=== sconklin is now known as sconklin-gone [00:14] kees, can you point me at the bug and patch so we can look it over === bjf is now known as bjf[afk] [00:17] apw: bug> https://bugzilla.kernel.org/show_bug.cgi?id=16294 patch> https://lists.ubuntu.com/archives/kernel-team/2010-July/011457.html [00:17] bugzilla.kernel.org bug 16294 in Video(DRI - Intel) "[Q35 bisected] hang at init of i915 driver" [High,New] [00:35] what does this mean in /proc/cpuinfo ? [00:35] address sizes : 36 bits physical, 48 bits virtual [00:43] jk-: any bright ideas on getting my t91mt touchscreen working? === andreas_ is now known as anoteng [05:10] apw and ogasawara can ya confirm this bug (it's about the ubuntu 10.04 configuration and the newest kernel releases, along with a fix) https://bugzilla.kernel.org/show_bug.cgi?id=16325 [05:10] bugzilla.kernel.org bug 16325 in Input Devices "i8042 input regression" [High,New] [05:21] What does the verb "to checkpoint to" mean? As in /usr/src/linux-source-2.6.32/drivers/platform/x86/thinkpad_acpi.c [05:30] bullgard_: 'save', i guess [05:31] bullgard_, another guess would be the intended destination. [05:49] jk-: I will accept your proposal as a working hypothesis. -- Thank you for your help. [05:50] dandel: I did ask for the meaning of a verb. I know the meaning of the noun "checkpoint" in programming. === cooloney_ is now known as cooloney|dell [07:20] ls [08:18] make nconfig feels a lot nicer than make menuconfig [08:44] cking, \o [08:44] * apw yawns [08:44] * abogani waves [08:45] * cking yawns, i was coding at 5.30am [08:45] * smb waves back in generic directions [08:45] Gooooooood morning kernel team! [08:45] cking, bad cking [08:45] * apw waves [08:46] Daviey, Just replied to your request. Its upstream now, though in two parts. But I have split the patch in your name [08:46] The kernel patch mentioned in: https://lists.ubuntu.com/archives/kernel-team/2010-July/011460.html has now been applied to linus' 2.6 tree.. Does this change anything that i should be doing? [08:46] smb, heh [08:46] Daviey, Only nearly beat you to it. :) [08:47] smb, Thanks.. appreciated your work on this [08:57] Good morning smb, apw :) [08:57] RAOF, morining [09:28] apw: could you remove me as admin of kernel-team ML? I seem to have misplaced the password to do it myself. [09:28] amitk, i may be able to [09:32] * apw waves to lag-mobile [09:33] apw: thanks === lifeless_ is now known as lifeless [09:44] amitk, ok i think you should no longer get spammed by the list [10:05] super [10:05] moin === lifeless_ is now known as lifeless === amitk is now known as amitk-afk [10:52] hmm, no lag and no mpoirier yet [10:53] ogra, lag has network issues [10:53] ah [10:53] * ogra has kernel issues :P [10:53] We are the team with a kernel problem. :-P [10:54] well, it pays off, doesnt it ? [10:54] Literally. :) [10:54] heh [10:58] argh ! [10:58] why is the omap4 metapackage named linux-*ti*-omap4 [10:58] * ogra sighs [11:03] * ogra guesses he has to wait until tgardner gets up for that to be fixed :/ [11:14] ogra, wassup? [11:16] * apw pokes ogra [11:16] apw, well, a) the package doesnt follow the general naming scheme we have [11:16] and b) a 2.6.35 meta package depends on a 2.6.34 image [11:17] see #ubuntu-devel === lag is now known as Guest50263 [11:32] lag_, so you are back and stable ? === lifeless_ is now known as lifeless [11:32] lag_, we found i little prob with the omap3 kernel yesterday, seems the NAND driver vanished between lucid and maverick [11:39] ogra: No, I am using my mobile phone ... Eeek! [11:39] ouch [11:39] ogra: I can't help you with NAND - BBXM doesn't have any [11:40] you dont have the C4 ? gah, i always forget who has what [11:40] i'll wait for mpoirier [11:40] i suspect its just a config skew [11:43] ogra: No problem - sorry I couldn't help === lifeless_ is now known as lifeless === amitk-afk is now known as amitk === lifeless_ is now known as lifeless [12:59] ericm|ubuntu: dude! [13:00] amitk: I am here, :) [13:01] ericm|ubuntu: I'd like you to introduce you to yao who works for CodeSorcery and is based in Beijing [13:03] er, CodeSourcery [13:11] sorcery and sourcery are not entirely