| === smb` is now known as smb | ||
| ppisati | moin | 08:34 |
|---|---|---|
| smb | morning | 08:39 |
| * apw yawns | 09:13 | |
| === yofel_ is now known as yofel | ||
| === henrix_ is now known as henrix | ||
| ppisati | brb | 09:56 |
| henrix | apw: hi! could you please help me with the linux-overlay file? I added some locals there last friday and i still can't see the updates on the matrix | 10:42 |
| henrix | apw: CVEs 2013-0216 and 2013-0231 should have been updated already. 2013-0217 has been added a few minutes ago. | 10:43 |
| apw | henrix, sure | 10:45 |
| henrix | apw: cool, thanks. this is not urgent, of course ;) | 10:45 |
| apw | CVE-2013-0216 has no break-fix: information ? | 10:48 |
| henrix | apw: no, they all have a patch: section only | 10:48 |
| apw | retired/CVE-2012-0055: break-fix: 549ffd5a07c9098a6a1c7a82e7fd731a132fc1a0 local-2012-0055 | 10:48 |
| henrix | gah! so that's it | 10:48 |
| apw | so there is an example how you might represent a 'local' fix | 10:48 |
| apw | obviously the first sha1 can be '-' as well | 10:49 |
| henrix | apw: yep, got it. sorry to bother with this ;) | 10:49 |
| henrix | thanks a lot | 10:49 |
| apw | henrix, np, it is not really your fault, it is an esoteric thing which i have never correctly documented | 10:49 |
| apw | henrix, also .. you would need those to be different | 10:50 |
| apw | :title('UBUNTU: SAUCE: netback: correct netbk_tx_err to handle wrap around.', ['local-2013-0216']) | 10:50 |
| apw | :title('UBUNTU: SAUCE: xen/netback: shutdown the ring if it contains garbage.', ['local-2013-0216']) | 10:50 |
| ogra_ | heh | 10:50 |
| apw | henrix, so they likely should be like 'local-2013-0216-1' and -2 | 10:50 |
| * ogra_ read nethack | 10:50 | |
| apw | ogra_, heh :) | 10:51 |
| apw | now that would be something to have in the kernel, a complete nethack on the console | 10:51 |
| henrix | apw: ah, the 'local-...' thing has to be different. ok, makes sense. any convention for that naming? | 10:51 |
| apw | henrix, same for both the recent ones you added | 10:51 |
| ogra_ | "finally steam included in the kernel, just modprobe your favorite game !" | 10:51 |
| apw | henrix, local-<cvenumber>[-<patch number>] i suppose | 10:51 |
| henrix | apw: ack, thanks | 10:52 |
| apw | henrix, but ... as long as it starts local- i think you can do what you want | 10:52 |
| henrix | cool | 10:52 |
| apw | henrix, i really should get with security and move that file into their repo, it makes little sense for it to be separated | 10:53 |
| henrix | apw: yeah, it would make sense to keep it in the same place. | 10:53 |
| infinity | henrix: I see all the fsnotify changes have made a comeback in this SRU round. Are we sure the backport is complete this time? :) | 11:22 |
| henrix | infinity: i hope so :) | 11:23 |
| henrix | infinity: we did a lot of testing and couldn't reproduce any of the issues we've seen before. so, confidence is high | 11:23 |
| henrix | (famous last words, i know...) | 11:23 |
| infinity | henrix: Fingers crossed, then. I ask because I'll need to install the current SRUs on sagari to validate my two bugs for you guys, and I'd prefer not to be installing something crap. :P | 11:27 |
| infinity | henrix: But I guess I can install, verify the two bugs, and roll back. | 11:27 |
| infinity | Well, not that you'll lose any sleep if I don't verify the HVC console bug. | 11:28 |
| henrix | infinity: heh, right. i've been running these kernels myself (Q and P) and haven't seen any issue. | 11:28 |
| infinity | But the tg3 one is actually a pretty awful bug that's not PPC-specific. | 11:28 |
| henrix | infinity: i may not lose my sleep, but i'll be ping'ing you soon if you don't verify them :) | 11:28 |
| infinity | henrix: Yeah, maybe I should say "*I* wouldn't lose any sleep if I lied about verifying the console bug, but the tg3 one is dreadfully important". | 11:29 |
