=== JanC_ is now known as JanC === ericm-Zzz is now known as ericm === kamal is now known as kamal-away [03:48] pgraner, nevermind, seems to have fixed itself === hrw|gone is now known as hrw [08:19] * apw waves [08:20] hey apw [08:20] morning amitk hows it going? [08:21] fleshing out specs and pushing pending patches upstream [08:21] you? [08:22] * smb good mornings the channel [08:24] * amitk good mornings smb [08:24] hi amitk [08:24] .. and adds "morning" to webster's dictionary as a verb [08:26] apw and amitk smb, good morning guys, what's up [08:26] cooloney, I am (more or less) :-P [08:26] amitk, heheh [08:27] smb, moin [08:27] cooloney, just fighting aufs2 and the liveCDs [08:27] smb: do we still have the kernel debug package? [08:27] apw, still? [08:27] smb, the fighting is now to ensure i've got and tested the right image [08:27] cooloney, you mean on ddebs.ubuntu.com? [08:27] so if we wanna to debug a module, we need a .ko with debug info, right? [08:27] i think its fixed [08:28] cooloney, symbols are often helpful [08:28] The only problem might be I did not upload lately [08:28] apw: will union-mounts happen? [08:28] They would be there with the next security update [08:28] amitk, too soon to tell, i am waiting on testing of it [08:28] smb, i think lucid is the only one missing isn't it? [08:29] just due to luck [08:29] smb: since our kernel package including the module which were --strip-debug, how's the debug kernel package? [08:29] apw, Would need to check. I changed another one to the new format [08:29] apw, Might have been karmic [08:29] cooloney, Its packaged without strip [08:29] smb, yeah but last i looked there was .31 ones there [08:30] cooloney, http://ddebs.ubuntu.com/pool/main/l/linux/ [08:30] smb: yeah, so that's for debugging. but i never use that before. this question comes from TI guys [08:30] cooloney, But if you need Lucid, it will take one or two more days [08:31] cooloney, The package basically contains the unstrippeped modules and the full vmlinux [08:31] smb: thanks a lot. cool, fully understand now [08:31] smb: if someone wanna debug the modules, he can install this debug kernel package, right? [08:32] cooloney, right, if they don't vanish again, which should not be the case, hopefully... [08:32] * amitk recovers from hours of reading the "suspend blocker api" thread on LKML [08:33] An api to block suspend or one to find them? [08:35] smb: an api to block suspend after enabling "automatic aggressive suspend" (called opportunistic suspend by Google) [08:35] apw: is there any way to build udebs from a source package? [08:35] apw: i can use sbuild to get udebs. is there another way? [08:36] cooloney, skip_dbg=false [08:36] i tried 'fdr binary-arch', but no udebs. [08:36] smb: skipdbg=false for udebs? [08:36] err udebs [08:36] * smb clear his glasses [08:36] smb: i think that is for debug package, right? [08:36] cooloney, Yeah, forget what I said [08:37] cooloney, I would not thinks so because you need the modules to be build to create packages with the modules referenced [08:37] cooloney, ok udebs are currently only enabled if you have /CurrentlyBuilding [08:38] apw: no idea what's /CurrentlyBuilding [08:39] apw: a directory which i need to create manually? [08:39] cooloney, Its a file [08:39] its possible that if you build binary-arch binary-udebs it will do it [08:39] but if you make /CurrentlyBuilding in the chroot, it will also build it [08:40] cooloney, sudo touch /CurrentlyBuilding in your chroot [08:40] * cooloney finds using sbuild to native build my ti-omap4 source package is very fast in tyler.mills [08:40] * apw adds _documenting_ and cleaning up the controls for all releases to his TODO [08:40] just 1 hr [08:41] cooloney, that sounds acceptable to me [08:42] smb and apw, yeah, does that work for cross compiling? [08:42] 'fdr binary-arch binary-udebs' works for cross compiling? [08:44] the problem is if linux-tools is enabled it does not work [08:44] so yuo'll need do_tools=false at least to cross compile i think [08:44] apw, ok, no problem, i will try that. [09:09] cking, hey ... when filing bugs on iso testing, is there any special magic for that? [09:09] ie. how do identify the bugs so they can be found [09:13] apw, there