[02:11] sconklin: I thought I had verified that patch already [02:28] jjohansen: I was unable to boot with the kernel you built me; details on launchpad [05:28] Stupid question: why does lucid-backports have linux kernels up to .35, but e.g. "compat wireless" modules up to .38? [05:29] Oh, I see, it looks like it's the driver taken from the .38 tree and compiled against .32 === smb` is now known as smb [07:42] morning [08:24] smb morning [08:25] apw, before my vacation I traced down a bug in PulseAudio that could be a gcc bug. Do you know how to proceed with that? [08:26] diwic, normally one would file a bug against gcc, they will want the minimum code fragment that causes the miss-compile and a good description of how it is miss compiled [08:27] against gcc-4.5 or whatever i guess [08:27] how do we know it is a compiler problem ? [08:27] apw: problem goes away when you change the compiler [08:27] as unfortuanatly code which is relying on undefined side effects of some code forms can then change if you change the compiler [08:28] twb: all other parts of the environment the same? [08:28] twb, one would really want to find the exact bit of code whihch is being miss built [08:28] apw, is bug 789031 filed against the right package? [08:28] Launchpad bug 789031 in gcc-defaults "Built-in ASM constraint "rm" does not work" [Undecided,New] https://launchpad.net/bugs/789031 [08:28] apw: granted [08:29] diwic now that even sounds familiar, i wonder if something in the kernel ht the same issue [08:29] diwic, i think i'd expect it to be gc-4.5 or gcc-4.6 depending on the version you are using [08:30] apw, actually in pulseaudio the problem was worse, it allocated one "r" and one "rm" into the same register [08:31] apw, but I couldn't get that done to a simple example [08:31] so i think other than i think you want to do the right version, it looks good. i would poke doko so he is aware, he is likely to recognise it if its come by already as a fix [08:32] ok [08:33] * diwic pokes in ubuntu-devel === jk-- is now known as jk- === Quintasan_ is now known as Quintasan [09:27] jjohansen, you around? Was given your name for the person to contact for possible interview series? [09:27] akgraner: hey, of course I am around, what are you doing up so late [09:28] jjohansen, late it's early here I'm about to start my day :-) [09:28] akgraner: ouch that is early [09:28] nah just sleep decided to hid from me so I thought I would get some work done :-) [10:29] hi, I upgraded my server from 9.04 to 10.04 (via 9.10); the server is connected in my lan via wlan(wpa2); ... [10:29] the problem now is that the speed downloading from the server to the lan is down from ~700kb to ~100kb, while uploading to it is mostly not affected (~700kB). ... [10:29] I tried: [10:29] a) using a wired lan connection->normal speed, [10:29] b) booting with kernel 2.6.28-19-generic -> wlan speeds are back to normal [10:29] c) booting with kernel 2.6.31-14-generic -> wlan speeds still degraded [10:29] question now is: what changed between the kernels and where do I start to fix this? any help would be highly appreciated [10:35] has someone tried to run a natty kernel in lucid? [10:37] jazz2: what wifi chipset? [10:39] intel wifilink 5100 agn [10:39] ohsix, where do I look for it [10:39] i think it's something like Realtek [10:40] ohsix, zd1211rw [11:08] cooloney: http://kernel.ubuntu.com/git?p=roc/ubuntu-lucid.git;a=commitdiff;h=fc58ea3748869d2fa811dff26aa0fa07490b22b2 [11:08] cooloney: i can't find any branch where this patch has been applied [11:08] cooloney: is it in use at all? [11:09] * smb -> lunch [12:09] * ppisati out to get some food [14:54] smb: your comment in https://bugs.launchpad.net/ubuntu/+source/linux/+bug/681083 - sorry I'm not up on all the virt stuff - does his test imply that the bug has been fixed in the XenMotion case, or only that the kernel doesn't fail in a major way? [14:54] Ubuntu bug 681083 in linux "Ubuntu Crashes/Freeze on XenMotion" [Medium,In progress] [14:56] sconklin, Whatever they exactly do in xenmotion I actually do not know. But one part is running generic kernels (-pae in i386) as domU guests. [14:56] So when he took that 2.6.35 kernel I provided and installed it into his Lucid guest, the Maverick kernel got tested [14:57] sconklin, Probably another argument could be that this patch is now added for the next 2.6.32 (may or may not get picked for .35 as well) longterm tree [14:58] So sooner or later it will come. But I heard you are slightly frustrated by longterm trees. :) [15:02] BenC: I see your test results in the bug - It's that the SRU process requires that someone actually test the kernel in -proposed, and it looks like you tested a test build to validate the patch. [15:03] sconklin: Ok, I'll test the actual build [15:03] thanks, much appreciated [15:23] hello [15:35] sconklin: Ok, confirmed...where do I add my two-cents? [15:36] add a comment and I'll set the tag. Thanks! [15:41] * BenC has forgotten the bug report [15:41] stand by [15:41] Found it in scrollback [16:05] apw ... [16:13] apw_ ? [16:13] apw ?? [16:39] * smb wonders whether apw exploded... [16:39] smb, trying out a new irc client was all [16:40] apw, Ah. :) [17:39] * smb -> eod [17:45] apw: tangerine seems to be hanging on