/srv/irclogs.ubuntu.com/2010/06/29/#ubuntu-kernel.txt

RAOFogasawara: Hah.  Excellent timing for my email then!  Yeah, I'll wait for them to get back :)00:01
ogasawaraRAOF: cool.  just didn't want you think it was getting ignored.00:01
jjohansen-> back on later00:47
=== kamal is now known as kamal-away
=== KB1JWQ_ is now known as KB1JWQ
=== kraut_ is now known as kraut
bullgard5What is the function of the phy0 process on my Ubuntu 10.04 computer?09:03
smbbullgard5, Just a guess but phy* usually relates to network cards, so I would suspect a certain network driver to create it.09:20
bullgard5smb: I see. Thank you for commenting.09:24
chetnickHello everyone. I would like to add support for few things in kernel without losing current configuration. I am on the default Ubuntu install/kernel right now. How do I do that? Would make oldconfig do the thing?13:31
=== sconklin-gone is now known as sconklin
mpoirierogra ?15:37
bjf##16:07
bjf## Kernel team meeting in two hours16:07
bjf##16:07
manjobjf, the meeting does not show on the canonical cal16:38
manjobjf, does it show for you ? 16:38
bjfmanjo, yes it does16:42
manjobjf, strange16:44
ogrampoirier, whats up ?16:51
mpoirierogra: I had questions but got in touch with Sebastien @ TI - all good.16:53
ograoki16:53
=== cking is now known as cking-afk
cking-afkback in 1517:35
=== cking-afk is now known as cking
bjf##17:47
bjf## Kernel team meeting in 14 minutes17:47
bjf##17:47
=== kamal-away is now known as kamal
ogasawaratgardner, smb, apw:  just fyi , cjwatson  moved the kernel upload permissions to be attached to the ubuntu-kernel-uploaders team.  If this breaks anything (either upload permissions for any of us, or the ability to target bugs to releases), then please let him know18:30
smbogasawara, ok, will do as soon as I notice anything18:31
ogasawaratgardner: you actually might not be affected as you're motu I think18:31
smbogasawara, sconklin, bjf FYI, if yo not already know this: http://people.canonical.com/~ubuntu-archive/pending-sru.html18:32
sconklinnow ~that~ is a handy thing18:32
smbsconklin, If you see a red bug number at the side of a kernel package, run! :)18:33
tgardnersmb, the SRU verification team must hate the kernel packages.18:34
smbtgardner, They surely do. Even with the hundreds of patches hidden in one tracking bug, we cause quite a bit a work...18:35
* bjf has to run an errand will be back asap18:35
=== bjf is now known as bjf[afk]
=== bjf[afk] is now known as bjf
manjocking, around ? 19:23
manjocking, interested in uefi banter?19:23
* tgardner lunches19:25
keessure is nice to have a 64-way builder :)19:25
smbkees, Surely beats my 4-way :-P19:26
manjocking, ?19:32
ckingmanjo, I'm kinda tied up doing some OEM work and putting the kids to bed19:37
jjohansen-> lunch20:07
=== ayan_ is now known as ayan
=== bjf changed the topic of #ubuntu-kernel to: Home: https://wiki.ubuntu.com/Kernel/ || Maverick Kernel Version: 2.6.35 || Ubuntu Kernel Team Meeting - July-06 - 17:00 UTC
shadeslayerhi,is there a way to delay the boot,and to get it to go step by step? 20:39
shadeslayerlike each process is loaded one at a time and not getting it multi threaded20:39
keesshadeslayer: there is not yet.  I really want something like this, though, it's very painful to debug at the moment.21:03
shadeslayer:)21:03
shadeslayerkees: were discussing a problem over at #ubuntu-x,its related to something between the boot and the part where gdm has to start up21:04
shadeslayerX starts up,but g/kdm hangs21:04
keesew21:04
keessmb: okay, so you have your local u-c-t branch, with one commit, and you want to merge the recent upstream u-c-t changes?21:05
smbright. this is my current local head.21:05
smbrevno: 279621:06
smbcommitter: Stefan Bader <smb@maximegalon>21:06
smbbranch nick: ubuntu-cve-tracker21:06
smbtimestamp: Tue 2010-06-29 18:55:13 +020021:06
smbmessage:21:06
smb  Assigned smb to CVE-2008-725621:06
ubot2smb: mm/shmem.c in the Linux kernel before 2.6.28-rc8, when strict overcommit is enabled and CONFIG_SECURITY is disabled, does not properly handle the export of shmemfs objects by knfsd, which allows attackers to cause a denial of service (NULL pointer dereference and knfsd crash) or possibly have unspecified other impact via unknown vectors.  NOTE: this vulnerability exists because of an incomplete fix for CVE-2010-1643. (http://cve.mitre.21:06