is a field in the bug tracker as per test case to enter the bug number - this is all I do [09:13] cking, ok thanks [09:14] but there may be some tags that I don't know about [09:15] All the notes say is: "Please report any bugs you identify in Launchpad and report on the success or failure of the test in the tracker: https://launchpad.net/ubuntu/" [09:15] apw, you biting the QA bullet - cool! [09:15] cking, dipping a toe in the pool [09:16] see how many bugs you can find - I aim for 4 each spin - usually it ends up as stupid ones (e.g. broken eye candy) [09:20] apw, I've added the link bjf mentioned in the reply to https://wiki.ubuntu.com/KernelTeam/MeetingHowTo for easier reference, hope that's OK [09:21] ericm_, ok [09:22] * cking goes to get some H/W from the loft - back in 5 === cwillu_ is now known as cwillu [09:50] apw: just a quick feedback, 'fdr binary-arch binary-udebs' works fine with cross compile [09:50] it will give us all the udebs as well as kernel image and header packages in parent dir [09:52] https://bugs.launchpad.net/ubuntu/+source/linux/+bug/588643 => if somebody feels checking :) [09:52] Launchpad bug 588643 in linux (Ubuntu) "Lockup with stacktrace (native_smp_send_reschedule) (affects: 1) (heat: 6)" [Undecided,New] [09:59] cooloney, good === smb is now known as smb-afk [11:42] Some small queryion, could bug https://bugs.launchpad.net/ubuntu/+source/linux/+bug/588643 be caused by an error in BIOS ? [11:42] Launchpad bug 588643 in linux (Ubuntu) "Lockup with stacktrace (native_smp_send_reschedule) (affects: 1) (heat: 8)" [Undecided,New] === lag is now known as lag-afk [11:55] * apw returns [11:56] moin === thunderstruck is now known as gnomefreak === lag-afk is now known as lag [13:05] hmm [13:06] looking at https://edge.launchpad.net/ubuntu/+source/linux/2.6.34-5.13/+build/1768097 we dont seem to build any omap udebs at all anymore === smb-afk is now known as smb [13:13] ogra, it seems unlikely we ever did [13:13] we did with the topic branch in lucid [13:14] (and we need them for netinst images) [13:14] there seem to also be issues with the DSS code on a beagle XM [13:14] ogra, yep but omap in maverick is a complete rebuild [13:14] i know [13:14] ogra, its a shame noone ever tests things when they go in, only the day before the milest [13:15] i'm just examining the omap3 kernel now and note down issues i find :) [13:15] i dont test because A1 is due :) [13:15] we dont build A1 images for arm [13:15] its just coincidence :) [13:16] apw, btw, would there be a chance to not have the kernel produce an oops if it doesnt find its rootfs device and instead have it spit out a userfriendly error msg ? === cking is now known as cking-afk [13:17] http://paste.ubuntu.com/443294/ is what i typically get if there is no rootfs device [13:17] ogra, i thought it dropped to busybox [13:17] not if you dont have an initramfs [13:18] see the bottom of the paste [13:18] then its meant to tell you which things you have [13:18] i seem to remember (from my pre ubuntu times) that it properly panicked and gave a useful error [13:18] ogra, given that shows the card appearing the instant before it explodes i suspect this is not cause it doesn't find root [13:19] i'm only using initramfs since ubuntu :) [13:19] when my root is missing i see a panic which ends with a list of the filesystems and devices available [13:19] weird [13:19] i get the panic since i use arm [13:19] actually i got used to have it but thinking about it i remember it was not always that way [13:20] from the panic i'd say that a timer is firing to flash the cursor, but the framebuffer didn't initialise [13:20] yes, there are definately other issues with that kernel on the XM board i'm justy trying [13:20] [ 2.365386] Waiting for root device /dev/mmcblk0p2... [13:20] [ 2.385223] mmc0: new SD card at address aff7 [13:20] [ 2.390167] mmcblk0: mmc0:aff7 SU02G 1.89 GiB [13:20] [ 2.395141] mmcblk0: p1 p2 [13:20] but i see the oops across all arm arches [13:21] from that you can even see that the disk was appearing, so its not at all clear its anything to do with root at all [13:21] i also saw it on imx51 before or on dove when experimenting with SDs without rootfs and initrd [13:21] notice it dies here: [13:21] [ 