=== cking is now known as cking-afk === yofel_ is now known as yofel === cking-afk is now known as cking === Guest61100 is now known as ogra [16:36] has anyone already complained about latest natty kernel upgrade with respect to broadcom wireless and nvidia? [16:42] hallyn, Sort of Henrik was running into it (at least broadcom) when he was around Andy [16:43] hallyn, So nvidia has the same wrapper problem? [16:44] smb: don't know if it's the same. on first login it told me i couldn't run unity, but then i coudl run it by hand, and on another attempt to instal lkernel, I got: [16:44] run-parts: executing /etc/kernel/postinst.d/nvidia-common 2.6.37-12-generic /boot/vmlinuz-2.6.37-12-generic [16:45] /etc/kernel/postinst.d/nvidia-common: line 8: [: too many arguments [16:45] sound like nvidia-detector is bogus i guess [16:45] yeah that at least for a starter sounds like something not correctly catched in the script [16:46] oh, haha - i have intel graphics. weird [16:46] how to do a make install for all kernel modules? [16:47] hallyn, So at least that would not be your problem with unity then. ) [16:47] :) [16:47] smb: at any rate, i'm having to tether right now... should i file a bug, or is Henrik on top of it? [16:47] For the broadcom problem afaik he had a patch ready and just needed to bring it on [16:47] well lemme reboot one more time :) [16:47] oh, cool [16:48] thanks [16:52] make install does not install my radeon kernel modulewhat can I do? [18:55] hello. here's something i don’t understand concerning wireless regulatory domains. my regdomain is as such: http://paste.ubuntu.com/552544 but i’m still getting this least common denominator world domain: http://paste.ubuntu.com/552526 [18:56] is this an ubuntu specific thing? on my openwrt i get the officially approved channel configuration. [18:57] I'm getting: "fatal: The remote end hung up unexpectedly" when I try to git a copy of the kernel. Is the git server messed up? Or am I doing something wrong? [18:58] towolf: what wifi hw? [18:58] iwlagn 5000 series [18:58] on natty [18:59] ah hm, dunno; does it have an eeprom that sets the region? cuz those can get in the way [18:59] could it be that the crypto verification of the regdb doesn’t work at all? this is just a hunch though. [19:00] ohsix, shouldn’t be. and if at all it should be overridden. [19:00] if you can see it in iw get it's already been applied i think, you can check in dmesg for crda's/the kernels message that it loaded the regulatory info [19:00] yes, it did apply it, but that isnßt reflected in the channel list. [19:01] read the regulatory compliance stuff on the kernel.org wireless section; generally the regdb complements the eeprom and the compliance [19:01] ohsix, on my openwrt the eeprom says US, but then openwrt overrides that and sets DE. [19:02] they have hacked up stuff :] and i dunno if they even use crda [19:03] yes, they do. and there i can even use /sbin/crda, on natty it fails with -22, which i’ve read is due to failed signature verification. [19:05] hrmph, that sucks [19:07] i wish they'd use openssl instead of gcrypt in the ubuntu package, so you can just drop in your own signing certificate instead of having to rebuild it [19:08] ohsix, how would i do that? and by the way, the regulatory.bin has identical SHA1 on kernel.org, on natty, and in openwrt. somethign else is going awry [19:13] essentially "iw reg set" has no effect, but it doesn’t fail, and that puzzles me. [19:13] try iw reg set US and look in dmesg; it should show it changing the regdomain [19:19] yes. it shows it. but then "iw phy0 list": no change at all. still world regdomain. [19:19] that might be an eeprom thing, crda doesn't change what the card does; it just applies channel policies on top of it [19:19] can i query eeprom to see what’s in there? [19:19] i don't have an iwlwifi device in front of me, but there's usually a tool to poke at the eeprom parameters [19:19] ooh. you might be right and this is iwlwifi specific. just cross-checked with b43 driver. [19:19] and that works. [20:25] <\sh> guys, is there an easy way (which could be done e.g. via break=premount into initramfs) to debug a kernel network module, and why so ever it take ages to initialize? [21:04] apw: any estimate on when bug 693401 will hit hardy-updates? [21:04] Launchpad bug 693401 in linux (Ubuntu Hardy) (and 1 other project) "hardy kernel lacks support for ICH10 controller in Intel Server System SR1600UR (affects: 1) (heat: 238)" [Wishlist,In progress] https://launchpad.net/bugs/693401 [21:07] lamont, am looking into it [21:07] ta [21:14] lamont, it looks like it has been applied to the tree and we are at the beginning of an SRU cycle, it "should" hit -updates in 2 weeks [21:17] awesome [21:17] that's the blocker for 2 reasonably phat ppa builders [21:41] for someone in my loco channel (she /quit), with the latest maverick kernel she said, the kernel didn't recognize her phone when plugged in with a usb cable, and there was no mention of usb dmesg (http://paste.ubuntu.com/552589/). when she tried with the previous kernel it worked. [21:51] anyone know how that could be? [22:06] wrt bug #396286 exactly what is the point of not attempting to backport ext3/ext4 fixes to Lucid? [22:06] Launchpad bug 396286 in linux (Ubuntu) (and 1 other project) "[Geode LX] [ION603] kernels >= 2.6.31 fail to boot [initramfs] (affects: 2) (heat: 15)" [High,Won't fix] https://launchpad.net/bugs/396286 === BenC_ is now known as BenC [22:34] smb: should linux-ec2 in -proposed be promoted to -updates yet? [22:36] kees, I think I did test that ok. So I would promote it [22:37] smb: okay, thanks [22:47] did you guys see my question? what should i ask for the next time she comes online, more than ubuntu-bug linux? i don't think she's the bisecting type… unless the problem persists [22:57] virtuald, what bug are you referring to? [22:58] jfo: no bug# [22:58] then I am not sure to what you are referring [22:58] 22:41 < virtuald> for someone in my loco channel (she /quit), with the latest maverick kernel she said, the kernel didn't recognize her phone when plugged in with a usb cable, and there was no mention of usb dmesg (http://paste.ubuntu.com/552589/). when she tried with the previous kernel it worked. [22:58] 22:51 < virtuald> anyone know how that could be? [22:59] virtuald, not without a bug and the relevant logging (as supplied by 'ubuntu-bug linux)) [22:59] sorry, that first end-parens should have been a ' [22:59] ok [23:00] virtuald, that way we can see what information there is from her attempt to connect the phone [23:00] it is possible there is some error [23:00] * virtuald googles if there's some command line option to turn on usb debugging [23:00] also, if her phone charges from the usb cable, I would be interested to see if it does that when plugged in [23:01] virtuald, it would be helpful as well if she did a lsusb from the command line and attached that data [23:01] 8] [23:02] that lsusb should be done when the phone is attached [23:02] preferably [23:04] * hallyn hopes that the upload to fix broadcom chips will be hitting natty archive soon... [23:06] hallyn, me too :-/ [23:09] hallyn, I am told you should hassle tseliot to make sure it is happening :) [23:10] * hallyn wonders where to find tseliot :) [23:14] JFo: do you have a workaround? [23:17] JFo, why do you unassign people from bugs if you mark them fix released ? that steals their LP karma ... [23:23] then LP karma is broken, becasue I remove them after I set it Fix Released [23:23] hallyn, I don't [23:23] JFo, but why do you remove them at all ? [23:24] because once a bug is 'closed' there shouldn't be an assignee as far as I have been told [23:25] well, its one thing that can turn people down ... its credit thats gone then [23:26] if you are a developer that really cares about karma...... [23:26] if its fix released why would it matter if there is an asignee [23:26] many newcomers in the community do care about karma [23:26] but they are not likely to be the ones fixing any of the kernels bugs [23:27] why ? [23:27] seriously? [23:28] were I to be a developer working on kernel bugs, I can't honestly say that karma would be the reason I would work bugs for Ubuntu. [23:28] why wouldnt a teenie who wants to be a developer not file and fix a kernel bug he finds a patch for in some other distro for example [23:28] karma is a thing to encourage people to contribute [23:29] you and i might not care about it much ... community people do [23:39] I guess I just don't see a Ubuntu Kernel Dev, Ted T'so or some Upstream level developer being all excited because his karma went up a few points. [23:39] I supppose I am hindered by the high level of ability needed [23:40] in order to solve the vast majority of these [23:40] I could be wrong, and to be honest, I have almost no energy around the issue [23:40] I'm just perplexed that it is, in fact, an issue [23:41] I don't mean to imply that karma is useless, far from it [23:41] i'm not talking about a ted t'so :) [23:42] in other packages I am sure there is a bit of energy around it [23:42] I just never have seen some teenager fixing kernel bugs. [23:42] i'm talking about the teenage guy who does one of his first contributions and is excited [23:42] I seriously doubt they are contributing in the kernel here rather than on the main kernel [23:43] but like I said, maybe I am wrong [23:43] well, apparently the kernel team has its own bug policies (emmet just told me) so do as you like, i just noticed it and found it weird [23:43] the bug control team has no policy to unassign closed bug [23:43] s [23:44] yeah, I work waaaaay outside of that [23:44] well, time for a beer