overlapping fields ... [13:16] * apw waves to yao [13:17] apw: hello, [13:17] welcome to our little club :) [13:18] apw: thanks, my pleasure... [13:19] ams_cs: IMO sourcery and sorcery is one and the same for you compiler writers :-p [13:19] anything which takes longer to build than the kernel is a black art by definition [13:21] apw: black art doesn't need compiling - that's done in assembler [13:21] so very true [13:35] * lag_ has his internet-net in tow and it going to bag some cloud (going to buy an internet dongle) === lifeless_ is now known as lifeless === lag is now known as Guest82967 [13:40] lag_, lag just reappeared [13:43] =:-0] [13:43] But I have my internet-net ready and everything! [13:43] * lag_ takes his jeans off and puts back on his 'comfy pants' [13:44] too much detail [13:44] I was just uploading pictures to Photobucket - you don't want the link? [13:47] *slap* [13:48] lag_, you sound like you are under water [13:48] * cking has all sorts of issues with maverick - wifi not working, X chewing up CPU, machine running at 85 degrees C [13:48] cking, is that a real maverick install or just the kernel ? [13:49] just done a full upgrade and regrettingit [13:49] apw: Yeah, I think the internet is still not quite right [13:49] But at least it's up [13:49] yeah i expect they have some damn string papered over the crack [13:50] And not wet it yet? [13:50] right, they always forget to spit on it [13:50] but it'll short out if it gets wet [13:50] Yum [13:51] tgardner, yeah but its pretty while its working :) [13:52] cking, there is no going back. now you have to fix it :) [13:52] hrm, when I stop moving the cursor it vanishes [13:53] urgh [13:53] cking, thats the new surprise feature. its always a surprise when it reappears [13:53] very amusing [14:01] cking: that's "unclutter" [14:01] the name of the program which does that, that is.. [14:03] that's an intentional feature? [14:07] yao, hi - nice to meet you [14:08] ericm|ubuntu: me too, good evening, :) [14:08] yao, so you are the 2nd employee here in China of CodeSourcery? [14:08] yao, did you ever meet Jie? [14:08] ericm|ubuntu: right, jie si the first one, [14:09] ericm|ubuntu: nope, :( [14:09] yao, cool - guess we will find chances to meet together [14:09] ericm|ubuntu: Yes, we should, [14:10] yao, so are you working on CS toolchain? [14:10] ericm|ubuntu: I see your photo in linaro wiki yesterday, [14:10] ericm|ubuntu: yes, [14:11] yao, cool - now I know who to harass should there be issues in toolchain ;-) [14:12] ericm|ubuntu: oh, absolutely, === arun__ is now known as arun === cnd` is now known as cnd [15:07] * cking may go offline, CPU is at 85 degrees and I'm not happy about it overheating [15:08] Neither the CPU I guess [15:08] heh [15:08] it's smokin' [15:14] I'm looking at Bugs #561926, #425264 & #550401 concerning Thinkpad X31 suspend/resume issues. Would anyone in #ubuntu-kernel be able to offer a fix for this issue? [15:14] Launchpad bug 561926 in linux (Ubuntu) "Thinkpad X31 fails to resume after suspend (affects: 6) (heat: 62)" [Undecided,Incomplete] https://launchpad.net/bugs/561926 [15:14] Launchpad bug 425264 in linux (Ubuntu) "[IBM 2673PXG] Thinkpad X31 suspend/resume failure (affects: 6) (heat: 38)" [Undecided,Triaged] https://launchpad.net/bugs/425264 [15:15] that's better [15:19] komputes, I think there are some patches coming via stable updates in Lucid that might help this. You could have them try the Lucid kernel in http://ppa.launchpad.net/kernel-ppa/pre-proposed/ubuntu (if it ever builds) [15:21] tgardner: thanks tim, so this is something that will be included in maverick kernels [15:21] komputes, yep [15:21] tgardner: sweet, thanks === sconklin-gone is now known as sconklin [15:32] Hrm, Lenovo "Rest in Peace". It's dead jim [15:36] cking, did you manage to let the smoke out of your Lenovo? [15:36] it's not a happy bunny [15:36] normally the thermal trip kills it at 80 degrees C, it got to 95 and it's not happy [15:37] ho hum [15:37] Time for the fridge? [15:37] time to get reserve H/W out of the box [15:39] df [15:45] heh, plenty of H/W to chose from [15:47] cking, testing your machines to destruction