[00:29] <pwnguin> is there a reason bug #137686 isn't marked fix released?
[11:18] <elmargol> Hibernation almost works here. It terminates everything and copies the ram to the sawp. The only problem is the laptop (dell inspiron 9400) does not power off. I have to hold the power key in order to do this. Resuming works perfect!
[11:18] <elmargol> Is there a way to fix this?
[13:18] <abogani> tjaalton: Hi Timo! Are you around?
[13:19] <laga> elmargol: have you seen the freedesktop.org page about power management?
[13:19] <tjaalton> abogani: hey, seems that I am ;)
[13:20] <abogani> tjaalton: Sorry to bother you about lrm... :-)
[13:20] <abogani> Is there a chance to have nvidia closed drivers updated to 173.08 (actually marked by Nvidia as Beta) in 8.04.1 ?
[13:20] <tjaalton> abogani: doubtful, but there should be a stable version soonish
[13:21] <tjaalton> that version has some pretty severe bugs
[13:21] <abogani> :-(
[13:21] <laga> elmargol: http://people.freedesktop.org/~hughsient/quirk/quirk-suspend-index.html
[13:22] <tjaalton> abogani: phoronix had something about it, let me check
[13:28] <tjaalton> abogani: ok here it is: http://www.phoronix.com/scan.php?page=news_item&px=NjQzMA
[13:29] <tjaalton> 2D slowdowns
[13:30] <abogani> gosh!
[13:31] <tjaalton> hm, I really should update the lrm for intrepid
[13:31] <mjg59> laga: That's not usefully relevant in this case
[13:31] <laga> mjg59: it was worth a try. :)
[13:32] <abogani> tjaalton: Some -rt users report problem with nvidia stable version and GLX that stop to works! :-(
[13:32] <tjaalton> abogani: yep..
[13:51] <alex_joni> hi abogani
[13:52] <abogani> alex_joni: Hi Alexandru
[13:53] <alex_joni> abogani: I talked to BenC about pushing -rtai stuff to ubuntu
[13:53] <alex_joni> and I don't think it's important enough (enough users) that it's worth the effort
[13:55] <abogani> alex_joni: Is it your opinion or the BenC one?
[13:56] <alex_joni> abogani: well.. I'm not sure if I can support it once it's pushed into ubuntu
[13:56] <alex_joni> (building/testing on each new release, etc)
[13:58] <abogani> I understand well your doubts.
[13:58] <alex_joni> abogani: anyways, I have a repo for the -rtai stuff
[13:58] <alex_joni> so if you're interested, feel free to use it
[13:59] <abogani> Thanks! It is very good news! 
[13:59] <abogani> URI?
[14:00] <alex_joni> abogani: http://www.linuxcnc.org/hardy/
[14:01] <abogani> Great!
[14:02] <SEJeff> alex_joni, Do you use rtai for robotics? How does the latency compare to a mainline kernel if you don't mind my asking?
[14:02] <alex_joni> SEJeff: we get about 10-20 usec latency on rtai
[14:02] <SEJeff> alex_joni, As opposed to?
[14:02] <alex_joni> the -rt is still quite away from that afaik
[14:03] <SEJeff> Yeah I've used Xenomai, but it is more about pretty code than low latency
[14:03] <alex_joni> Xenomai shouldn't be much worse than rtai
[14:03] <alex_joni> but it depends on the skin you use
[14:03] <abogani> ~10 times more
[14:04] <alex_joni> abogani: you mean -rt .. right?
[14:05] <abogani> Yes. In any case with CPU Isolating (or called CPU Shielding) you can achieve very similar performance.
[14:05] <SEJeff> alex_joni, Well which personality do you suggest for best case low latency
[14:06] <alex_joni> SEJeff: I didn't try xenomai enough to say
[14:06] <alex_joni> but the best latency I've ever seen was with RTAI on a SMP intel system
[14:06] <abogani> I always use POSIX skin.
