[01:24] <syldeb35> does someone know Bug #316405
[01:24] <ubot3> Malone bug 316405 in linux "Kernel OOPS during initialization SAA7134 jaunty kernel 2.6.28-4.generic" [Undecided,New] https://launchpad.net/bugs/316405
[01:25] <JesperHansen> Is this the right channel to ask about broken/missing frequency scaling?
[01:29] <Kano> rtg: could you update ndiswrapper to 1.54?
[01:33] <johanbr> JesperHansen: sure
[01:37] <JesperHansen> ok, I have a eeePC 900 with this readout http://pastebin.mozilla.org/620365
[01:40] <JesperHansen> only prob. is it stays at 900 MHz after I replaced the default eeePC linux distro with a xubuntu
[01:41] <JesperHansen> I am looking at the eeepc-acpi-scripts package, but
[01:43] <JesperHansen> its setting some impossible requirements
[01:44] <JesperHansen> Its wants to remove acpi-support, powermanagement-interface, ubuntu-desktop, xubuntu-desktop
[01:46] <JesperHansen> this is also thrown: http://pastebin.mozilla.org/620369
[01:46] <johanbr> It doesn't seem to support standard SpeedStep.
[01:46] <johanbr> You might need a custom kernel module.
[01:47] <johanbr> And for the second error, looks like you're mixing repos from different sources.
[01:48] <JesperHansen> yea, I am considering what I am missing though, since the kernel I am running right now is a custom build. Neither this nor ubuntu's supported it (is a 2.6.28 kernel on 9.04)
[01:48] <IntuitiveNipple> JesperHansen: What does this report: for node in /sys/devices/system/cpu/cpu0/cpufreq/*; do if [ -f $node ]; then echo "$node $(cat $node)"; fi; done
[01:49] <JesperHansen> no, wait 8.10 with a few packages from 9.04 to make it happy
[01:50] <JesperHansen> IntuitiveNipple: nothing, /sys/devices/system/cpu/cpu0/cpufreq/ doesn't exist either.
[01:50] <IntuitiveNipple> well that's a big clue then :D
[01:52]  * JesperHansen doesn't get the clue :P
[01:52] <IntuitiveNipple> can you show the relevant settings from the .config the kernel was built with? (egrep 'FREQ|IDLE|GOV' .config)
[01:53] <IntuitiveNipple> JesperHansen: if that node doesn't exist there's no frequency-scaling governer installed, hence why it stays at 900Mhz
[01:56] <JesperHansen> just gotta figure out which of the two folders I compiled to the current kernel
[01:58] <IntuitiveNipple> did you set a custom version signature?
[01:58] <JesperHansen> http://pastebin.mozilla.org/620377 here
[02:03] <IntuitiveNipple> OK, can't see anything wrong there
[02:04] <johanbr> The processor in the eee doesn't support standard speedstep, so frequency scaling is not going work.
[02:08] <JesperHansen> johanbr: any ideas on how getting it to work?
[02:09] <johanbr> You may want to try these kernels: http://www.array.org/ubuntu/
[02:09] <JesperHansen> There's a package called eeepc-applet that things like in /proc/eee/.., but does nothing really. 
[02:10] <IntuitiveNipple> JesperHansen: http://kerneltrap.org/index.php?q=mailarchive/linux-kernel/2008/11/23/4311824/thread
[02:14] <JesperHansen> johanbr: possible to see those patches in 2.6.28?
[02:14] <JesperHansen> IntuitiveNipple: thanks, looking it through
[02:17] <johanbr> No idea. I'm not a kernel developer.
[02:21] <JesperHansen> weird, http://www.array.org/ubuntu/status.html?model=eeepc-900 lists that a module called asus_eee is in 2.6.27-7generic, but I cant see it
[02:28] <JesperHansen> well, thanks anyways. I'll look into it :)
[03:04] <JesperHansen> How long should git clone git://git.array.org/array/ubuntu-intrepid.git take? Its super duper slow on a eeePC
[03:05] <IntuitiveNipple> The transfer off a full kernel tree is about 350M
[04:41] <JesperHansen> Trying to compile a 2.6.28 kernel with the eeepc modules on that page
[04:56] <JesperHansen> Will be back with a report on it in... 7-8 hours when its complete :P
[05:30] <dholbach> hello my Kernel friends :)
[05:30] <JesperHansen> what do you think we are, modules :)?
