=== lifeless_ is now known as lifeless [09:05] eep, first victim of yama/ptrace_scope changes: wine will no longer run Rosetta Stone [09:09] .. not that it ran that well in the first place, but would indicate that wine uses ptrace() on non-child processes [09:11] jk-: you can turn yama off [09:11] yep [09:35] cooloney: did you look at the patches that Tony posted on linux-omap for SMP kernel on UP? [09:36] I asked him to post patches so you or someone else can test it [09:36] amitk: yeah, man [09:36] amitk: i tested rmk's version [09:36] and tony posted a fixing just now [09:36] i am going to testing that on my board [09:37] cool [09:37] amitk: so we a close to boot smp kernel on beagle, i think. [09:37] cooloney: hopefully :) [09:38] amitk: but the patch changes several important low level assemble line code. [09:38] need more testing and review [09:38] cooloney: Tony said that he doesn't think we need to rewrite locking primitives during runtime like is done on x86 [09:40] amitk: no sure about that. spin_lock version is different in smp and up, i think. === Nicke is now known as 18VABDTK5 === maxb_ is now known as Guest7510 === amitk is now known as amitk-afk === amitk-afk is now known as amitk === Guest7510 is now known as maxb === maxb is now known as Guest10006 === maxb_ is now known as maxb === lamont` is now known as lamont === zul_ is now known as zul === JayFo is now known as JFo === sconklin-gone is now known as sconklin === bjf[afk] is now known as bjf === 18VABDTK5 is now known as Nicke_ [16:29] not traffic shaping, just saturatign my uplink [16:29] the tests that failed were obviously ones that were not for ext4, but were not correctly excluded from the tests [16:31] Hm, ok. I think I had at least one about aio but those failed before and after. [16:31] * smb goes to reboot the test machine [17:04] is the -rt kernel being added to main that anyone knows of? [17:05] rather -lowlatency and -preempt [17:05] I am being asked about them [17:06] apw or tgardner ^^? [17:07] JFo, preempt is an official flavour for Lucid, not for Maverick [17:07] ah, I see [17:08] anything for the -lowlatency? [17:08] JFo, not in maverick [17:08] nor lucid [17:08] tgardner, thank you [17:09] tgardner, is there any rationale for not including them? [17:09] I have been asked that specifically [17:10] the -rt guys are providing a kernel AFAIK, so it was duplicated functionality. [17:10] cool, thanks again tgardner === ivoks is now known as ivoks-away [20:44] ps ax is hanging over here, known issue? kernel or elsewhere? [20:45] fwiw it gets to 1751 ? S 0:00 udisks-daemon: polling /dev/sr0 [20:46] sabdfl: what does "strace ps ax" show it's stuck on? [20:47] read(6, "Name:\tchromium-browse\nState:\tD ("..., 1023) = 786 [20:47] close(6) = 0 [20:47] open("/proc/1898/cmdline", O_RDONLY) = 6 [20:47] read(6, [20:47] ew [20:47] sabdfl, maverick? [20:47] tgardner: very :-) [20:48] cat /proc/version_signature [20:49] Ubuntu 2.6.35-16.22-generic 2.6.35.2 [20:49] ~/ $ ls /lib/modules/2.6.35-16-generic/updates/dkms [20:49] hid.ko hid-magicmouse.ko vboxdrv.ko vboxnetadp.ko vboxnetflt.ko [20:50] sabdfl, lots of stable fixes since then. I hate to give you the runaround, but can you install just the -proposed kernel and try again? [20:50] sure, url? [20:50] one sec [20:50] didn't realise we had -proposed on maverick [20:51] sabdfl, doh, I was thinking Lucid. never mind. [20:51] :-) [20:51] i think about lucid often, too :-) best release EVAR [20:51] though we do have stable updates in the pipe for Maverick [20:51] how's .35 treating us? [20:52] not too bad, but there is a known mm regression in the version you are running. I wonder if thats it [20:53] ogasawara should