[14:06] <alex_joni> with isolcpus, and RTAI only on one core
[14:06] <alex_joni> on one system it had something around 4-5 usec latency
[14:06] <SEJeff> alex_joni, latency for network or applications? Intel nics with e1000 have the lowest latency on the market
[14:07] <alex_joni> SEJeff: I'm only talking about realtime apps
[14:07] <elmargol> laga, thx for the link
[14:07] <SEJeff> alex_joni, Ah, I work at a stock trading company as a Linux guy. Lots of -rt interest
[14:08] <alex_joni> SEJeff: for that, -rt might be better
[14:08] <SEJeff> alex_joni, -rt is higher latency, but more deterministic than the normal kernel
[14:08] <SEJeff> One of the reasons for the rtai / xenomai split is because rtai only cares about the lowest latency. It would be huge to have easier to use rtai kernels
[14:09] <alex_joni> SEJeff: might be a bit OT for #ubuntu-kernel.. shall we move to another channel?
[14:10] <SEJeff> alex_joni, Sure where?
[14:10] <alex_joni> #rt-and-stuff :)
[14:13] <xhaker> can someone help me with -server ?
[14:13] <xhaker> i can't get it to boot on a pentium M
[14:13] <xhaker> found a bug filed already
[14:14] <xhaker> bug #222253
[14:14] <xhaker> but there doesn't seem to be much attention
[14:15] <xhaker> rtg: ^
[14:28] <xhaker> subscribed ubuntu-kernel-server
[14:35] <xhaker> subscribed ubuntu-kernel-team also, u-k-s doesn't seem to be used for bugs :)
[14:36] <rtg> xhaker: does the -generic kernel boot?
[14:37] <rtg> I can't tell from the LP report what is being attempted, -server or -generic.
[14:40] <xhaker> rtg: it's server.. whatever ubuntu-server-i386.iso installs
[14:41] <rtg> xhaker: do you have one of the VIA C3 procs?
[14:41] <xhaker> and yes, -generic worked
[14:41] <xhaker> rtg: pentium M
[14:41] <rtg> xhaker: ah, thats what I wanted.
[14:41] <xhaker> it seems specific to -server
[14:42] <rtg> xhaker: the big diff might be PAE. I'll be back in a bit....
[14:42] <xhaker> rtg: ok
[14:59] <tjaalton> rtg: I'll do a bugfix lrm for hardy-proposed first, then try to get it to build against intrepid. is that ok?
[14:59] <tjaalton> (sorting out the libwfb mess & authatieventsd.sh)
[15:00] <rtg> tjaalton: thats fine. I'm not in a big hurry for Intrepid LRM since I'm kind of buried under SRUs at the moment.
[15:00] <tjaalton> rtg: heh, same here
[15:55] <xhaker> rtg: CONFIG_X86_PAE=y on -server only
[15:55] <rtg> xhaker: yes
[15:55] <xhaker> is there any boot option to disable it so i can see if it boots?
[15:56] <rtg> xhaker: no, its a feature of the page table that is hardwired into the mm.
[15:57] <rtg> xhaker: what is it about the -server kernel that is attractive for the VIA C3 cpu?
[15:57] <xhaker> rtg: people are using the server ISO to setup their home servers.. mine runs pentium M but people buy via c3 cpus for the low Wattage
[15:58] <xhaker> what other way to setup a lean and mean home server than to install ubuntu-server flavour?
[15:58] <rtg> xhaker: you can get the same result using the alternate install and choose 'command line only'. It uses the desktop kernel, but does not install X/
[16:00] <xhaker> maybe true, but not what a user would expect
[16:00] <xhaker> even and admin
[16:01] <rtg> xhaker: it seems to me to be a matter of RTFM. In any event, the PAE setting for the kernel isn't going to change.
[16:03] <xhaker> rtg: allow me to disagree.. as a user setting up a x86 home server, the server ISO and server kernel seem to be the right choice
[16:04] <JanC> maybe there should be a "home server" edition  ;)
[16:04] <xhaker> which cpus boot -server and which cpus don't? should the server ISO install -generic for those cpus? should there be a -server-pae kernel flavor?
[16:05] <xhaker> JanC: ^
[16:06] <rtg> xhaker: looks guys, this is open source. you are free to develop your own flavour. I've got enough on my plate managing the flavours we do support.