2.490692] LR is at cursor_timer_handler+0x34/0x38 [13:21] that sounds very suspicious [13:22] hmm [13:26] * ogra unpacks a rootfs to the SD [13:29] ogra, that callback seems to be the cursor flasher ... every half second [13:30] i think there is a bug in the omap fb which is leaving the cursor running when it fails to init [13:31] yeah, you are right, i even get the oops with a valid rootfs in place [13:31] the DSS/frambuffer code is a mess on omap i guess we will need a bunch of patches from the linux-omap tree to make it work [13:32] * apw can't hear you :) [13:32] (it doesnt work at all on the touchbook or zoom2) [13:32] i am somewhat intriqued who tested the image and what on, cause it was tested [13:32] and apparently also on the beagle XM [13:32] i suspect just an official beagle is the only one supported [13:32] it likely works on one of the older beagles [13:32] XM is official [13:33] zoom2 is official too but will need patches [13:33] we agreed to support all these devices [13:33] XM will be official at release, i am sure up to now we've not even had one [13:33] AI touchbook would be a nice to have since its an actual ARM netbook out there [13:33] apw, we have two in the company since about two months :) [13:33] ogra, and who has them ? [13:34] i bet noone who is doing the kernel work [13:34] amitk had one for kernel, not sure where it went [13:34] i have one as well [13:34] and i think there were even more [13:34] plars and jamiebennet used to have one each iirc but i'm not 100% sure [13:35] i am sure about the ones amitk and i have though [13:35] shame that its eric doing the kernel work... ho hum [13:35] heh [13:35] erm, you mean bryan [13:35] possibly, the disconnect is the same [13:35] he is doing omap4 atm [13:36] and he is doing it awesome :) without having HW at all i got a fully working kernel for my blaze board :) [13:36] then i suspect noone is doing it [13:36] ho hum [13:36] i know mpoirier and lag are supposed to do omap [13:36] under leadership of cooloney and amitk [13:37] but cooloney is busy with omap4 for the special 10.07 release [13:39] apw: ogra: The HW situation should improve soon(ish). We should probably revisit our HW allocations and get unused boards into hands of people that need them [13:39] yeah [13:42] Now that there is some action here, could somebody tell me if http://dupondje.be/DSCF1025.JPG can be caused by a broken BIOS? [13:46] dupondje, hard to say as thats not the first panic === cking-afk is now known as cking [13:47] dupondje, if you could capture more of the first panic then we could say for certain - the trace before the apic timer interrupt would be handy [13:48] dupondje, the panic you do have implies a CPU is offline when we are not expecting it to be so, if its repeating then we are truly confused [13:48] I need full hd tv :P [13:49] used mainline kernel, and there it gives 'BUG: cpu soft lockup' ... :s [13:49] 2.6.34 as mainline kernel btw [13:52] Keybuk, hey ... did you get to do any testing on the init args kernels ?? [13:53] to bad its not possible to scroll when the stacktrace is displayed [14:00] apw: not yet [14:01] apw: I will be doing that today [14:11] apw: while I can see items vanishing off my todo list, I feel like I'm not accomplishing much at the moment [14:11] too much catch-up [14:17] http://webcache.googleusercontent.com/search?q=cache:EGSLAZNT6VwJ:www.pubbs.net/kernel/200907/83651/+"no+console+suspend"&cd=1&hl=en&ct=clnk&gl=uk [14:17] kernel/printk.c:__setup("no_console_suspend", console_suspend_disable); [14:18] Keybuk, i can only sympathise. i had 5 items on mine, i've done 4 of them, but still have 6 to do for alpha-1 ... go figure [14:32] Keybuk, do you know if you need special mount support for union-mounts ? [14:35] yes [14:36] Keybuk, damn, i'll go look for those then [14:36] I have that ready [14:37] Keybuk, ahh, if you have something i can slurp up i'll get it into the PPA as well [14:37] Keybuk, then i can go do some testing myself [14:40] Keybuk, actually it looks to be only two patches if i am reading Valerie's page correctly ... match your expectations? [14:44] that would require me getting some time to actually sort the package out [14:49] BOOT_IMAGE= root= ro console=tty1 