[12:48] <BenC> infinity: ping
[02:02] <Goink> I am having timing problems on a dual core amd64. Does anybody know how i can fix it?
[02:02] <BenC> what sort of timing problem?
[02:03] <BenC> clock running twice speed?
[02:04] <Goink> Clock iss ssspeeding, bbut at  all sorts oof sspeeeds, it warrries aa  lot, and i get ddooouuubble orr mmmoree   kkkeey  pressess
[02:04] <BenC> boot with noapictimer
[02:04] <BenC> add it to /boot/grub/menu.lst in the kopt
[02:04] <BenC> and rerun update-grub
[02:05] <Goink> Thank you! I'll try that!
[02:05] <BenC> let me know if it works
[02:05] <Goink> Will do
[04:30] <infinity> BenC : Pong.
[06:26] <BenC> infinity: with -8.0, I am doing a lrm 2.6.15.3-1
[06:26] <BenC> new acx firmware and filenames for that firmware
[06:26] <BenC> -8.10 I mean
[06:27] <BenC> just a heads up
[07:03] <fabbione> morning
[07:04] <BenC> hey fabbione
[07:04] <fabbione> hey Ben
[07:04] <fabbione> BenC: I got a regression somewhere, but i think it's udev related
[07:05] <fabbione> midiC$something devices are now created in /dev instead of /dev/snd
[07:05] <fabbione> that's one
[07:05] <fabbione> the airo module for pci cards is problematic
[07:05] <fabbione> it is loaded properly but it doesn't work at the first shot
[07:05] <fabbione> i need to unload and reload to make it going
[07:06] <fabbione> reproducible quite often. what do you want me to do to try debugging it?
[07:06] <BenC> airo cs works perfectly for me, that's weird
[07:06] <fabbione> airo cs yes
[07:06] <fabbione> this is Pci
[07:07] <BenC> not sure, is it a regression?
[07:07] <fabbione> yes..
[07:07] <fabbione> it also tends to change interface name
[07:07] <BenC> want to start recompiling kernels? :)
[07:07] <fabbione> like at the first time is eth2
[07:07] <fabbione> unload -> reload -> eth0 ?
[07:07] <fabbione> yeah sure i have no problems with that
[07:07] <BenC> not sure about that
[07:07] <fabbione> recompilng it's not an issue
[07:08] <fabbione> one of the pci ethernet (marvell) always shows up as eth1
[07:08] <fabbione> even if alone on the system..
[07:08] <BenC> ok, you'll need git
[07:08] <fabbione> it's kind of weird
[07:08] <BenC> and compile out of the tree
[07:08] <fabbione> yeah
[07:08] <fabbione> will try that during the week
[07:08] <BenC> git-log, find hte sha1's for "Linux v2.6.1[432] "
[07:09] <BenC> git-branch 2.6.14 <sha1>
[07:09] <fabbione> i know it works on .12
[07:09] <BenC> do that for 2.6.1[432] 
[07:09] <BenC> git-checkout 2.6.13
[07:09] <fabbione> yup
[07:09] <BenC> build, and if it breaks, good, if not try 2.6.14
[07:09] <fabbione> eheheh
[07:10] <BenC> once you find work/nowork
[07:10] <BenC> then start using git-bisect
[07:10] <fabbione> what does that do?
[07:10] <BenC> you could just start out with 2.6.12 and HEAD
[07:10] <BenC> git-bisect makes your life a lot easier :)
[07:10] <BenC> git-bisect start
[07:11] <BenC> git-bisect good 2.6.12
[07:11] <BenC> git-bisect bad HEAD
[07:11] <BenC> it will bisect the commits, and checkout at the middle point
[07:11] <BenC> you build, you test, and you come back and do "git-bisect <good|bad>"
[07:12] <BenC> it wil bisect again, and you keep doing that cycle until it tells you "This commit broke it"
[07:12] <fabbione> ah cooooolness!