[05:30] <dholbach> did anybody complain already that they just get a black screen when usplash is done with the new kernel (intel graphics, Lenovo X61s)?
[10:18] <dholbach> did anybody complain already that they just get a black screen when usplash is done with the new kernel (intel graphics, Lenovo X61s)?
[11:13] <apw> dholbach, which kernel?
[11:13] <apw> not heard anything from anyone as yet
[11:13] <dholbach> 2.6.28-7.20
[11:13] <dholbach> with 2.6.28-6.17 it's all good
[11:14] <apw> poop.  nothing of the sort reported
[11:14] <dholbach> I'll try again in a bit and see if anything changed
[11:14] <apw> i assume the machine is up?
[11:14] <apw> can you flip to other VT's
[11:14] <dholbach> let me do another update and see if anything fixed it
[11:14] <dholbach> apw: no
[11:14] <apw> is X server running ... etc
[11:14] <apw> so machine was proper dead.  hmmm
[11:15] <dholbach> apw: all black, if I logged in (though I couldn't see anything) the hd was doing some work
[11:15] <apw> ok ... so machine is up in some sense.  be good to get openssh-server up and installed
[11:15] <apw> sounds a lot like backlight is off, which is bad
[11:16] <dholbach> let me try again
[11:16]  * dholbach does an upgrade first
[11:17] <dholbach> brb
[11:19] <dholbach> apw: same happening now - I'm logged in via SSH
[11:19] <apw> worth trying to turn on the screen
[11:19] <apw> something like xset dpms force off
[11:20] <apw> then ... on
[11:20] <apw> also worth attaching to X and seeing what if anything it is doing
[11:20] <apw> if its in an EAGAIN loop from select/ioctl, then check the FD thats on (it may be DRI)
[11:20] <dholbach> erm
[11:21] <dholbach> if I do that, it turns off the screen on the machine I'm logging in from?
[11:21] <dholbach> (used ssh -X because of  xset:  unable to open display "" )
[11:23] <dholbach> /var/log/Xorg.20.log:(EE) [drm] Could not set DRM device bus ID.
[11:23] <dholbach> /var/log/Xorg.20.log:(EE) intel(0): [dri] DRIScreenInit failed. Disabling DRI.
[11:23] <dholbach> /var/log/Xorg.20.log:(EE) No input driver/identifier specified (ignoring)
[11:23] <dholbach> /var/log/Xorg.20.log:(EE) config/hal: NewInputDeviceRequest failed
[11:23] <dholbach> ^ is that something I should worry about?
[11:32] <dholbach> apw: ^
[11:33] <apw> dholbach, use DISPLAY=:0 xset ...
[11:33] <dholbach> ah
[11:33] <apw> isn't that .20 your ssh one?  ie the fake one for -X
[11:34] <apw> as in your DISPLAY is :20.0 or something over ssh _X
[11:34] <apw> -X
[11:34] <dholbach> daniel@miyazaki:~$  DISPLAY=:0 xset dpms force off
[11:34] <dholbach> No protocol specified
[11:34] <dholbach> xset:  unable to open display ":0"
[11:34] <dholbach> daniel@miyazaki:~$ 
[11:35] <apw> oddness
[11:35] <apw> try :0.0
[11:35] <apw> though i would expect it to mean the same
[11:36] <dholbach> same happening
[11:38] <apw> ok, then lets ocnfirm which log the X server is using
[11:38] <apw> fd 2 for it should be connected to the right log
[11:38] <apw> so ls -l /proc/<pid>/fd/2 should tell us that
[11:39] <apw> also what intel h/w do you have, my laptop here is intel
[11:39]  * apw goes download the kernel
[11:42] <dholbach> apw: you were right .20. was irrelevant - it was an old one
[11:42] <dholbach> sorry for confusion there
[11:42] <dholbach> apw: 00:02.0 VGA compatible controller: Intel Corporation Mobile GM965/GL960 Integrated Graphics Controller (rev 0c)
[11:44] <apw> whats the right log say, and what was it called?
