| Notch-1 | hi, any news about the loop module? still =y? | 00:02 |
|---|---|---|
| MTecknology | Notch-1: it is by default | 01:29 |
| MTecknology | Notch-1: I'm running w/o it | 01:29 |
| MTecknology | 1.5hr later.. | 01:30 |
| geoff918 | I recently tested my system, which works, and I can't add it to this page: xbRiAnNa o7x | 02:04 |
| geoff918 | oops | 02:04 |
| geoff918 | this page: https://wiki.ubuntu.com/KernelTeam#Contacts | 02:04 |
| geoff918 | https://wiki.ubuntu.com/KernelTeam/Grub2Testing | 02:05 |
| geoff918 | geeze... | 02:05 |
| geoff918 | Anyway, I have an Acer Aspire 5610 | 02:05 |
| Notch-1 | MTecknology: yeah, i'm recompiling, again :D | 02:07 |
| MTecknology | Notch-1: I think I'm done tweaking for now - the kernel size is down to 2.2MB | 02:08 |
| Notch-1 | this really sucks, i could use a clean default normally-upgradable kernel... | 02:08 |
| Notch-1 | i don't want to tweak | 02:09 |
| MTecknology | Notch-1: why aren't you using that? | 02:10 |
| Notch-1 | i need to use loop-aes.. | 02:11 |
| Notch-1 | my life was seriously better when it was =m | 02:13 |
| MTecknology | did you file a bug about it? | 02:13 |
| Notch-1 | yeah, it's an old story... | 02:14 |
| Notch-1 | in few words, some people care too much about some microseconds at boottime | 02:15 |
| MTecknology | you mean me?s | 02:16 |
| Notch-1 | i don't think so, since you seem to don't know the problem yet, but i don't know so much... look here if you want to know... https://bugs.launchpad.net/ubuntu/+source/loop-aes-source/+bug/342902 | 02:19 |
| ubot3 | Malone bug 342902 in loop-aes-source "Build error: ‘struct bio’ has no member named ‘bi_hw_front_size’" [Medium,Triaged] | 02:19 |
| Notch-1 | beside, the original real compiling bug, indipendent of the the kernel configuration, was never fixed because this thread ended in nothing. so even if i recompile the kernel without loop, i still need to download the loop-aes source from sourceforge because the repositorys one is bugged... there are 2 unhandled problems with loop-aes since jaunty | 02:25 |
| Notch-1 | i hope that you used the correct version for karmic, there were 2 or 3 bugged ones, i don't have the versions now... this will solve one problem, at least... | 02:28 |
| MTecknology | sorry - internet died | 02:30 |
| MTecknology | I didn't even know if that sent | 02:30 |
| MTecknology | Notch-1: I meant the millisecond thing - I now have a 10sec boot across the area I care about | 02:31 |
| * MTecknology hoping that I didn't remove something I need | 02:31 | |
| Notch-1 | ah you did that for the boottime? :D | 02:34 |
| MTecknology | hm? | 02:36 |
| Notch-1 | you recompiled the kernel to speed up the boot? i will never do that :D anyway i was talking about the default configuration, the people that make decisions that should be best for everybody... sigh... | 02:41 |
| MTecknology | sorry :( | 02:42 |
| MTecknology | how is it that you're recompiling it? | 02:42 |
| Notch-1 | i told you, this freakin loop module... | 02:43 |
| MTecknology | how is it that you're recompiling it? - not what are you trying to do | 02:43 |
| MTecknology | I just pulled the git branch, grabbed the default .config (tossed it together from 3 .configs), make menuconfig, make all modules_install install && update-grub | 02:44 |
| Notch-1 | ah | 02:44 |
| Notch-1 | not exactly, it's very hard to find the original formula :D | 02:44 |
| MTecknology | the ubuntu kernel is rolled out differently, but that's because it reaches a larger base | 02:45 |
| Notch-1 | anyway yes with git, the exact current version | 02:45 |
| Notch-1 | with AUTOBUILD=1 skipabi=true skipmodule=true fakeroot debian/rules binary-debs | 02:45 |
| Notch-1 | after a menuconfig and a clean off course | 02:45 |
