=== jjohansen is now known as jj-afk === bjf is now known as bjf[afk] [04:49] I think I have a regression between 2.6.36-rc7 and -rc8 regarding wifi, not sure where to writeup the bug or what would be needed [04:49] anyone avail to help? [04:52] also what is drm-intel-next ? [07:59] ikepanhc: hi, I tested ideapad-laptop. Almost all special buttons worked. what did not was the wifi/BT switch and the Play/Pause, Stop, FF, REW buttons [08:03] papillon81: I guess the brightness/volumn control, touchpad switch, works fine [08:04] papillon81: those key enabled because of BIOS (DSDT) or they will generate an keycode [08:05] papillon81: they are not enabled by ideapad-laptop yet [08:05] ikepanhc: yes, they work [08:05] IIRC, the Camera switch is also working without ideapad-laptop [08:05] ...let me check [08:05] papillon81: ya [08:06] papillon81: the camera key just power on/off the camera, you can try to press the key and use dmesg to check [08:07] papillon81: usb camera will disappear and reappear [08:10] ikepanhc: well, it looks like most of it works without the driver already [08:10] the play/pause... keys do not as they produce no key event [08:11] wifi produces an event, but no connection/disconnection happens [08:12] papillon81: how you see the event? from dmesg which says unknown scancode? [08:12] ikepanhc: yes [08:13] papillon81: oh, some ideapad model will report scancode when hotkey pressed, but not all of the ideapad [08:17] BRB === smb` is now known as smb [08:34] Morning smb [08:34] Hi ikepanhc :) [08:34] lag, morning [08:34] good morning .eu [08:34] :) [08:35] Are you packed and ready for UDS [08:35] me? not yet [08:36] I am used to pack 6hrs before departure [08:36] and hope nothing missed [08:36] s/hope/pray for/ [08:41] * abogani waves all [08:44] :) [08:54] * apw waves to ikepanhc [08:55] ikepanhc, can you point me to the oem public trees [08:55] git://kernel.ubuntu.com/hwe/oem-master.git [08:55] ikepanhc, thanks :) [08:56] apw: :) [09:00] By the way the natty-meta package contain a typo in README file at line 6. [10:12] * smb spots gkh signs === diwic is now known as diwic_afk === diwic_afk is now known as diwic [13:38] boing [13:40] Not so much "boing" than "whoosh" (or how would a erupting volcano sound?) [13:41] eh? [13:42] * smb is causing confusion by merging two statements hours apart on irc [13:46] apw, are you working on a 2.6.36 final for natty? I can do it if you're busy. then we can leave natty alone for a few weeks. [13:52] when put 2.6.36 in ubuntu? [13:52] when put in ppa? [13:54] tgardner, will do it :) [14:07] JFo: [14:07] lag [14:07] :) [14:07] What do we do? [14:07] Where do we go? [14:07] no idea [14:07] clueless [14:07] was wondering that myself [14:07] as the rest of us [14:07] Good job we had this test :) [14:08] indeed [14:08] This could have been the real thing [14:08] smb: No [14:08] smb: We have a video conf call [14:08] lag, "We" have not [14:08] :) [14:08] lol [14:09] lag, you see that ping? [14:10] just invited you lag [14:11] JFo: I would like to compile my own kernel as MTecknology posted on his latest blog post. Are you or is anyone able to walk me through this or point me to a guide which tells me the best way to do this on Ubuntu? [14:12] MTecknology: btw, I can't find the .config for kernel options on http://profarius.com/content/what-you-need-do-after-installing-ubuntu-1010 [14:12] komputes, there should be some information in the /Kernel wiki pages at wiki.ubuntu.com/Kernel [14:12] let me know if you have trouble finding them [14:13] JFo: ok, wel thanks to ogasawara and the LPI201 study guide, I know the basics. But I will try my best to do this without being too much of a pain. :-) [14:14] heh, no problem :) === tgardner is now known as tgardner-afk === bjf[afk] is now known as bjf === tgardner-afk is now known as tgardner === BenC__ is now known as BenC === diwic is now known as diwic_afk [16:30] the maverick distro master branch has been updated with the merge of the