[11:46] <dholbach> [    0.003639] (WW) The directory "/usr/share/fonts/X11/cyrillic" does not exist.
[11:46] <dholbach> [    0.201686] (WW) intel(0): libpciaccess reported 0 rom size, guessing 64kB
[11:46] <dholbach> [    0.909985] (WW) intel(0): Register 0x61200 (PP_STATUS) changed from 0xc0000008 to 0xd000000a
[11:46] <dholbach> [    0.910021] (WW) intel(0): PP_STATUS before: on, ready, sequencing idle
[11:46] <dholbach> [    0.910041] (WW) intel(0): PP_STATUS after: on, ready, sequencing on
[11:46] <dholbach> [    0.910070] (WW) intel(0): Register 0x71024 (PIPEBSTAT) changed from 0x00000206 to 0x00000246
[11:46] <dholbach> [    0.910113] (WW) intel(0): PIPEBSTAT before: status: VSYNC_INT_STATUS SVBLANK_INT_STATUS VBLANK_INT_STATUS
[11:46] <dholbach> [    0.910175] (WW) intel(0): PIPEBSTAT after: status: VSYNC_INT_STATUS LBLC_EVENT_STATUS SVBLANK_INT_STATUS VBLANK_INT_STATUS
[11:46] <dholbach> [    0.941704] (WW) intel(0): DRI2 requires UXA
[11:46] <dholbach> no errors at all
[11:46] <apw> ie what was its name
[11:47] <dholbach> /var/log/Xorg.0.log
[11:47] <apw> so it really should be :0
[11:49] <dholbach> blindly logging in on tty1 and typing 'xset dpms force off' does not seem to do much
[11:55] <apw> dholbach, and force on?
[11:55] <apw> can you tell if you have backlight?
[11:56] <apw> also if you think you are in and have a terminal
[11:56] <apw> echo "$DISPLAY">/tmp/FOO
[11:56] <apw> and see which terminal you are really on,
[11:58] <dholbach> apw: it creates /tmp/FOO for the blindly-logged-in user, but it's empty
[11:58] <dholbach> force on doesn't do anything either
[11:59]  * apw blinks
[12:00] <apw> what variables does env >/tmp/FOO show
[12:00] <apw> are there any GNOME ones in there?
[12:00] <apw> and indeed what does 'tty >/tmp/FOO' say
[12:01] <dholbach> apw: http://paste.ubuntu.com/116426/
[12:01] <dholbach> not sure there's anything exciting there
[12:01] <apw> and tty?
[12:02]  * apw bets its ttyN
[12:02] <apw> i think you are not logged in under X there
[12:02] <apw> i think you are in on the VT-1 console login (TERM=linux)
[12:03] <dholbach> apw: ah yes, sorry
[12:03] <dholbach> I should have said that
[12:03] <dholbach> I can blindly log into gnome and try to xset something there
[12:04] <apw> its $DISPLAY off the gnome thats interesting
[12:04] <apw> and xset there of course
[12:04] <apw> the $DISPLAY will tell us where we should talk to on ssh
[12:06] <dholbach> apw: from the SSH session I can xset now
[12:06] <dholbach> excuser-moi
[12:07] <dholbach> when I "off", the screen flickers shortly and goes from black to even-more black for a sec :)
[12:07] <dholbach> nothing in X log, syslog or .xsession-errors though
[12:08] <apw> so backlight is likely on then
[12:08] <apw> as it got blacker when turned off
[12:08] <apw> and X is working cause you can type
[12:08] <apw> what does xrandr say in ssh
[12:08] <dholbach> Screen 0: minimum 320 x 200, current 1024 x 768, maximum 1024 x 1024
[12:09] <dholbach> VGA disconnected (normal left inverted right x axis y axis)
[12:09] <dholbach> LVDS connected 1024x768+0+0 (normal left inverted right x axis y axis) 246mm x 185mm
[12:09] <dholbach>    1024x768       50.0 +   85.0     75.0     70.1     60.0*    40.0  
[12:09] <dholbach>    832x624        74.6  
[12:09] <dholbach>    800x600        85.1     72.2     75.0     60.3     56.2  
[12:09] <dholbach>    640x480        85.0     72.8     75.0     59.9  
[12:09] <dholbach>    720x400        85.0  
[12:09] <dholbach>    640x400        85.1  
[12:09] <dholbach>    640x350        85.1  
[12:09] <apw> is it that low res?  a t60?
