[02:41] amitk: hey === dhaval is now known as dhaval_away === dhaval_away is now known as dhaval [05:08] ssbar [05:08] bar [05:08] eek === dhaval is now known as dhaval_away === dhaval_away is now known as dhaval === dhaval is now known as dhaval_away === dhaval_away is now known as dhaval [11:11] hi, when do you merge 2.6.24.6? [11:12] 2.6.25.1 is the interesting target :D [11:13] i mean for hardy [11:13] security issues if you dont know [11:14] http://kernel.ubuntu.com/git?p=ubuntu/ubuntu-intrepid.git;a=commitdiff;h=c1838f1600cc7196363a23da1f76869600e17da1 [11:14] can you show me where you see 0x1430, 0x4748 in upstream? [11:15] Kano: check the kernel-team mailing list. smb posted a list of patches that are being considered from the stable series [11:19] with a pure 2.6.25 it definitely does not work [11:21] Kano: I am not sure, wait for rtg to come online or ask on the mailing list [11:21] well i can compile your git code too, then i know it, but it did not work with standard 2.6.25 [11:22] will you revert the usb gpl only code? [11:27] or do you just drop the usb modules from avm? === gnomefre1k is now known as gnomefreak [15:33] tjaalton: I just upload LRM -17.35 with a simple ABI bump. [15:39] rtg: ok, cool [15:58] tjaalton: you have a minute? [15:58] tjaalton: is there a way for X to work around nvidia crashing X? [16:01] tjaalton: speaking of LRM, can you build a package for Intrepid? See http://ppa.launchpad.net/kernel-ppa/ubuntu for headers and stuff. I'm going to announce the Intrepid daily build pretty soon. [16:02] rtg: can new drivers be added to .1? [16:03] I am thinking specifically about ov51x-jpeg instead of the ov511 that we have in LUM [16:05] ...even with the evil in-kernel jpeg [16:05] compression [16:08] rtg: no problem, I'll probably have time to do it later this evening [16:08] gnomefreak: dunno, how does it crash? [16:09] tjaalton: scrolling on a page well 1 confirmed page and all crashes have 8600 or 8800 nvidia card with our drivers [16:09] tjaalton: see bug 224966 [16:09] tjaalton: im wondering if xul is involed atm but i will ping asac about that later when i have something to base it on [16:10] noone can give me a backtrace just a couple of errors [16:11] gnomefreak: I have a 8600GT.. :) [16:12] tjaalton: can you please get me one lol [16:12] they are cheap ;) [16:14] amitk: perhaps. it is going to change current jpeg compression behavior in the general case? [16:17] rtg: hi, does anybody want to add 2.6.26.6 to hardy kernel? [16:18] 2.6.26? [16:18] gnomefreak: yep, crashes here too [16:18] Kano: only on a case by case basis. [16:18] err 2.6.24.6 [16:18] Kano: I knew what yuou meant :) [16:18] tjaalton: thought it would something with nvidia drivers have never been good with 6200 and up cards [16:18] can you tell me how to merge that correctly? when i get an error that it cant be merged? [16:19] Kano: well, you have to resolve the merge issues. git pretty much tells you what to do when the merge fails. [16:20] well then the file hals ===== commented parts, you mean those i guess [16:21] Kano: right. Start your search for '<<<', which is where the offending entries begin. [16:21] well then i commited the change, i guess i made it right then [16:22] for .5 there was only 1 wrong file [16:22] did not try .6 yet [16:22] gnomefreak: it doesn't crash here (Geforce 7300) however I'm using my own (customised) packages based on the lrm [16:23] rtg: will you revert the usb gpl only patch? [16:23] from 2.6.25 [16:24] Kano: I do not know about this issue. [16:24] try to compile avm usb modules [16:25] like fwlan [16:25] you have got those in lrm [16:25] and every other avm usb module [16:26] Kano: I haven't the foggiest idea what you are talking about. To my knowledge the Hardy LRM has no build issues. [16:26] rtg: 2.6.25 is in your intrepid git [16:27] Kano: for which we don yet have an LRM package. [16:27] well does not matter, the drivers will not compile [16:27] tjaalton: is just now starting to think about LRM for Intrepid. [16:28] for testing i added a patch for fglrx 2.6.25 (x64) and a revert patch for usb [16:31] Kano: you're on your own there. Its gonna be July before I get really interested in Intrepid problems. [16:32] using fglrx is stupid enough [16:33] maks_: no flame-baits, please [16:47] rtg: also it seems that the dmraid in lum does