| MTecknology | sounds harder than what I did | 02:46 |
| Notch-1 | what you mean? | 02:46 |
| MTecknology | did you wind up making a .deb? | 02:46 |
| Notch-1 | hard it's not good, i prefer soft :D | 02:47 |
| Notch-1 | you mean just rebuild the kernel and replace it by hand? | 02:48 |
| MTecknology | When you finished creating the kernel, did you have a .deb package you could install with dpkg? | 02:48 |
| Notch-1 | yes | 02:48 |
| MTecknology | then you went the hard route ;) | 02:49 |
| MTecknology | do you have ccache installed? | 02:49 |
| Notch-1 | i understand, but wich is the softer one? | 02:49 |
| MTecknology | softer? | 02:49 |
| MTecknology | I mean that you went through more effort than you needed to if you're making a kernel for your system only | 02:50 |
| Notch-1 | ah ccache, i read about it... | 02:50 |
| MTecknology | You should install it - it makes a difference when you compile | 02:51 |
| MTecknology | What proc do you have? | 02:51 |
| MTecknology | how many cores | 02:51 |
| Notch-1 | softer=better | 02:51 |
| Notch-1 | just one | 02:51 |
| MTecknology | 1cpu with 1core ? | 02:51 |
| Notch-1 | this is not the real problem, i recompile on an old notebook, i just don't want to | 02:52 |
| MTecknology | After you finish doig make menuconfig, then run this command and you will have your new kernel built and installed --> export CONCURRENCY_LEVEL=3; export MAKEFLAGS="HOSTCC=/usr/bin/gcc CCACHE_PREFIX=distcc"; make all modules_install install && update-grub | 02:52 |
| Notch-1 | i don't want to speed up the recompile, i just wish to not have to at all... it's not a matter of time | 02:52 |
| MTecknology | You can copy the .config from the existing kernel too.. | 02:53 |
| Notch-1 | noted, thanks | 02:53 |
| MTecknology | I'm tryiing ot find the command I used | 02:53 |
| Notch-1 | but i don't know if i will use it, i want the simplest way possible... | 02:54 |
| MTecknology | that is the simplest | 02:54 |
| MTecknology | You can just use "make all modules_install install && update-grub" if you want | 02:54 |
| Notch-1 | anyway my grub can't be reached from my booted system, i'm inside a kexec :D | 02:55 |
| MTecknology | Here's what I used to get the .config cat debian.master/config/config.common.ubuntu debian.master/config/amd64/config.common.amd64 debian.master/config/amd64/config.flavour.generic > .config | 02:55 |
| MTecknology | I've been building the kernel from my home directory :P | 02:55 |
| MTecknology | the git branch is at ~/Downloads/ubuntu-kermic | 02:56 |
| MTecknology | s/e/a/ | 02:56 |
| Notch-1 | :D | 02:56 |
| Notch-1 | i'm not bash :D | 02:56 |
| MTecknology | I learned how to compile a kernel on Gentoo | 02:56 |
| MTecknology | You can ommit && update-grub | 02:57 |
| Notch-1 | commit && update-grub to recompile? | 02:57 |
| MTecknology | This part uses 3 compile processes which is good for 2 cores; it uses gcc to compile, and it sets the compile to use ccache which helps emilinate redundant processor calls -> export CONCURRENCY_LEVEL=3; export MAKEFLAGS="HOSTCC=/usr/bin/gcc CCACHE_PREFIX=distcc"; | 02:59 |
| MTecknology | This part compiles the kernel, compiles the kernel modules, and installs them where they need to go -> make all modules_install install | 03:00 |
| MTecknology | this part updates grub configs so the newly installed kernel will be used update-grub | 03:00 |
| MTecknology | this part updates grub configs so the newly installed kernel will be used -> update-grub ** | 03:00 |
| MTecknology | hopefully that all made sense | 03:01 |
| MTecknology | This can help you too - https://wiki.ubuntu.com/KernelTeam/KernelGitGuide | 03:02 |
| MTecknology | I'm going to go watch tv w/ my gf - bbl | 03:02 |