| henrix | heh, ok. i don't need to know that, do i? | 11:29 |
| infinity | Well, if I verify the important one, I'll be doing the other at the same time, so disregard any claims of lying. ;) | 11:30 |
| infinity | Sadly, I no longer have direct access to the hardware, so I'll have to puppet GSA through it all. | 11:30 |
| apw | henrix, to be absolutly correct, for linux-lowlatency tracking bugs i need to 1) get the title version right, and 2) switch upload-to-ppa to InProgress once all packages are in the PPA and building | 12:56 |
| henrix | apw: if that was a question, the answer is 'yes' :) | 12:57 |
| apw | henrix, heh yeah it was, i was wondering how it knows whether i need a meta or not, and am assuming thats why i move it InProgress after upload not before | 12:58 |
| henrix | apw: for derivative kernels, the bot doesn't know if a -meta is required or not (i.e., you need to manually set the task if there's an abi bump) | 13:00 |
| henrix | herton: ^^ | 13:00 |
| henrix | herton: not sure if there's some specificity for the lowlatency kernels... | 13:00 |
| apw | herton, henrix, see i am confused again :) | 13:01 |
| herton | apw, just get the title version right, upload-to-ppa can be set to fix released. But with the right version the bot should move tasks without problem now, it just ignores upload-to-ppa for now | 13:01 |
| apw | herton, and it will handle -meta without my brain | 13:02 |
| herton | apw, correct | 13:02 |
| henrix | interesting... | 13:02 |
| apw | sweet, and upload-to-ppa i can do waht i like with, ie Inprogress it when i start making them | 13:02 |
| henrix | i know that for -ec2 kernels it didnt handle the -meta... | 13:02 |
| apw | Fixreleased now cause i have uploaded it and it is building | 13:02 |
| herton | henrix, if it didn't handled it's a bug, I have to check | 13:03 |
| apw | herton, https://bugs.launchpad.net/kernel-sru-workflow/upload-to-ppa/+bug/1117449 and https://bugs.launchpad.net/kernel-sru-workflow/upload-to-ppa/+bug/1118282 | 13:03 |
| ubot2 | Ubuntu bug 1117449 in linux-lowlatency (Ubuntu) "linux-lowlatency: 3.2.0-38.38 -proposed tracker" [Medium,In progress] | 13:03 |
| ubot2 | Ubuntu bug 1118282 in linux-lowlatency (Ubuntu) "linux-lowlatency: 3.5.0-24.23 -proposed tracker" [Medium,In progress] | 13:03 |
| dhanasekaran | Hi Guys, Where i can find kernel release notice please guide me | 13:03 |
| apw | herton, in theory those are 'minimum andy input' done on them. i will watch them to see what sb does to them | 13:04 |
| herton | apw, ack | 13:04 |
| apw | herton, i wonder if henrix fixed the title on his to the right version | 13:04 |
| herton | apw, probably yes, I'll take a look to see what happened | 13:04 |
| henrix | herton: about the -ec2, the bot creates the -meta task but if there's not ABI bump, then i need to manually set to to 'invalid' | 13:05 |
| herton | henrix, actually I placed a code in the bot that should in theory set it to invalid too | 13:06 |
| apw | herton, ok sb just did something to the bugs, it just slamed the prepare-package to in progress and by the changelog preparer .. nice, will watch the meta now which was behind it timewise | 13:06 |
| henrix | herton: hmm, ok. i'll pay attention to that next time. thanks | 13:06 |
| herton | apw, yep, seems sb is doing things as supposed, meta I'll follow on it too | 13:07 |
| zequence | apw: Uploading to PPA. Is that meant for the Ubuntu Studio kernel team to do later? | 13:07 |
| zequence | Or, could we not do that? | 13:07 |
| zequence | To our own PPA, that is | 13:07 |
| apw | zequence, currently they are coming through the main kernel team PPA, so currently you cannot do | 13:07 |
| dhanasekaran | Hi Guys, Where i can find kernel release notice please guide me | 13:08 |
| zequence | dhanasekaran: Is this what you need? https://lists.ubuntu.com/#Package+Upload+and+Automatic+Notification+Lists | 13:09 |