not work, at least Rabiddog has problems [16:48] the module is there but it is not recognized [16:49] Kano: which lum? Hardy or Intrepid? [16:49] hardy [16:49] i would like to test it myself, but i only have got 2 similar drives, not 3 [18:03] rtg: http://lkml.org/lkml/2008/5/2/205 [18:03] will be needed for the server kernel [18:04] mdomsch: backported for Hardy I assume? [18:04] rtg, yeah [18:05] soon as ingo pushes it to linus for .26 and -stable picks it up [18:05] mdomsch: I'm working on SRUs, so this is a good time to git 'er done. [18:06] I'm filing a bug in LP now [18:06] mdomsch: I'll watch the .7 stable updates for this (when they appear). [18:06] to track [18:06] hmm, I read on lwn that there wouldn't be .7 [18:06] or did I misread? [18:06] mdomsch: oh, thats right. [18:07] mdomsch: well, even if it goes into 25.y it ought to work. [18:07] mdomsch: lemme know the LP number and I'll assign myself to it so I don't forget. [18:09] Is CONFIG_VERSION_SIGNATURE an ubuntu specific config option? [18:09] SEJeff: yeah, I think so. [18:10] rtg, Ok so in Re: the btrfs failing miserably on ubuntu thread... could we do like SUSE and use CONFIG_VERSION_SIGNATURE instead of CONFIG_SUSE? [18:11] SEJeff: that doesn't mean the AA patches exist. Intrepid won't get them for awhile. I still think the CONFIG search is the best approach. [18:12] rtg, LP 225811 [18:12] Ok so how about an explicit test for the hardy kernel version? If it is a user built kernel, they will know enough to apply a patch manually? [18:14] SEJeff: I don't spend a lot of time supporting user built kernels. Its not that I don't care, I just don't have time. So, if they know enough to apply the patches, then more power to them. Otherwise, they are on their own. [18:14] rtg, something like: http://pastebin.com/m64b6a4dc [18:15] rtg, That was my point. You are agreeing with me. If it is a stock ubuntu hardy kernel, that would catch it. A stock hardy kernel has apparmor applied [18:15] SEJeff: that definitely works for Hardy. [18:16] Ok well then I'll respond to that thread, Chris can put that in btrfs, and we can happily test the next get linux filesystem. Thanks [18:16] SEJeff: np. [18:18] rtg, One last thing. Would it be a better idea to do a version check for each official ubuntu kernel with apparmor? [18:22] SEJeff: I can't remember when the AA patches appeared. with Feisty? You know, I'm still not sure why CONFIG_SECURITY_APPARMOR isn't sufficient? If the AA patches have been applied, then the Ubuntu kernel won't compile unless CONFIG_SECURITY_APPARMOR=y. So, I can guarantee the AA code is compiled. [18:23] Is that because it is such an invasive patch and doesn't ifdef everything out? [18:24] SEJeff: pretty much. I think remove_suid() is a good example. The prototype is changed, but not guarded by an AA ifdef. [18:25] SEJeff: it would take long to verify that for sure. [18:25] s/wwould/wouldn't/ [18:25] rtg, Alright, I'll do a few compile tests with CONFIG_SECURITY_APPARMOR disabled on hardy kernels [18:25] rtg, Thanks [18:25] SEJeff: ok, let me know if I'm completely wrong. [18:52] rtg, compile test without APPARMOR builds, tried it twice. make -j8 is a great thing [18:52] It gives some nasty warnings so I'm not sure how well it would work [18:56] re: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/220493 [18:56] does anyone know what changed from kernel 2.6.24-12 to 2.6.24-16 that affected dmraid? [19:00] SEJeff: drat. All I can say is that Ubuntu isn't gonna apply the AA patches without also compiling them. I think its safe to assume CONFIG_SECURITY_APPARMOR will always be 'y' [19:00] And upstream will likely also agree with that. It is a safe assumption to assume the user is smart if the case is otherwise [19:01] SEJeff: The user best be smart :) [19:01] The following error "ERROR: device-mapper target type "raid45" not in kernel" is occuring [19:02] Rabiddog: Might be just a naming problem [19:02] rtg: do you recall which bug number is about the key stuck scheduler related bug? [19:03] i'm seeing 188226 but that doesn't appear to be the one i saw before [19:03] Rabiddog: I just got a PPA ready for testing https://launchpad.net/~stefan-bader-canonical/+archive [19:03] smb, any idea how to fix it? my raid array is no longer recognized, in gutsy and buddy of mine compiled a new kernel for me [19:03] k