again? [15:48] not by choice [15:48] I think my lenovo's days are looking numbered [15:48] was that running maverick ? [15:48] yes, for about 10 mins [15:49] cooked it [15:49] X was maxing out one CPU [15:49] not had any issues with stuff running maverick kernels overheating ... [15:49] cking: my friends laptop runs at ~100c and HP says it's "normal" [15:49] hp is full of doo-doo if they think anything will live long at 100c [15:50] komputes, my lenovo usually hits thermal trip at 80 [15:50] agreed [15:50] I don't want a machine that can boil water [15:50] ;-) [15:50] * cking will get the screwdriver out over the weekend and poke around [15:52] my last HP laptop had some weird multi-threaded ACPI setup, and it took a while before a kernel came out that could understand that === bjf[afk] is now known as bjf [15:53] I smelled really hot metal once, and I think it got to 105 degrees before it shut itself down. It still runs, though. [15:53] it's when one sees blue puffs of smoke you know you are in trouble [15:55] bonkers ... how can that heat hope to stay away from ones groin ... scorching [15:56] it's a laptop not a nut cooker [15:57] cking, :) [16:05] apw, hmm, seems we still get a linux-ti-omap4 binary ... http://ports.ubuntu.com/ubuntu-ports/pool/main/l/linux-meta-ti-omap4/ [16:07] ogra, the source package is still names ti-omap4, but the binaries it produces should be just -omap4. I don't see any incorrect 901.2 packages yet 'cause I think its still pending a build. [16:08] ogra, none of the binary packages have even been NEWd so the ones there are likely old [16:08] pgraner, ping [16:08] apw, oh, indeed, i'm confused because the former ones also ended up in universe [16:08] tgardner: What's the difference between doing "git diff HEAD..FETCH_HEAD" and "git diff HEAD FETCH_HEAD"? [16:08] ogra, indeed only the source is in there, the binaries are not there at all [16:08] i was just reacting on the new dir that showed up suddenlx [16:09] lag_, i would expect them to be identicle [16:09] tgardner, the build finished a while ago [16:09] What's the .. all about then? [16:09] tgardner, https://edge.launchpad.net/ubuntu/+source/linux-meta-ti-omap4/2.6.35.901.2/+build/1850527 [16:09] it works like a set operator [16:09] ogra, oh. I would have expected the binary packages to end up in main [16:09] lag_, the and .. forms are synonymous [16:10] sorry for the confusion, there is indeed a linux-omap4 [16:10] tgardner, that needs some admin fiddling for new packages [16:10] tgardner, but its all sorted [16:10] kk - thanks [16:10] ogra, the binaries should end up in the source packages directory right? and that has ti- in it still as its the branch name which is not changes [16:11] lag_, the first form is the original form, the second is there as that has become the common form for other commands, and it allows understanding of the ... form as well [16:11] apw, right, NEW is the issue here [16:11] ogra, indeed, and they are new packages cause they changed names, so all is as expected [16:12] yep [16:12] confusion was on my side [16:12] my brain is fighting the heatwave here === kamal-away is now known as kamal [16:22] cking: pong [16:34] pgraner, earl send me a couple of laptops but they are stuck in customs - I got a demand for £143.09 to pay the tax and release them [16:35] do I pay this and we extract the expense from earl later or what? [17:30] hi [17:31] what's the correct way to commit a kernel patch to ubuntu? [17:34] m4tr1x, normally we look at patches sent tot he kernel-team email list, details can be found in the FAQ: https://wiki.ubuntu.com/Kernel/FAQ#Can I get a patch included in the Ubuntu Kernel? [17:35] ok [17:35] apw, make-kpkg has a few problems on ubuntu 10.04 and making sure the i8042 driver is built into the kernel in the latest torvalds git repo. [17:35] thanks [17:35] i fix the problem with marvel 9123 sata3 controller [17:36] m4tr1x, sounds excellent, thanks for looking at it ... is there a bug report open for that ? [17:36] i think yes [17:37] dandel, yeah not many of the kernel team use the make-kpkg so i can believe it could have bit-rotted for our use [17:37] dandel, whats the