[00:39] so, i just send the patches as attachments to the list? === asac_ is now known as asac [01:23] laga, could it be that patch 1/6 is missing? [01:23] it's stuck in moderation [01:23] i forgot to add it, then sent another mail which is stuck now [01:23] 5/6 has a wrong subject line - it's past 2am here :) [01:23] G'night [01:24] laga, ah ok. yeah I figured that out. Und auch die Uhrzeit. ;-) [05:49] hello, I have this problem: I installed Ubuntu Server for sparc, I have 2 disk (hda, hdd) both with four partitions, the partitions 1º are only unused 8 Mb partitions, the partitions 2º are about 512 Mb, and I created a Raid1 device for the swap, the partitions 4º about 19Gb in both disk are for a RAID1 device, wich contain a ext3 filesystem (including /boot), ubuntu instalation was ok, until the instalation of silo, where it says there was an === thegodfather is now known as fabbione [15:31] * BenC likes the new ABI blacklisting [15:31] I can't remember the last time we've been able to do 4 mid-development kernel uploads without bumping the ABI === _fs is now known as fs [17:06] BenC: have you seen the aufs patch? ;) [17:08] laga: Yeah, I'll review it over the weekend [17:08] laga: thanks...Have you tested that it actually works? [17:09] yes. [17:09] not on a live disk, but it boots my diskless client [17:10] it looks like i'm not supposed to set config options to "n", they need to be undefined instead. i'll send another patch to fix that. [17:13] laga: to make it easy, run "debian/rules updateconfigs" [17:13] that's what I'll end up doing afterwards anyway [18:40] Hi, I just attached a patch to launchpad bug #242448, I hope it's helpful [18:56] kirkland: ' Hi, I just attached a patch to launchpad bug #242448, I hope it's helpful' [18:56] rtg: adamwood: hiya, let me reboot ... messages here are logged by my proxy [19:02] adamwood: hi, do you have a kernel built that I can try? [19:02] rtg: or perhaps you have one? [19:02] kirkland:Hi, I have a 2.6.24-21-generic .deb if you want it [19:03] kirkland: no, I was just looking at his patch. part of it at least is already in Hardy, so I'm not sure what he's got for sure. [19:03] adamwood: please [19:03] kirkland:give me a min to upload it somewhere [19:04] adamwood: which arch? [19:04] adamwood: I'm going to install a new kvm [19:04] i386 [19:04] k [19:06] adamwood: I'm totally confused by your explanation in the bug report. you are referencing commits that are already in Hardy. How did you assemble your patch? [19:07] rtg:the commits are from around the 2.6.25-rc3 tag so they're not part of the Ubuntu- tags as far as I could tell. [19:08] adamwood: they are from the Hardy tree. How can they be 2.5.25-rc3? [19:09] rtg:something that confused me too, but running git tag -l on the hardy tree lists a lot of extra tags I was surprised to see. [19:10] hmm, I wonder if someone pushed all of their fetch info from upstream. [19:10] rtg:I first downloaded the ubunut-hardy tree a couple of days ago so I don't really know, I just muddled through ;-) [19:15] adamwood: I'm seeing the same one freshly cloned ubuntu-hardy trees [19:15] s/one/on/ [19:15] ogasawara:I hope that means I'm looking a little less stupid ;-) [19:15] adamwood, ogasawara: lemme see if I can clean things up. the LP message makes more sense. [19:16] adamwood: heh, I thought I was going crazy for a while [19:18] kirkland: http://www.xephi.co.uk/files/linux-image-2.6.24-21-generic_2.6.24-21.41_i386.deb [19:19] rtg:sorry if I've given you some extra work there, i just assumed the full tree was supposed to be included [19:21] adamwood: t'aint your fault. [19:22] rtg, ogasawara, adamwood: the one could have been me when I pushed the re-worked head [19:25] adamwood: for future reference, the preferred way to backport a patch is to 'git cherry-pick', then resolve conflicts, etc. That way you preserve the provenance of the upstream patch. [19:26] rtg: thanks, I thought something like that which is why I linked to the sources, but I doubt applying the second patch I listed would be a good idea. [19:26] adamwood: you're right about the second one. [19:28] smb_tp: assume the Hardy repo is offline for a bit. [19:28] rtg: there's only a 1 line change of any use in it anyway. Hence my patch just to keep things simple even though I knew it wasn't the best way to handle it [19:34] adamwood: you don't happen to have an amd64 deb, do you? [19:35] kirkland: not ready made but I could probably build one tomorrow [19:36] adamwood: okay, let me get a fresh vm [19:38] kirkland: I'll be going afk shortly (it's been a long day ;) ) but it can't be too hard to find my e-mail address on launchpad if you need me to build a different arch [19:38] adamwood: okay, thanks [19:38] adamwood: i'll update the bug status [19:39] kirkland: let's hope i'm not wrong then [19:41] * adamwood is back [21:37] hi there, i would like to know why "all_generic_ide" is not working with hardy. i really need it to make my old pc working. thanks [21:38] soreiser: it works, at least worked with server version [21:38] soreiser: what's the error you're getting? [21:38] on some systems I used to use all_generic_ide, but then switched to irqpoll to make them work [21:39] soreiser: have you tried -- generic.all_generic_ide=1 [21:39] ? [21:39] Tallken, alex_joni just tried with ubuntu-server and ubuntu (8.04) and simply it does not work (i can see many I/O errors from dmesg loading). it works on xubuntu 7.10 instead :-/ [21:39] Tallken no, should i try that too? [21:39] why not? [21:40] soreiser: add "generic.all_generic_ide=1" and hope for the best [21:40] Tallken, alex_joni i forget to say that it does not work only with livecd/install cd [21:40] on grub it works [21:40] ah [21:41] don't know then :/ [21:41] ah and "generic.all_generic_ide=1" if i dont mistake i have already tried also this, some time ago [21:41] the livecd boot system acts weird now and then [21:41] also all_generic_ide=1 <-- about this i'm sure [21:41] soreiser: try the irqpoll option too [21:41] can't hurt [21:43] alex_joni does is serves to my ide-problem? :-/ 'cause i have an old motherboard with the 34gb limit block and i need to get rid of it (all_generic_ide does :)) [21:44] soreiser: probably not [21:44] the problemTallken how can i install hardy-server on my old pc if that parameter not woks? is there any workaround? [21:45] alex_joni irqpool has to do with hdd too? 'cause that pc is now down and i wuold like to be sure that it works to dont wait half hour for it to wake up uselessly ;) [21:45] soreiser: I was bitten by this: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/206635 [21:47] this is what i get if all_generic_ide is not added: http://launchpadlibrarian.net/6454122/A%3A%5Cdmesg.txt [21:48] see under, all those I/O errors [21:48] alex_joni [21:49] alex_joni excuse me, have you read when i wrote the problem is only for the livecd? have you any idea/workaruond? [21:50] ah, i forgot: on xubuntu 7.10 the parameter worked, on ubuntu 7.10 did not (livecd both i mean) what do you think about? [21:50] soreiser: I would still try to add irqpoll and all_generic_ide as boot params for the LiveCD [21:51] alex_joni both? [21:51] or just irqpool? [21:51] *poll [21:51] yes.. did you read the bug link I pasted? [21:51] both [21:53] sorry no i didn't read it so much, 'cause i wanted to search the laucnhpad report i already did sorry ;) ok i'll try it tomorrow then if i'll got troubles i will return here :) [21:54] ah alex_joni: ah, i forgot: on xubuntu 7.10 the parameter worked, on ubuntu 7.10 did not (livecd both i mean) what do you think about? <-- [21:55] soreiser: not familiar with xubuntu's livecd [21:55] and neither with ubuntu's 7.10 [21:57] alex_joni okey thank you anyway i'll try it tomorrow thanks again :) [22:00] np.. [22:01] ;) [22:14] kirkland: when it is done building http://ppa.launchpad.net/timg-tpi/ubuntu will have a kernel with the ecryptfs patch from upstream e4465fdaeb3f7b5ef47f389d3eac76db79ff20d8 [22:15] rtg: awesome, i have had a hell of a time getting an i386 vm up and running [22:15] kirkland: I'm outta here a bike ride. have a good weekend. [22:15] rtg: nice! [22:16] rtg: i bought a road bike yesterday [22:16] rtg: Canondale [22:16] mine's a canondale too, but its a 15 year old F700 mountain bike. still works good. [22:17] the thing I like about mountain biking is that there are no cars to run me over. [22:17] rtg: i hear ya, this thing still scares me shitless [22:18] kirkland: like when you forget to unclip at a stop light? [22:18] rtg: yeah, and doing 50mph down hill (well, i haven't done that yet) [22:18] rtg: but it will scare me [22:19] kirkland: yeah, well be careful. gotta go. [22:45] can anyone here explain to me how key presses on the keyboard get mapped to keycodes by the kernel? [23:00] godlygeek: There's a table in drivers/input/atkbd.c [23:00] godlygeek: There's a table in drivers/input/keyboards/atkbd.c [23:05] maybe i'm missing something... [23:06] what's the distinction between keycode and scancode? [23:06] scancode is what the hardware generates. keycode is what it gets turned into. [23:06] keycodes are things in /usr/include/linux/input.h [23:06] ah, ok. i had them flipped in my head... [23:07] i'm going to try very hard to figure out how to get a sysrq button press on my macbook... [23:07] because not being able to use magic sysrq is absolutely killing me... i'd love to be able to cleanly unmount my disks when i crash... heh [23:12] one last thing.. if two key presses generate the same scan code (with showkey -s), this means that they look exactly the same to the kernel, since showkey -s does absolutely no translation... right? [23:13] They should, yes [23:13] ok. mac docs say that fn-shift-f11 is printscrn on windoze, but that definitely doesn't do anything different from regular-old f11 here. [23:14] Oh! USB is different. [23:14] Entirely different. [23:14] Showkey will probably lie to you when it comes to the raw scancodes, since there aren't any [23:14] They're already abstracted by the HID protocol [23:14] i take it the internal keyboard in mac laptops are USB, then? [23:14] never knew that. [23:16] Yup [23:16] ok. well, pointers to what i should read up on to try to further my goal, then? [23:17] Not sure. I'm really not sure how the sysrq stuff is implemented, I'm afraid. [23:17] well, from what i can see, whenever the kernel gets a keycode it checks if it's for SYSRQ, if it's down, whether alt was pressed, etc, and waits for a new keycode for function to do... [23:18] so, that all makes perfect sense, except that my keyboard never gives it any scan codes that match the keycode for SYSRQ [23:18] so, it ought to be no different from the general question of "how do i make key XXX send keycode YYY" [23:19] except that it needs to be done on the lowest level possible, of course. :) [23:50] does anyone know why CONFIG_MD_RAID10 is not selected in hardy's kernel image?