| zequence | dhanasekaran: You could add a filter to your mail for excluding anything else but kernels | 13:10 |
| dhanasekaran | zequence: thanks, i will do | 13:10 |
| herton | henrix, is linux-ec2 - 2.6.32-350.59 the current one for this cycle? | 13:12 |
| herton | henrix, nevermind, now I saw it's from January | 13:13 |
| henrix | herton: its bug #1119764 | 13:13 |
| ubot2 | Launchpad bug 1119764 in linux-ec2 (Ubuntu) "linux-ec2: <version to be filled> -proposed tracker" [Medium,In progress] https://launchpad.net/bugs/1119764 | 13:13 |
| henrix | version hasn't been set yet | 13:13 |
| zequence | dhanasekaran: You can also subscribe to this. You'll get notified whenever a new release is being prepared https://launchpad.net/ubuntu/+source/linux | 13:13 |
| herton | henrix, so it's not prepared yet correct? The sb behaviour I'm talking should only happen since last week, when I commited the changes | 13:13 |
| henrix | herton: ah, ok. so that's why i haven't seen that yet :) | 13:14 |
| dhanasekaran | zequence: I Believe it's very hard, to find release notice. for newbies | 13:16 |
| zequence | dhanasekaran: Released kernels will of course appear when you do apt-get update, so in that sense, I don't agree. And I'm a mere newbie myself, you could say. If you subscribe to the LP bug, you can add a filter for it right there. Mail filtering is more or less required when subscribing to high volume mail lists, which is hard to avoid in this case :) | 13:20 |
| dhanasekaran | zequence: I got change log from /usr/share/doc/linux-image-3.2.0-37-generic# ls -ltr | 13:21 |
| dhanasekaran | total 28 | 13:21 |
| dhanasekaran | -rw-r--r-- 1 root root 1292 Jan 24 10:28 copyright | 13:21 |
| dhanasekaran | -rw-r--r-- 1 root root 23414 Jan 24 11:11 changelog.Debian.gz | 13:21 |
| zequence | dhanasekaran: Ah, so you were looking for the changelog? | 13:23 |
| dhanasekaran | zequence: changelog release notice look like same are difference ? | 13:26 |
| zequence | dhanasekaran: I misunderstood you. I thought you wanted to be noticed of releases | 13:27 |
| * rtg__ works on Raring -rc7 rebase | 13:28 | |
| zequence | apw: I take it you are working out the procedure and I'm not required to do anything for the bug report as I was assigned to Prepare-package? | 13:28 |
| dhanasekaran | zequence: I requirement is I have two different version of kernel installed in my server. But I want know what the specific changes done and modified newer version of kernel. | 13:29 |
| apw | zequence, indeed, we have added some tooling to make those just work | 13:29 |
| apw | you are getting assigned to the ones you did as they complete rather than you having to care | 13:29 |
| apw | herton, so .. is sb hourly ? | 13:31 |
| herton | apw, I belive it's on a 15min cron run, brad runs it | 13:32 |
| herton | *believe | 13:32 |
| apw | herton, ok, well i'll give it a while, but if so i think it has ignored my -meta, its not expecting the version to be the smae form is it | 13:33 |
| herton | apw, no, it just checks the abi part | 13:33 |
| herton | apw, let me run manually here | 13:33 |
| * apw idly wonders how it maps emails to launchpad ids | 13:34 | |
| zequence | dhanasekaran: You could use aptitude too. aptitude changelog <package> | 13:35 |
| herton | apw, it checks the launchpad ids with launchpad api, each upload has a property saying who created the package | 13:36 |
| herton | apw, running manually it seems did the right thing | 13:36 |
| apw | herton, ahh i see | 13:37 |
| apw | herton, odd ... lack of patience on my side then ? | 13:38 |
| herton | apw, perhaps bjf set the bot to run in intervals greater than 15 min, I thought it was each quarter of an hour but perhaps is more | 13:38 |
| zequence | There's also apt-listchanges. It shows the changelog when doing an upgrade | 13:38 |