builds (not just in schroots), who do we hit up to look into it [17:46] jjohansen, whats the symptoms [17:46] cnd what you doing to tangerine? [17:46] apw, fakeroot is hanging in the kernel trying to get some semaphore [17:46] apw: it just hangs, sforshee reported and I duplicated [17:47] sforshee, anything specific, which sema ? [17:47] apw, not sure [17:47] i can bounce it if thats useful [17:47] you'll see a bunch of faked-sysv processes in uninterruptible sleep [17:48] /proc//wchan shows call_rwsem_down_write_failed [17:48] apw, yeah it prolly needs to be bounced when cnd is done [17:49] ok well let me know when he is done, if he can be anythiong other than done [17:49] sforshee, given its actual makes which are spinning, i'd say he is not making progress [17:50] yeah they are dead i recon [17:53] sforshee, i don;'t think cnd is making progress [17:53] votes for bouncing ?? [17:53] jjohansen, sforshee ^^ [17:53] +1 [17:53] +1 [17:54] in progress [17:55] bjf, could you accept my natty nomination on bug #767192 ? [17:55] Launchpad bug 767192 in linux "Wireless flaky on Acer Aspire 5100 after installing Natty" [Medium,In progress] https://launchpad.net/bugs/767192 [17:55] thanks apw [19:16] anyone around that can accept my natty nomination on bug #767192 ? [19:16] Launchpad bug 767192 in linux "Wireless flaky on Acer Aspire 5100 after installing Natty" [Medium,Fix released] https://launchpad.net/bugs/767192 === yofel_ is now known as yofel === Quintasan_ is now known as Quintasan [21:41] Hi there, are there any plans to publish a kernel update for natty based on a recent -stable (> 2.6.38.6) kernel version? [21:43] I am asking because I am affected by bug #746860, which has already been fixed upstream [21:43] Launchpad bug 746860 in linux "System sporadically freezes during suspend to RAM" [Undecided,Confirmed] https://launchpad.net/bugs/746860 [21:49] Specialist: the natty kernel currently in -proposed is current thru 2.6.38.8 but has two known regressions that we will have to resolve before we can fix it and spin another release [21:50] https://bugs.launchpad.net/ubuntu/+source/linux/+bug/792013 is the tracking bug for that, and has other bugs linked from there for the regressions [21:50] Ubuntu bug 792013 in linux "[Regression] linux: 2.6.38-10.44 -proposed tracker" [Medium,In progress] [21:50] sconklin: thanks! i'll see whether those affect me and will probably just take the kernel from -proposed if they don't [21:52] Specialist: if either of them affect you in a way you can reproduce, we could use help bisecting kernels to locate the problem. I think our main reporter is gone until Monday . . . [21:53] sconklin: ah cool, I didn't know -10 in -proposed carried all the way through 2.6.38.8 [21:53] Specialist: sorry, I got that wrong. The release in -proposed has up through 2.6.38.7 [21:53] * vanhoof has a few bugs that will get closed out then :) [21:53] ah rats :D [21:54] we've temporarily stopped taking upstream updates because of the load of regressions we got in the last batch [21:55] vanhoof: anything you really care about, open an SRU bug even though it's in a stable update, and get it on the kernel list for acks. [21:55] sconklin: np, as long as it's > .6, it's fine ;-) i'll let you know if i see anything unusual comparable to the two regressions with the -proposed version [21:55] sconklin: is the bisect progress documented somewhere? [21:55] vanhoof: sorry to push things into that path, but we're trying not to take another multi-week train wreck [21:56] Specialist: hold in - herton has been doing those ^^ [21:57] sconklin: well .8 will make its way in at some point [21:59] Specialist: for bug 793796, the bisect progress is on last comment on bug report [21:59] Launchpad bug 793796 in linux "2.6.38-10 panic after ejecting drive" [Undecided,Confirmed] https://launchpad.net/bugs/793796 [22:00] it's waiting for the reporter to test the next step [22:01] the other one (bug 794096) is on a bisect also with the reporter, he will only be able to test next bisect next week... [22:01] Launchpad bug 794096 in linux "SMTP and posting to a web-form time out (probably due to netfilter changes)" [Undecided,In progress] https://launchpad.net/bugs/794096 [22:02] this last one is weird, as reporter already tested with the most likely changes which could cause the issue reverted [22:03] and there isn't much info yet which could point to real issue [22:04] I tried to reproduce both here but without success, so if there is anyone else being able to reproduce the same and able to help would be great [22:58] hi, which config option is the problematic one that now a newer module-init-tools is required [22:59] rc2 worked fine with lenny, rc3 does not [23:00] but even with squeeze 3.0rc3 works,so you do not need that extreme depend [23:00] i lower it in my install scripts anyway [23:00] this just takes useless time [23:58] Kano, its there becase we need it in the oneiric tree where the versions are correctly specified, and its getting copied to the mainline builds. i'll fix it when i have time [23:59] as you have noted you can actually fix it yourself as you make your own packages [23:59] no i use dpkg-deb [23:59] http://kanotix.com/files/fix/mainline/install-daily.sh [23:59] take a look