| rtg | apw, pushed raring rebased on v3.7-rc4, build testing now | 12:48 |
|---|---|---|
| cking | rtg, I'll be happy to give it a test spin on various boxes | 13:06 |
| rtg | cking, I had a docs build failure that I'm still tracking down, but the kernel binary builds OK if you want to give it a go | 13:08 |
| cking | ack | 13:09 |
| cking | rtg, have you any .debs that I can download? | 13:10 |
| rtg | cking, working on it. the docs failure stops the build before the kernel binaries are packaged. should have them in 10 mins or so (on gomeisa) | 13:11 |
| ricotz | rtg, hi, i ran into having broken input (keyboard/mouse) earlier (raring master-next with rc4 merged) | 13:13 |
| rtg | ricotz, I've only _just_ [ushed -rc4. are you sure it wasn't an earlier version ? | 13:14 |
| rtg | *pushed | 13:14 |
| ricotz | rtg, i merged it myself yesterday | 13:15 |
| rtg | ricotz, well, if you have a kernel version that does work, then how about starting a bug report and working with jsalisbury to bisect the issue ? | 13:16 |
| ricotz | rtg, i am not sure if this is something real, i guess i will try to reproduce it with the new rebased branch | 13:16 |
| ricotz | also having nvidia-blob around might be doing this | 13:17 |
| ricotz | just wanted to mention it, but you probably not running it yet | 13:17 |
| rtg | nope, haven't tested it. I've asked Alberto to have a look | 13:18 |
| ricotz | patched nvidia works fine with the rc3 | 13:19 |
| rtg | cking, I turned on signed modules. do you have an external module you could build and install to make sure the kernel complains ? Isn't there something in fwts ? | 13:46 |
| cking | rtg, I will give it a spin | 13:47 |
| rtg | cking, we decided to not make requiring signed modules the default, but there some Xen folks that would like to use it. | 13:48 |
| cking | rtg, the skarkbay laptop works so much batter with that 3-7rc4 kernel | 14:52 |
| rtg | cking, haswell ? | 14:53 |
| cking | rtg, yep | 14:53 |
| rtg | huh | 14:53 |
| rtg | I guess thats a good thing | 14:53 |
| cking | well, I get get a desktop now that works, and 3D games render OK w/o it dying | 14:54 |
| * ogasawara back in 20 | 15:39 | |
| bjf | rtg, got that -rc4 kernel somewhere handy | 16:16 |
| bjf | ? | 16:16 |
| rtg | bjf, gomeisa.buildd:~rtg/ukb/raring/amd64/master-next | 16:17 |
| apw | bjf, i need to drop an patch for the linux-lts-quantal package on the branch see any issues with that? | 16:37 |
| bjf | apw, you need to commit a patch to quantal master-next branch ? | 16:37 |
| apw | bjf, nope specially to the lts-quantal version in precise, to the branch specific machinary, to enabled signed kernels there | 16:38 |
| bjf | apw, ah! i see no problem there | 16:39 |
| bjf | apw, you are free to maneuver | 16:39 |
| apw | bjf, thrusters to one half indeed | 16:39 |
| bjf | rtg, how come only the -generic headers were built and not the -all? | 16:41 |
| rtg | bjf, likely because of the way I built. lemme check | 16:41 |
| apw | bjf, herton, does shankbot know about when -signed is needed yet. if not we likely need to codify that too | 17:46 |
| bjf | apw, i believe it is already on heron's todo list | 17:46 |
| apw | purr'fec | 17:46 |
| apw | we probabally talked about it over beer, and that is what i am blaming for me not remembering | 17:47 |
| infinity | bjf, herton: While we're talking shankbot wishlists, it's getting unwieldly for people who aren't intimately familiar with the process to hunt down all related/dependant packages. Maybe the description should list them all, so we don't miss any on promotions? (linux, lbm, meta, signed, etc) | 17:47 |
| infinity | bjf, herton: Granted, that can be worked out from the prepare-* tasks (well once there's a prepare-signed), so maybe it would be duplicate info, I dunno. | 17:47 |
| infinity | None of this is a huge deal while I'm the one doing the AA side of all of this, cause I know what's up, but I imagine it'll make someone's head explode if they have to cover for me on vacation. | 17:48 |
| bjf | infinity, i'm not sure what you'd like to see .. a wiki page explaining the relationships? | 17:48 |
| infinity | bjf: No, a list in the description of "these are the packages involved in this SRU". | 17:48 |
| bjf | infinity, ah, ok, that seems reasonable | 17:49 |
| infinity | bjf: Since the bug is (technically) only against linux (or linux-flavour), it's not immediately obvious that "promoting this also means promoting meta, lbm, signed...) | 17:49 |
| bjf | infinity, wouldn't they get "the idea" when we pocket copy those packages? | 17:50 |
| infinity | bjf: Well, I tend to do all the PPA copies too. But, even if that weren't the case, you/me doing the PPA copy today, and someone else doing the promotion in two weeks, is a fair bit of disconnect. | 17:51 |