| zequence | dhanasekaran: ^ | 13:38 |
| apw | herton, *shrug* ... if it works for you i guess it works :) | 13:40 |
| dhanasekaran | zequence: my question release notice and change log same? or different? | 13:42 |
| apw | dhanasekaran, i would not expect them to be the same, for me release notes tell you what has changed which you need to worry about, changelog tells you everything changed | 13:43 |
| herton | apw, yep, looks fine and I commited things hehe, will keep a look if something doesn't go as planned | 13:43 |
| zequence | dhanasekaran: English is not my native language, and I often mix up terms, but I usually say changelog. I misunderstood your sentence for something like: to be notified fo changes, like when a new release is out. | 13:44 |
| zequence | Or, rather, to be notified of a new release | 13:45 |
| apw | herton, does (can) shankbot handle the 'non series' kernel specific thingy ? | 13:45 |
| apw | herton, like linux-lowlatency (Ubunt) when we are targetting quantal | 13:45 |
| herton | apw, you mean linux-lowlatency uploaded to development series used in quantal? | 13:46 |
| apw | https://bugs.launchpad.net/kernel-sru-workflow/upload-to-ppa/+bug/1118282 | 13:47 |
| ubot2 | Ubuntu bug 1118282 in linux-lowlatency (Ubuntu) "linux-lowlatency: 3.5.0-24.23 -proposed tracker" [Medium,In progress] | 13:47 |
| apw | herton, there are two linux-lowlatency tasks there, one for raring and one for quantal | 13:47 |
| apw | the raring one is just an artifact of the bot not accepting the nominations | 13:48 |
| apw | once the nomination is accepted, the non-nomination is invalid | 13:48 |
| herton | apw, ah ok, it just ignores the development tasks. But I could make the bot or create-release-tracker to set it to invalid | 13:48 |
| apw | herton, maybe after it has been nominated ... dunno, not sure | 13:55 |
| herton | apw, yes, there are two cases, if who is creating the tracking bug has permissions or not to nominate (upload rights in fact). probably I'll have to do something also in the bot to automatically fix this | 13:57 |
| apw | herton, though in theory the bot runner currenlty does | 13:57 |
| herton | apw, yep | 13:58 |
| * herton -> lunch | 14:10 | |
| timblechmann | hi, after building a 3.8-rc kernel, i cannot build the dkms modules: it complains with: ERROR (dkms apport): kernel package linux-headers-3.8.0-rc7 is not supported | 14:13 |
| timblechmann | is this a known problem and is there any workaround? | 14:13 |
| timblechmann | (same message with virtualbox and nvidia dkms drivers) | 14:14 |
| rtg__ | timblechmann, you're a little ahead of the process. give it a day until the -rc7 kernel is in the archive. | 14:15 |
| timblechmann | rtg__: i have a custom configuration, similar to the -low-latency kernels, but with some tweaks ... so in a way i prefer to build myself (have been doing this with make-kpkg for some time) | 14:21 |
| infinity | timblechmann: If you're building yourself, you also need to install the headers yourself, so dkms can find them. | 14:22 |
| zequence | timblechmann: Just out of interest. What are you using your kernel for, and how does it differ from -lowlatency? | 14:23 |
| rtg__ | apw, ogasawara: pushed -rc7 rebase. haven't tagged it yet since I wanna make sure the perf tool builds for all 3 arches. | 14:27 |
| ogasawara | rtg__: ack | 14:27 |
| apw | rtg__, thanks | 14:29 |
| timblechmann | infinity: linux-headers-3.8.0-rc7 *is* installed | 14:29 |
| timblechmann | that's why i'm here | 14:29 |
| timblechmann | usually installing image/header and dkms works out of the box ... but with 3.8 it seems to fail for some reason | 14:29 |
| timblechmann | zequence: don't remember the details ... but i have a configuration which i sometimes use with -rt kernels ... | 14:31 |