no_console_suspend quiet splash [14:50] lag, it looks ok to me [14:50] remove quiet and splash too [14:52] mumble fail [14:53] It sounded like you were on a CB radio and you were transmitting via a piece of string prior to dampening [14:54] lag, my audio is screwy after a suspend/resume cycle - dunno why [14:56] lag, console=tty is the normal default. [14:56] cking, talk to a kernel dev [14:56] :) [14:56] "physician, heal thyself" [15:00] "Hoist by me own petard." [15:02] JFo: Are you a fan of the great WS? [15:03] I am [15:04] "Man does not live by bread alone" [15:07] moin yall [15:07] hi bjf === ericm_ is now known as ericm-Zzz [15:22] moin bjf [15:22] from bug 571378 [15:22] Jun 1 16:42:23 warthog libvirtd: 16:42:23.833: error : udevStrToLong_ui:73 : Failed to convert '008' to unsigned int#012 [15:22] Launchpad bug 571378 in linux (Ubuntu) "C-Media USB Headphone fails to enummerate (affects: 2) (heat: 10)" [Medium,Triaged] https://launchpad.net/bugs/571378 [15:22] looks like some udev failure? [15:23] JFo, well that has libvirtd in it as well [15:23] ah yes [15:23] i'd tend to blame that first, and let them blame udev :) [15:23] heh [15:24] who owns that bit? [15:24] JFo, moin, since we have maverick daily isos now, i'll work on producing daily maverick test isos (shouldn't take more than a minute or two) [15:24] or rather those bits [15:24] bjf awesome [15:30] anyone know if emerald.pgraner is among the living? [15:35] * cking notes that suspend/resume on maverick alpha 1 is very speedy on his old dell 6400 [15:38] apw: updating my laptop so I can install your test kernels [15:39] is the Lucid bacport kernel going to be based on .34 or .35? I was thinking the latter. [15:39] Keybuk, i've run maverick kerenels on lucid no bother [15:40] apw: laptop is at a broken mid-maverick point ;) [15:40] updating to get it to boot again [15:41] Keybuk, ooosp :) [15:42] Keybuk, the 'text only' plymouth like screen (with four dots) is that part of plymouth as a package |? [15:42] plymouth-theme-ubuntu-text iirc [15:46] bjf: I'm working on it [15:46] bjf: had to bandage a cut finger [15:47] * ogasawara bails to dentist apt, back in an hour [15:47] JFo, the LTS backport kernel is tracking Maverick [15:48] thought so [15:49] smb, https://wiki.ubuntu.com/Kernel/Tagging [15:52] Hi all, anyone know the reasons behind the move of autoconf.h and utsrelease from the /usr/src/linux-headers-2.6.34-5-generic/include/linux directory? [15:52] manjo, box is on it's way - probably get to you by Friday [15:52] It caused VMware's tools script to die - vmware-config.pl [15:52] Took a bit of digging but found this: https://blueprints.launchpad.net/ubuntu/+spec/kernel-maverick-misc [15:53] Any comments? [15:58] hi folks, I'm bck from the conference and should be in the normal swing again [16:05] woo hoo! welcome back sconklin [16:06] lag: +CONFIG_USB_SERIAL=y [16:06] +CONFIG_USB_SERIAL_CONSOLE=y [16:06] CONFIG_USB_SERIAL_CH341=y [16:06] CONFIG_USB_SERIAL_PL2303=y [16:07] I'm going to do it via the menu system [16:07] * lag is looking [16:07] lag, I added them to debian.master/config/config.common.ubuntu [16:10] cking: Why CH341? Surely you mean CONFIG_USB_SERIAL_FTDI_SIO? [16:10] lag, you need to chose the appropriate driver ;-) [16:11] I thought there were only two in the UK? [16:11] FTDI and PLxxxx [16:13] lag, well, what ever is appropriate - I've got CH341 and PL2303 for some reason [16:13] JFo, something is busted building the maverick test isos (was working fine for lucid), am debugging [16:15] k [16:24] cking, thanks for the box [16:25] manjo, it requires a US power lead - it did not come with one [16:25] cking, I can get that [16:25] and I formatted it up to boot Karmic, so you can see it will work ;-) [16:26] apw, you worked on some of the fan cntrl stuff, are we missing any modules for the sensors? (lucid/maverick) ? [16:27] apw, what sensor modules need to be loaded ? [16:27] apw, when I run pwmconfig it says at the end /usr/sbin/pwmconfig: No sensors found! (modprobe sensor modules?) [16:29] apw, sensors-detect tell me what modules .. never mind [16:30] manjo, yep sensors-detect ... but even then sometimes there are none still [16:37] JFo: thanks for commenting my bug, by the latest upsteam