| braernoch | Is there any way to check the value of symbols exported by built-in kernel objects at runtime? | 00:10 |
|---|---|---|
| shawn1 | I've been trying to find a place to get some help on an issue I'm having with the ubuntu kernel | 00:56 |
| shawn1 | lspci's not detecting one of my computer's components | 00:57 |
| RAOF | shawn1: Your dual-core GPU? | 00:57 |
| shawn1 | it detects it as an unclassified Non-VGA device when I use detection method intel-conf2, but none of the other methods will detect it. | 00:58 |
| shawn1 | haha | 00:58 |
| shawn1 | you remember me! | 00:58 |
| shawn1 | Yes, | 00:58 |
| RAOF | It probably *is* a non-VGA device :) | 00:58 |
| RAOF | What exactly are you trying to do? | 00:58 |
| shawn1 | It's a dual-core GPU | 00:59 |
| shawn1 | well | 00:59 |
| shawn1 | dual-GPU card, I should say | 00:59 |
| shawn1 | but it's only detecting one of the GPUs | 00:59 |
| RAOF | Right. And what are you trying to do? | 00:59 |
| shawn1 | lspci -A intel-conf2 shows: 00:08.0 Non-VGA unclassified device: Device aebc:75c5 (rev 40) | 00:59 |
| shawn1 | none of the other methods show anything on that address. I'm trying to do distributed computing. | 01:00 |
| shawn1 | CUDA apps | 01:00 |
| RAOF | Ok. So your problem is that the nvidia binary driver doesn't see both GPUs? | 01:00 |
| shawn1 | no | 01:00 |
| shawn1 | lspci doesn't | 01:00 |
| RAOF | And why is that a problem? | 01:01 |
| shawn1 | I've talked to NVidia - they said that if it doesn't show up at the lspci level, that there's nothing that the NVidia driver can do | 01:01 |
| shawn1 | I told them that I ended up finding it on intel-conf2 and they said that if one of the other detection methods located it, that it was probably a kernel bug | 01:02 |
| RAOF | Ah. Ok. So, your problem is that the nvidia binary driver isn't seeing both GPUs, and the probable proximal cause is that lspci also doesn't see both gpus. | 01:02 |
| RAOF | Ok. | 01:03 |
| RAOF | Now, have you filed a bug? | 01:03 |
| shawn1 | Well, I'm not great in my knowledge of how the linux system works, but that's what it sounded like he was saying | 01:03 |
| shawn1 | Well, the linux guy just told me that, but it's been a bug that's out there for awhile, so I figured I'd check to see if there was a fix I was missing. There are a lot of threads about people having problems with GTX 295s in ubuntu and I haven't found any resolutions yet. | 01:04 |
| shawn1 | Where would I file the bug? | 01:04 |
| RAOF | You'd (a) first try running a Precise livecd, since that's got a newer kernel. Maybe it's already fixed there ☺ and then (b) from the livecd, run “ubuntu-bug linux”. Also attach the ‘lspci -A intel-conf2’ output. | 01:06 |
| shawn1 | Precise livecd? | 01:07 |
| shawn1 | how's that different from a normal livecd? | 01:07 |
| shawn1 | oh | 01:08 |
| shawn1 | Precise is the newest version? | 01:08 |
| RAOF | Precise is the current development version, yes. | 01:08 |
| RAOF | You can find daily livecd builds here: http://cdimage.ubuntu.com/daily-live/current/ | 01:08 |
| RAOF | Also, Alpha 2 will be released on Thursday, so you could try that (but it's not going to be significantly different to the current livecd) | 01:09 |
| shawn1 | ....How dangerous is alpha to my computer? | 01:09 |
| RAOF | If you don't install it, very little. | 01:10 |
| shawn1 | I've got....like a 6 core, a GTX 295 and a GTX 460SE | 01:10 |
| shawn1 | oh, yes | 01:10 |
| shawn1 | so | 01:10 |
| shawn1 | if I run it off the livecd and the kernel works with the GPU, then I should know by checking the lspci, right? | 01:11 |
| RAOF | Right. | 01:11 |
| RAOF | Well, that would tell if the lspci output is correct, at least. That doesn't guarantee that it'll work, but given what the nvidia people have said it would strongly suggest that it would work. | 01:12 |