| infinity | timblechmann: Is there an error above that one? That's only supposed to trigger if the build failed. | 14:32 |
| infinity | timblechmann: That's apport saying it won't report a bug because that package isn't a distro package. But it needs to have had a reason to want to report a bug in the first place. | 14:33 |
| infinity | rtg__: \o/ perf. | 14:33 |
| zequence | timblechmann: If you think there is something that would be good to improve for the -lowlatency config, let me know. I'm the kernel guy for Ubuntu Studio atm | 14:33 |
| rtg__ | infinity, I noticed you promoted libaudit-dev | 14:33 |
| infinity | rtg__: Yeah, I poked Leann and Andy about it, I would have poked you, but you insensitively log off IRC every night so I can't abuse you. :) | 14:34 |
| rtg__ | infinity, indeed :) | 14:34 |
| zequence | timblechmann: I've been meaning to do some testing, to see if we could imrpove performance in different ways, so any good info about that would be helpful | 14:34 |
| timblechmann | zequence: will need to double-check | 14:35 |
| timblechmann | infinity: the error message of the nvidia-current module is something along the lines that i cannot determine the kernel version | 14:36 |
| infinity | timblechmann: Kay, that's the real failure, the apport message is a red herring. | 14:36 |
| ppisati | brb | 14:37 |
| timblechmann | infinity: regarding virtualbox: it cannot find generated/bounds.h | 14:37 |
| timblechmann | ... which is in /usr/src/linux-headers-3.8.0-rc7/include/generated/bounds.h | 14:37 |
| infinity | timblechmann: Kay, both of those seem like a potential issue on your end, mind you, not dkms. | 14:38 |
| timblechmann | infinity: so ... it works out of the box with earlier kernels, but with 3.8 it stops working ... any idea, why? | 14:39 |
| infinity | timblechmann: Well, it works with our 3.8 packages too, AFAIK. So, no idea why your headers seem to be unfindable. Missing a build symlink? | 14:40 |
| infinity | timblechmann: As in /lib/modules/3.8.0-rc7/build -> /usr/src/linux-headers-3.8.0-rc7 | 14:41 |
| timblechmann | infinity: /lib/modules/3.8.0-rc7/build -> /media/hd3btr/sources/linux-3 | 14:41 |
| timblechmann | which is the location of my source tree | 14:42 |
| timblechmann | (at the correct git revision) | 14:42 |
| infinity | timblechmann: You just said it was /usr/src/linux-headers-3.8.0-rc7 :P | 14:42 |
| infinity | timblechmann: source tree != installed post-build headers. | 14:42 |
| timblechmann | infinity: i'm not hacking the kernel every day ... but afaict, dkms does a preparation-step to generate the headers | 14:43 |
| timblechmann | infinity: again ... that has been the case for the last years and it worked fine | 14:44 |
| infinity | dkms doesn't generate headers, it just looks where you tell it to. | 14:44 |
| infinity | Now, there may be a curious bug in make-kpkg, if your build symlink is wrong, but I don't think we've updated it in ages, have we? | 14:45 |
| timblechmann | infinity: ah | 14:45 |
| timblechmann | dkms does `make prepare-all' | 14:45 |
| timblechmann | which is not a make target | 14:46 |
| timblechmann | infinity: or rather: the dkms installer for the nvidia module does a `make clean' and `make prepare-all' step | 14:49 |
| infinity | win 37 | 14:50 |
| === nightwis- is now known as nightwish | ||
| === henrix is now known as henrix_ | ||
| === henrix_ is now known as henrix | ||
| smb | tseliot, I bet you are aware of that already but I was not so pleased to see fglrx-updates in Precise render my media pc unusable (due to AMDs weird support policy) | 15:55 |
| dhanasekaran | Hi Guys, what are patches applied to the current kernel | 15:56 |
| dhanasekaran | [RFC,3/3] x86,smp: auto tune spinlock backoff delay factor I want to know the patch applied or not | 15:56 |
| tseliot | smb: err... was your card dropped? | 15:56 |
| smb | tseliot, Yeah, one of those unfortunate onboard chips they moved into "legacy" | 15:56 |