[07:12] <BenC> good/bad depends on whether your test suceeded or failed
[07:12] <fabbione> yeps
[07:12] <fabbione> that was clear
[07:13] <BenC> it's so simple I feel I have to overexplain it :)
[07:13] <fabbione> ahah
[07:13] <BenC> it's what I used to track down the sparc64 oops on my e3k
[07:15] <fabbione> yeah it makes sense
[07:15] <fabbione> assuming that when you bisect you can actually build
[07:17] <BenC> yeah, you can bump up/down the commit it stops at if you hit a bad spot
[07:18] <fabbione> i see you are brave enough to readd misdn
[07:19] <fabbione> it was removed because upstream sucks
[07:19] <BenC> doko requested it
[07:19] <fabbione> at least it did at that the time
[07:19] <BenC> it's compiling cleanly
[07:19] <fabbione> ok
[07:19] <fabbione> they must have started again
[07:19] <fabbione> it was OOPSorama in hoary
[07:19] <BenC> heh, didn't say I tried using it :)
[07:19] <fabbione>  .../usb/media/podxtpro/.tmp_versions/podxtpro.mod  |    2 
[07:20] <fabbione> i think there is a leftover in git
[07:20] <fabbione> yeah
[07:20] <fabbione>  create mode 100644 drivers/usb/media/podxtpro/.tmp_versions/podxtpro.mod
[07:21] <BenC> thanks, removed
[07:21] <fabbione> no problem.. i just woke up and bored to death as you can see ;)
[07:22] <BenC> heh, try perusing bugs for 3 days
[07:22] <BenC> I think I've gone through half of them
[07:22] <BenC> bad thing is, if I stop now, I don't think I'll remember where I left off
[07:22] <fabbione> ahah
[07:23] <fabbione>     [UBUNTU:debian]  Add ABI files (missing hppa and sparc still)
[07:23] <fabbione>  <- they are in the archive now
[07:23] <fabbione> if you want i can push them
[07:23] <BenC> I have them
[07:23] <fabbione> up to you
[07:23] <fabbione> ok
[07:23] <BenC> just need to commit it
[07:23] <BenC> my abi grabbing script is smooth
[07:24] <BenC> I need to put it in the debian/bin/
[07:25] <fabbione> that's a good idea
[07:27] <fabbione> hmmm smoke :)
[08:33] <infinity> BenC : Which firmware are you grabbing for acx_pci stuff?... I've been all over the interweb and seen many conflicting reports on what we should be shipping.
[08:33] <infinity> BenC : Also, I already rolled a .3 with updated nvidia and ati drivers, so we may as well just do this all at once.
[08:34] <infinity> BenC : And I'm going to send out a mail to -devel/-users today to have people test vga16fb at 640x400 with kernels I've tossed on people.ubunut.com.  Would it be worth holding off on -8.10 until I gte reports back from that?
[02:57] <BenC> infinity: sorry, went to bed
[02:57] <BenC> infinity: but yeah, I can hold off. I'll get you a tarball of the firmware I have
[02:58] <BenC> infinity: bad thing is there is a table of firmware, different cards need different versions, but the driver (even the new one) isn't smart enough right now to load different versions for different cards
[02:58] <BenC> the new driver is smarter than the old one though
[02:58] <BenC> like the RADIO16.BIN thing is totally broken with the new driver
[02:59] <BenC> you have to have a pristine firmware tree (no firmware of other names that it looks for) in order for things to load right
[03:00] <BenC> so you really can't have acx100 and acx111 firmware sitting together
[03:43] <BenC> infinity: I have an idea
[03:44] <BenC> infinity: so wake up! :)
[03:59] <infinity> About to send mail to -users and -devel about the vga16fb thing, so I'm here for a few minutes.
[03:59] <infinity> What's your stellar idea?
[03:59] <infinity> (Can we not just fix acx_pci to DTRT about firmware, whatever TRT is?)
[04:00] <infinity> Map certain filenames to PCI IDs or something?
[04:00] <infinity> (or something less crude, if that's possible)
[04:00] <infinity> BenC : flashy noisy nick alert.
[04:13] <BenC> hey
[04:13] <BenC> I have a fw tarball
[04:14] <BenC> several different versions for acx100/usb acx100/pci and acx111/pci
[04:14] <BenC> we'll have a default (newest version)
[04:14] <BenC> and the module will take firmware_ver=X.X.X to use a different version if they need to
[04:15] <BenC> if they have do that to get it to work, they'll be instructed to send us lspci/lsusb output so we can code it into driver
[04:15] <BenC> infinity: flash back
[04:16] <BenC> I'll send you the tarball when I'm done
[04:22] <infinity> Anyonw want to give this a quick read?
[04:23] <infinity> http://cerberus.0c3.net/~adconrad/vga16fb.txt
[04:23] <infinity> BenC : Does acx_pci even support USB devices?... Or do we now have an acx_usb or something?
[04:23] <infinity> BenC : Also,want to read over that mail before I hit "send"?
[04:23] <BenC> we have acx_usb
[04:24] <BenC> but, the new driver is usb/pci combined
[04:24] <BenC> yeah
[04:24] <BenC> it's just acx.ko
[04:24] <infinity> Anyhow, however you work the firmware is up to you, since you have to maintain that driver.
