=== _Purple_ is now known as _Purple_away === _Purple_away is now known as _Purple_ [01:03] hi. sorry for being thick, but the script did not output any results when finished. am I supposed to fill them in manually? === _Purple_ is now known as _Purple_away === _Purple_away is now known as _Purple_ [05:49] hm, there is no ext4 support in intrepid, right? [05:51] apparently, there is... [07:20] JanC: there is support in the installer, but not as defualt [07:23] amitk: in intrepid? (ext4 wasn't officially in the kernel before 2.6.28) [07:24] not officially supported I mean (there was ext4dev) [07:25] JanC: ohh. Sorry. Not in Intrepid [07:26] except for ext4dev [07:27] even then, i find it's kinda dicey [07:27] there is a bug report against gparted in intrepid not supporting ext4... [07:28] and the gparted version in intrepid doesn't support it, but the one in jaunty does [07:29] but as ext4 isn't supported in intrepid, I don't really see the point of that bug ;) [07:30] I don't see us backporting ext4 to intrepid [07:31] amitk: at least the -server kernel in 8.10 has -ext4dev enabled (didn't check with -desktop kernel) [07:34] but that's not really the same as officially supported [07:34] its dev [07:35] not even the kernel officially supported that module ;) [08:52] Can I draw anyone's attention to https://launchpad.net/bugs/330824 [08:52] Malone bug 330824 in linux "Soft lockups (freezes) when deleting files from ext4 partitions on 2.6.28" [Undecided,Confirmed] [09:23] hi all [09:25] Because there isn't so much time until release I just wanted to point to Bug #350137 . Ath5k shipped with Jaunty seems to be really buggy. Crashes system, looses packets and so on. Backport module seems to work fine so I think it should be replaced. [09:25] Malone bug 350137 in linux "Jaunty ath5k freezes system and looses packets" [Undecided,New] https://launchpad.net/bugs/350137 [09:25] There also seem to be several similar bug reports. [09:43] in relation to the suspend/resume testing, I have a thinkpad X40 at home which has generally done suspend/resume reliably since I've had it, but after upgrading it to jaunty it won't suspend. The various debugging guides seem to assume that suspend works and resume fails - what debugging can I do to figure out what's breaking it? [09:44] it does a bit of disk activity, the blinking suspend light never starts, and then it locks the screen [10:01] Ng: Do you have an Intel graphic card? [10:01] Unggnu: yes [10:01] it's an 855GM [10:01] this might be the problem, at least for me [10:02] 8xx are really bad supported in newer drivers afaik [10:03] Ng: Do you have a standard xorg.conf? [10:03] Unggnu: the machine is at home, but I would think so, it was re-installed with hardy and my gf uses it, so it's unlikely I'd have fiddled with that stuff [10:05] Ng: You could try suspend with vesa to be sure I guess. [10:07] I'll give it a go and get a bug filed I guess [10:09] Ng: there are several already, search for 855 and intel. [10:09] But of course only if they fit === asac_ is now known as asac [11:26] bug #292381 [11:26] Malone bug 292381 in linux "[hardy] kernel 2.6.24-21-generic and gcc version mismatch" [High,In progress] https://launchpad.net/bugs/292381 [11:26] aw. no bot :\ [11:26] heh. snap. [11:28] apw, i was wondering if there was any news on that bug. cjwatson said a USN was going to be announced, but I certainly havent seen one (and the bug status remains the same) [11:30] hrm no i have lost track of that one a bit. will bring it up today [11:31] thanks. [12:09] hi [12:09] since I've reinstalled the cpu freq applet starts with the processor on performance mode, do you know how to change it to ondemand mode? (I can change it in the applet but next time I reboot it goes to performance mode again) [12:10] BTW this happens on jaunty 32 bits [12:10] perhaps try cpufreq-set [13:23] apw, I finally got around to testing your patches for the jax10 on real hardware; https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bug/280669 [13:24] Malone bug 280669 in linux "DMA mode and driver jax10" [Low,Triaged] [13:32] maxb, the lbm regressions with lbm_cfg80211_reg=EU etc, were you able to test on both jaunty and intrepid? [13:33] apw, What where the base for both test packages? Will we probably have to retest Intrepid with the minimum fix applied there? [13:34] smb_tp, yeah we will. hense hastling maxb about it as he has something to test with :) [13:34] i t [13:34] he tested the jaunty update which is basically the same as intrepid with my patch, but not i think on intrepid [13:35] Ok, cool. Yeah I think it is best to leave compat-wireless at its current state with intrepid. While newer version give