security release. Thanks to Brad for doing this. [16:33] komputes: ? [16:33] oh [16:33] komputes: http://profarius.com/sites/profarius.com/files/kernel.config [16:33] komputes: keep in mind that it's created by a psycho [16:34] komputes: It wasn't listed at the bottom of the page for you? [16:34] MTecknology: wow, right on time. cheers and congrats on the engagement [16:34] nope [16:36] komputes: thanks :) [16:36] I'll look into why. [16:37] MTecknology: do you recommend any particular build method from here for doing this: https://wiki.ubuntu.com/Kernel/Dev [16:37] MTecknology: if you can add the commands you used to your post, I would find it extremely helpful. [16:38] komputes: make menuconfig; make all install [16:39] komputes: I said phsycho :P If you're looking to do it the 'right' way, then you probably want to not follow my advice :P [16:39] MTecknology: where to I place the config, can I see your config file in curses menu view? [16:39] I'm experimenting, no worries [16:40] ls -a [16:41] MTecknology: are you asking for a listing of the kernel source dir? [16:41] komputes: no [16:41] k [16:41] komputes: that's where the .config is [16:42] err. where? [16:42] top level of the kernel source? [16:43] sorry, I'm a virgin at this [16:44] many attempts, never success [16:49] JFo: MTecknology: "make menuconfig" returns the error "Requires ncurses-devel" but the correct package name is libncurses5-dev [16:50] komputes: actually.. ncurses-dev [16:50] oh, ok [16:52] MTecknology: ncurses-dev is purely vityual pounting to libncurses5-dev ;) [16:52] virtual* [16:53] should we report to make a change to the error (from ncurses-devel to ncurses-dev) [16:53] komputes: it's not an error [16:53] komputes: other package managers use the virtual package ncurses-devel, for some that is the package [16:54] ok [16:54] hey. so, i'm trying to figure out what changed between 'linux-image-2.6.32-308-ec2 2.6.32-308.16' and 'linux-image-2.6.32-309-ec2 2.6.32-309.18' [16:54] You're not dealing with something that's ubuntu-centric, it's linux-centric :) [16:54] looking a the changelog for the newer version http://changelogs.ubuntu.com/changelogs/pool/main/l/linux-ec2/linux-ec2_2.6.32-309.18/changelog [16:54] it does not even include 2.6.32-308.16 [16:54] understood. [16:57] i'm guessing that the entries for 2.6.32-309.17 and 2.6.32-309.18. but programmatically, that would be difficult to determine since i can't locate the source version in the changelog [16:57] smoser, Some upload versions can get eradicated when security replaces a version in proposed [16:57] it seems that that entry got dropped [16:57] the changelog entry got dropped though [16:57] both those versions did exist [16:58] http://changelogs.ubuntu.com/changelogs/pool/main/l/linux-ec2/linux-ec2_2.6.32-308.16/changelog [16:58] If it did not leave proposed it did not officially exist [16:58] it did leave proposed [17:00] (it made it into an image created 20100923 that do not include -proposed in their sources.list) [17:01] smoser, Hm, right. .16 would actually have been a security release [17:01] i'm guessing its just human error, and also that only the changelog entry for 2.6.32-308.16 was dropped, and not actually the changes. [17:03] If I am not wrong and the .16 was a security update, then those changes should normally not get dropped when rebasing to a newer version of the master kernel [17:04] 308.16 was a security release. [17:04] Yes, so I am quite confident that this is just an error within the changelog [17:04] i'm not trying to be accusing, i'm just wanting to make sure that there isn't a hole in the process that dropped those security fixes. [17:05] and ideally, that in the future, even the changelogs wouldn't get dropped [17:05] i'm trying to write a tool that shows me what differed between two builds, and want to grab new changelog entries. [17:06] Right. The rebasing itself is safer than the changelog because that needs manual intervention to get it right [17:06] right. humans