| dhanasekaran | auto tune spinlock backoff delay factor? It's support ubuntu kernel | 15:57 |
| tseliot | smb: sorry about that. AMD don't have proper release notes for their driver releases... | 15:57 |
| smb | tseliot, Just that the previous version worked and the update removed it was, slightly unexpected. ;) For now I moved back to fglrx (release) | 15:58 |
| tseliot | smb: that's probably your best bet. Eventually you'll have to move to radeon though | 15:58 |
| dhanasekaran | Guys we have spinlock issue in our ubuntu kernel auto tune spinlock backoff delay factor I want to know the particular patch applied or not https://patchwork.kernel.org/patch/1904881/ | 15:59 |
| ogra_ | yeah, who needs acceleration on a media PC anyway | 15:59 |
| rtg__ | apw, ogasawara: raring 3.8.0-6.11 tagged, pushed, and uploaded | 15:59 |
| smb | tseliot, Yeah, since the legacy driver lags behind in supported X and kernels... | 15:59 |
| tseliot | smb: yes, exactly my point | 15:59 |
| smb | ogra_, For better energy management maybe. :) | 15:59 |
| ogra_ | pfft ... its not like you play movies or games on it .... errr :) | 16:00 |
| apw | rtg__, thanks | 16:00 |
| smb | ogra_, naaah :-P | 16:01 |
| smb | tseliot, btw, there is sort of a bug about that which has reached a hellish heat. ;) bug 1058040 | 16:01 |
| ubot2 | Launchpad bug 1058040 in fglrx-installer (Ubuntu) "fglrx-installer not working with AMD Radeon/Mobility Radeon HD 2000-4000 cards in Quantal" [Wishlist,Triaged] https://launchpad.net/bugs/1058040 | 16:01 |
| apw | dhanasekaran, that patch does not appear to be applied even to linus' tree, so it is doubtful we have it | 16:02 |
| tseliot | smb: my hands are tied. The legacy driver doesn't support Quantal's xserver ABI | 16:03 |
| smb | tseliot, I realized that when looking at that during the weekend. Just found the heat number amusing... :) | 16:04 |
| apw | dhanasekaran, though it really isn't going to affect small machines by my reading, where small is pretty big in 'our' world | 16:04 |
| tseliot | smb: I'll add a comment to the bug report... | 16:04 |
| dhanasekaran | apw: We have 32 core machine, we have issue with spinlock issue | 16:05 |
| dhanasekaran | apw: How to apply with ubuntu kernel | 16:05 |
| dhanasekaran | apw: I have go with vanilla kernel? with ubuntu base | 16:07 |
| dhanasekaran | apw: please guide me.. How to resolve this.. | 16:07 |
| infinity | dhanasekaran: It's not in the vanilla kernel, that was his point. | 16:09 |
| apw | dhanasekaran, i am supprised you would be seeing any sort of issues with that small a system but, if you want to test the patch you would have to apply them and build your own kernel as they are not yet in even linus' master tree so none of the trees we maintain would have the relaevant patch for testing | 16:10 |
| ogra_ | https://help.ubuntu.com/community/Kernel/Compile | 16:10 |
| apw | dhanasekaran, he does imply that the thing is a tunable, so there may be existing tunables in your kernel which may allow you to confirm this is the fix for your issue before you go to the effort | 16:10 |
| dhanasekaran | apw: please look https://patchwork.kernel.org/patch/1904881/ | 16:12 |
| stgraber | sforshee: hey, what's the status of bug 1098216? wasn't that supposed to make it to the Ubuntu kernel weeks ago? :) | 16:12 |
| ubot2 | Launchpad bug 1098216 in linux (Ubuntu) "Regression in brightness control on Lenovo Thinkpad x230" [Medium,Triaged] https://launchpad.net/bugs/1098216 | 16:12 |
| sforshee | stgraber, I got distracted by other stuff, and also spent a little time looking to see if there was a more generic way to quirk for this since we're up to 6 models now | 16:13 |
| stgraber | sforshee: oh, I see some activity in the upstream bug, right | 16:13 |
| sforshee | yeah | 16:13 |
| sforshee | I'm trying to get the new models tested | 16:13 |