| shawn1 | so the only file I need is the precise-desktop amd64.iso, right? | 01:12 |
| shawn1 | well I know that I wouldn't get to test it out except under circumstances where the driver was installed.... | 01:12 |
| shawn1 | This was a Christmas gift and I bought a new power supply to be able to power everything, so I hope it works! | 01:13 |
| shawn1 | if not, I may have to sell it off, pay a little more, and settle for a GTX 580 | 01:14 |
| shawn1 | (Since I'm using these exclusively for distributed computing, the GTX 295 is preferrable to the GTX 580 because while the DirectX version isn't as recent, the raw processing power is higher) | 01:15 |
| shawn1 | The 11.10 kernel seems to handle it better than the 11.04. The 11.04 kernel wouldn't even let my machine start up.... | 01:16 |
| shawn1 | the xserver, I mean | 01:16 |
| shawn1 | Well, thank you! I'll wait for that to download and then I'll see what happens when I run it. I'll write my results here, too, in case anyone else is wondering about multi-GPU cards in Ubuntu. | 01:21 |
| RAOF | Yeah, you just need precise | 01:27 |
| RAOF | -desktop-amd64.iso | 01:27 |
| shawn1 | yeah, that's what I'm getting. 40% downloaded. I just have a wireless connection here. | 01:48 |
| === CharlieMike is now known as ayan | ||
| shawn1 | The .ISO's too big for my CD, apparently | 03:34 |
| RAOF | Ah, yeah. You'll need a usb stick. | 03:37 |
| shawn1 | I'm guessing larger than my 128MB one, lol | 03:37 |
| RAOF | :) | 03:37 |
| shawn1 | I don't suppose that there's a 'live quarter-inch-cartridge' image, is there? | 03:38 |
| shawn1 | =] | 03:38 |
| RAOF | HEh. | 03:39 |
| shawn1 | Well....It looks li ke it's time for a Walmart run. See you later! | 03:40 |
| shawn1 | ROAF | 04:49 |
| shawn1 | Hello from 12.04 | 04:49 |
| shawn1 | unfortunately, it would appear that the kernel is still having the same issue with this GPU | 04:49 |
| shawn1 | but I'm installing a current driver just to make sure | 04:50 |
| shawn1 | @ROAF | 05:25 |
| shawn1 | It didn't work | 05:25 |
| shawn1 | any suggestions? | 05:25 |
| RAOF | shawn1: Ok. Now we go to stage 2, which was “…and then (b) from the livecd, run “ubuntu-bug linux”. Also attach the ‘lspci -A intel-conf2’ output.” | 05:26 |
| shawn1 | how do I attach that? | 05:26 |
| shawn1 | ugh | 05:29 |
| shawn1 | didn't have a launchpad account | 05:29 |
| RAOF | Heh. | 05:35 |
| RAOF | You can just run ‘lspci -A intel-conf2’ then copy the terminal output into a text file and attach that. | 05:36 |
| shawn1 | got it | 05:44 |
| shawn1 | I just didn't know what you meant because I've never done a bug report before | 05:44 |
| RAOF | That's ok. | 05:48 |
| shawn1 | bug report sent. Thanks!! | 05:49 |
| shawn1 | Will they send me some sort of response if it gets fixed? | 05:50 |
| shawn1 | or do I just kind of have to find out | 05:50 |
| shawn1 | ?* | 05:50 |
| shawn1 | erg | 05:52 |
| shawn1 | I have to leave | 05:52 |
| RAOF | You'll get | 05:52 |
| shawn1 | thanks for all of your help, though!! | 05:52 |
| shawn1 | okay | 05:52 |
| shawn1 | thanks!! =] | 05:52 |
| RAOF | You'll get notifications when something happens to the bug. | 05:52 |
| ohsix | RAOF: i keep trying to get shawn1 to post his dmesg :D there are a few bios pci enumeration modes you can try in case the bios does something funny | 06:19 |
| RAOF | His dmesg should now be attached to that bug :) | 06:20 |
| ohsix | right, it's been a few days now though | 06:20 |
| RAOF | Yeah, it was difficult to get information out of them. | 06:20 |
| ohsix | i think the intel-conf2 thing might be a red herring | 06:22 |
| === himcesjf2 is now known as himcesjf | ||