[12:09] <dholbach> X61s
[12:09] <apw> oh baby one
[12:10] <dholbach> 1024x768 is correct
[12:10] <dholbach> yeah :)
[12:10] <apw> i think kirkland has one of those too, must get him to try also
[12:10] <apw> now dri was enabled wasn't it
[12:10] <apw> so lets try turning that off
[12:11] <apw> Option "NoDri"
[12:11] <apw> in the device section of xorg.conf, eliminate the kernel
[12:11] <apw> which we believe it is
[12:12] <dholbach> apw: no dice :-/
[12:12] <dholbach> apw: the strange thing is, I don't see a tty either
[12:12] <dholbach> when I stop gdm I still don't see anything
[12:12] <apw> very very strange
[12:13] <apw> if you disable GDM, fix the init script to prevent it starting X ?
[12:13] <dholbach> I could try booting with nosplash or however it's called
[12:13] <dholbach> I'll try gdm first
[12:14] <apw> its odd for sure
[12:16] <dholbach> apw: the same, even without gdm running on start
[12:16]  * dholbach tries booting without usplash
[12:16] <apw> good plan indeed
[12:16] <dholbach> how do I regenerate all initramfs-es? could it be something with those that's broken?
[12:17]  * dholbach has no clue whatsoever :)
[12:17] <tjaalton> just edit the grub cmdline
[12:17] <tjaalton> remove 'splash'
[12:17] <apw> yeah to turn that off is easy
[12:17] <dholbach> tjaalton: yep, just did it - let's see what happens
[12:17] <dholbach> I see boot messages! :-)
[12:17] <dholbach> I see GDM!
[12:17] <apw> if you reinstall the kernel package it will rebuild those
[12:17] <tjaalton> there you go :)
[12:18] <apw> so uspash is buggering something, hmm
[12:18] <mjg59> Are you in a KMS world for Intel yet?
[12:18] <tjaalton> no..
[12:18] <mjg59> Ok, so usplash will still be using vesa
[12:18] <tjaalton> so jaunty should be the last release with usplash
[12:18] <apw> tjaalton, what will we do instead?
[12:18] <mjg59> It ought to be falling back to text mode
[12:18] <tjaalton> apw: plymouth
[12:18] <mjg59> I mean, it ought to be switching back to text mode on exit
[12:19] <mjg59> dholbach: If you switch back to a text console and run usplash by hand, does it exit cleanly and give you a console again?
[12:19] <tjaalton> apw: it was discussed at UDS, but since KMS is still immature, it didn't make any sense to have it for jaunty
[12:20] <tjaalton> besides it needs .29 and only supports intel atm
[12:20] <dholbach> mjg59: no, leaves me in a black world again :-/
[12:21] <tjaalton> dholbach: hmm, maybe if you regenerate the -6 initramfs.. it might get broken too
[12:21] <tjaalton> usplash was updated recently
[12:21] <dholbach> mjg59: exit code is 0
[12:21] <tjaalton> blame kees :)
[12:22] <dholbach> tjaalton: just regenerated the intramfseseses
[12:22] <mjg59> dholbach: Yeah, sounds like usplash is failing to restore textmode in that case
[12:22] <tjaalton> dholbach: have you tried the working kernel since?
[12:22] <mjg59> dholbach: Does typing blind vbetool vgamode 3 work?
[12:22] <dholbach> tjaalton: hang on, trying with the new initramfs
[12:23] <tjaalton> competing questions :)
[12:23] <dholbach> no, that doesn't make it work
[12:23]  * dholbach tries mjg59's suggestion next :)
[12:23] <tjaalton> dholbach: so, the kernel version that used to work is now broken as well?
[12:24] <tjaalton> I'm quite sure it's the changes to usplash that broke it
[12:24] <dholbach> mjg59: it gives me the tty back and says "function not supported" :)
[12:24] <dholbach> tjaalton: hang on, I'll try rebooting into 2.6.28-6.17
[12:25] <tjaalton> dholbach: did you regenerate the initramfs for it?