looking at that [19:04] PPA stand for what? [19:04] Rabiddog: Personal Package Archive [19:04] I see...does it have a fix for that in it? [19:05] Rabiddog: IOW someplace to put stuff to try. I hope so. MAinly I just made sure the module is named in a more standard way [19:05] calc: I think it was this one: http://kernel.ubuntu.com/git?p=ubuntu/ubuntu-hardy.git;a=commit;h=4b7c68904bf9ada8c4770ce5927e8ec71769ed92 [19:05] Well I'm willing to test, I'll add those sources to apt [19:05] * Rabiddog flips screens [19:05] rtg: yea :) [19:06] bug 218516 was the one i saw before :) [19:06] Rabiddog: Great. Let me know whether this helps [19:06] a user was mentioning they were bitten by what sounds like that bug [19:06] gimme a few gotta manually copy the sources [19:06] rtg: is it fixed in hardy-proposed/hardy-updates yet? [19:06] calc: its an SRU in -proposed. I'll upload -meta in a bit (which should complete the whole ABI bump process) [19:07] rtg: ok [19:14] smb [19:14] Rabiddog: yo [19:14] inux-ubuntu-modules-2.6.24 - 2.6.24-16.23ubuntu4 <--- is this the package I should upgrade? [19:14] Rabiddog: yes [19:14] k [19:14] sec [19:19] rebooting the box now :) [19:21] The following error "ERROR: device-mapper target type "raid45" not in kernel" is still occuring, smb [19:21] hmm however [19:21] Rabiddog: Thats bad [19:22] Rabiddog: does lsmod show a dm-raid45 or dm_raid45 [19:22] ? [19:22] rtg: LP 225811 if you'd take it please [19:23] sec...it gave the error but my hdd array loaded [19:23] checking lsmod [19:24] Rabiddog: likewise "dmsetup targets" should list it as well. If yes, maybe the msg is just an initial warning before loading the module [19:28] shows dm_raid45 [19:29] is that bad or good, smb [19:30] Rabiddog: I would count is as good. If dmsetup shows dmraid45 as target as well, well then it should work [19:30] sec checking that [19:31] Rabiddog: And one final thing: "dmsetup table" should list something with dmraid45 as well (if that is used) [19:32] dmsetup targets shows just "raid45" [19:32] raid45 as well [19:32] for tables [19:32] Rabiddog: Not linear? [19:32] huh [19:33] Rabiddog: linear is a built in target of device-mapper. Should normally be there [19:33] sec [19:34] yep its there first line [19:34] forgive me I may have compiled kernels but I'm still a bit of a noob :) [19:34] Rabiddog: So that looks good setup wise. No prob. [19:35] Rabiddog: So before, there were the messages and the array was not created and now we still have the message but it is [19:35] ok.... now this adventure broke my nvidia drivers for sum reason....lol [19:36] smb, k [19:36] thanx...is that all that needs testing smb? [19:38] Rabiddog: Hm, strange. I guess one thing. I would put some updates to bud 22ß403. Maybe you could add some coments from your test [19:38] bug 220493 [19:40] doh tat links to bugzilla of mozilla [19:41] Rabiddog: launchpad. https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bug/220493 [19:43] k as soon as I get my video drivers working, dang thing is launching a bad video mode when I log into gnome [19:43] can't figure out how to delete the default setting [19:49] * Rabiddog sighs and runs off to another channel to figure out his next issue [20:52] smb: I added a comment https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bug/220493/comments/4 [20:55] Rabiddog: Ok, thanks. Hopefully this helps others as well. And then maybe that qualifies as simple quirk... Let's see [20:56] k [20:58] btw. the dmraid should be patched too [20:59] in 3 positions [20:59] init script, then the module should be added to initramfs, then it should be loaded in initramfs [21:02] looks at smb [21:02] Kano: That would be a different story (domain). [21:04] heh, I forgot I still have have modprobe -Q dm-raid4-5 in my initramfs script [21:05] Rabiddog: That was the old name. Should not do anything now [21:05] ah k === thegodfather is now known as fabbione [21:07] bah dc ftw [21:13] hmm === Rabiddog_Biteme is now known as Rabiddog [21:13] hmm [21:14] woot :) [21:14] thanx for your help smb [21:14] Rabiddog: ur welcome :) [21:15] and now u know it works except for the quirk [21:18] Rabiddog: Yup. Thanks for trying [21:21] btw. the dm name is correct [21:21] no need to change that [21:22] dm-raid4-5 is the name from the official patch too [21:22] Kano: But that was the only change