/srv/irclogs.ubuntu.com/2011/04/08/#ubuntu-kernel.txt

Sarvatt_vanhoof: ping00:07
fairuzmorning07:37
=== smb` is now known as smb
=== skaet is now known as skaet_afk
=== yofel_ is now known as yofel
=== diwic is now known as diwic_afk
=== diwic_afk is now known as diwic
=== herton_ is now known as herton
=== _LibertyZero is now known as LibertyZero
tgardnerogasawara, 'just say no' email sent14:17
ogasawaratgardner: ack14:17
ogasawarasforshee: bug 508516, care if I just mark that Won't Fix?14:21
sforsheeogasawara: d'oh, I meant to do that bug guess I forgot. Please do.14:22
ogasawarasforshee: cool, thanks.  just cleaning up some bugs on the release meeting agenda.14:22
* ogasawara back in 2014:29
=== sconklin-gone is now known as sconklin
jan_dHello! I am new to this channel... yesterday I installed & upgraded kubuntu-natty, and I am now running into this bug in a most severe way. As I was under the impression that this but had been fixed in 2.6.38 final, I am now wondering how it still happens in natty (with 2.6.38-8). this is the bug: http://www.mentby.com/Group/linux-kernel/mass-udp-flow-reboot-linux-with-realtek-rtl-8169-gigabit.html 14:43
jan_dhas anybody else noticed this?14:43
jjohansenjan_d: the rrl8192 driver is not very stable, I have been suffering through it this week14:44
jjohansenjan_d: I am not sure I would attribute the errors I am seeing to this specific bug14:45
jan_dok14:46
jan_dI can undestand that14:46
jan_dthat bug report however details my problems, which is why I cited it14:46
jan_din 10.10 I had no such problem, ever14:46
jan_dso Now I was wondering if I should go back to Maverick + Kubuntu-PPA, instead of testing Natty14:47
jan_dwhich would be a pitty, but if there is no immediate (or short-term) solution, then I will have no choice14:47
jan_dluckily, it's not a production machine14:47
jjohansenjan_d: well that is going to depend on if you can live with the bug until it gets fixed14:48
jan_dnormal internet usage does not trigger a crash, but if I want to do image processing or mathematical processing, it's going to crash every 5min14:48
jjohansenjan_d: if you need stability and you aren't getting it with natty and were with maverick then the choice seems obvious14:48
jan_dyep14:49
jan_dok, so I take it that there is no already-existant solution14:49
jan_dWould it make sense to file a bug with (k)ubuntu? Seems like upstream...14:49
jjohansenjan_d: can you boot a maverick livecd/usbkey and double check which driver is being used14:50
jan_dic14:50
jan_dyou thing I should maybe try 8168 instead?14:50
jjohansenjan_d: not atm, I haven't followed through on the bugs to closely yet, but I plan to, as I am experiencing them14:50
jjohansenjan_d: hrmm maybe, and yes you should file a bug14:51
jan_dok, will do when I get home tonight. Thank you for now.14:51
fairuzhey jjohansen: 14:53
jjohansenhi fairuz14:53
fairuzjjohansen: doing good?14:53
jjohansenfairuz: heh, lets just say its been an interesting week.14:56
fairuzjjohansen: good to hear that14:56
ogasawaracooloney: bug 608312, looks like you had some patches to fix this, did they ever get pushed upstream?15:15
cooloneyogasawara: oh, yeah, i posted that patch to upstream15:17
ogasawaracooloney: did it get picked up?15:17
cooloneyogasawara: i don't think so, not much review about that15:18
ogasawaracooloney: so I assume we would need it for Natty?15:18
ogasawaracooloney: if so, I'll leave it in your hands to get it tested and submitted to the kernel-team mailing list.15:20
=== skaet_afk is now known as skaet
cooloneyogasawara: yeah, i will take care of that. 15:23
ogasawaracooloney: awesome, thanks.15:23
=== nik0 is now known as niko