smbOh shut up21:06
keesheh21:06
keeswe should train ubot2 to only respond to CVEs if we specifically say !CVE....21:07
smbThat might be helpful21:07
keessmb: alright, so if you do "bzr pull lp:~ubuntu-security/ubuntu-cve-tracker/master/" what happens?21:07
keeswho controls ubot2?21:07
smbbzr: ERROR: These branches have diverged. Use the missing command to see how.  21:08
smbUse the merge command to reconcile them21:08
keesokay21:08
smbkees, maybe someone in IS does?21:08
keesso, now if you do "bzr missing" it should say 1 (yours) and 1 (marc's)21:08
smbThat says nothing as my change is pushed to my remote 21:09
keesoh, sorry, "bzr missing lp:~ubuntu-security/ubuntu-cve-tracker/master/"21:09
smbBasically I bzr branches from lp:~canonical-kernel-team...21:09
* kees nods21:10
* smb waits for bzr21:10
smbok says I am missing one change21:10
shadeslayerkees: btw if you have any idea,head over to ubuntu-x :P21:10
keesshadeslayer: I don't; I had been fighting bug 597075 which caused a much earlier hang21:11
ubot2Launchpad bug 597075 in linux (Ubuntu) (and 1 other project) "2.6.35 hangs at boot due to regression in i915 or intel_agp (affects: 2) (heat: 12)" [High,Triaged] https://launchpad.net/bugs/59707521:11
keessmb: alright, so "bzr merge lp:~ubuntu-security/ubuntu-cve-tracker/master/" should pull down marc's change and apply it.21:11
shadeslayerkees: ow21:11
keessmb: and when you go to commit, you'll make a "merge" commit.21:11
shadeslayerkees: when you have time,bug 59304121:12
ubot2Launchpad bug 593041 in linux (Ubuntu) "New 2.6.35 kernel keeps toggling bluetooth radio and wireless radio (affects: 1) (heat: 192)" [Undecided,Triaged] https://launchpad.net/bugs/59304121:12
shadeslayerits very annoying21:12
smbkees, Well it pulls down the changes, yes and I can commit them, but then there is no evidence about those changes made by him21:12
smbI think21:12
* shadeslayer didnt know it had 192 heat21:12
keessmb: it's visible if you do "bzr log --include-merges"21:14
Kanohi, why does my i5 run with 1200 mhz with lucid?21:15
Kanothen i installed powernowd and it went to 333421:15
Kanotested with load21:15
smbkees, How obvious. :-P I am beginning to feel the urge of apw-like rants...21:16
keessmb: now you know how I felt learning git when I already knew bzr.  :P  *head explodey*21:16
keestgardner: btw, I learned how to do a moving rebase when maverick gets rebased.21:16
keesgit rebase --onto maverick commit-of-old-maverick-head-in-my-maverick-topic-branch my-topic-branch21:17
smbkees, Did you have to concentrate as much as I need to not to type git everytime I mean bzr. :)21:17
keessmb: I still type "bzr" instead of git about 1/4 of the time.  :)21:17
smbkees, hehe21:18
keesthough I'm now getting very angry that "bzr log" doesn't automatically use the $PAGER env var.  :P21:18
smbRight that is very useful. As it is rare that you want to know about the first comit21:18
keesno kidding :)21:19
keesheh, maximegalon.21:20
smbkees, Oh well, I still don't know how to override the local hostname with something sensible. :)21:21
kees~/.bazaar/bazaar.conf21:21
kees[DEFAULT]21:21
keesemail = Kees Cook <kees.cook@canonical.com>21:21
keeslaunchpad_username = kees21:21
smbOhhh, no BZR_EMAIL? :-P But thanks21:21
keesand you can do individual overrides in ~/.bazaar/locations.conf21:22
keeshm, looks like both $EMAIL and $BZR_EMAIL are respected, actually (so says "man bzr")21:22
jaminclooking for some help tracking down the cause of an oddity... changing only the system's kernel version results in libvirt being unable to write to VM disk images unless the images are owned by root.  I've ruled out apparmor interactions as apparmor is uninstalled and the behavior persists21:22
jdstrandjaminc: is this on maverick?21:23
smbkees, Ok, guess i try to implement something of that in my scripts, as they also take some of that21:23
jdstrandjaminc: actually, I was thinking it was something else, but you ruled out apparmor21:23
jamincjdstrand, not on maverick, but I guess I could give it a shot... Lucid's stock kernel works fine, but mainline's 2.6.34 and 2.6.35 exhibit the problem21:24