| Notch-1 | yes yes, thanks, i understand but i already know that, i messed up my last question, sorry :P | 03:03 |
| Notch-1 | i was thinking that you were saying that on gentoo is different | 03:04 |
| Notch-1 | enjoy, see you | 03:06 |
| Notch-1 | and thanks for the tips, next time i'll try to speed it up, maybe it will be less painful.. | 03:08 |
| === Whoopie_ is now known as Whoopie | ||
| === akgraner_ is now known as akgraner | ||
| MTecknology | Notch-1: gentoo actually lets you pull the kernel sources using the package manager - some things are different - but what I told you is universal across most any distro | 04:09 |
| MTecknology | now... what is it in the kernel that I don't have that makes the system run cooler | 05:48 |
| marcriera | hello everybody. does anyone know why karmic koala release candidate is shipped with a vanilla kernel? | 10:52 |
| marcriera | pgraner: hello, i have a question about kernel versions on 9.10. I was wondering if when the final release comes it will come with the actual 2.6.31 or it will switch to another kernel release. Many thanks. | 11:32 |
| marcriera | to myself : http://en.wikipedia.org/wiki/Linux_kernel#Versions . I understand that the even-odd stile is not working any more. so 31 is not vanilla. i guess ubuntu will remain on it for a while. | 11:42 |
| pgraner | marcriera: I'm not sure what you mean. 9.10 has 2.6.31.4 plus some backported fixes. | 11:55 |
| pgraner | marcriera: it was announced back May that 9.10 will ship with 2.6.31 | 11:56 |
| marcriera | pgraner: i'm sorry i bother you. I work with a group of researchers and the like the 2.6.31 improvements, and i was not sure ubuntu will stay with 2.6.31 when it passes from RC to final release. I was still thinking about the even-odd vanilla-stable stile releases. | 11:57 |
| marcriera | pgraner: thanks for you clarification. We will migrate to 9.10 , starting tomorrow , and start using the 2.6.31 for developing pourposes. thanks again. | 11:59 |
| === apw` is now known as apw | ||
| === csurbhi1 is now known as csurbhi | ||
| dtchen | oh, oops | 15:57 |
| dtchen | bah, wrong channel | 15:57 |
| RainCT | Hi | 16:39 |
| RainCT | Touchpad and WLAN don't work anymore on Karmic with the new kernel (-14), the previous one (-13) still works. Any idea? | 16:40 |
| ogasawara | RainCT: can you tell me the specific kernel version where it's failing (cat /proc/version_signature) | 16:44 |
| ogasawara | RainCT: some rfkill patches went in recently | 16:44 |
| ogasawara | RainCT: and are you on a Dell? | 16:45 |
| RainCT | no, ASUS EeePC 1005HA | 16:45 |
| ogasawara | RainCT: hrm, can you open me a bug using "ubuntu-bug linux" | 16:46 |
| ogasawara | RainCT: that'll get me all your specific hw info etc | 16:46 |
| RainCT | no connection :) | 16:46 |
| ogasawara | RainCT: no wired? | 16:46 |
| RainCT | awesome, USB mouse doesn't work either o_O | 16:47 |
| RainCT | ogasawara: connecting it to the router with an Ethernet cable doesn't seem to do anything either | 16:50 |
| RainCT | (ethernet didn't work on Jaunty, but did with previous Karmic kernels) | 16:51 |
| ogasawara | RainCT: hrm, what about saving to a USB and then attach to a launchpad bug | 16:51 |
| ogasawara | RainCT: want to see dmesg, sudo lspci -vnvn, and the /proc/version_signature | 16:51 |
| RainCT | ok I think that'll work (running ubuntu-bug just said "no such file or directory: /proc/asound/cards" btw) | 16:52 |
| RainCT | (and someone with another eeepc just confirmed he has the same problem with today's update) | 16:53 |
| dtchen | you've got some serious resource contention, then | 16:53 |
| ogasawara | RainCT: if you can, point any others with the same issue to your bug in lp | 16:55 |