be releasing an update with a couple of days [20:53] is it always chromium it blocks on, or is it at random? [20:54] kees, 'mm: fix page table unmap for stack guard page properly' is what I suspect [20:55] tgardner: what sha is that? [20:55] tgardner: I think we have it in the -security kernels, but I wanted to double-check [20:55] in the maverick repo its 6106dafae6d0ff50c82bb544b3b999562238504b [20:56] sabdfl: While hung, writing W to /proc/sysrq-trigger should give you the in-kernel backtrace for the blocked task [20:57] sabdfl, you could also reboot to Ubuntu-2.6.35-14.20. Thats based on 2.6.35 _before_ any stable updates (which actually destabilized things a bit) [20:57] yup [20:57] mjg59: hello there! nice to see you again [20:58] so, echo W > /proc/sysrq-trigger ? [20:58] Yeah, as root [20:58] echo W | sudo tee /proc/sysrq-trigger [20:58] tgardner: still have it around, iirc, so yes can roll back [20:59] sabdfl, remember you have to hold the shift key just prior to grub getting control or you'lll miss the grub menu [20:59] so, that's put stuff in dmesg? [20:59] Yup [21:00] lots of blech at http://pastebin.ubuntu.com/480605/ [21:01] sabdfl: Hm. Sure you did w? And ps is still blocked in read? [21:01] mjg59, anon_vma_prepare, wasn't this what Linus was seeing? [21:02] Yeah, looks like something's very unhappy there [21:03] * jjohansen -> lunch [21:03] I'm pretty sure thats the stack guard regression [21:04] mjg59: just tried again, ps ax in one terminal tab, and the echo W magic in another [21:05] ps doesn't get unblocked [21:05] Yeah, ps is pretty solidly dead at that point [21:05] One of those scheduling while atomic errors probably left a lock held [21:08] sabdfl, I think your best bet for now is to use the older kernel until ogasawara get the next stable update uploaded. [21:08] okdokey [21:08] sabdfl: hoping to upload this afternoon actually. just have a few more patches I wanted to review and possibly apply. [21:09] other than the vbox bits, which i am not using so doubt they are active, i'm on standard bits with dell m1330 [21:09] so i wouldn't be surprised if others see the same thing [21:09] thanks leanne [21:10] sabdfl: I think I've got a test kernel with a fix for that bug 480605, just a sec [21:10] Launchpad bug 480605 in telepathy-haze (Ubuntu) (and 2 other projects) "Repeated subscription request from same user (affects: 12) (dups: 2) (heat: 78)" [Medium,Triaged] https://launchpad.net/bugs/480605 [21:11] bah, not bug, the pastebin [21:11] ok, will let you guys know if the next update fixes it [21:11] sabdfl: bug 618846, http://people.canonical.com/~ogasawara/lp618846/ [21:11] Launchpad bug 618846 in linux (Ubuntu) "Kernel 2.6.35.2 reports "scheduling while atomic" (affects: 1) (heat: 6)" [Medium,Fix committed] https://launchpad.net/bugs/618846 [21:13] will reboot to prior kernel. thanks for the help, hope the report was helpful, nice to see you mjg59 [21:28] * ogasawara drops to grab lunch, back in an hour [22:03] can someone provide a pointer to the differences between ubuntu server kernel and desktop kernel? i have found this: https://help.ubuntu.com/10.04/serverguide/C/preparing-to-install.html but was wondering if there are any other differences [22:06] joshhunt, there are no source code differences. the only differences are in some of the config options. [22:07] tgardner, ok thanks. that's what i suspected, but wasn't sure. [22:09] tgardner, is there a place where i can download the kernels manually. instead of going through apt, etc? === sconklin is now known as sconklin-gone [22:45] joshhunt: I tend to do it through Launchpad. https://edge.launchpad.net/ubuntu/+source/linux === yofel_ is now known as yofel