| * infinity shrugs. | 17:52 | |
| infinity | It's not a huge deal, and I'd like to think I'll train someone a bit before they cover for me anyway. | 17:52 |
| infinity | And, to me, the relationships are obvious. | 17:53 |
| bjf | infinity, i'm not disagreeing, i'm just thinking about it | 17:53 |
| infinity | But I can see how the list on sru-report.html would be mind-bending to someone who's not used to reading it. ;) | 17:53 |
| infinity | Actually, maybe we could just make the report more clever. | 17:53 |
| bjf | i was thinking that as well | 17:54 |
| infinity | If it could group things in related units, so the base/meta/lbm/signed were always together. | 17:54 |
| * rtg -> lunch | 19:08 | |
| herton | apw, yes shank bot should be covering the signed packages only for quantal and later now | 19:17 |
| herton | we will need to update this though on the backport for precise and may be precise when it's done there too | 19:18 |
| infinity | apw: When is the first kernel upload for raring going to happen, BTW? Trying to decide if I should care about copying qunatal SRU kernels to raring, or just leave it be. | 19:22 |
| infinity | ogasawara: ^ | 19:23 |
| ogasawara | infinity: I say just leave it be, we just rebased to v3.7-rc4 today so I suspect we'll attempt an upload of this soon baring anything catastrophic | 19:24 |
| infinity | ogasawara: Shiny, thanks. | 19:24 |
| infinity | ogasawara: If some dreadful security vuln shows up in the next SRU cycle and R still doesn't have its own kernel, I'll reevaluate. :P | 19:25 |
| bjf | rtg, rc4 working ok on my macbook air | 19:27 |
| bjf | rtg, also on SandyBridge HEDT | 19:36 |
| jdstrand | jjohansen: hey, so this linux-lowlatency, is that a supported kernel (ie, it gets USNs)? | 20:03 |
| jjohansen | jdstrand: no, its community supported | 20:04 |
| jdstrand | jjohansen: so, how does that work with the process? | 20:04 |
| jdstrand | ie, the cadence process | 20:04 |
| jjohansen | jdstrand: its just not part of the process, they rebase on our kernels | 20:05 |
| jdstrand | oh I see. it is entirely driven by them. cool | 20:05 |
| jdstrand | jjohansen: thanks | 20:06 |
| arges | jsalisbury, hi | 20:15 |
| jsalisbury | arges, hey | 20:27 |
| arges | jsalisbury, were you able to re-open the upstream bug for bug 922906 | 20:27 |
| ubot2` | Launchpad bug 922906 in linux (Ubuntu) "Kernel Oops - BUG: unable to handle kernel NULL pointer dereference at 0000009c; EIP is at __ticket_spin_lock+0x8/0x30" [High,Triaged] https://launchpad.net/bugs/922906 | 20:27 |
| arges | jsalisbury, i was going to start investigating this a bit more | 20:28 |
| jsalisbury | arges, I wasn't able to, I didn't have the appropiate premissions | 20:28 |
| arges | jsalisbury, ok should i reopen after i take a look at it a bit? | 20:29 |
| jsalisbury | arges, I noticed Ben Hutchings requested that it get reopened as well, but never got a respnse | 20:29 |
| arges | i mean file a new bug not re-open | 20:29 |
| jsalisbury | arges, sure, that may get it moving again | 20:29 |
| jsalisbury | arges, there were a couple of discussions on LKML about it as well, but none ever went anywhere. | 20:31 |
| jsalisbury | arges, I think it was left off around March of 2011 | 20:32 |
| jsalisbury | maybe april | 20:32 |
| arges | ok i'll get more data first | 20:32 |
| jsalisbury | arges, I think henrix and apw also looked at this for a while and there was even a test kernel. | 20:34 |
| jsalisbury | arges, Actually there was a fix upstream on fsdevel: http://marc.info/?t=130806561400010&r=1&w=2 I don't think it was ever picked up in mainline though. | 20:37 |
| arges | jsalisbury, yea i'll have to dig on this a bit more | 20:40 |
| * rtg bugs out | 20:51 | |
| bjf | ogasawara, i just boot tested rc4 on two server systems, it's up on both however, the emerald ridge system has no video and lots of backtrace in the dmesg | 21:11 |
| ogasawara | bjf: ack | 21:12 |
| === yofel_ is now known as yofel | ||
| akhileshss | Hi, I am trying to add a new system call to kernel 3.6.5. I followed the steps here: http://goo.gl/f0hk7 . But I hit the error "arch/x86/built-in.o: …: undefined reference to "sys_foo". I am trying to put new sys call foo. Can anyone help me figure out what is missing ? | 21:38 |
| bjf | ogasawara, i tested rc4 on 7 systems, only the one had any issues | 22:01 |
| ogasawara | bjf: ok good to know. are you able to pull the logs from that emerald ridge? | 22:02 |
| bjf | ogasawara, sure, it's up and running if anyone wants to also poke at it | 22:02 |
| ogasawara | bjf: I'll sync with rtg in the morning about debugging it | 22:03 |
| bjf | ogasawara, it's "statler" in the lab | 22:04 |
| ogasawara | bjf: ack | 22:04 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!