[12:25] <dholbach> tjaalton: yes, for both
[12:25] <tjaalton> ok
[12:25]  * dholbach gets more tea in the meantime
[12:26] <dholbach> ... if only booting was faster! ;-)
[12:26] <tjaalton> sucky bios vendors ftl
[12:26] <dholbach> tjaalton: yay... now both kernels don't work with "splash" :-)
[12:26] <tjaalton> dholbach: touché
[12:26] <tjaalton> so bug kees :)
[12:27] <mjg59> dholbach: When the screen is black?
[12:28] <dholbach> mjg59: right after usplash it's flickering a bit, but stays black - or what is your questions?
[12:28] <dholbach> question
[12:29] <mjg59> dholbach: You run usplash, it turns the screen black. You then do vbetool vgamode 3 - what happens?
[12:31] <dholbach> mjg59: it gives me the tty back and says "Function not supported"
[12:32] <mjg59> dholbach: So it makes the screen turn back on?
[12:32] <dholbach> mjg59: yeah
[12:32] <mjg59> If so, it's a usplash bug - it should be doing exactly the same thing
[12:33] <dholbach> thanks a bunch mjg59
[12:33] <dholbach> mjg59: I passed the message on to kees
[12:34] <dholbach> thanks mjg59, tjaalton, apw - the world definitely needs more people like you :)
[12:35] <tjaalton> dholbach: heh
[14:28] <Ng> smb_tp: hey
[14:29] <smb_tp> Ng, Hi
[14:29] <Ng> smb_tp: did the sparc fix we were discussing the other day with fabbio make it into the archive yet? :)
[14:29] <Ng> istr it was heading for -proposed?
[14:29] <smb_tp> Ng, Into -proposed, yes
[14:30] <Ng> aha, excellent
[14:30]  * Ng goes hunting
[14:31] <Ng> linux-image-2.6.24-23-sparc64_2.6.24-23.49_sparc.deb ?
[14:32] <smb_tp> Ng, this should be the right version
[14:32] <Ng> smb_tp: great, thanks
[14:32] <smb_tp> Np
[15:42] <Keybuk> smb_tp: could you join #ubuntu-meeting ?
[16:08] <pgraner> sconklin: book the tix... arrive Thurs AM (or Wed PM your choice), leave Fri PM. Stay near the Airport
[16:09] <pgraner> sconklin: also book a rental car
[16:09] <pgraner> -EWRONGCHANNEL
[16:42] <maco> when a patch that ubuntu has committed stops being needed because it's accepted upsteam, is it automatically removed by you guys, or do you need to be told it's been added upstream?
[16:47] <smb_tp> maco, We will normally detect this as long as the upstream version is not completely different. In that case a hint is always appreciated.
[16:47] <maco> ok
[16:51] <dholbach> hiya
[16:51] <dholbach> what is our answer to people with a broadcom 4312 card in intrepid?
[16:51] <rtg> dholbach: LRM works, though its a blob
[16:51] <dholbach> rtg: ok.. I'll check out what's happening here
[16:55] <dholbach> rtg: seems it's not stock ubuntu and it wasn't there - thanks again
[16:56] <rtg> dholbach: what do you mean its not stock ubuntu? linux-restricted-modules contains the broadcom device driver.
[16:57] <dholbach> rtg: it's an Ubuntu derivative who ship their own lrm-eeepc-something
[16:57] <dholbach> rtg: it's called "easy-peasy"
[16:57] <dholbach> ha
[16:57] <rtg> oh, ok
[20:03] <CarlFK> how do I disable cpu scaling? 
[21:04] <Kano> hi rtg , i mailed you about qc-usb, i fixed the compile error but could not test it yet, but as it worked maybe before it should do too. i only added the new driver because i tested that first and it was working well.
[23:25] <IntuitiveNipple> Do we have a plan to fix the ehci/uhci load order bug # 296710 that causes USB2 (480mbs) devices to run at USB1 speed (11mbps) ? I just got stung by this copying a 23G file-system image from an external USB drive, wondering why it was taking so long. (dd reported 7.9MB/s instead of the usual 29MB/s)