| JabberwockyA19 | Is there a list of backports made from kernel 3.3 ? I would like to know if 12.04 will have hdmi audio support for nvidia/amd open source drivers. | 08:53 |
| RAOF | JabberwockyA19: By and large the list is "none". | 10:22 |
| JabberwockyA19 | iceroot told me about the changelog send with each deb package in the kernel-ppa, I just grep'ed for hdmi and audio. | 10:30 |
| JabberwockyA19 | only found updates in kernel 3.3-rc1 | 10:30 |
| diwic | JabberwockyA19, afaict there are no backports for open source hdmi audio | 10:53 |
| JabberwockyA19 | I'm trying to move away from proprietary drivers, but I'm not going to run 3.3-rc1 on the work's pc just yet. Can't wait to test it out at home though. | 10:57 |
| * ppisati -> back in 10mins | 11:40 | |
| apw | cking, would you be able to cast an eye over chinstrap:~apw/stddev to see if it is indeed what i intended | 11:53 |
| * cking looks | 11:54 | |
| apw | apw@dm$ stddev 2.218344 2.222308 2.222326 2.218320 2.214328 | 11:54 |
| apw | 0.00298821300445593 | 11:54 |
| elmo | apw: the result matches numpy.std() for whatever that's worth | 12:03 |
| apw | elmo, so i am as inaccurate as numpy | 12:05 |
| cking | 0.0029882130044559338661 | 12:10 |
| cking | double precision | 12:10 |
| cking | 0.002988 single precision | 12:12 |
| cking | apw, did you find your sigma char? σ | 12:35 |
| apw | cking, thanks | 12:36 |
| cking | libreoffice states "STDEV: Estimates the standard deviation based on a sample". Heh, it should not estimate but instead calculate the stdev | 12:38 |
| apw | cking, perhaps it is using RANDOM() | 12:39 |
| cking | perhaps the GUESS() function | 12:39 |
| cking | apw, LibreOffice calculates the "sample std dev" where as we are calculating the "population std dev". The latter applies if the samples are a random sample drawn from a larger parent population. | 12:47 |
| apw | cking, so some other stddev than the one we are using then ? | 12:48 |
| cking | yep, for N samples, the "sample std dev" divides by (N-1) where as the one we are looking at divides by N | 12:49 |
| cking | since our population is a complete set and NOT randomly selected from a larger set, we can safely say dividing by N is OK for the "population std dev" which we are using | 12:50 |
| apw | gah | 12:53 |
| cking | heh, but with N being large it won't make much difference. Anyhow, you want to see how much variance there is and either calculation gives you a gut feeling of high or low sigma. | 12:54 |
| cking | apw, this may be your friend: http://easycalculation.com/statistics/standard-deviation.php | 12:56 |
| === rsalveti` is now known as rsalveti_ | ||
| === rsalveti_ is now known as rsalveti | ||
| sforshee | jsalisbury, re bug 923512, see http://marc.info/?l=linux-wireless&m=132798971403989&w=2 | 14:30 |
| ubot2` | Launchpad bug 923512 in linux "ath9k wireless stopped working after kernel upgrade to 3.2.0-12" [High,Confirmed] https://launchpad.net/bugs/923512 | 14:30 |
| ogasawara | apw: can I mark "generate config review for omap4" as DONE? I recall you resolved that at the rally. | 14:40 |
| === bigbash is now known as zz_bigbash | ||
| apw | ogasawara, i'll have to check, not sure if i did or not on the latest info ... will add it to todays todo | 14:44 |
| ogasawara | apw: ack, thanks. | 14:45 |
| ogasawara | jsalisbury: "document kernel team workflow" (from https://blueprints.launchpad.net/ubuntu/+spec/other-p-bug-workflows)... that got resolved at the rally too? | 14:45 |
| jsalisbury | ogasawara, yes, that was taken care of at the rally | 14:54 |
| jsalisbury | sforshee, thanks | 14:55 |
| ogasawara | jsalisbury: k, I'm gonna mark it DONE then. thanks. | 14:55 |
| jsalisbury | ogasawara, cool, thanks. | 14:55 |