| sforshee | sorry for the slow progress | 16:13 |
| mainerror | Hello there. | 16:15 |
| apw | dhanasekaran, i did look, that patch isn't in any kernel yet it is still only a "for discussion" patch | 16:15 |
| apw | dhanasekaran, no vanilla kernle has this patch therefore runing those will not help you | 16:15 |
| sforshee | stgraber, I'm going to shoot the patch upstream today to see if anyone has any better suggestions, I'll be sure to send it for raring sometime this week | 16:16 |
| stgraber | sforshee: no problem, I'm using the cmdline argument here, just wanted to make sure it wasn't blocked on something and was still going to make it in our kernel soon :) | 16:16 |
| mainerror | Is there any ETA on the integration of RC6 in Raring? | 16:16 |
| dhanasekaran | apw: thanks for you clarification. How to resolve my spin lock issue kill my server machine.. | 16:17 |
| infinity | mainerror: rc6 is already in raring. | 16:17 |
| mainerror | Whoa! Really? | 16:17 |
| mainerror | :) | 16:17 |
| infinity | mainerror: If you meant rc7, rtg just merged it, and is working out some final kinks. | 16:17 |
| apw | dhanasekaran, i have no idea what your issue even is, if you think this patch might resolve your issue, you could try adding this patch yourself and building a kernel for comparison | 16:18 |
| apw | infinity, unless he means the other rc6 the graphics one | 16:18 |
| mainerror | rc6 is just fine for me. I own a Samsung Series 9 notebook and the samsung-laptop drivers were a blocker for me. | 16:18 |
| mainerror | Great news then. Thanks! | 16:18 |
| apw | mainerror, is this the brickage laptop? | 16:18 |
| mainerror | Yea. | 16:18 |
| apw | is it in bios or efi mode ? as the issue only occurs in efi | 16:19 |
| mainerror | I'm aware, yea. | 16:19 |
| mainerror | In EFI. | 16:19 |
| apw | the safest thing to do is switch to bios | 16:19 |
| apw | not that the issue is actually linux there at all | 16:20 |
| apw | but actually a bug in the efi bios | 16:20 |
| mainerror | It is a bug in the BIOS yea. | 16:21 |
| mainerror | I know that in rc6 samsung-laptop has just been blacklisted which temporarily fixes the brickage when booting in EFI mode. | 16:21 |
| infinity | In theory, mind you, the current version of samsung-laptop no longer tickles the bug. | 16:21 |
| mainerror | Oh? | 16:22 |
| infinity | (or, as you say, doesn't load in EFI mode) | 16:22 |
| infinity | (which doesn't tickle the bug) | 16:22 |
| mainerror | Ah yea. | 16:22 |
| apw | i think you are referring to the same work-around in both cases | 16:22 |
| infinity | But... With a bug like that, I'm with Andy. Just run in BIOS mode. | 16:22 |
| apw | though you can do the same thing in windows it seems too | 16:22 |
| infinity | Yeah, Matthew wrote the Windows PoC as a way to wake people up to it not being a "Linux problem". | 16:22 |
| infinity | It's trivial to do from any OS, really. | 16:23 |
| mainerror | Then there's this other strange issue. Apparently those notebooks are a thermal mess. Watching two or three YouTube videos or doing other graphic intensive stuff will trigger a kernel panic. | 16:23 |
| mainerror | I wasn't able to pin it down completely yet. | 16:23 |
| mainerror | I believe it is a thermal problem of some sort. | 16:24 |
| infinity | It could just be a dead fan. | 16:24 |
| mainerror | The odd thing is that it doesn't happen on Windows 8 (default OS on that thing). | 16:24 |
| apw | mainerror, well ... it really can | 16:25 |
| apw | mainerror, you can kill your box in windows in about 10s | 16:25 |
| apw | windows just gets lucky, and ... well it would as they tested that before releaseing it | 16:25 |
| dhanasekaran | apw: check third fix it's look like spinlock solution http://git.kernel.org/?p=linux/kernel/git/torvalds/linux.git;a=commit;h=39923134885759405e64cb9491a273a2bcd356ed | 16:26 |