tgardnerbjf, do you know if there a repo for the tests that Q/Q is running?16:42
tgardnerQ/A*16:42
bjftgardner, depends on which tests you are talking about, i know there is for the security tests, for the others I don't know 16:43
bjftgardner, i'm talking to them right now, you want me to ask?16:43
tgardnerbjf, I guess the security tests will do for now16:43
bjftgardner, one sec16:43
tgardnerbjf, no rush16:43
bjftgardner, not sure if that's all the regression tests, but i think the security is in there       bzr branch lp:qa-regression-testing16:48
tgardnerbjf, tahnks16:48
* tgardner has fat fingers this morning. must be all the snow....16:48
bjfsconklin, ^ in case you needed that as well16:49
sconklinbjf: thanks16:49
=== herton is now known as herton_lunch
tgardnerlaunchpad timeouts suck17:32
=== herton_lunch is now known as herton
keesbjf: uhm, what the heck is going on? why did QA test 2.6.35-22.33 instead of 2.6.35-28.50 ?18:03
bjflooking18:03
=== sforshee is now known as sforshee-lunch
bjfkees, nice catch, i missed that, i just asked, they tested the right kernel, https://wiki.ubuntu.com/QATeam/KernelSRU-maverick-2.6.35-28.5018:06
keesbjf: well, I think it's still a kvm bug18:38
bjfkees, you were not able to reproduce it, right ?18:40
keesbjf: correct. also, in those new results, there are no aslr failures. they tested the wrong kernel before, afaict.18:40
keeshggdh: https://wiki.ubuntu.com/QATeam/KernelSRU-maverick-2.6.35-28.50 shows no aslr failures.18:42
hggdhkees: no, it does not. I tried multiple times, last one succeeded18:42
bjfkees, i might buy that there is a kvm bug lurking in there, but looking at the commits that went into that release, it's difficult to see that it was caused by one of them18:43
bjfkees, unless it was your commit :-)18:44
keesbjf: I mean a kvm _host_ bug -- i.e. not the kernel.18:44
keesbjf: we've seen evidence of all kinds of weird rare memory corruption when running under kvm.18:45
keesbjf: it just happens that the aslr collision test is at about the same level of brutality to uncover such host bugs18:46
keesbjf: here's an example of such a kvm host bug: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/69402918:48
bjfkees, yes, sounds like something someone should look into18:49
keeshggdh: how could that have been a typo, the kernel that was tested actually reported itself as -22.18:50
* tgardner --> lunch18:51
keeshggdh: http://reports.qa.ubuntu.com/reports/kernel-sru/home/ubuntu/sru-kernel-test/maverick-2.6.35-22.33-server/kvm-amd64/qrt-kernel-aslr.txt18:51
hggdhkees: a typo running the test, no upgrades were installed18:52
keesbjf: anyway, I don't see any of these issues as needing to hold back the 2.6.35-28.50 kernel18:52
keeshggdh: oooooh, okay18:52
keeshggdh: well, in that regard, the test worked. :) -22 had that flaw. :)18:52
hggdhkees: heh. At least something worked :-)18:52
kees:)18:53
=== sforshee-lunch is now known as sforshee
hggdhBTW, updated kernel SRU tests tarball created & published19:00
* bjf -> lunch19:23
=== bjf is now known as bjf[afk]
vanhoofherton: heya19:48
vanhoofherton: you still around19:48
hertonvanhoof: hi, yes19:48
vanhoofherton: quick q on a bug you chimed in on19:49
vanhoofhttps://bugs.launchpad.net/ubuntu/+source/linux/+bug/72094919:49
vanhoofherton: just noticed this: https://bugzilla.kernel.org/show_bug.cgi?id=27402#c719:49
vanhoofherton: wondering if we should do the same in natty19:50
hertonvanhoof: I think is a sensible thing to do, as ath3k doesn't seems to handle it ok, and by reports should be working with btusb before19:51
vanhoofherton: its awesome it made its way through -stable :D19:51