| sforshee | jsalisbury, np | 14:55 |
| === bladernr_afk is now known as bladernr_ | ||
| ogasawara | jsalisbury: you going to get a test kernel built with the patch reverted for 923512? I'd like to know the outcome asap. I can help get it built if you're tied up with other things. | 15:01 |
| jsalisbury | ogasawara, yes I posted the first bisected kernel, which worked fine. I'm building the next kernel now. | 15:02 |
| jsalisbury | ogasawara, this is one of the bugs I was having an issue bisecting, re the email I sent. I found a workaround by bisecting in one tree and building in another. I still haven't found the root cause, but I'll investigate further today. | 15:03 |
| ogasawara | jsalisbury: and by "next kernel" do you mean a kernel with the patch reverted that sforshee mentioned? or are you just continuing with the bisect? | 15:04 |
| ogasawara | anyone else on the team have hw which uses ath9k? | 15:05 |
| jsalisbury | ogasawara, sorry, with the patch sforshee mentioned reverted. I want to bisect seperatly to also figure out the mainline-build-one issue. | 15:05 |
| ogasawara | jsalisbury: ack, thanks. | 15:05 |
| HFSPLUS | ....................../´¯/) | 15:12 |
| HFSPLUS | ....................,/¯../ | 15:12 |
| HFSPLUS | .................../..../ | 15:12 |
| HFSPLUS | ............./´¯/'...'/´¯¯`·¸ | 15:12 |
| HFSPLUS | ........../'/.../..../......./¨¯\ | 15:12 |
| HFSPLUS | ........('(...´...´.... ¯~/'...') | 15:12 |
| HFSPLUS | .........\.................'...../ | 15:12 |
| HFSPLUS | ..........''...\.......... _.·´ | 15:12 |
| HFSPLUS | ............\..............( | 15:12 |
| HFSPLUS | ..............\.............\... | 15:12 |
| HFSPLUS | ....................../´¯/) | 15:12 |
| jussi | !ops | 15:12 |
| ubot2` | Help! lamont, zul, T-Bone, mdz, or jdub | 15:12 |
| HFSPLUS | ....................,/¯../ | 15:12 |
| HFSPLUS | .................../..../ | 15:12 |
| HFSPLUS | ............./´¯/'...'/´¯¯`·¸ | 15:12 |
| HFSPLUS | ........../'/.../..../......./¨¯\ | 15:12 |
| HFSPLUS | ........('(...´...´.... ¯~/'...') | 15:12 |
| HFSPLUS | .........\.................'...../ | 15:12 |
| HFSPLUS | ..........''...\.......... _.·´ | 15:12 |
| HFSPLUS | ............\..............( | 15:12 |
| HFSPLUS | ..............\.............\... | 15:12 |
| HFSPLUS | !ops | jussi | 15:13 |
| ubot2` | jussi: Help! lamont, zul, T-Bone, mdz, or jdub | 15:13 |
| HFSPLUS | ....................../´¯/) | 15:13 |
| HFSPLUS | ....................,/¯../ | 15:13 |
| HFSPLUS | .................../..../ | 15:13 |
| HFSPLUS | ............./´¯/'...'/´¯¯`·¸ | 15:13 |
| HFSPLUS | ........../'/.../..../......./¨¯\ | 15:13 |
| jussi | Pici: | 15:13 |
| HFSPLUS | ........('(...´...´.... ¯~/'...') | 15:13 |
| HFSPLUS | .........\.................'...../ | 15:13 |
| HFSPLUS | ..........''...\.......... _.·´ | 15:13 |
| HFSPLUS | ............\..............( | 15:13 |
| HFSPLUS | ..............\.............\... | 15:13 |
| Pici | uh | 15:13 |
| jsalisbury | ogasawara, a test kernel with SHA1 7a532fe7131216a02c81a6c1b1f8632da1195a58 reverted is building on tangerine now. I'll update the bug as soon as the build is done. | 15:16 |
| ogasawara | jsalisbury: thanks. lets put it on the hot list discussion this morning and see if anyone else has ath9k they can help test with and speed things along | 15:17 |
| jsalisbury | ogasawara, ok, thanks. | 15:17 |
| ogasawara | jsalisbury: assuming reverting fixes the issue, I'd like to upload but it may be too late. I have to check with the release team. | 15:18 |
| jsalisbury | ogasawara, ok. | 15:18 |
| ogasawara | jsalisbury: ah, hallyn can reproduce. I'd ping him when you have a test kernel. | 15:27 |
| jsalisbury | ogasawara, will do | 15:27 |