[16:07] <xhaker> rtg: hey man, i agree it's hard to support everything.
[16:08] <xhaker> rtg: but please direct me then to a document that explains what hardware does -server support
[16:09] <soren> xhaker: You have to accept that it's rather ironic that people expect the same kernel images for the *really* big servers, and the tiny, tiny media boxes or whatever.
[16:09] <soren> ...and then -generic for anything in between.
[16:10] <soren> That's hardly sensible.
[16:11] <xhaker> soren: i completely agree with you. that's not what i'm saying. i'm not a debian-installed guy, but couldn't it install -generic for the machines that won't even boot after install?
[16:11] <xhaker> soren: that would fix it.
[16:12] <rtg> xhaker: I'm not sure if there is a doc that describes the HW requirements of the -server flavour. However, it is codified in the kernel config, so you could certainly examine that.
[16:12] <xhaker> s/installed/installer
[16:12] <soren> xhaker: Probably, yeah.
[16:13] <xhaker> rtg: the features page states the -servel kernel is PAE enabled. so yeah, people that know their hardware is not compatible will stay away
[16:13] <rtg> soren: it seems like this issue has come up before for folks attempting to run the lice CD on a 486 CPU.
[16:13] <rtg> s/lice/live/
[16:14]  * soren eeks
[16:16] <xhaker> in the wiki ServerFaq page
[16:16] <xhaker> Do I have to choose the kernel for my system on my own?
[16:16] <xhaker> The installer is capable of recognizing your cpu and install the "best" kernel for it. 
[16:18] <xhaker> help.ubuntu.com rather
[16:18] <rtg> xhaker: is the page about dapper server? https://help.ubuntu.com/community/ServerFaq
[16:20] <xhaker> was linked from http://www.ubuntu.com/products/whatisubuntu/serveredition/documentation
[16:22] <xhaker> the system requirement https://help.ubuntu.com/8.04/serverguide/C/preparing-to-install.html#system-requirements
[16:25] <rtg> xhaker: I don't see anything in there that allows you to choose a kernel.
[16:27] <xhaker> rtg: what?
[16:31] <rtg> xhaker: " The installer is capable of recognizing your cpu and install the "best" kernel for it." Isn't that what you were looking for?
[16:32] <xhaker> After seeing that on the dapper ServerFaq, i was searching for something to contradict that on the Hardy documentation
[16:34] <xhaker> did not find any
[16:37] <rtg> xhaker: that looks like a feature that has been removed (at least on the server install). I haven't played with the i386 alternate install, but it closely resembles the Debian installer so the feature may still be there.
[16:38] <xhaker> it's just that i've just wasted 500mb+ of bandwidth, time: downloading, burning, installing. And now i have a non working system. And might have to download the alternate i386 install
[16:39] <xhaker> rtg: does the -server iso ship other images other than -server and -virtual?
[16:40] <rtg> xhaker: not to my knowledge, but I'm not an expert on the installer stuff.
[16:44] <xhaker> https://bugs.edge.launchpad.net/ubuntu/+source/linux-source-2.6.22/+bug/151942 marked invalid, not reassigned
[16:49] <xhaker> wikipedia to the rescue: PAE is provided by Intel Pentium Pro and above CPUs (including all later Pentium-series processors except the 400 MHz bus versions of the Pentium M), as well as by some compatible processors such as Athlon and later models from AMD.
[16:50] <xhaker> i was affraid celerons would be bit by this too, but it seems they're not
[16:54] <xhaker> let me go try install generic on my server then
[16:54] <xhaker> :D
[17:06] <smb> Hi all, this would be the time of the kernel-team irc meeting, sort of... But currently most of us somhow busy. So maybe just a reminder to put stuff that should be thought of for intrepid onto the wiki.
[17:07] <amitk_> link to wiki for listing UDS topics: https://wiki.ubuntu.com/KernelTeam/UDS/May2008
[17:07] <smb> Otherwise, are there any questions?
