing-device-settings | 00:07 |
| Sarvatt | has a bunch more touchpad options in it | 00:07 |
| aarcane_ | so, a day or two ago there was a big poulsbo-build-fest going on in here, while I was away.. what ever came of that ? | 01:43 |
| RAOF | aarcane_: That looks to have ended up in a merry X server crash. | 01:45 |
| aarcane_ | RAOF, I figured it would have culminated in many of them to date. I'm wondering more if it's making any appreciable forward progress, as I've got a nice system with a poulsbo graphics chipset :( | 01:47 |
| * RAOF suggests a rethinking of the calibration for “nice system” :( | 01:47 | |
| RAOF | But there appears to be significant forward progress. | 01:48 |
| aarcane_ | RAOF, if you want to know why I say nice, google up the T91SA | 01:49 |
| aarcane_ | RAOF, anyway, I'm glad to hear that the progress is forward. | 01:49 |
| libv | bjsnider: i lost my job at suse, and while this did mean a bit of a career and personal setback, it now does allow me to hang out the dirty laundry properly | 04:53 |
| libv | bjsnider: this is just a fraction of the stories/crap i have on bridgman | 04:53 |
| libv | it is high time that he stops spewing his lies and bullshit and that he finally leaves free software the hell alone. | 04:54 |
| virtuald | libv: do you think you have some abillity to "sense" what other people are thinking? | 05:19 |
| Bernardo | good morning | 08:08 |
| lucazade | morning | 08:08 |
| RAOF | Aloha. | 08:08 |
| Bernardo | lucazade: did you see the launchpad group created by sloanua? | 08:09 |
| lucazade | yes | 08:09 |
| lucazade | he did everything by his own.. i believe | 08:09 |
| Bernardo | yes, that is what I found strange | 08:09 |
| lucazade | hehe | 08:10 |
| Bernardo | but we can always keep both repositories in sync... :) | 08:10 |
| lucazade | more is better | 08:11 |
| Bernardo | yep. I'll try to upload the latest changes I did to xpsb-glx | 08:11 |
| Bernardo | interesting that disabling all acceleration the driver is stable | 08:11 |
| lucazade | on kde is it stable? | 08:12 |
| Bernardo | I thought it was stable for you with AIGLX, Render and Composite disabled | 08:12 |
| lucazade | a little bit better... gdm is stable... desktop crash after a while | 08:13 |
| lucazade | icons are visible | 08:13 |
| lucazade | nothing more | 08:14 |
| Bernardo | ah | 08:14 |
| Bernardo | :( | 08:14 |
| Bernardo | well, noiw we have to start filling the code with debug statements | 08:15 |
| lucazade | i don't have a great experience in this | 08:15 |
| Bernardo | we have to start from our backtraces | 08:16 |
| Bernardo | see where the failing functions are being called | 08:16 |
| lucazade | ok | 08:16 |
| Bernardo | I was wondering if we can get a more detailed analisys qwith | 08:17 |
| Bernardo | with | 08:17 |
| Bernardo | with valgrind or even going full debug with gdb | 08:17 |
| lucazade | i really don't know | 08:17 |
| RAOF | What parts are open-source, and thus amenable to debugging? | 08:18 |
| Bernardo | most stuff, you have a binary firmware blob, and xpsb-glx has 2 binary libs | 08:19 |
| Bernardo | correct me if I'm wrong, lucazade | 08:20 |
| Bernardo | unfortunately I think the problem is most likely to be in those libs and the code around them. | 08:21 |
| RAOF | Well, disabling acceleration should keep you out of xpsb-glx. | 08:21 |
| lucazade | and it still crash | 08:21 |
| RAOF | Well, that should be debuggable. | 08:21 |
| lucazade | ** was looking for Imagination support (designed the gma500 HW) http://www.imgtec.com/forum/forum_posts.asp?TID=610 .. but is is all about intel | 08:23 |
| Bernardo | yep... should be | 08:23 |
| Bernardo | lucazade: can you register at that forum and ask for any help they can give on this effort to port the drivers? | 08:25 |
| lucazade | i was going to do (even with my childish english) | 08:25 |
| Bernardo | I'm going to try to follow this guide - http://www.x.org/wiki/Development/Documentation/ServerDebugging#Debuggingwithonemachine | 08:33 |
| Bernardo | have to reboot into lucid, bbl | 08:33 |
| lucazade | Bernardo switching migrationheuristic from greedy to smart i get another segfault | 09:19 |
| lucazade | i knew MigrationHeuristic to greedy caused icon corruption | 09:20 |
| lucazade | bryce miCopyRegion was a common issue during lucid development? | 09:36 |
| lucazade | bryceh miCopyRegion was a common issue during lucid development? | 09:37 |
| RAOF | I recall that being associated with a bug. | 09:37 |
| lucazade | http://www.mail-archive.com/ubuntu-bugs@lists.ubuntu.com/msg2167716.html | 09:37 |
| lucazade | it the same backtrace on gma500 | 09:38 |
| lucazade | or really close to it (i have libfb.so instead of radeon_drv.so) | 09:39 |
| RAOF | Hm. That's fixed in a PPA, but should be SRU'd. | 09:43 |
| lucazade | Bernardo: selecting gnome emergency from gdm session type is far more stable | 10:34 |
| lucazade | 5min uptime! | 10:34 |
| lucazade | (i've also tried 2.6.31 on lucid and IgnoreABI but nothing changed) | 10:34 |
| lucazade | glxgears: 1249 frames in 5.0sec (like in karmic) | 10:55 |
| fkefer | is there a way to force X using a certain driver (vesa) when booting??? | 12:40 |
| fkefer | failsafeX does not work but freezes the machine | 12:40 |
| === Bernardo|away is now known as Bernardo | ||
| Bernardo | hi | 16:16 |
| === Bernardo is now known as Bernardo|away | ||
| === Bernardo|away is now known as Bernardo | ||
| Bernardo | I just tried bryceh ppa to see if the backtrace we get pointing to miDoCopy in psb was also fixed by it, but alas, it was not to be | 18:52 |
| * Bernardo isn't sure if it is running X with gdb that is awfully slow, or if the system has crashed already | 19:00 | |
| furan | are you ssh'd in for the gdb session? | 19:13 |
| furan | that's what I do | 19:13 |
| Bernardo | furan: yes, but I am still waiting for the desktop to show | 19:57 |
| Bernardo | What usually takes a couple of seconds had taken over one hour until now | 19:58 |
| furan | Bernardo: so break in and look at stacks? | 20:50 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!