| * ogasawara back in 20 | 15:30 | |
| jsalisbury | ** | 15:48 |
| jsalisbury | ** Ubuntu Kernel Team Meeting - Today @ 17:00 UTC - #ubuntu-meeting | 15:48 |
| jsalisbury | ** | 15:48 |
| === yofel_ is now known as yofel | ||
| jsalisbury | ogasawara, the test kernel is done building for bug 923512, uploading to people now. | 16:29 |
| ubot2` | Launchpad bug 923512 in linux "ath9k wireless stopped working after kernel upgrade to 3.2.0-12" [High,Confirmed] https://launchpad.net/bugs/923512 | 16:29 |
| jsalisbury | hallyn, There is a test kernel for bug 923512, which is s dup of your bug 924015. When you have a chance can you test the kernel? | 16:32 |
| ubot2` | Launchpad bug 923512 in linux "ath9k wireless stopped working after kernel upgrade to 3.2.0-12" [High,Confirmed] https://launchpad.net/bugs/923512 | 16:32 |
| ubot2` | Launchpad bug 924015 in linux "Wireless drivers broken with latest precise kernel (dup-of: 923512)" [High,Confirmed] https://launchpad.net/bugs/924015 | 16:32 |
| jsalisbury | hallyn, The kernel can be downloaded from: http://people.canonical.com/~jsalisbury/lp923512/ | 16:32 |
| hallyn | jsalisbury: love to. it'll be a bit before i'm out of meetings | 16:32 |
| hallyn | (meaning i can't reboot) | 16:32 |
| hallyn | i'll start downloading now | 16:33 |
| jsalisbury | hallyn, cool, thanks! | 16:33 |
| ogasawara | jsalisbury: did you revert that commit from the latest precise, or did you do an upstream build with it reverted? | 16:35 |
| jsalisbury | ogasawara, upstream build with it reverted. | 16:36 |
| jsalisbury | ogasawara, should I also build from the precise tree? | 16:36 |
| ogasawara | jsalisbury: for this case, it would be better to revert it from precise and build, as then it would be identical to what we'd upload. | 16:37 |
| ogasawara | jsalisbury: I can do that real quick and have it built by the time hallyn is done with this meetings | 16:37 |
| jsalisbury | ogasawara, cool, thanks. | 16:37 |
| ppisati | jjohansen: bug 893147 | 16:38 |
| ubot2` | Launchpad bug 893147 in linux-ti-omap4 "CVE-2011-4131" [Medium,Fix committed] https://launchpad.net/bugs/893147 | 16:38 |
| jsalisbury | ogasawara, I'll update the bug with a comment that the kernels I uploaded are from an upstream build | 16:38 |
| ppisati | jjohansen: never mind, it's in master-next... | 16:41 |
| tgardner | jsalisbury, make deb-pkg -j64 INSTALL_MOD_STRIP=1 | 16:48 |
| ogasawara | hallyn: if you could give that a test that'd be great -> http://people.canonical.com/~ogasawara/lp923512/amd64/ | 17:15 |
| hallyn | ogasawara: wait, i should use that instead of jsalisbury's? | 17:16 |
| ogasawara | hallyn: I'd prefer that. the one that jsalisbury built was an upstream kernel. I built precise with the patch reverted. | 17:16 |
| ogasawara | hallyn: and I'd prefer you test something identical to what we'd upload | 17:17 |
| ogasawara | hallyn: I'd like to squeeze this in before the Alpha if we have time | 17:17 |
| hallyn | ok | 17:17 |
| ppisati | tgardner: do you know PARTUUID? | 17:18 |
| tgardner | ppisati, not sure what you're asking. are you talking about /dev/disk/by-uuid ? | 17:19 |
| hggdh | sconklin: how is the kernel SRU code doing? | 17:35 |
| skaet | ogasawara, https://bugs.launchpad.net/ubuntu/+source/linux/+bug/924400 - will the upload address this? or is it something new? | 17:49 |
| ubot2` | Launchpad bug 924400 in linux "kernel NULL pointer dereference at 00000000000001f0" [Undecided,Confirmed] | 17:49 |
| ogasawara | skaet: this upload would only contain a fix for 923512, I'd have to take a look at 924400 | 17:49 |
| ogasawara | skaet: I'm still waiting on test confirmation though for 923512 | 17:50 |
