=== rikai_ is now known as rikai === macd_ is now known as macd === emma_ is now known as emma [06:18] hello bug about amilo resolved the amd74xx bug was due to CONFIG_IDEPCI_SHARE_IRQ to NO [06:19] you have to set it to no [06:19] and cdrom is ok [06:19] how can ia do a simple patch for ? === superm1 is now known as superm1|away === maks is now known as maks_ [07:35] hi [07:47] benje: first find the bug at launchpad, then add the info there [09:45] did somebody try that next kernel? i updated it to kernel git but still rt2x00 is broken and i get dmi bios error that it wants acpi=force [10:35] which alsa is in intrepid? [10:36] please say 1.0.17 ;) [10:36] thats in 2.6.27, in 2.6.26 is 1.0.16 [11:38] Kano: but afaics ubuntu uses a separate ALSA from the kernel anyway, 'cos it's in LUM [11:38] Ng: only 2.6.24 [11:38] there is no lum for 2.6.26/7 [11:38] oh [11:38] bah [11:39] 1.0.17 has a bunch of new hardware support :/ [11:39] then delete buildin alsa and compile it externally [11:39] modprobe -l|awk /snd-/|xargs rm -f [11:39] then the kernel ones are gone away, compile alsa manually [11:40] I already rebuild LUM with 1.0.17 for my laptop, but there are plenty of users for whom that isn't a realistic option :) [11:40] perhaps lbm could contain 1.0.17? [11:49] https://lists.ubuntu.com/archives/kernel-team/2008-August/002867.html [11:50] aha [11:50] I'll wait for ben to wake up then :) [14:22] TheMuso: your zinc account is done [14:22] Ng: ? [14:22] BenC: aha :) [14:22] BenC: I was wondering about the possibilities of alsa 1.0.17 getting into intrepid [14:22] my laptop is entirely silent with .16, as are some others aiui [14:22] Ng: it's highly possible....not for Alpha4, but shortly after [14:23] Ng: is that a regression? [14:23] no, it was never supported in hardy, it required new code in hda_intel to work [14:23] I've been building my own LUM with 1.0.16 removed and 1.0.17 put in ;) [14:25] I've encountered at least one other person with newer hardware than .16 supports, so my guess is that several hda_intel variants are new enough to only be in .17 [14:25] obviously this will always be the case, but if we can sanely do it, then I'd be very happy to see it :) [14:27] Ng: Can you test 2.6.27-rc2 deb's from http://kernel.ubuntu.com/pub/next/2.6.27-rc2/ ? [14:27] Ng: It's got LUM built-in (like intrepid) [14:29] sure [14:30] should it JustWork with hardy userspace? [14:30] BenC: Thanks. [14:31] Ng: I'm actually going to be looking at backporting important driver fixes/updates from 1.0.17 to hardy's 1.0.16 alsa-driver code. [14:32] TheMuso: aha, well I'll probably be running Intrepid in the next month if I stick to my usual schedule, but I'm sure there are people who will be sticking with hardy for a lot longer who'd appreciate something like that [14:33] Ng: Yeah, supporting newer sound hardware for hardy, at least for a while, is something that IMo needs to be done. [14:34] TheMuso: well my suggestion would be to start with hda_intel - my impression is that that covers more current hardware than any other driver. I could be wrong, that's just my guess ;) [14:34] Ng: Yes thats one of the ones being considered. [14:34] :) [14:35] Ng: it's a good guess :) [14:35] Almost anything coming out today is based on HDA, whether it's AMD or Intel chipset [14:35] it's a bit weird that there are so many variants though [14:35] IMO weird is an understatement. [14:35] Ng: it's more than weird, it's hugely irritating [14:36] Ng: Multiple different codecs that can be attached, and then vendors can wire them all up in different ways [14:36] Ng: the main thing is that HDA just defines a way to access the codec...it doesn't have a built-in way to define the pin-outs, so they have to be hardcoded :/ [14:36] Oh, there's a definition [14:36] It's just often ignored [14:37] The BIOS is supposed to provide a table [14:37] ahhh. by which I mean "ugh" ;) [14:37] mjg59: that's funny...I assumed it didn't because more often than not, new hardware