| mainerror | I was able to take a picture when a kernel panic occurred though. http://ubuntuone.com/0OzMcDHB5mIyKHpCmdIFe1 | 16:27 |
| apw | dhanasekaran, that looks like a completely different patch which avoids some lockups | 16:27 |
| apw | dhanasekaran, and it applies to arm cpus, is your 32cpu job arm based? | 16:28 |
| dhanasekaran | apw: no Intel cpu.. wrong search for me. | 16:30 |
| chadbyoung | boot process is hanging - please take a look it you have a minute. http://pastebin.com/1HviMyHR | 16:38 |
| chadbyoung | using an Odroid-pc board and trying to boot Ubuntu on a SATA disk | 16:38 |
| === henrix is now known as henrix_ | ||
| === henrix_ is now known as henrix | ||
| * ppisati -> gym | 17:05 | |
| infinity | rtg__: Why the NO_LIBPYTHON for perf? Is the feature useless? The build-dep certainly wouldn't hurt? | 17:30 |
| rtg__ | infinity, its a wrapper around the perf call. I didn't think adding yet another build dependency was really necessary. tools/perf/python/twatch.py is the only thing that requires it. | 17:32 |
| rtg__ | I think | 17:33 |
| rtg__ | maybe I'm all wet. | 17:33 |
| infinity | tperf just looks like PoC code to talk to the perf python module. | 17:33 |
| infinity | Which I assume is C, and in the parent directory? | 17:34 |
| infinity | Maybe? | 17:34 |
| rtg__ | not sure, still checking | 17:34 |
| infinity | script/python | 17:34 |
| infinity | Has a bunch of scripts, and a C module. | 17:35 |
| infinity | There's a perl module too, do we build that? | 17:35 |
| rtg__ | I think the only binary that gets packaged is perf itself. | 17:36 |
| rtg__ | looks like there has been some changes since the last major kernel version. | 17:36 |
| infinity | This stuff all so desperately wants to not be in the kernel source. | 17:41 |
| infinity | We should be building a proper python module package, a proper perl module package, all the cute GUI crap, etc. | 17:42 |
| infinity | But that has no business in a kernel build. | 17:42 |
| rtg__ | infinity, agreed. so NO_LIBPYTHON prevents some python support libraries from getting built. | 17:42 |
| * rtg__ -> lunch | 18:37 | |
| === rtg__ is now known as tgardner | ||
| === tgardner is now known as rtg__ | ||
| * ppisati -> dinner | 19:47 | |
| penguin42 | I just moved bug 1114673 to triaged; reporter confirmed it still happens on upstream - should I dupe bug 1070256 back to it? | 20:08 |
| ubot2 | Launchpad bug 1114673 in linux (Ubuntu) "mount.cifs occasionally causes GPF error/kernel panic when mounting at boot [crypto_larval_kill+0x2b/0x90]" [High,Triaged] https://launchpad.net/bugs/1114673 | 20:08 |
| ubot2 | Launchpad bug 1070256 in linux (Ubuntu) "mount.cifs occasionally causes GPF error/kernel panic when mounting at boot [crypto_larval_kill+0x2b/0x90]" [High,Confirmed] https://launchpad.net/bugs/1070256 | 20:08 |
| penguin42 | (1114673 was originally a dupe of 1070256 but the reporter of 1114673 was actually responsive and provided logs and tried upstream on his bug) | 20:09 |
| apw | that doesn't seem wholy unreasonable | 20:10 |
| penguin42 | done | 20:12 |
| penguin42 | the other one I'm not sure what to do with is bug 1093217 they've got a boot hang that appeared on 3.2.0.30 - they've tried a bisect but either something odd is happening or they've gone down a weird path | 20:15 |
| ubot2 | Launchpad bug 1093217 in linux (Ubuntu) "Ubuntu 12.04 10-20min boot delay (From 3.2.0.29->3.2.0.30) [Lenovo IdeaPad Z580]" [High,Incomplete] https://launchpad.net/bugs/1093217 | 20:15 |
| Kano | hi, what was the dl url for the 3.8-6 kernel? | 20:45 |
| ohsix | man you seriously can't figure anything out :[ | 20:46 |
| Kano | it is not in main yet | 20:49 |
| === henrix is now known as henrix_ | ||
| * rtg__ -> EOD | 21:35 | |
| === kentb is now known as kentb-out | ||
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!