| apw | RainCT, can you also attach a dmesg output from the -13 (clearly marked as from the working one) | 16:56 |
| ogasawara | RainCT: and the USB mouse regression, is that also between -13 and -14? | 16:56 |
| RainCT | I believe so, haven't tried it with -13 | 16:57 |
| dtchen | RainCT: err, wait. is this issue reproducible with linux-backports-modules-alsa-2.6.31-14-generic purged? | 16:57 |
| apw | RainCT, and is it directly connected or via a hub | 16:57 |
| RainCT | directly. USB sticks aren't showing up as /dev/{s,h}d* either | 16:58 |
| RainCT | dtchen: trying that now | 16:58 |
| * apw wonders if udev is still running | 16:58 | |
| RainCT | ps shows 1 upstart-udev-bridge and 3 udevd | 16:59 |
| apw | on the usb sticks, does anything appear at the end of the dmesg when you shove them in? | 16:59 |
| RainCT | dtchen: postrm gives "Exec format error" | 17:00 |
| === solarion_ is now known as Solarion | ||
| RainCT | (disregard someone else having the same problem, he asked for a bug number but hasn't tried updating yet.. doing so now) | 17:02 |
| RainCT | dtchen: actually the files in /var/lib/dpkg/info/linux-backports-mod../* are all empty | 17:03 |
| RainCT | fsck complained when I rebooted after the update so it may be because of file corruption | 17:04 |
| dtchen | RainCT: ext4 for /var ? | 17:05 |
| RainCT | yes | 17:06 |
| dtchen | RainCT: what version of upstart? | 17:06 |
| RainCT | dtchen: 0.6.3-9 | 17:07 |
| dtchen | RainCT: regardless, it sounds eerily like the sync() issue that Keybuk mentioned for ext4, although it should have been worked around in that version of upstart that you have installed | 17:07 |
| dtchen | RainCT: http://pastebin.com/d3d01c4bf has the contents of /var/lib/dpkg/info/linux-backports-modules-alsa-2.6.31-14-generic.postrm | 17:08 |
| Keybuk | that was only a guess | 17:12 |
| Keybuk | I took the sync() out ages ago, and only noticed corruption recently | 17:12 |
| Keybuk | so I put it back on the basis lots of things in the kernel source said THOU SHALT SYNC! | 17:12 |
| Keybuk | it's entirely possible we have an O M G ext4 bug in there | 17:12 |
| Keybuk | at some point, we should probably mention it to slangasek | 17:14 |
| dtchen | i know i've been able to trigger it in 2.6.32-rc1, but uh, that was 2.6.32-rc1 | 17:16 |
| Keybuk | rtg: have you been naughtily backporting patches again? :p | 17:18 |
| RainCT | OK, reinstalling the kernel fixed everything | 17:18 |
| ogasawara | RainCT: sweet, so no regressions now? | 17:18 |
| RainCT | ogasawara: nope :) | 17:18 |
| rtg | Keybuk, I love being naughty, but I'd like to know what for? | 17:18 |
| Keybuk | rtg: ext4 | 17:19 |
| rtg | not recently. | 17:19 |
| rtg | perhaps in stable, lemme check | 17:19 |
| Keybuk | colin says he's getting reports of grubenv being zero sized too | 17:19 |
| Keybuk | which would bit the "not written before reboot" pattern too | 17:19 |
| Keybuk | (and I've seen that!) | 17:19 |
| RainCT | WLAN is worse than using the latest drivers (even with backports-modules-wireless installed), but that isn't a regression from previous Ubuntu kernels (and it's certainly better than Jaunty, where it didn't work at all) | 17:20 |
| rtg | Keybuk, "xt4: Don't update superblock write time when filesystem is read-only" is the most recent in the changelog | 17:20 |
| Keybuk | right, it won't be that one | 17:20 |
| rtg | thats also the last touch in git | 17:21 |
| Keybuk | hmm, that's annoying | 17:21 |
| rtg | git log fs doesn't show anything either. I thought maybe jbd might have been messed with | 17:22 |
| dtchen | rtg: / apw: thanks very much for the wireless and alsa backports, BTW | 18:25 |
| rtg | dtchenno problem. did the ALSA stuff actually help? | 18:26 |
| rtg | dtchen: ^^ | 18:26 |
| dtchen | rtg: some people so far | 18:27 |