[04:24] <infinity> I just hope it works "out of the box" for the majoriy of users, and requires tweaking for a tiny subset.
[04:28] <infinity> Oh, while I'm asking them to give me output proving they're not doing something dumb, "lsmod | grep fb" might be nice too.
[04:31] <infinity> Updated.
[04:37] <BenC> yep :)
[04:37] <infinity> Ignore the few spelling mistakes (timstamp -> timestamp, etc), I just spellchecked in Thunderbird.  That's all fixed.
[04:38] <BenC> btw, ia64 has vga16fb and vesafb now
[04:38] <infinity> Just want to know if the general procedure gets the point across and if you think it'll generate useful reports from otherwise not-always-useful-but-well-meaning users.
[04:38] <infinity> Sure, but our users don't run ia64.  You can test it for me there, if you want. :)
[04:38] <BenC> I'll also test it on my nvidia geforce4 with tvout
[04:38] <BenC> can't yet, waiting for lamont to get my elilo that supports initramfs :)
[04:39] <infinity> I thought he already uploaded it to Debian..?
[04:39] <infinity> Or was he just talking about it?
[04:39] <infinity> Anyhow.  Text looks sane?
[04:39] <BenC> yeah
[04:39] <infinity> Sent.
[04:40] <infinity> I hope I don't get moderated into oblivion on -users... I can never remember what address I have subscribed there.
[04:40] <infinity> I really should un-sub from all the lists and re-sub with adconrad@ubuntu.com
[04:40] <infinity> (subscribed when I was a contractor and didn't have the @u.c email yet)
[04:41] <BenC> if you wait 5 minutes, I'll tell you my results
[04:41] <infinity> My brain shot circuited and saw that as:
[04:41] <infinity> s/shot/short/
[04:42] <BenC> lol
[04:43] <BenC> so where did the handdrawn dapper usplash screen come from?
[04:44] <infinity> mjg59's handiwork.
[04:44] <infinity> Brilliant, isn't it?
[04:44] <BenC> it's not going to be the official one, is it? :)
[04:44] <infinity> Also, I don't expect this to change your tvout situation at all, but I'll be pleasantly surprised if it does.
[04:45] <infinity> No, it was just a quick hack to shrink the artwork.
[04:45] <infinity> I'll put a proper one back in in the next upload.
[04:45] <infinity> Damn, I did get moderated to -users.
[04:46] <BenC> here goes nothing
[04:59] <BenC> weird
[04:59] <infinity> I don't like the sounds of that.
[04:59] <BenC> I don't even see the usplash screen, but my system stays in 640x400-70, and the console works great
[04:59] <BenC> before vga16fb would break even console
[04:59] <infinity> Did you have usplash 0.1-25 installed?
[04:59] <BenC> wonder why I don't see it
[04:59] <infinity> If you have -24 or earlier, your usplash will be black.
[05:00] <infinity> Maybe I should have made that more explicit in the mail. :)
[05:02] <BenC> I have 0.1-25
[05:02] <BenC> even reran update-initramfs -u to make sure
[05:02] <infinity> Are you getting a black screen until usplash exits, or n ousplash at all, or..?
[05:03] <infinity> What do you get if you run "usplash" as root from a console (and wait around for it to timeout and exit)?
[05:04] <BenC> black screen
[05:04] <infinity> Rather curious about console output after it exits, if any.
[05:04] <BenC> seems to timeout on boot
[05:04] <BenC> after it exits I get normal scrolling output
[05:05] <infinity> Yeah, I mean if you run it manually right now.
[05:05] <BenC> this system is only partially upgraded to dapper (just what the kernel needs)
[05:05] <BenC> anything you can think of that I might need to upgrade too?
[05:06] <BenC> how long does it take to timeout?
[05:06] <infinity> Not off the top of my head.  If usplash is using the right (read: mjg59's goofy temporay) artwork, and usplash is 0.1-25, it SHOULD work.
[05:06] <infinity> Timeout if 15 seconds.
[05:06] <infinity> s/if/is/
[05:06] <BenC> it timed out, and my screen stayed black
[05:06] <infinity> Roughly.  The timer seems a bit goofy on some systems.
[05:07] <BenC> can't get console back now
[05:07] <infinity> chvt to X and back?
[05:07] <infinity> Anyhow, I think this means you and I have a date tomorrow to look into this more.
[05:08] <infinity> Interesting that I've fixed your console, but broken usplash.
[05:08] <infinity> Halfway there, I guess.