vanhoofogasawara: sounds like you're already putting together the final natty upload?19:52
ogasawaravanhoof: I am, fixing up some build failures and then will get it packaged19:53
ogasawaravanhoof: is there a last minute patch we need?19:54
vanhoofits not end of the world, but does impact a number of machines destined for 11.04 cert19:54
vanhoofI think it could be SRU'd19:54
slangasekogasawara: could you pull in the patch from bug #750585 as part of this upload?  Just got FFe approved today19:55
vanhoofI just got a list of natty cert failures, oh about 3 hours ago ;)19:55
slangasekit would be a big help for multiarch to have that in the archive19:55
ogasawaravanhoof: if you can get a patch to the list in the next hour or so, I can probably squeeze it in.  otherwise we'll target for SRU.19:56
ogasawaraslangasek: care to send it to the kernel-team@lists.ubuntu.com just so we can get proper Ack's on it?  I'll be sure to pull it in before I upload.19:58
slangasekogasawara: hmm, strange process.. ok :)19:59
vanhoofherton: do you know where gustavo might have submitted the revert? ... I'm not seeing it in linux-next20:01
hertonhmm no idea, let me check here20:02
hertonvanhoof: found it, it is on his tree at git://git.kernel.org/pub/scm/linux/kernel/git/padovan/bluetooth-2.6.git20:03
hertonhttp://git.kernel.org/?p=linux/kernel/git/padovan/bluetooth-2.6.git;a=patch;h=8390679da56a903b916a5d86423fcc872814d99a20:03
vanhoofherton: think its reasonable to pull in for natty since it breaks things?20:05
ogasawaratgardner: I'm gonna disable rts_pstor staging driver on armel and powerpc, it's throwing me multiple build errors20:06
hertonvanhoof: yes, it's a small change for only these devices, and regression20:08
vanhoofherton: mind sending it to k-t?  If you're tied up I can torture manjo :)20:10
hertonnp, I can send20:11
vanhoofherton: awesome, thank you20:12
* vanhoof loves the final day fire drill :)20:13
tgardnerogasawara, ack20:15
tgardnerogasawara, I'm handling the multi-arch patch20:19
ogasawaratgardner: ack20:19
ogasawaratgardner: I just pushed the rts_pstor config change to master-next20:19
tgardnerogasawara, ok, pushed the multi-arch patch. I'll get a test build going on my emerald, but it looks pretty simple.20:22
ogasawaratgardner: ack20:22
vanhoofogasawara: herton sent it along to k-t20:23
vanhoofherton: i appreciate it :)20:23
* vanhoof makes mental beer note20:24
ogasawaravanhoof: thanks, will review it20:24
tgardnerherton, is this from upstream, or from a maintainers repo?20:25
hertontgardner: from one of bluetooth maintainers tree20:25
herton git://git.kernel.org/pub/scm/linux/kernel/git/padovan/bluetooth-2.6.git20:25
=== bjf[afk] is now known as bjf
hertonI forgot to add that in the patch I sent to kernel-team ML20:27
tgardnerherton, vanhoof: so this patch effectively drops support for that part, right?20:27
vanhooftgardner: it drops support for that part, and fixes others as a result20:27
tgardnervanhoof, so you'd consider it a regression fix?20:27
vanhooftgardner: from my point of view definately20:28
vanhooftgardner: we have machines needing 11.04 cert that are broken because of this20:28
tgardnervanhoof, alright.20:28
vanhoofand ive not seen any actual real work come through wanting that part20:28
vanhooftgardner: if its sketchy to you, we can wait for SRU20:28
vanhoofbut it seemed like a quick win to me20:28
vanhooftgardner: i got a laundry list of bugs mid-day today against 11.04 cert :\20:29
tgardnervanhoof, right on time :)20:29
vanhooftgardner: yeah seriously :)20:29
tgardnerherton, do you think we'll need this oneiric, or will upstream solve the issue?20:31
hertontgardner: I expect this to be pulled in for 2.6.39, so will end up automatically on oneiric I think20:31