kernel ? you mean the latest daily build on kernel.ubuntu.com? Or compile a fresh one from kernel.org ? [16:38] dupondje, there is a MainlineKernels page on the wiki [16:38] we build them so you don't have to :) [16:38] * JFo goes to find the link [16:38] well i'll try it when i'm home :) [16:39] tgardner, https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bug/475641 [16:39] Launchpad bug 475641 in linux (Ubuntu) "pwmconfig does not work after upgrade to 9.10 on TYAN server (affects: 2) (heat: 12)" [Medium,Triaged] [16:39] 2.6.34 gave 'BUG: soft lockup ...' :) [16:39] didn't try 2.6.35 yet [16:39] hmmm [16:40] dupondje, here is the link to all the info: https://wiki.ubuntu.com/KernelTeam/MainlineBuilds?action=show&redirect=KernelMainlineBuilds [16:40] let me know how it goes :) [16:41] i'll do, and take some 'screenshots' ;) [16:45] cool [16:48] its not possible to scroll up right ? cause if the error is to long, can't see everything then :s [16:54] dupondje, right no up scrolly [16:55] gotto ask full hd tv then :) === kamalmostafa is now known as kamal [17:14] cking: Are you around? [17:18] yep [17:20] read 2675 modules : new(121) missing(3) [17:20] EE: Missing modules (start begging for mercy) [17:20] make: *** [module-check-generic] Error 1 [17:21] lag, gimme 5 - I've got to sort out the gas men again [17:21] k [17:23] lag, build with skipmodule=true I believe [17:29] lag, fakeroot debian/rules .... skipmodule=true [17:32] cking: Trying [17:32] :) [17:33] yet another rune to remember [17:33] :( [17:34] I already skip the abi test [17:34] Why doesn't it allow you to have any built-ins? [17:35] eh? skipmodule=true should work [17:37] see line 104 of debian/scripts/module-check - this is where it skips the missing modules check [17:39] I mean, why doesn't it allow built-ins by default [17:39] Would anyone know why kdump wouldn't be working on a HP Z600 workstation? [17:39] Or how to go about debugging why it isn't working? [17:39] lag, oh, I see. dunno [17:52] bjf, are you around [17:52] akgraner, present [17:53] can you take a look at the the Kernel Team Meetings listed on the Fridge and let me know if the series is correct [17:53] if it is not can you let me know so I can get it fixed immediately :-) [17:54] akgraner, is it on the front page? [17:55] akgraner, give me a URL to what you want me to look at [17:56] * manjo needs food.. goes to look in the kitchen === sconklin is now known as sconklin-lunch [18:00] bjf, oh sorry let me get you the link [18:00] bjf, http://fridge.ubuntu.com/calendar [18:02] akgraner, and from the calendar I do what? the calendar looks fine :-) [18:02] bjf, nothing if all the times are correct [18:03] akgraner, 1700 UTC is the correct time, so that looks good to me [18:03] apw, smb is balcony open ? [18:03] manjo, only if you have beer [18:04] ok noted that Tuesday -1700 UTC is the time for all Kernel Team Meetings unless I am told otherwise - thanks! [18:04] akgraner, np [18:04] * bjf will brb === hrw is now known as hrw|gone [18:41] mpoirier_, just to say i am still working on the omap udebs, its being truculant [18:42] apw: truculant is definitely a new word for me :) [18:42] and likely spelt wrong by me of course [18:43] no, i just looked it up - it is correct. [18:44] mad as it looks awful spelt that way [18:44] anyhow its being truculant, but i think i've nearly got it licked [18:45] fabulous. [18:45] in the mean time sbuild worked for me. [18:45] I can finally move on. [18:45] thanks for the tips. [18:46] mpoirier_, awsome [18:46] it took a bloody long time, even on tyler.mills. [18:47] Keybuk, hey ... how did your testing go ? [18:47] mpoirier_, arround an hour ? [18:47] is there a way to disable all the udebs except the one you are interesed in ? [18:47] no, more than that. I'll 'time' it next time. [18:48] if you are considering that to make things faster you won't gain anything, as the whole kernel being built is a pre-requisite to generate any udebs [18:48] mpoirier_, once i have this lot fixed I should be able to cross compile only one flavour and generate udebs for it [18:48] which for me here takes about 20 mins [18:48] ok, it will be even faster on tyler. [18:49] it's worth the wait. [18:50] if i haven't killed it by