are incompetent [17:06] :) [17:07] Agreed :) But in general it could get difficult to compare changelogs [17:07] Its already sometimes confusing for the master tree. But rebase branches are worse [17:08] MTecknology: My docs says Kernel configuration settings are stored in a file named .config. Historically, this file was saved within your kernel source to /usr/src/linux or /usr/src/linux-kernel-version, but this is no longer the case. Older applications (based on the standard libc) required /usr/src/linux, but the introduction of a new library, glibc, eliminated that dependency. So I simply placed it in the linux-2.6.35-MTeck/ dir and r [17:08] an "make menuconfig" and I'm now going through all the changes you made. Very cool :-) [17:09] Especially cool with split screen terminal [17:10] thanks smb . just kindly take note of my "please try to get changelog entries correct" request. [17:10] komputes: there's also make xconfig, but I think menuconfig is better [17:10] smoser, note taken [17:10] :) [17:11] MTecknology: I'm aware of these ways to do it: make config, make menuconfig, and make xconfig [17:11] MTecknology: menu is still my favorite [17:20] MTecknology: few questions about disabled configurations: pass kernel param to init, LZO vs Gzip kernel compression, POSIX msg queue, BSD process accounting, export task, namespaces - how can i better understand what these are and what is the result of disabling them? [17:21] komputes: look at the help for it [17:21] komputes: then look at the code or online [17:22] MTecknology: ok, help wasn't working, but I found it - not quite english, but I kind of understand [17:23] MTecknology: do you remove upstart from your system and simply use init? [17:23] komputes: nope, upstart is still there [17:24] MTecknology: and disabling CONFIG_INIT_PASS_ALL_PARAMS does not mess that up? [17:27] komputes: I didn't say upstart still worked the way it's supposed to :P [17:30] muahahaha [18:09] [19:30] hi there [19:30] what's sane way to load custom DSDT table in ubuntu 10.10? [19:31] DSDT table for my laptop is broken (thermal and battery methods are incorrect) and so I need to load my custom DSDT [19:31] anarsoul, as i recall things there is no offical way to do that, as you can brick your laptop [19:31] anarsoul: Incorrect in what way? [19:32] i believe the approved approach is to quirk round the deficienies [19:32] mjg59: thermal provides incorrect info about CPU temperature (sometimes stick at 70C and fan works all this time) [19:33] and battery method provides some heuristic info about capacity instead of get this info from smart controller [19:33] so it's dangerous to use laptop on battery with default DSDT table [19:33] apw: Well, approved approach is for people to either get the bugs fixed in Linux or, if Windows has the same behaviour, return the hardware for being broken [19:33] as it can kill battery because it can't detect correctly when it's broken [19:34] mjg59: windows has the same behaviour [19:34] mjg59, well put [19:34] and I can't return this hardware [19:34] anarsoul: Your hardware's broken, then! [19:34] it's 3year old laptop [19:34] There's an interface in debugfs that lets you override individual ACPI methds [19:34] mjg59: wrong answer [19:34] it works with fixed DSDT table [19:35] mjg59: yeah, I know, it does not fit [19:35] Then you get to build your own kernel [19:35] as it can override only methods [19:35] but I need to override OperationRegion aswell [19:36] mjg59: I remember there was a patch that allowed to load custom dsdt from initramfs [19:36] why it was removed? [19:36] anarsoul: It got rejected upstream [19:36] so what? [19:36] anarsoul, because people would load random dsdt's into their system [19:36] ubuntu keeps much their custom patches [19:36] anarsoul, where someone on a bug suggested a mod for a different system [19:37] why not to keep another one? [19:37] anarsoul, and as you can do genuine dammage to your system [19:37] anarsoul, it was seen as