[17:09] <smb> Anyways, we hang around here, in any case. So, back to work it is. ;-)
[19:24] <noelferreira> my keyboard keys sometimes get stuck and other times don't work. Anyone knows about this huge bug : http://pastebin.com/m78e61aa4
[19:32] <noelferreira> my keyboard keys sometimes get stuck and other times don't work. Anyone knows about this huge bug : http://pastebin.com/m78e61aa4
[19:33] <laga> noelferreira: don't repeat yourself like that, it won't help :)
[19:33] <laga> try reporting your bug in laucnhad
[19:33] <laga> and talk to people in #ubuntu-bugs maybe
[19:33] <laga> if nobody answers in here
[19:34] <noelferreira> thanks
[20:28] <psufan> hi we are running ubuntu-server 6.06 lts on some blades hosting vm's and I need to patch against the 2.6.15 nfs crash and burn bug
[20:28] <psufan> what are the standard steps to get the source with buntu's existing config options
[20:29] <psufan> is it still the old patch --verbose < file.patch thing and 
[20:29] <alex_joni> psufan: apt-get source linux-image-generic
[20:29] <psufan> how do I compile etc so I can put the image on the other 9 blades
[20:29] <alex_joni> then look in debian/ for ubuntu patches
[20:29] <alex_joni> and put your patch there
[20:29] <psufan> no way just to get the ubuntu sources
[20:29] <alex_joni> the config is also there
[20:29] <psufan> with the existing .config
[20:29] <alex_joni> maybe apt-get install linux-source-'uname -r'
[20:30] <psufan> hmm
[20:30] <psufan> yeah that might be it
[20:32] <psufan> question
[20:33] <psufan> if this http://www.howtoforge.com/kernel_compilation_ubuntu
[20:33] <psufan> is the right way to do it
[20:33] <psufan> how do I ensure it's what is in the existig 2.6.15-57 package
[20:33] <psufan> I wish you guys would have just patched this in 6.06
[20:34] <psufan> :/
[20:38] <psufan> err sorry 6.10
[20:39] <psufan> what is this git stuff
[20:39] <rtg> psufan: start here: https://wiki.ubuntu.com/KernelMaintenance
[20:40] <psufan> yeah yeah I saw the standard party line "blah blah we don't blah"
[20:40] <psufan> oh my bad
[20:40] <psufan> not the same thing this time :)
[20:40] <noelferreira> thanks
[20:41] <noelferreira> my keyboard keys sometimes get stuck and other times don't work. Anyone knows about this huge bug : http://pastebin.com/m78e61aa4
[20:41] <psufan> wow this is confusing as hell and I remeber hacking a 1.1.x kernel to run on 1.6 meg of ram
[20:42] <psufan> ugh all I want to do is apply a nfs fix to the existing 2.6.15-51 source
[20:45] <rtg> psufan: your can do it in 3 easy steps: get the source first using 'git clone git://kernel.ubuntu.com/ubuntu/ubuntu-dapper.git  ubuntu-dapper'
[20:46] <rtg> install the build tools 'sudo apt-get install linux-kernel-devel devscripts fakeroot libncurses5-dev'
[20:46] <rtg> do the build: 'cd ubuntu-dapper; debuild -b'
[20:49] <psufan> oh shit
[20:49] <psufan> looks like when that pxe boot vm that setup a automated install of ubuntu 6 it screw some stuff up
[20:49] <psufan> apt-get is getting stuff from dapper when it's edgy
[20:49] <psufan> ugh
[20:50] <psufan> nm
[20:50] <psufan> we are using 6.06lts sevrer
[20:50] <psufan> damn you people and you stupid names
[20:50] <psufan> just give me freaking version 1 etc
[21:03] <pwnguin> psufan: you can always try debian stable. that never changes on ya ;)
[21:17] <psufan> na it's set in stone now I can't change it
[21:22] <noelferreira> my keyboard keys sometimes get stuck and other times don't work. Anyone knows about this huge bug : http://pastebin.com/m78e61aa4
[22:01] <noelferreira> my keyboard keys sometimes get stuck and other times don't work. Anyone knows about this huge bug : http://pastebin.com/m78e61aa4
[22:17] <blueyed> Can somebody please look at bug 217504, which has a simple fix.
[22:17] <blueyed> ?
[22:17] <blueyed> https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bug/217504