more support they tended to break stuff sometimes [13:35] Especially since we now have different policies about crda [13:35] yep concur. will prepare an lbm with the patch only for intrepif (lbm-2.6.27) and see if we can find a tester [13:35] Thumbs up :) [13:59] smb_tp, one last aspire one rfkill test kernel for you if you have the time to test it === thunderstruck is now known as gnomefreak [13:59] the upstream for the driver turned up in my bug :) and asked for the final patch which this hopefully is [13:59] apw, sure [14:00] Yeah, I saw the mails [14:00] :) [14:00] they are just uploading now, so about 5 mins at: http://people.ubuntu.com/~apw/lp319825-jaunty/ [14:00] LEt me fire up the aspire [14:05] smb_tp, uploaded [14:06] apw, ok, downloaded [14:15] Ok, wireless is there. Is it supposed to print something? [14:19] no it prints something if you have sw rfkill enabled, that lack of a messsage to that effect and working wireless is a good resule [14:19] ok, sounds good [14:49] * apw is starting to get the feeling that wireless is less reliable in jaunty -11 kernels than it was in -8 [14:49] intel wireless === foka_ is now known as foka [17:30] rtg: so I'm confused again about this whole PAE thing [17:31] Keybuk: how so? [17:31] why can't it be compiled into our generic i386 kernel and enabled if the processor supports it? [17:32] Because that would involve somebody writing the code to support that [17:32] it's just a code issue, right [17:32] I heard tell that there was no way to do that [17:32] but that appears to be false [17:32] Keybuk: AFAIK its not a dynamic selection, e.g., its a statically compile option. I think PAE adds another level of page table complexity. [17:33] The code isn't conditional. You'd need something like the SMP/UP rewriting [17:36] ok, so it's a Linux kernel limitation [17:37] not a "there's no way to detect whether the processor supports PAE" issue [17:37] since the processor clearly has a "pae" flag [17:37] Keybuk: pretty much [18:20] apw: heads up on regression bug 350491 [18:21] Malone bug 350491 in linux "hibernation should be disallowed from the desktop when the installed kernel does not match the running kernel" [High,Triaged] https://launchpad.net/bugs/350491 [18:21] mdz, t [18:21] so we don't check any more then [18:21] mdz: is that a regression? [18:22] kees: yes it is, we fixed that in feisty [18:22] ah, okay, I didn't realize that. [18:23] kees: the kernel already touches a flag file when it's upgraded, we just stopped checking for it [18:23] probably in getting rid of powermanagement-interface [18:23] * kees nods [18:24] should be easy enough to add back [18:24] mdz: should that be mile-stoned? [18:25] kees: probably [18:25] yep its not a big job, just good to catch that its missing [18:26] apw: bug 350609 may be another example of a WARNING which doesn't justify bug reports [18:26] Malone bug 350609 in linux "WARNING: at /build/buildd/linux-2.6.28/kernel/power/main.c:177 suspend_test_finish+0x7c/0x80()" [Undecided,New] https://launchpad.net/bugs/350609 [18:26] apw: I was wondering if we should simply exclude WARNINGs in the apport hook [18:26] mdz thats another which really should not be a warning at all [18:27] most warnings in the kernel real things to report in general [18:27] I almost wrote it, then realized that because the oops text isn't very well delimited, I would need to duplicate a bunch of code from kerneloops [18:27] this one really is not a panic level issue, there is simply no category for it [18:27] a message noone reads, or something which explodes your world are the options [18:27] apw: triaged accordingly [18:28] hrm. perhaps it should be turned into something apport detects, but it also need to have a variable time for once its been seen and nothing can be done [19:06] earthquake here === ivoks_ is now known as ivoks === mcasadevall is now known as NCommander [19:25] How to add a module in initramfs? [19:26] peerless: add it to /etc/initramfs-tools/modules [19:27] Where should I place the modules? [19:27] It will find them from the existing installed modules via the depmod info [19:28] I am just trying a test module built out of kernel [19:28] Figuring out how to include that in initramfs [20:16] copy_modules_dir [20:16] there is a function by that name..trying it [22:49] apw: I do have a seldom-used intrepid partition on my Aspire One, I can make specific lbm tests on request [23:25] do we know what kernel is going into karmic? sorry if this is the wrong place to ask [23:35] I've no idea how final it is, but see the description at http://kernel.ubuntu.com/git?p=ubuntu/ubuntu-karmic.git;a=summary [23:38] hmm ok, thanks maxb [23:38] gotta run for now