| dtchen | trying to get everyone to test it | 18:28 |
| dtchen | at least jerone's problem with the dells is progressing | 18:28 |
| dtchen | alsa-side it's fixed, just need to twiddle the right PA bits now | 18:28 |
| apw | rtg i thought the alsa in there was alsa-stable, and so pretty close to .31 | 18:37 |
| rtg | apw, its whateveris in alsa-drivers snapshot 1.0.21 | 18:38 |
| rtg | which is what dtchen suggested I use | 18:38 |
| apw | fair enough | 18:39 |
| dtchen | apw: pretty far from .31 | 18:39 |
| dtchen | it's much closer to .32 | 18:40 |
| dtchen | so it's 2.6.32 + quirks and fixes from 20091012 | 18:40 |
| dtchen | i.e., no core changes from .32 | 18:41 |
| apw | dtchen, cool | 18:43 |
| apw | dtchen, is it me or did the last updates trigger some alsa control lossage agian... i had to go enable a bunch of capture things on dell | 18:44 |
| dtchen | apw: meaning alsa-utils upload? | 18:47 |
| dtchen | apw: it shouldn't have, but there're too many reports of further regressions | 18:48 |
| apw | last couple of times i've updates things alsa things have gone haywire, no idea if kernel or userspace much | 18:48 |
| apw | i do seem to have a bunch more alsa controls than last time i looked | 18:48 |
| dtchen | apw: in your case, is it mixer settings being muted? | 18:49 |
| apw | i had a couple of those muted, and a couple of captures without the red CAPTURE L R at the bottom | 18:49 |
| apw | never seen that CAPTURE thing before ... | 18:49 |
| dtchen | apw: hmm, is it just capture elements being muted or playback elements as well? | 18:51 |
| apw | streaching my memory now, likely just capture elements | 18:51 |
| apw | i think the speaker mute was me | 18:52 |
| dtchen | if it's just capture elements, that's likely the symptom that jerone described | 18:53 |
| dtchen | particularly if you've a model affected by the latest quirk changes (HP dv series) | 18:53 |
| === bjf is now known as bjf-afk | ||
| RainCT | dtchen: (ah, installing the alsa backports thing worked here to get the microphone working on the eeepc 1005ha) | 19:37 |
| dtchen | RainCT: yes, as expected | 19:37 |
| dtchen | that's part of the massive mic-switching infrastructure that was dumped into 2.6.32 | 19:38 |
| dtchen | hence why lbm is a good idea for newer HDA | 19:38 |
| rtg | apw, we should go through our list of UBUNTU patches (once again) and figure out which ones should go upstream and/or stable. | 20:02 |
| apw | yep we should next week? | 20:03 |
| rtg | seems about right | 20:03 |
| apw | excellent | 20:04 |
| rtg | I'll wait until you are bleary eyed Wed night, after a few beers. then we can do it. | 20:04 |
| apw | heheh | 20:04 |
| === bjf-afk is now known as bjf | ||
| h00k | I'm having terrible kernel panics on my laptop, I was wondering where linux-crashdump actually dumps useful information to? | 21:31 |
| ogasawara | h00k: the following might help https://wiki.ubuntu.com/KernelTeam/CrashdumpRecipe | 21:37 |
| h00k | ogasawara: ahha, thank you! | 21:38 |
| h00k | ogasawara: it must not be able to figure out what is going on because I don't get anything with apport :( | 21:38 |
| h00k | standby | 21:39 |
| h00k | lemme play around and see what I can do | 21:41 |
| h00k | am I assuming that when it freezes and the numlock/capslock light flash that its a kernel panic? | 21:44 |
| h00k | yeah, forcing the kernel oops is not having my laptop reboot itself. | 21:45 |
| h00k | Yeah, I'm not having any luck :( I really hope that other people aren't having this problem, I'd like to get as much info to the developers as I can | 21:57 |
| === Keybuk_ is now known as Keybuk | ||
| === akgraner_ is now known as akgraner | ||
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!