problem with i8042 ? [17:37] latest kernel git doesn't include it at all. [17:37] however, i tested against 2.6.35rc1, 2.6.35rc2 and so forth [17:37] dandel, ?? [17:38] it's not built at all [17:38] https://bugzilla.kernel.org/show_bug.cgi?id=16325 (i made a kernel report about it) [17:38] for maverick kernels or for the mainline kerenls [17:38] bugzilla.kernel.org bug 16325 in Input Devices "i8042 input regression" [High,New] [17:38] mainline [17:38] i used make-kpkg to make it easier to clean up after each bisect (but it took an insane time to build) [17:39] apw, https://bugzilla.kernel.org/show_bug.cgi?id=14543 [17:39] bugzilla.kernel.org bug 14543 in Serial ATA "AHCI SATA: hard link reset on controller power management change" [Normal,New] [17:39] :D [17:39] m4tr1x, sounds bad ... thanks [17:39] i already identified the patch which is causing the chaos though: namely commit: 0b28bac5aef7bd1ab213723df031e61db9ff151a [17:39] make sure you mention that in the bug report === cking is now known as cking-afk [17:40] dandel, i wonder why our maverick kernels are not affected [17:40] being 35-rc3 based [17:40] anyways, how do i get make-kpkg to let me customize the config params then. [17:40] https://bugs.launchpad.net/ubuntu/+source/linux/+bug/539467 [17:40] Launchpad bug 539467 in pm-utils-powersave-policy (Ubuntu Lucid) (and 4 other projects) "SATA link power management causes disk errors and corruption (affects: 19) (heat: 139)" [Undecided,Fix released] [17:41] dandel, if i am fair i never have used it [17:41] dandel, i use the normal build system as i care if that works for my uploads [17:42] either way, the patch changes the default setup by explicitly disabling i8042 input driver whenever the X86_MRST flag is set [17:43] dandel, we have _MRST turned on [17:43] debian.master/config/config.common.ubuntu:CONFIG_X86_MRST=y [17:43] debian.master/config/config.common.ubuntu:CONFIG_SERIO_I8042=y [17:43] and seem to have 8042 enabled too ? [17:44] yes, and i8042 is required for the test machine. [17:44] without it, i have no mouse and keyboard [17:44] right, but i have both turned on in a 35-rc3 kernel there i believe [17:44] it's post -rc3 commit. [17:45] dandel, ahh ok, so we are about to get mooshed by it ... hrm [17:45] do yo have a fix, or just the offending commit ? [17:46] no fix, just unroll the offending commit [17:46] i'm trying to find the i8042 driver in the config file of default kernel t ho [17:47] dandel, i thought this was familar [17:48] we hit the bug he is tryign to fix, namely that the serial must be =y if atkbd=y [17:48] and i'm not able to find the i8042 driver in the input devices on the latest config. [17:48] but we fixed the config, but his change seems wrong [17:49] i'm expecting it to be under device drivers -> input device support -> keyboards (at keyboard has a star) [17:50] You can;t modify the i8042 settings unless CONFIG_EMBEDDED is on, IIRC [17:50] however, i8042 is missing. [17:50] yep, and its not allowed to be if the other one is on under the new [17:50] config as he did it [17:51] either way, it's best to know of the bug and squash it before you get the whack from it. [17:51] yep, thats just plain the wrong change as far as i can see [17:51] did you only report it in the bugzilla or also email randy ? [17:51] bugzilla only [17:51] either way it's a high priority bugfix. [17:52] be best to email him direct and CC: lkml and mention the issue, feel free to copy our team list too [17:52] It's been discussed upstream already [17:52] mjg59, ahh ok [17:52] And a patchset to replace it with a boot-time check has been psoted [17:52] ok so hopefully we can suck that up instead [17:53] the if statements that that guy setup explicitly makes it impossible to even set that as a loadable driver also. [17:53] mjg59, i won't ask how you can keep abreast of the whole of lkml [17:53] dandel, yeah we need to find the upstream discussion [17:53] apw, where's the conig option for embedded at? [17:54] dandel, i normally just edit the configs with VI [17:54] oh and sad part, the patch is from oracle ><; [17:55] the fix for that should have been in code, not kconfig. [17:55] dandel, heh