[05:08] <BenC> yeah, better atleast
[05:08] <BenC> ok, good night
[05:25] <zul> heylo
[05:26] <zul> BenC: apparently i suck
[05:26] <BenC> how so?
[05:26] <zul> gitweb for me was working fine i just cant spell..now i can do some work
[05:28] <BenC> lol
[05:31] <zul> i should have a couple of small things for you to push by wednesday
[05:33] <BenC> ok
[05:33] <zul> start slow as i always say ;)
[05:35] <zul> hmm...that sucks richard pryor is dead
[05:40] <makx> BenC: is ide-generic still modular in your builds
[05:40] <makx> ?
[05:45] <makx> don't find the CONFIG_IDE_GENERIC settings in your debian/config tree
[05:45] <imrabti> Hello
[05:46] <makx> aah modular still.
[05:46] <imrabti> how to desactivate The CD ROM Reader In Ubuntu 
[05:46] <makx> why imrabti?
[05:46] <imrabti> Because It doesn't work 
[05:47] <imrabti> And often The Text mode in Ubuntu give an error Message
[05:47] <zul> what error message?
[05:47] <imrabti> Wait
[05:52] <imrabti> [429475.993000]  No reference position found (media may be upside down) -- (as 0x06,ascp = 0x00) [4294750.993000]  The failed "Read cd/dvd capacity" packet 
[05:52] <imrabti> it not stop to give this message 
[05:52] <imrabti> in text mode
[05:52] <imrabti> I must desactivate the CD reader 
[05:53] <zul> BenC: is the i386/config the base config now or what is the scoop?
[05:54] <BenC> imrabti: rmmod ide_cd
[05:54] <BenC> zul: base config for i386, yes
[05:54] <imrabti> yes 
[05:54] <BenC> each arch has a base config, and then additional config.$flavour
[05:55] <zul> ok so if i do a CONFIG_EFI_PARTITION=y in i386/config it will appear in the 686 kernels as well correct?
[05:55] <imrabti> it works 
[05:55] <BenC> yeah, it will appear in all of them
[05:55] <imrabti> Thank a lot 
[05:55] <imrabti> man 
[05:55] <imrabti> I'm very Happy 
[05:56] <BenC> imrabti: no problem
[05:56] <imrabti> it doesn't give the message
[05:56] <zul> BenC: sweet.
[05:56] <BenC> zul: put it in alpha order or it will get moved next time I run debian/bin/oldconfig
[05:57] <BenC> just to avoid uneeded deltas
[05:57] <BenC> well, it's already in there, isn't it?
[05:57] <BenC> just needs to be enabled
[05:57] <zul> yeah i just enabled it in my tree
[05:57] <BenC> nope, it isn't
[05:58] <BenC> there it is
[05:58] <zul> in my tree just enabled it
[06:29] <imrabti_> Please How Can Desactivate CD Rom Reader Forever
[06:29] <imrabti_> when i use the command rmmod ide_cd 
[06:30] <imrabti_> when i restart the computer I find that the cd rom reader is still here
[06:31] <imrabti_> Please can anyone help me 
[06:31] <BenC> this is a kernel channel, you should really try #ubuntu
[06:31] <imrabti_> Ok 
[06:32] <imrabti_> You don't have any idea.
[06:34] <imrabti_> that could help me to solve this problem 
[06:34] <BenC> not that, just that, as the topic says, this channel is for discussing kernel development only
[06:34] <imrabti_> ok thanks 
[06:51] <trevilor> is there a quick and dirty way to get a list of features that differ in ubuntu specific kernels from the common kernel source?
[06:59] <imrabti_> Please Could juste Tell me How te Completly desactivate The CD drive 
[07:00] <imrabti_> i serched a lot but anyone give a ansewer
[07:14] <imrabti_> Please
[07:15] <imrabti_> you tell me how to desactivate it with rmmod ide_cd 
[08:44] <imrabti> Please 
[08:44] <imrabti> I searched a lot 
[08:44] <imrabti> and i don't find anything
[09:43] <dilinger> oh freenode
[09:43] <dilinger> how i despise you
[09:43] <dilinger> --- Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( http://freenode.net/faq.shtml#privmsg )
[09:43] <dilinger> jbailey__: code of conduct signed
[10:43] <czr> can I use the git in breezy/universe to get the kernel source too?
[10:44] <czr> I'm looking for the root cause for why ext2online doesn't work on breezy kernel (using LVM2 LVs underneath)
[11:23] <BenC> czr: see topic
[11:26] <czr> BenC, the topic didn't answer my question. problems solved though, I built my own git from sources
[11:26] <BenC> yes it does
[11:26] <BenC> it says to install git-core
[11:26] <czr> it says to get git-core for dapper
[11:26] <BenC> apt-get install git-core
[11:27] <czr> anyhow, problem solved.
