[00:33] apw or ogasawara. why was the ndiswrapper patch pulled from 2.6.36 buil\ds? [00:34] jewsucanuse: dunno, I myself don't touch the 2.6.36 mainline builds. [00:34] jewsucanuse: and I suspect apw is already sleeping as it's almost 1am his time [00:34] not the mainline, the natty builds. [00:34] jewsucanuse: natty builds? [00:35] yah. [00:35] 2.6.36-0.2 [00:35] heh, I knew the natty tree was open, didn't know we were building kernels yet [00:35] jewsucanuse: tgardner put the initial natty tree together so would have to ask him [00:35] I don't think we are officially yet [00:36] * [Config] Disable aufs, dmraid-4.5, ndis-wrapper [00:36] tim gardner [00:36] https://launchpad.net/ubuntu/+source/linux/2.6.36-0.1 [00:36] right none of that is surprising, we will revisit those at uds [00:36] jewsucanuse: unfortunately tgardner is away on holiday for the next week, but I recall he sent an email saying he'd disabled a few items because they possibly broke the build? [00:36] jewsucanuse, that's usually because they don't apply cleanly right now and need some love, so rather than hold everything up they were disabled [00:37] okay. i see. [00:37] I know we are looking at overlay fs to replace aufs, whether that happens, well we will have to wait and see [00:37] jewsucanuse, what ogasawara said :-) === jjohansen is now known as jj-afk [01:03] back on later === bjf is now known as bjf[afk] === ivoks-afk is now known as ivoks === ivoks is now known as ivoks-afk [08:17] hi, i have writtent a program client program to find the load ... 'clnt' .. when i do 'strace clnt address', it says "strace: clnt: command not found" ... i have already compiled including debugging information [g3] .. [08:17] anything special to do , when we want to use strace ?? [08:18] Is clnt in $PATH? You probably need to run it as “strace ./clnt address” [08:19] RAOF, perfect ... that was the problem .. thanks :-) === smb` is now known as smb [08:25] smb: I notice that Maxim has been CC'ed in that 'patch' [08:25] smb [08:26] smb: The answer is, yes the kernel will now hang again [08:26] lag, Joy [08:27] Maybe one should intervene to that thread then... at least bring up the issue [08:28] I will [08:28] Next on my TODO list [08:29] lag, cheers [08:38] smb: Okay, their changes 'should' not re-introduce the bug [08:39] smb: They are not _moving_ the code back, they are _duplicating_ it [08:39] smb: The code in pm_notify will still be doing the removing [08:39] ah, because it is done first? [08:40] Yeah [08:40] Ok, ugly but may work [08:40] pm_notify runs _before_ any suspend functions [08:40] Yeah, it is ugly [08:40] lag, Thanks for checking this [08:40] smb: np [09:11] smb: http://paste.ubuntu.com/513678/ [09:15] I've narrowed down the firewire issue to being a regression between kernels, and not an inherent problem with the new firewire stack [09:16] now busy with a full set of tests and reporting back to Stefan Richter for more discussion [09:17] sounds to be on a good track then. :) [09:19] how about the other regression though, I found bug 660315 [09:19] Launchpad bug 660315 in linux (Ubuntu) "U232-P9 USB Serial adapter not working in Ubuntu 10.10 (affects: 1) (heat: 6)" [Undecided,New] https://launchpad.net/bugs/660315 [09:19] works in 2.6.32-24-generic, broken in 2.6.35-22-generic [09:20] mct_u232 module... how to know if something is changed between then and now? [09:20] I'd like to do bissection and know what change broke it but I don't know how to do that [09:21] Its possible to do a rough initial approach, using the mainline kernels [09:21] http://kernel.ubuntu.com/~kernel-ppa/mainline/ [09:22] That has builds for 2.6.33, .34 and .35 (and the rc versions too) [09:22] great === diwic is now known as diwic_afk === diwic_afk is now known as diwic === _LibertyZero is now known as LibertyZero === amitk is now known as amitk-afk === amitk-afk is now known as amitk [13:51] apw: have you ever faced this problem with git? https://pastebin.canonical.com/38684/ [14:44] smb, hmm, there is a lbm-alsa-distro-flavour package, are you saying it doesn't build properly? [14:45] diwic, no I am saying it would probably not have build correctly if it had been enabled. But it has not been enabled yet, so nothing to worry about [14:46] I just thought to fix the typo while I saw it [14:47] smb, are you saying this package doesn't exist? http://packages.ubuntu.com/lucid/linux-backports-modules-alsa-lucid-generic [14:48] diwic, I assumed so because the comment sayd "enable alsa after release" and do_alsa is set to false [14:49] smb, then you're very likely wrong ;-) might want to talk about it with bjf[afk] as he built lbm recently [14:50] diwic, Oh, we are talking about *Maverick* btw ;) [14:51] (sorry did not notice lucid in your link above, otherwise I could have told before) [14:53] smb, hmm, https://launchpad.net/ubuntu/+source/linux-backports-modules-2.6.35 <- it still exists there [14:54] Right, that would be the source package. Which produces (or does not produce) binary packages with extended names like net or input or alsa [14:57] smb, I'm just confused that things are building although there are typos. But as long as you have everything under control, I'm satisfied :-) [14:57] i e there *were* typos. [14:59] I agree about the potential of confusingness. :) It should be under control. As I said the expected effect of this would have been someone flipping the switch to "now build an lbm-alsa" and then going "heck, why does it still not build" === bjf[afk] is now known as bjf === ivoks-afk is now known as ivoks [15:39] jj-afk: ttx suggested i ask you for advice on https://blueprints.edge.launchpad.net/ubuntu/+spec/hardware-server-n-unprivileged-mounts - should it be called hardware-server or hardware-kernel (or other-server) ? [15:47] http://lkml.org/lkml/2010/10/12/279 [15:48] sweet [15:50] * ogasawara bails for appt, back on in a bit [15:58] bjf, https://launchpad.net/~firmware-testing-team/+archive/ppa-firmware-test-suite-dev [15:58] ^dev version, most current, lots more goodies in it [15:58] cking, thanks [16:03] hi [16:03] what happened to kernel PPA? [16:03] https://launchpad.net/~kernel-ppa/+archive/ppa [16:03] it's empty for some reason [16:03] I wanted to install backported maverick kernel on my lucid machine [16:04] smugglerFlynn, it is located here: http://kernel.ubuntu.com/~kernel-ppa/ [16:04] as there is a size limit on the LP PPA [16:04] which precludes us from using it [16:05] oh, thanks a lot, JFo [16:05] my pleasure smugglerFlynn [16:09] hm, how can I add this repo to apt? (tried "deb http://kernel.ubuntu.com/~kernel-ppa/ lucid main") [16:09] smugglerFlynn, the backports kernel is now in lucid offical repo, in -proposed currently [16:12] apw, you around today, been a bit quiet. [16:14] apw, found it, thanks ^-) === ivoks is now known as ivoks-afk === diwic is now known as diwic_afk === jj-afk is now known as jjohansen [17:11] hallyn: I don't really seeing it called hardware-* as it really doesn't have anything to do with hardware [17:12] hallyn: then again looking at the tracks, and since its about mounting hardware- it is [17:13] hallyn: how much kernel work is involved? I haven't followed user mounts recently === ivoks-afk is now known as ivoks [18:48] switching venues be back on in an hour === bjf is now known as bjf[afk] [20:03] jjohansen: the user mounts is mostly kernel work, but really 'mounting devices' isn't necessarily relevant, bc then you worry about all of the possibly exploitable holes in fs reading code [20:04] jjohansen: so that's why at first bind mounts and sysfs/proc would be the most likely to be allowed [20:04] (and clean tmpfs) [20:04] so, should it be 'other-security-n-unprivileged-mounts' maybe? [20:04] * hallyn shrugs [20:05] hallyn: thats what I was expecting. yeah I think I would do other-security- [20:05] jjohansen: ok, thanks [20:06] it'd rock to get that done :) === ivoks is now known as ivoks-afk [20:46] jjohansen, did you fix an MCA with EC2 recently ? I recall you did something of that sort ... by backporting some patches ... [21:02] manjo: I reverted a patch [21:03] manjo: it was to do with load reporting on the Lucid kernel, and I could have applied the upstream version of Chase's patch or just done a revert, as they work out to the same thing [21:07] jjohansen, :) ack === bjf[afk] is now known as bjf [21:20] is there any way to see how many transactions are queued up in jbd2? [21:21] dpkg has been blocked for several minutes here, and it seems to be because it did a sync() and jbd2 is busy creating a huge number of files [21:22] /proc/fs/jbd2/*/info has some statistics but not the information I'm looking for [21:23] perseus:[/sys/fs/ext4/sda1] cat delayed_allocation_blocks [21:23] 37467 [21:23] and decreasing slowly === yofel_ is now known as yofel [21:37] Is it possible that for some reason options you add to /etc/modprobe/alsa-base.conf are ignored on Maverick? I've got the feeling my system does. [22:02] Aaargh! I wrote 'option' rather than 'options'! That was a waste of one to two hours trying to find a solution... [23:13] sconklin: around? [23:13] yes [23:14] I don't have anything for you [23:14] although if you want to rebase the ec2 branches, I can point you at the branches we have for the master [23:14] sconklin: okay just checking [23:14] sconklin: sure