| skaet | ogasawara, ok, thanks. Let slangasek and myself know the results of the test confirmation. We'll be doing a respin this afternoon, so want to pick up your regression fix if possible. | 17:51 |
| ogasawara | skaet: indeed, I'm hoping hallyn can get us his results soon | 17:52 |
| skaet | :) | 17:52 |
| hallyn | ogasawara: all good | 18:01 |
| ogasawara | hallyn: sweet! thank you!! | 18:01 |
| hallyn | what pray tell was the bug? | 18:01 |
| ogasawara | hallyn: bad patch from v3.2.2 stable | 18:02 |
| ogasawara | commit b4a82a0a2e32777267b2c997fa55b90056447a40 | 18:02 |
| ogasawara | Author: Felix Fietkau <nbd@openwrt.org> | 18:02 |
| ogasawara | Date: Sat Jan 14 15:08:34 2012 +0100 | 18:02 |
| ogasawara | ath9k_hw: fix interpretation of the rx KeyMiss flag | 18:02 |
| ogasawara | skaet: ^^ | 18:03 |
| njin | bug 924400 | 18:03 |
| ubot2` | Launchpad bug 924400 in linux "kernel NULL pointer dereference at 00000000000001f0" [Undecided,Confirmed] https://launchpad.net/bugs/924400 | 18:03 |
| hallyn | great, thanks, glad to be back :) | 18:03 |
| ogasawara | skaet: ok for me to upload? | 18:03 |
| njin | can someone tell me under wich tag send this upstream? | 18:04 |
| === himcesjf1 is now known as himcesjf | ||
| skaet | ogasawara, sorry, yup. | 18:30 |
| * ogasawara uploads | 18:30 | |
| === jdstrand_ is now known as jdstrand | ||
| skaet | ogasawara, has anyone else reproduced the kernel oops that nijin is seeing with today's images? bug 924400 | 18:40 |
| ubot2` | Launchpad bug 924400 in linux "kernel NULL pointer dereference at 00000000000001f0" [Undecided,Confirmed] https://launchpad.net/bugs/924400 | 18:40 |
| tgardner | apw, smp: Start up non-boot CPUs asynchronously | 18:47 |
| * tgardner -> lunch | 18:53 | |
| ogasawara | skaet: not that I know of | 18:57 |
| skaet | ogasawara, ack | 18:57 |
| * jsalisbury --> lunch. Be back in ~1hour. | 19:13 | |
| tgardner | bjf, do you know how to update a mini-iso image in cobbler after the initial import ? | 19:43 |
| === dduffey is now known as dduffey_afk | ||
| bjf | tgardner: nope | 20:10 |
| jono | hey | 20:13 |
| jono | is anyone seeing bugs in Precise regarding wireless randomly de-authenticating - I filed https://bugs.launchpad.net/ubuntu/+source/linux/+bug/911059 | 20:14 |
| ubot2` | Launchpad bug 911059 in linux "Intel wireless randomly drops connection" [Medium,Incomplete] | 20:14 |
| jono | I filed it a little while back | 20:15 |
| jsalisbury | jono, There were a couple of requests in that bug to test some kernels. Do you think you'll be able to do that? | 20:19 |
| jsalisbury | jono, basically the latest Precise and latest mainline kernels. | 20:20 |
| jono | jsalisbury, I will test that - which kernel? | 20:20 |
| jono | I am running latest Precise now | 20:20 |
| jono | and I still get the bug | 20:20 |
| jsalisbury | jono, ok, the latest mainline can be found at: http://kernel.ubuntu.com/~kernel-ppa/mainline/v3.3-rc1-precise/ | 20:21 |
| jono | jsalisbury, do I just install the headers and image for my arch, reboot and select that kernel? | 20:22 |
| jsalisbury | jono, yes. you probably don't need the headers, just the image for your arch. | 20:22 |
| jsalisbury | jono, further details on mainline kernels can be found at: https://wiki.ubuntu.com/KernelMainlineBuilds | 20:25 |
| jono | jsalisbury, downloading now | 20:26 |
| jsalisbury | jono, cool. It would also be great to know the last version of the kernel that did not have this issue, so we can bisect. | 20:27 |
| jono | jsalisbury, I have had this since I upgraded to Precise | 20:27 |
| jono | jsalisbury, ok, rebooting now into the new kernel | 20:27 |
| jsalisbury | jono, can you see if you boot back into the latest oneiric kernel the issue goes away. The latest oneiric kernel can be found at: https://launchpad.net/~canonical-kernel-team/+archive/ppa/+build/3061070/+files/linux-image-3.0.0-15-generic-pae_3.0.0-15.25_i386.deb | 20:28 |