[11:28] <BenC> if you're building with the 2.6.15 configs, you'll have to be running dapper anyway, atleast a little
[11:28] <BenC> breezy kernel-package will not like some things either
[11:29] <BenC> so unless you are doing straight "make" and installing the vmlinuz and modules by-hand, you'll run into trouble
[11:29] <czr> well, I wasn't going to build kernels actually
[11:29] <czr> wanted to check whether the ext2online resizing patches are present in the kernel sources or not
[11:29] <BenC> the git repo is not for breezy
[11:29] <BenC> breezy kernel is in baz
[11:29] <czr> I noticed. baz?
[11:29] <BenC> or just "apt-get source linux-source-2.6.12"
[11:30] <czr> ok, thanks
[11:30] <BenC> you'll need to enable deb-src line in your sources.list
[11:30] <czr> yeah, I've been using debian for about 6 years, I think I'll manage ;-)
[11:31] <BenC> then why didn't you do apt-get source to start with? :)
[11:31] <czr> someone just pointed to the wiki page so I thought that is the way in ubuntu :-)
[11:32] <czr> is there a nice list of patches applied versus vanilla anywhere?
[11:32] <BenC> debian/patches/
[11:32] <czr> thanks
[11:32] <BenC> also, there's a list in /usr/share/doc/linux-image-2.6.12-9*/
[11:33] <czr> -10 now I think
[11:33] <BenC> FYI, Ubuntu isn't that much different from debian, so whereever you look there, chances are it's the same way on ubuntu
[11:34] <czr> maybe I'm going the wrong way about it though. the problem is that ext2online doesn't work on breezy over LVM2
[11:34] <czr> and it works on RHEL4u2
[11:35] <czr> so I should really look at RH kernel sources first to locate the stuff that they put there
[11:35] <BenC> does it need a patch to work?
[11:35] <dilinger> yes
[11:35] <dilinger> rhel has had a patch in there for a while
[11:35] <czr> I think so
[11:35] <dilinger> it was merged into upstream at some point, i forget when
[11:35] <BenC> I'm pretty sure we don't have that patch
[11:35] <czr> well, ext2online fails on breezy, so breezy doesn't
[11:35] <BenC> unless it's in upstream, then dapper would probably have it
[11:36] <BenC> what is ext2online?
[11:36] <czr> although I'm not sure why ext2online fails. All I get is 'resize failed while in kernel' :--)
[11:36] <czr> grows ext2/3 fs:s while they're mounted
[11:36] <dilinger> it's an old patch by andreas dilger
[11:36] <czr> comes in ext2resize-package (breezy/univ)
[11:37] <czr> dilinger, you sure about the upstream merge?
[11:38] <dilinger> positive
[11:38] <dilinger> i remember being very pleased when it happened
[11:38] <czr> great
[11:39] <czr> RH had some problems getting it into RHEL4 :-) only got there in update2
[11:39] <czr> which kernel will dapper be using?
[11:39] <czr> -15?
[11:39] <BenC> yes
[11:40] <czr> hmm. would the universe end if I'd build a -15 from vanilla sources manually and install it in breezy?
[11:40] <czr> does udev need changes, etc?
[11:43] <BenC> you may be missing some devices on your desktop
[11:43] <BenC> but generally it should work
[11:44] <BenC> it doesn't take a whole lot to install the dapper kernel
[11:44] <BenC> it only updates a few packages
[11:44] <BenC> udev mainly, and it will remove hotplug, since it's obsolete
[11:44] <czr> ah, will all the hotplug helper scripts go too?
[11:44] <BenC> yeah, but udev handles it mostly
[11:45] <BenC> may want to add "auto eth0" to /etc/network/interfaces though
[11:45] <czr> yes, I wondered about that in some distros which have both udev and hotplug scripts
[11:45] <BenC> 2.6.15 allowed udev to be cleaned up a lot, and hotplug to just disappear
[11:45] <czr> ah, it's using hotplug. good idea
[11:46] <czr> well, I'll give it a spin soon
[11:46] <czr> the ext2online is really nice feature
[11:56] <dilinger> http://www.kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.git;a=blob;h=6104ad3105077ec7a14e5da4f24a261d116035a0;hb=e4f5c82a92c2a546a16af1614114eec19120e40a;f=fs/ext3/resize.c
[11:57] <dilinger> i wonder if he's still working for cfs
[12:03] <czr> ah, thanks dilinger