too dangerous [19:37] apw: you can damage your system by overriding acpi method via debugfs [19:37] anarsoul, indeed that is also true [19:38] but that was the reason the patch wasn't maintained [19:39] too easy for a naieve user to follow instructions to try and modded dsdt on the wrong laptop [19:41] apw: oh, it's easy to do "sudo dd if=/dev/urandom of=/dev/mem" [19:42] anarsoul: If you're competent to modify ACPI tables then you're competent to rebuild your kernel [19:45] anarsoul, yep you can shoot self in the foot a number of ways, but people seemed to like loading random dsdts that fixed things for other people [19:45] mjg59: yeah, but it takes my and cpu's time [19:45] anarsoul, its all a balance [19:46] mjg59: and I can't react on each security update [19:59] hi all [19:59] quick question [20:00] is echo 1 > /proc/sys/kernel/panic_on_oops supposed to work? [20:00] cuz it's not panicing :) [20:04] ppetraki, try /proc/sysrq-trigger [20:05] MTecknology: at the end of the compilation, I get many errors relating to ndiaswrapper. Do you know which path in the menu config I need to go to disable the ndis (module?). [20:05] tgardner, Tim, the goal is to panic on some mysterious oops we've been seeing, the systems are unattended [20:05] komputes: press / and search [20:06] ppetraki, well, I think you can do that from user space. see Documentation/kdump/kdump.txt in the kernel tree [20:07] tgardner, I'll take a closer look, thanks [20:08] MTecknology: found it but once i press exit i do not know the path to get to it [20:08] MTecknology: it brings me the help page only - can't turn it on/off [20:08] komputes: you need to follow the path it gives you [20:09] MTecknology: awesome, thanks [20:32] * ogasawara lunch [20:53] MTecknology: OK! :-) I finally finished compiling, no errors. I had to change compression (gzip) and strip out rtl and ndis networking extras. So what now? [20:54] komputes: you installed it? [21:29] MTecknology: well I did "make all install" not quite sure what that accomplishes, but it looks like it finally compiled without errors [21:29] update-grub2 [21:29] reboot [21:29] you should read the Makefile [21:29] orly? nice... [21:30] don't remove the generic kernel until you really know what you're doing and know how to fix things when it breaks [21:30] no, i don't inted to remove the ubuntu stock kernels at all [21:31] and this is a scratch box [21:31] MTecknology: I'm reading the makefile, you actually understand this? [21:31] :P [21:48] MTecknology: rebooted, kernel won't boot properly, when i run it in recovery mode I quickly see an error regarding VBIOS table [21:48] it makes two short beep noises [21:49] komputes: yup, that's the fun in compiling your own kernel :) [21:49] komputes: look on the gentoo install manual and it'll help you start to understand it better [21:49] MTecknology: will do. thanks for all your help today, what an adventure [21:49] komputes: np, make sure to enjoy [22:10] hi. I am taking an OS class in university, we did some kernel manipulation( adding a system call), after installing the new kernel, for every boot it says " starting up Uncompressing Linux... Ok, booting the kernel". does that means any problem, any way to avoid this uncompressing stage? [22:15] MTecknology: I'm guessing it's one of the first three here: http://www.google.com/search?q=gentoo+install+manual [22:16] savasci: it's normal, and you want to leave it that way. The kernel in memory has a lot of zero-filled data, and compressing it makes it a lot smaller on disk [22:16] http://www.ibm.com/developerworks/library/l-linuxboot/index.html [22:57] hi. where can i find out more about ubuntu dropping support for things below i686? [22:57] it seems the lubuntu folk are pretty upset about this. === BenC__ is now known as BenC [23:56] ogasawara, just responded on that e-mail you forwarded [23:56] sorry for overlooking that :-/ [23:56] JFo: no worries. not sure how much help he'll turn out to be. [23:57] every bit helps :)