then [18:51] apw: it hasn't happened yet [18:52] apw: your initargs kernel didn't build, so there's no deb to try [18:53] Keybuk, crap will sort that out [18:54] did the "was used" patches end up in the maverick kernel? [18:54] Keybuk, nope not yet [18:54] * apw adds to todo [18:55] k, if you want to go via ppa first, that's fine :) [18:55] yeah sounds like a plan === sconklin-lunch is now known as sconklin [19:14] kro, off to grab a bit of lunch [19:14] err ok [19:31] -> lunch === jjohansen-afk is now known as jjohansen [20:15] ogasawara, you abuot ? [20:15] apw: yep [20:15] yo ... are you doing the -rc1 rebase 'next' ? [20:15] as i am reminded it will need an aufs update before it will build right [20:15] apw: was just starting [20:15] so as soon as you have the base rebase, perhaps you can push it somewhere i can get to it [20:15] apw: just squashing debian commits as we speak [20:16] apw: ack [20:16] and i'll do the update for aufs in the morning on top [20:16] wicked :) [20:50] ogasawara, bug 587444 looks interesting [20:50] Launchpad bug 587444 in linux (Ubuntu) "Wrong dependency in linux-headers-2.6.34-x-generic (affects: 4) (heat: 20)" [High,Triaged] https://launchpad.net/bugs/587444 [20:50] JFo: will take a look in a sec [20:50] kro, np [20:50] sigh* [20:51] JFo: ah, that seems one for tgardner as it's the maverick lts backports [20:51] ah right you are [20:52] I can live with that [20:52] sorry to interrupt you :) [20:52] JFo, wtf is that marked high? its not even a released package [20:52] heh [20:53] I apparently hit the wrong one [20:53] meant for medium [20:53] second time I have done that today [20:53] first one I caught as it happened [20:53] JFo, nominate for Lucid and assign me so I don't forget [20:54] kro, will do [20:54] why does that keep happening [20:54] ah, it is tab completing [20:55] yep, automatic nick completion was checked [20:55] sorry about that kro [20:58] tgardner, done [21:02] cnd, looks like you could get a nice chunk of change to resolve bug 576601 [21:02] Launchpad bug 576601 in linux (Ubuntu) (and 1 other project) "[MacBookPro 7,1]mcp89 sata link reset fails, no disks detected (affects: 79) (heat: 462)" [High,Triaged] https://launchpad.net/bugs/576601 [21:02] http://www.cofundos.org/project.php?id=187 [21:04] JFo: heh, upstream looks like they're getting involved [21:05] cool [21:09] JFo: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/588643 => was able to get a dmesg output when it got locked [21:09] Launchpad bug 588643 in linux (Ubuntu) "Lockup with stacktrace (native_smp_send_reschedule) (affects: 1) (heat: 8)" [Undecided,Incomplete] [21:09] excellent! [21:12] need to mark the bug as New again ? or you change status ? :) [21:14] hey all. [21:14] lets just pretend that i had a hardy xen domU running 2.6.24-24-xen [21:14] what would be the most reasonable way to get ext4 filesystem support for that pretend guest [21:19] dupondje, I have changed it [21:20] I was reading over the bug [21:21] any idea what could be causing it ? could it be an instable BIOS version ? [21:23] smoser, I don't think you can. ext4 did not exist for 2.6.24 IIRC [21:23] well, it does exist. its there. [21:23] at least ./Documentation/filesystems/ext4.txt [21:23] its just likely old and buggy [21:24] smoser, its was probably a preview and marked EXPERIMENTAL [21:24] yes [21:28] JFo: also notice the segfaults, seems like some apps segfault randomly also :s [21:30] dupondje, not sure what could be causing it. odd that some apps are segfaulting randomly [21:31] its totally random also [21:35] happens since I changed CPU and did BIOS upgrade ... but before it was a single core, now quad .. so [21:37] smoser: define support? You could build the module externally, but no one will provide any kind of support for ext4 on 2.6.24 [21:37] i need it to work. [21:38] i'm guessing reasonable ext4 support is not backported that far back. [21:38] the driver in 2.6.24 will probably compile and maybe even work. [21:39] right now our UEC images build system is running the aforementioned kernel version, and I want to build images with an ext4 filesystem, which requires mounting said filesystem. [21:39] I don't think reasonable ext4 support for 2.6.24 is viable [21:39]