doesn't work out-of-the-box [14:38] I think Linux has occasionally been poor at using the BIOS information, but also vendors are poor at putting anything in there [14:38] The Windows .inf files usually contain the pinouts [14:38] What /would/ be nice would be some way of using the .inf to teach the kernel [14:39] mjg59: Tim and I talked about doing that, but never got around to looking at it [14:39] sounds like something Wubi would be able to harvest pre-install [14:39] Yeah [14:39] but useless to those of us who stab windows in the face when deboxing a laptop ;) [14:39] If you've got the driver CD, there's still hope [14:40] deflowering I think is the term you were looking for :) [14:40] * TheMuso is personally not a fan of recovery partitions. [14:41] yeah in this laptop's case there were no CDs in the box, it's pure recovery partition, but naturally I stabbed that in the face too. I'm probably an atypical user though, and I like the idea of being able to steal information from .inf files more generally :) === superm1|away is now known as superm1 [15:24] Ng: let me know if that 2.6.27-rc2 kernel works for you...I plan on using sound/ out of that for intrepid/2.6.26 kernel [15:24] BenC: will do, I'll give it a try when I get home [15:28] BenC: where is the root partition remounted? Is that done in initramfs? In particular, I'm curious how file system options are applied after the initial mount. [15:30] rtg: remount is done in init scripts from the root partition [15:31] BenC: checkroot.sh ? [15:39] rtg: looks right [16:22] rtg, BenC: Sorry to bother you. Are there some chance to have an linux-restricted-modules-source package (as i requested in https://lists.ubuntu.com/archives/kernel-team/2008-August/002871.html)? [16:23] abogani: we're considering making it spit out dkms packages instead, which will get it working for any kernel installed (including -rt) === asac_ is now known as asac [16:25] BenC, Wow! it would be great! One less package to handle! :-) [17:17] <_gAri-> hi there, can you please help me out where can I find the ubuntu specific kernel patch in peaces that is applied to the vanilla kernel? I mean I only want to use parts of it, not totally [17:18] _gAri-: It doesn't exist [17:18] _gAri-: If you download the Ubuntu kernel git tree, you can extract individual patches from it with git [17:41] <_gAri-> can you give me a clue how to do that? [17:41] <_gAri-> I'm not a kernel developper, I just want to patch a vanilla kernel with the apparmor version ubuntu uses, and PAX or grsecurity. [17:42] <_gAri-> actually if I patch a vanilla kernel with apparmor myself, it says that the profiles are malformed or something like this [18:19] has anyone looked at bug 255635? [18:19] https://bugs.edge.launchpad.net/ubuntu/intrepid/+source/linux/+bug/255635 [18:43] superm1: Do you have any ideas about bug 257003? [18:52] bdmurray, yeah its been addressed in the latest git tree [18:53] bdmurray, i was holding off doing an upload as it should just be in the next driver release though [18:54] superm1: what can I do in the mean time? [18:54] bdmurray, manually remove any symlinks in /usr/lib32/fglrx [18:54] make sure to only remove the symlinks though [18:55] Where should I ask about m-a kernel source package uploads? Noticed that drbd8 module source isn't in hardy/intrepid [18:57] scratch that, it's in ubuntu-modules. Sigh. [18:59] mjg59: would you agree that if dsilvers tlsup driver works as advertised, we don't need any of the tosh workarounds in acpi-support? [18:59] His driver compiles, and seems to do what he said it does (works out-of-box with no /dev/ or /proc/ stuff to mess around with) [19:00] has hotkeys going through input-devpoll [19:00] real backlight driver, etc. [21:48] ogasawara: Is it possible to have a count of current/carried-forward/fixed on intrepid-buglist.html? [21:49] BenC: yup, I can just add it to the page [21:49] ogasawara: thanks [21:49] ogasawara: looks like we are really backed up [21:49] BenC, iwlagn (using intel 4965 agn) looks for iwlwifi-4965-2.ucode in the 2.6.27 kernel.. where do i get the firmware? === hyperair_ is now known as hyperair [21:50] hyperair: I need to upload latest lrm to intrepid...then you can get it from linux-restricted-modules-common [21:51] BenC, i see. thanks. any chance of it being backported to hardy? [21:52] hyperair: You can install the lrm-common from intrepid to hardy with no problems [21:53] i see [21:53] what about the kernel itself? [21:53] i got it from kernels.ubuntu.com or soemthing of that sort [21:54] make that kernel.ubuntu.com [21:55] hyperair: what about it? [21:55] any chance of it entering backports? [21:55] i'm looking specifically at this kernel because it's got .17 alsa drivers [21:55] and anything below that doesn't support lenovo y410 very well [21:56] BenC, When i could know if lrm switch (or no) to dkms? [21:56] hyperair: not to hardy, no [21:56] hyperair: we will likely have .17 alsa in intrepid though [21:56] abogani: it will switch, hopefully by alpha5 [21:56] hyperair: lrm for intrepid just uploaded, so give it a few hours [21:57] BenC, Is it already definitive? [21:58] hello. [21:59] abogani: it was already decided awhile back, I just haven't switched it out yet [21:59] BenC: BTW, i showed the intrepid kernel source tree to the aufs maintainer and he said he can't support it because some files are missing ;) [22:00] BenC, Ok. thanks! [22:00] BenC, thanks. [22:01] laga: what files? [22:03] BenC: stuff like Kconfig and Makefile. i'm having trouble with NFS right now, but i think i know what's missing [22:03] laga: lol...there is a Makefile/Kconfig, else the damn thing wouldn't build :) [22:04] laga: There was some stuff related to NFS in there, but it required ugly patches to the kernel, and I wasn't prepared to apply them (unless aufs maintainer is prepared to support those patches in case they break anything else) [22:05] indeed, but they are truncated. [22:05] laga: truncated to just what we need...that has nothing to do with the code [22:09] BenC: ugly patches? it should require the following patches: [22:09] http://kernel.ubuntu.com/git?p=ubuntu/ubuntu-intrepid.git;a=blobdiff;f=fs/file_table.c;h=0b6dbac60adda31cb1ba57c93a46c2cad71e7a1d;hp=83084225b4c3b3195520d5a7a441360046d6f230;hb=3b3cff7d20e3a13b89f77e641362170000ffc5c0;hpb=952058fa3a0ea802fb5e958a69119aeb01aae7a9 [22:09] http://kernel.ubuntu.com/git?p=ubuntu/ubuntu-intrepid.git;a=blobdiff;f=fs/namei.c;h=df4bd0d1515eb139648088a194e1311bc3dd4dc0;hp=01e67dddcc3d2033a2038c37f9f791ba0921a781;hb=3b3cff7d20e3a13b89f77e641362170000ffc5c0;hpb=952058fa3a0ea802fb5e958a69119aeb01aae7a9 [22:09] http://kernel.ubuntu.com/git?p=ubuntu/ubuntu-intrepid.git;a=blobdiff;f=include/linux/namei.h;h=33c7f23f716ef1009158ffe01690356e11a2fe3e;hp=24d88e98a62648180a553fdfb457837bcbcb5c00;hb=3b3cff7d20e3a13b89f77e641362170000ffc5c0;hpb=952058fa3a0ea802fb5e958a69119aeb01aae7a9 [22:09] Mythbuntu really needs these patches :) [22:09] it's mostly exported symbols [22:12] laga: Actually, it shouldn't even need that if we compile aufs into the kernel (which I would rather do) [22:15] okay, then i'm going to edit the config, rebuild the kernel and try again. [22:18] can i just edit debian/config/amd64/config or is that bad karma? === macd_ is now known as macd [23:20] laga: that should work [23:21] okay. [23:21] interesting, make distclean deleted debian/ [23:23] I am looking for someone to sponsor bug #192772, the debdiff is 6 weeks old, and another freeze is coming... [23:24] the same happened with the same package in Hardy, please don't let it happen again [23:30] tormod: I'll look into it [23:30] thanks! [23:39] BenC: Yup. [23:53] BenC: is there any reason not to restore Kconfig for aufs? it looks like it's doing some magic to set invisible config items like AUFS_BR_NFS. but i could also just hardcode these in config [23:56] laga: just add the ones we care about [23:56] The rest are pointless clutter if we aren't actually using them