its easy to say once you have messed up everyones machines :) [17:56] dandel, randy suggests reverting that patch regardless of whether they ahve a fix [17:56] so i suspect they will at least do that for -rc4 [17:57] that will be good, anyways, i think that the i8042 chips are super common for laptops. [18:00] dandal thanks for bringing it to our attention :) [18:00] found the thread and post. [18:00] jun 9th is when the patch hit the mailing list. [18:00] yeah that looks like a foundation for a solution [18:01] Actually the i8042 is super common for all intel based machines I know of [18:03] apw, also, this bug report is now fixed (just waiting for the patch to make it into mainline) https://bugzilla.kernel.org/show_bug.cgi?id=14736 [18:03] bugzilla.kernel.org bug 14736 in Config-Interrupts "Toshiba P305D unhandled ACPI Interrupt" [Normal,Resolved: code_fix] [18:04] as of 35-rc1 the suspend is working nicely (except for memory corruption errors on suspend/resume that is caused by radeon kms) [18:04] joy for bad biosen [18:05] yea, it's a pheonix bios [18:43] smb: for CVE-2010-1643 you have upstream marked as not-affected. should that be "released" instead? (there an upstream patch) [18:44] kees: mm/shmem.c in the Linux kernel before 2.6.28-rc3, when strict overcommit is enabled, does not properly handle the export of shmemfs objects by knfsd, which allows attackers to cause a denial of service (NULL pointer dereference and knfsd crash) or possibly have unspecified other impact via unknown vectors. (http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2010-1643) [18:44] *there's [18:44] I guess that would be better. I usually tried to mark it as released [18:44] kees, Sorry if that one went wrong [18:44] smb: yeah, everything else looks great. [18:45] smb: oh, no worries at all. all the rest are fantastic. :) [18:46] * kees loves "git tag --contains SHA1" [18:46] kees, Yeah. Hm, I am a bit confused atm. Give me a sec I have the feeling I did not work on that [18:47] smb: well, that upstream commit shows up in 2.6.28, so the rest of the triage on that CVE looks accurate. [18:49] kees, Something strange. I cannot find the update to that in the cve tracker [18:49] out branch [18:49] ? [18:49] Oh [18:50] I realize [18:50] Sorry [18:50] That one I did a few days ago [18:50] So its just not in the right sequence [18:51] kees, Its the one that is for some reason related to the 2008-... one [18:51] kees, That would also explain why I did things differently then. :-P [18:52] heh [19:04] I'm uploading a test build [19:05] * tgardner lunches [19:11] yes === MTecknology is now known as MTeck-rcider === MTeck-rcider is now known as MTecknology [22:01] * manjo out early [22:23] Please help me triaging the bug 566543. In this the synaptics module is not loaded after the kernel upgrade for ALPS touchpad. Wheather the support has finished or is it a new bug ??? [22:23] Launchpad bug 566543 in snort (Ubuntu) "RFE: Support for variables like $eth0_ADDRESS in debconf script (affects: 1) (heat: 52)" [Undecided,New] https://launchpad.net/bugs/566543 [22:24] Please help me triaging the bug 565543. In this the synaptics module is not loaded after the kernel upgrade for ALPS touchpad. Wheather the support has finished or is it a new bug ??? [22:24] Launchpad bug 565543 in xserver-xorg-input-synaptics (Ubuntu) "horizontal scrollbar does not work in VAIO VPCEB15EL touchpad. (affects: 9) (heat: 54)" [Undecided,New] https://launchpad.net/bugs/565543 [22:24] Sorry for the wrong bug no above... === sconklin is now known as sconklin-gone [23:44] argh [23:44] anyone here willing to help me figure out why my touchscreen won't work? [23:44] hey all [23:44] I need to get the digital in on a M-Audio Fast Track working and I saw http://alsa.opensrc.org/index.php/M-Audio_FastTrack_Pro [23:44] jk- was helping before, but no matter what I do the hid-mosart driver doesn't "attach" to the device. [23:44] I am not entirely sure what I need to do, can someone help? [23:45] it seems it is possible to turn it on if you pass args to the module [23:45] but I am not sure what to pass and whether the Lucid kernel includes the driver