| jono | jsalisbury, ok I am now running: | 20:30 |
| jono | Linux forge2 3.3.0-030300rc1-generic-pae #201201191835 SMP Thu Jan 19 23:51:25 UTC 2012 i686 i686 i386 GNU/Linux | 20:30 |
| jsalisbury | jono, it looks like you were running linux-image-3.2.0-7-generic-pae when you reported the bug. Did you try 3.2.0-12.20, which is the latest precise? | 20:35 |
| jono | jsalisbury, yep, I have been fully up to date in Precise | 20:35 |
| jono | now I am running: | 20:35 |
| jono | Linux forge2 3.3.0-030300rc1-generic-pae #201201191835 SMP Thu Jan 19 23:51:25 UTC 2012 i686 i686 i386 GNU/Linux | 20:35 |
| jono | and so far so good | 20:35 |
| jsalisbury | jono, hmm, interesting. | 20:36 |
| jono | jsalisbury, mind you, it seems to randomly drop the wireless, so it might happen in a bit | 20:36 |
| jsalisbury | jono, OK, it would be great if you can test the kernel for a while to see if the bug is really fixed. If it is, we can try to identify what commit actually fixed it. | 20:37 |
| jono | jsalisbury, absolutely, I will keep using this until it happens again | 20:37 |
| jono | and hopefully it won't happen again :-) | 20:37 |
| jono | this bug has been driving me nuts | 20:37 |
| jsalisbury | jono, thanks, and just to confirm, you ran with 3.2.0-12.20 and that version had the bug. | 20:38 |
| jono | jsalisbury, I have basically experienced the bug with every Precise kernel | 20:38 |
| jsalisbury | jono, ok, thanks. | 20:39 |
| jono | thanks for taking the time to help me with this jsalisbury | 20:39 |
| jsalisbury | jono, np. It would be great to get this bug fixed. | 20:39 |
| jono | yes indeed :-) | 20:39 |
| jono | jsalisbury, the bug just happened | 20:43 |
| jono | how can I get some useful debugging info to you? | 20:43 |
| jono | I guess I should any info in https://bugs.launchpad.net/ubuntu/+source/linux/+bug/911059 | 20:43 |
| ubot2` | Launchpad bug 911059 in linux "Intel wireless randomly drops connection" [Medium,Incomplete] | 20:43 |
| jsalisbury | jono, I think the next best thing to do is to have you test the latest oneiric kernel, and confirm the bug does not exist there. Then we can bisect between the latest oneiric and earliest precise. | 20:44 |
| jsalisbury | jono, can you test one more kernel, at: | 20:44 |
| jono | sure | 20:44 |
| jsalisbury | jono, https://launchpad.net/~canonical-kernel-team/+archive/ppa/+build/3061070/+files/linux-image-3.0.0-15-generic-pae_3.0.0-15.25_i386.deb | 20:44 |
| ohsix | jono: does the randomness "feel" like it may be heat related? | 20:45 |
| jono | ohsix, I don't think so | 20:45 |
| ohsix | ok, i've only seen that with broadcom devices anyways | 20:46 |
| jono | jono@forge2:~/Downloads$ sudo dpkg -i linux-image-3.0.0-15-generic-pae_3.0.0-15.25_i386.deb | 20:46 |
| jono | dpkg-deb: error: `linux-image-3.0.0-15-generic-pae_3.0.0-15.25_i386.deb' is not a debian format archive | 20:46 |
| jono | dpkg: error processing linux-image-3.0.0-15-generic-pae_3.0.0-15.25_i386.deb (--install): | 20:46 |
| jono | subprocess dpkg-deb --control returned error exit status 2 | 20:46 |
| jono | Errors were encountered while processing: | 20:46 |
| jono | linux-image-3.0.0-15-generic-pae_3.0.0-15.25_i386.deb | 20:46 |
| jono | jsalisbury, ^ | 20:46 |
| jono | oh hang on | 20:46 |
| jono | it is still downloading | 20:47 |
| jono | LOL | 20:47 |
| jsalisbury | jono, ok | 20:47 |
| jono | just really slowly | 20:47 |
| tgardner | jono, did you load with 11n-disable=1 to see if it had an impact ? 'sudo modprobe -r iwlagn;sudo modprobe iwlagn 11n_disable=1' | 20:47 |
| jono | tgardner, I haven't tried that | 20:48 |