I made between not working and working. [21:23] it never had a differnet name.. [21:23] Kano: Most other modules keep their names close to the target name. So autoload works. [21:24] well you need to put it in the initrd anyway [21:24] or you need to use another hd for booting [21:25] Kano: sure. Though I don't see this likely for Hardy after it is now released. [21:27] maybe fast enough for a service relase [21:27] as you want to support it serveral years ;) [21:27] * Rabiddog eyes the gnome Ui windows bug that cropping up [21:28] btw. the lum for intrepid does not compile, missing unionfs include [21:28] * Rabiddog eyes the hal mod and its bugs [21:29] can i copy the aufs folder to the hardy lum? [21:32] kano whats is lum btw? [21:34] also it was 100% wrong to drop my patch for the guitar, it does not work with your intrepid git [21:34] nothing in upstream [21:35] at least not the guitar [22:05] guitar? [22:05] in kernel? [22:05] the xbox guitar [22:06] Kano: why isn't it in the kernel.org? [22:06] http://kernel.ubuntu.com/git?p=ubuntu/ubuntu-intrepid.git;a=commit;h=c1838f1600cc7196363a23da1f76869600e17da1 [22:06] thats my patch [22:06] that seems like a simple patch to pass upstream [22:06] and the guitar does not work with 2.6.25 [22:07] 2.6.25 would work with the extra identifier though? [22:08] did not try yet, i only added some other patches to fix build problems [22:08] tested it before with the debian trunk kernel [22:08] I'm just saying if it works, you might as well get it included in kernel.org [22:09] its The Right Thing To Do [22:09] well i made it for 2.6.24 [22:09] and there it works [22:09] its a two line patch that almost certainly didn't change [22:10] it is cool to play frets on fire with that guitar [22:10] sure [22:11] but wherever possible I think the kernel team should be providing changes upstream as well [22:11] and this is... very possible I think [22:13] why is alsaconf removed in ubuntu? [22:15] wait [22:15] kano [22:15] it IS in linus's tree [22:15] Kano: known to break audio configuration and/or freeze the machine. [22:15] http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=blob;f=drivers/input/joystick/xpad.c;h=b29e3affb805a97126ccee39dcc867cd6e40dfe3;hb=b66e1f11ebc429569a3784aaf64123633d9e3ed1 [22:15] crimsun: well i need that [22:16] Kano: the correct method is to fix the ISA drivers. [22:16] crimsun: i need it to unload drivers when they have been loaded in wrong order [22:16] Kano: err... [22:17] you should never have to use alsaconf to reorder indices [22:17] so what do you use [22:18] you can either use the index module parameter for each ALSA driver, or you can use the slots parameter for snd.ko (requires alsa-kernel >=1.0.16) [22:18] options snd-emu10k1 index=0 [22:18] you'd place them in a modprobe conffile, e.g., options snd slots=snd-usb-audio,snd-emu10k1,snd-hda-intel [22:19] when this is set it still loads the other as 0 [22:19] that's why I recommend using snd.ko's slots parameter instead of each driver's index parameter. [22:20] but as hotfix alsaconf is really good to use [22:20] with slots, index 0 is reserved for snd-usb-audio, index 1 for snd-emu10k1, index 2 for snd-hda-intel [22:21] there's no real reason to use alsaconf for that. One can script the above sequence for an arbitrary set. [22:21] how about putting it in an extra package for those who really want it? [22:22] I would strongly recommend against doing so, but I certainly won't stand in the way of other core-devs changing debian/rules. [22:23] blah, I'll just write a script to do what you need [22:23] (and stash it in alsa-utils) [22:24] lol [22:25] crimsun, I saw your tip on the bug report for me I'll try it later [22:25] Rabiddog: err, sorry, which bug report? [22:25] the dmraid [22:26] regression [22:26] Rabiddog: hmm, I don't remember commenting... Which bug #? [22:26] regession is the wrong word as i patched it before for you ;) [22:26] lol [22:28] https://bugs.launchpad.net/ubuntu/+source/linux/+bug/220493 crimsun if crimson fox is u [22:29] Rabiddog: no, I'm strictly crimsun and/or appended underscore variants, sorry. [22:29] heh [22:29] there are far too few bugs associated with crimson-fox for that ;) [22:31] heh [22:40] woot samba fixed [23:54] Steve Langasek in here? [23:55] mkrufky: slangasek is in #ubuntu-motu for example [23:55] thank you