jaminchad to move to mainline to fix a periodic kernel crash/reboot that appears to be related to the wireless21:24
jamincchatting with the libvirt folks, indications are that the write failure is some sort of kernel issue.  Considering that simply booting to an older kernel alleviates the issue, I'm inclined to agree, but not sure how to pin it down...21:47
* ogasawara late lunch21:51
dandeli caught a regression between 2.6.35-rc1 and 2.6.35-rc5... however, i'm not sure what to classify the bug as.... the bug is that onboard keyboard and mouse on my laptop stopped responding to all input, however i can reboot the machine with a usb mouse.21:57
dandel** 2.6.35-rc3 (latest git revision)21:58
keeshow do I specify a branch in the request-pull ?22:01
keesoh, duh [end]22:02
smbkees, Yeah, sorry slow22:03
smbkees, the name of the branch refers to the HEAD of it22:03
keesI keep getting an error, though22:04
keeswarn: No branch of git://kernel.ubuntu.com/kees/ubuntu-maverick.git is at:22:04
keeswarn:   7bc505d: SAUCE: security: unconditionally chain to Yama LSM22:04
keeswarn: Are you sure you pushed maverick-yama there?22:04
keesoh man, it must be time for lunch22:04
keesI see the problem22:04
jjohansenkees: something like22:07
jjohansengit request-pull c2e8d51100f22433062a86ec36383b549b7eb191 git://kernel.ubuntu.com/jj/ubuntu-maverick.git atop22:07
jjohansenusing git:// is important22:08
keesjjohansen: yeah, I was aiming at the wrong repo.  *hang head*22:08
smbkees, Hah, this looks ok. Thanks for the hints22:09
keessmb: okay, great22:09
smbkees, Now it will get interesting if you ever pull things back. I *am* very confident of bzr as you see. :)22:11
keessmb: heh22:12
keessmb: this is why I like "bzr missing" it's kind of handy22:13
smbkees, I won't go into the how this likely can be done with git just the same (or is not needed that much). Its too late here...22:14
jamincis there a place to report bugs found in kernels from the mainline-ppa?22:22
jjohansenjaminc: https://bugzilla.kernel.org/22:24
jamincguess I'm a bit confused, are the mainline builds essentially just the kernel.org versions?22:26
jjohansenyes22:27
jjohansenjust built into a deb for testing22:27
* jaminc slaps himself22:27
jamincthank you22:27
* smb Thinks its time to call it a day22:33
jamincBug 16316 filed22:37
ubot2Launchpad bug 16316 in debian (and 1 other project) "ximian-connector: FTBFS: Cannot find evolution libs (heat: 1)" [Unknown,Fix released] https://launchpad.net/bugs/1631622:37
jamincubot2 got it wrong... https://bugzilla.kernel.org/show_bug.cgi?id=1631622:38
ubot2jaminc: Error: I am only a bot, please don't think I'm intelligent :)22:38
ubot2bugzilla.kernel.org bug 16316 in kvm "VM's started through libvirt can't write to disk image files unless the file is owned by root" [Normal,New]22:38
Kanobtw. i tested a lucid install, my i5-680 runs with 1200 MHz by default and does NOT scale up22:42
Kanosame with maverick22:42
Kanois that normal...22:42
=== arun__ is now known as arun
bullgard4What does the message "ath5k_pci 0000:02:00.0: registered as 'phy0'" mean in http://www.ureader.de/msg/12564876.aspx? (My Ubuntu 10.04 dmesg produces the same message.)22:49
mjg59It means that the atheros 5000-series wireless device at pci address 02:00.0 has been associated with the wireless core and given the name phy022:50
ograKano, its called green computing :P22:51
Kanono its called stupid22:52
ograwell, rather buggy i'd guess22:52
bullgard4mjg59: What programs will use the name 'phy0' thereafter? '~$ iwconfig' only lists »lo«, »eth0«, »irda« and»wlan0«.22:59
mjg59bullgard4: It's not a network interface23:06
mjg59iw will use the phy name for some things23:06
bullgard4mjg59: In order to establish this association you spoke about, does the kernel just run a process called 'phy0'? (My Ubuntu 10.04 computer does run a kernel process called 'phy0'.)23:08
mjg59No, that's just the kernel workqueue associated with the phy23:08
bullgard4mjg59: Thank you very much for your explanations.23:09
=== sconklin is now known as sconklin-gone
Kanoi just tested a few things, ondemand does nothing with that i5, performance is 100% and conservative works too23:38
Kanoit scales up/down23:38
Kanoon demand does definitely not scale23:38

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