tgardnerherton, ok, then I won't pick the patch for oneiric20:32
tgardnervanhoof, applied and pushed20:32
vanhooftgardner: herton: thanks guys!20:32
vanhoofi hear the beer is cheap in budapest :)20:33
tgardnervanhoof, $1.80 per pint as near as I can tell20:33
* ogasawara pulls and kicks off new test builds20:33
vanhooftgardner: i think i'm up to at least a 100 bucks w/ you then ;)20:33
tgardnervanhoof, of course the kernel team drinks for free with all the favours we've racked up.20:34
vanhooftgardner: i wonder if there is miller lite in budapest ;)20:34
tgardnerblech!20:34
elmotgardner: so, FWIW, LP are really working hard to eliminate the timeouts, you might want to ping lifeless next week, if you can tell him which pages are timing out, I suspect he'll fix them for you20:46
tgardnerelmo, its seemed transient20:46
tgardnere.g., it happened on several pages within a few minute period, then all started working again.20:47
bjfelmo, lots of timeouts for a number of us today20:47
bjfelmo, not specific to individual pages, also on qastaging 20:48
elmodid you guys happen to get OOPS numbers or remember any of the pages?20:48
elmo(the particular pages + a rough time would probably allow them to find the OOPSes)20:48
tgardnercan't remember for sure, but I think it was the generic 'try again later'20:49
elmohuh, ok20:49
tgardnerelmo, it was especially annoying because LP has been so well behaved lately. whatever you're doing seems to be the right thing despite this momentary outage.20:50
elmotgardner: oh, it's not me, all credit to lifeless and the LP team - but good to hear20:51
bjftgardner, https://bugs.launchpad.net/launchpad/+bugs?field.tag=timeout20:58
tgardnerbjf, whoa, lotta critical bugs there20:59
bjftgardner, looks like they are taking timeouts seriously 21:00
bryceh_tgardner, yeah LP team policy is that all timeout bugs are handled as criticals21:31
bryceh_tgardner, I find pretty much whenever I get one of those "try again later" pages, I copy the OOPS id into a bug report and send it in, and they take care of the rest21:32
bryceh_tgardner, part of the grand scheme as I gather is that they're slowly ratcheting down the timeout duration in order to improve performance21:33
tgardnerbryceh_, I guess they are making progress. its the first timeout in awhile.21:40
kristian-aalborgevening21:40
bryceh_tgardner, yeah things are a lot better21:41
tgardnerogasawara, EOD, I'm off to trudge through the snow. 21:41
ogasawaratgardner: cool.  I'm about to bounce for a bit too, the suns actually out today.21:41
kristian-aalborgI just ran strace and mpc update (because the latter halts"... I got "erestartnohand" (in caps), which seems to be a kernel error?!21:41
ogasawaratgardner: not much else to do for the kernel.  just waiting for some builds to finish, then will upload over the weekend sometime.21:42
tgardnerogasawara, we've about 5 inches since this morning. gonna make getting to the brewery a long walk21:42
tgardnerogasawara, k, I'll keep watch as well.21:42
bryceh_ogasawara, how's the little one?  Is he giving you better sleep finally?21:42
ogasawarabryceh_: he's doing great!  well, not so great on the sleeping part though :)21:43
bryceh_:-)21:43
ogasawarabryceh_: still wakes up about every 2hrs21:43
bryceh_yeah, something I'm not looking forward to for #2...21:44
hallynsmb`: hi, re bug 747090,22:03
hallynsmb`: wondering if i should be going to the kvm mailing list, or if you'd like to do it (or someone else from your team)22:03
hallynnm, mailing :)22:08
=== hallyn is now known as hallyn_afk
=== sconklin is now known as sconklin-gone

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!