| jono | how do I pass a kernel options at boot? | 20:48 |
| ohsix | is there anyone that mirrors ddebs? (getting a big -dbgsym, say for the kernel is quite an ordeal, also it would be very nice if there was a package just with vmlinux in it, and in a place where perf can find it :) | 20:48 |
| jono | or should I modprobe -r it? | 20:48 |
| tgardner | jono, it would go into /etc/modprobe.d | 20:48 |
| jono | tgardner, ok, I will try booting the Oneiric kernel first, see how that works and then try and do this | 20:49 |
| jono | brb, rebooting | 20:49 |
| jono | thanks folks | 20:49 |
| tgardner | jono, just modprobe it by hand for now to see if it makes a difference (once you've tested Joe's kernels) | 20:49 |
| jono | jsalisbury, ok I am now running: | 20:56 |
| jono | Linux forge2 3.0.0-15-generic-pae #25-Ubuntu SMP Mon Jan 2 19:40:15 UTC 2012 i686 i686 i386 GNU/Linux | 20:56 |
| jono | the package wanted me to install wireless-crda | 20:57 |
| jono | which it grabbed from Precise | 20:57 |
| jono | I am not sure if this will affect anything | 20:57 |
| jono | I will see if this bug nails me | 20:57 |
| tgardner | jono, you should be OK | 20:57 |
| jono | thanks tgardner | 20:57 |
| tgardner | wrt wireless-crda | 20:57 |
| tgardner | jono, if the bug continues, then try 'sudo modprobe -r iwlagn;sudo modprobe iwlagn 11n_disable=1' with any of these kernels. | 20:58 |
| jono | tgardner, I assume I don't need to reboot when I have run that command? | 20:59 |
| jono | just re-connect to the network | 20:59 |
| tgardner | jono, correct | 20:59 |
| jono | thanks tgardner | 20:59 |
| ohsix | is modprobe -r generally preferable over rmmod? | 21:00 |
| tgardner | its the same thing | 21:00 |
| tgardner | modprobe removed dependent modules | 21:00 |
| ohsix | (i know modprobe != insmod though) | 21:00 |
| tgardner | removes* | 21:00 |
| ohsix | ah, cool | 21:00 |
| cking | anyone tried to install the daily precise ISO on a netbook - this is taking forever | 21:02 |
| tgardner | cking, yesterday | 21:03 |
| cking | hrm, cursor won't move, it's kinda locked up | 21:03 |
| tgardner | hmm, maybe its gonna take along, long time | 21:04 |
| * cking leaves it for another 10 mins | 21:05 | |
| * tgardner is EOD | 21:10 | |
| cking | EOD is a good idea | 21:15 |
| === jsalisbury changed the topic of #ubuntu-kernel to: Home: https://wiki.ubuntu.com/Kernel/ || Ubuntu Kernel Team Meeting - Tues Feb 7th, 2012 - 17:00 UTC || If you have a question just ask, and do wait around for an answer! | ||
| jono | jsalisbury, still working fine, I think the Oneiric kernel might be fine | 22:30 |
| jono | which would suggest that the book does exist from a new kernel onwards | 22:31 |
| jono | I am going to keep testing this kernel to see if it happens though | 22:31 |
| jsalisbury | jono, ok, thanks. | 22:31 |
| jsalisbury | jono, maybe you can try the following tomorrow on a precise kernel: | 22:31 |
| jsalisbury | sudo modprobe -r iwlagn;sudo modprobe iwlagn 11n_disable=1 | 22:31 |
| jono | indeed, will do | 22:32 |
| jsalisbury | jono, thanks | 22:32 |
| jono | thanks jsalisbury | 22:32 |
| jsalisbury | jono, np | 22:32 |
| hallyn | ogasawara: that kernel you had me try, did it by chance not have the lxc reboot kernel patch applied? | 22:46 |
| ogasawara | hallyn: it should have | 22:48 |
| ogasawara | hallyn: you're talking about the "Add reboot_pid_ns to handle the reboot syscall" patch right? | 22:48 |
| hallyn | eah | 22:48 |
| hallyn | yeah | 22:48 |
| hallyn | ogasawara: never mind, separate tests confirm it should be working - sorry | 22:50 |
| hallyn | (so something else is goign wrong here) | 22:50 |
| ogasawara | ok cool | 22:51 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!