[01:30] is it too early to file bugs against 2.6.24? [01:36] nope [01:47] did l-u-m go away? [02:24] no, it just hasn't been uploaded yet. [02:29] ok === asac_ is now known as asac === \sh_away is now known as \sh === doko__ is now known as doko === luka74 is now known as Lure [15:30] BenC: ping some xen info for you http://pastebin.com/m2de00eee [15:44] http://ubuntuusers.de/paste/19628 Why does the timestamp jump suddenly in line # 213 to a smaller value? (5,352 s) [15:46] Because it's changing timesource [15:49] From what timesource to what other? === \sh is now known as \sh_away [20:10] moin [21:05] BenC: randal schwartz's tech talk on youtube, 44 minutes in... explains why one does not rebase branches that have been published... [21:05] just fyi [21:32] lamont: I understand the social aspect of not rebasing, but I also know the technical benefits that rebasing provides us :) [21:33] lamont: Our kernel tree really isn't for providing a patch base for other developers, like most projects...it's meant to publish our work [21:33] lamont: in most cases at least (there are people using our tree for other development) [21:46] BenC: right... and when I'm one of those people using it for other development, it scuks [21:46] lamont: what other development are you doing? [21:46] lamont: one good thing is, rebases are on a schedule (which will be posted soon on the wiki) [21:46] well, historically,it's been dealing with pushing hppa/ia64 patches back to ubuntu [21:46] ah, rebasing schedules will help [22:08] BenC: do you want to update the kernel git? there are some problems with current code, like i have got no dma with my dvdrw now [22:08] hardy git [22:10] Yes, I want to update our repo [22:10] also could you add a patch [22:10] that's an odd question considering we know we'll be following 2.6.24 until it is released :) [22:10] Kano: all patches need to be posted to kernel-team@lists.ubuntu.com [22:10] http://people.redhat.com/~heinzm/sw/dm/dm-raid45/ [22:11] I don't personally take in patches [22:11] i did for 2.6.22, no response [22:11] IIRC, that's because it was too close to gutsy release [22:11] http://archives.free.net.ph/message/20071023.131911.301ab4c4.en.html [22:12] Kano: that dm-raid is not a patch, it's a whole new module [22:12] Kano: likely should be in lum [22:12] if you like add it in lum [22:12] Kano: if you like, please check out the lum git, fix the new module into our source, and request a pull, or export the patch to kernel-team@ [22:13] i am not so good with that [22:13] also new ndiswrapper would be needed in lum [22:13] 1.50 is out and 1.45 is in lum [22:14] so you've checked out the lum source for hardy? [22:14] aufs should be added, unionfs dropped [22:14] BenC: i have even a live cd with your kernel [22:14] you obviously haven't done much research...everything you are asking for is outlined in the kernel roadmap, in blueprints on launchpad, etc [22:15] all of lum will be updated according to the kernel release schedule for 8.04 [22:15] http://kanotix.com/files/kernel/kernel-update-pack-generic-future/source/sec_perm-2.6.24.patch [22:15] but i needed that patch, to export somehing for aufs [22:15] aufs and unionfs fate have been outlined in a blueprint [22:15] Phillip has already started on this [22:16] i even patches unionfs and it did not work, aufs works [22:16] we know this [22:16] please see blueprints for our work on this [22:17] well you have blueprints, i have a working live cd [22:17] So do we [22:17] show me url [22:18] I'm not continuing with someone that questions my integrity, as if I'm some sort of liar [22:18] well all i say it is impossible with your current git code [22:19] Phillip has been working on this for several weeks, his work is not published because we don't want to spread incomplete work, and work that isn't based on what we have in the archive [22:19] Kano: have a good evening === reynaldo_ is now known as reynaldo [23:27] https://lists.ubuntu.com/archives/kernel-team/2007-December/001934.html [23:29] the server one has pae enabled [23:30] did you try it [23:31] As it should be in most cases, though there should be a non-pae server kernel. [23:31] well pae does not work in virtual box for example, so the default should not be pae enabled [23:32] Even if the PAE kernel did work, it was designed for a server, not a desktop. [23:32] for me thats just a name,nothing else [23:32] "PAE kernel" I mean server kernel. [23:33] I mean there is more to a server kernel than enabling PAE. [23:37] https://lists.ubuntu.com/archives/kernel-team/2007-December/001934.html [23:37] did you ever compare the 2 settings [23:38] I was going to show you a diff between the .config of both kernels, except that I am still installing ubuntu. [23:39] i already saw those, but why not compile your own kernel with pae enabled? [23:40] when you add it as custom kernel it is not that hard [23:41] I am suggesting this because processors with NX support are now common. [23:41] No but I am suggesting this because processors with NX support are now common. [23:41] i would prefer other settings like complete free of old ide code kernels, just like only one changed option ;) [23:42] Kano: the server kernel is 250 Hz instead of 1000 Hz and there are other differences about drivers [23:42] Exactly. [23:42] but compiling an extra kernel just for 1 option sounds a bit extreme... [23:42] and Virtualbox has issues with Ubuntu's tickless kernel anyway ;) [23:43] Agreed, I am recommending this because processors with NX support are now common. [23:43] well i made a live cd with debian 2.6.23-1-686 kernel and it could not boot, but it boots the 2.6.24 generic one ;) [23:43] anyway, I think the current -generic kernel is okay for most uses [23:44] Which processor does that computer have? [23:44] which? [23:45] yuhong: for what purpose do you want this? [23:45] I want to do this because processors with NX are now common. [23:45] how much ram do you have got is the better question [23:45] if it's a reasonably common purpose, and someone wants to take care of it, there have been community "maintained" kernels in universe for some time now [23:46] "how much ram do you have got is the better question" [23:46] well you dont need pae for less than 3gb [23:46] Kano: and also, why not run a 64-bit kernel? ツ [23:47] But you still need PAE for NX. [23:47] Because the NX bit does not fit in the non-PAE pagetables. [23:47] JanC: on desktop there are many drawbacks - at least with debian or ubuntu [23:48] Kano: that's actually not much of a problem on Ubuntu 7.10 anymore [23:48] it is not that easy to use a native firefox with java and flash [23:48] also you can not use win32 codecs [23:48] I know, that is why I suggest enabling PAE on NX capable processors. [23:48] flash works just fine [23:48] same time as java does? [23:49] and probably java plugin too, but I didn't test that [23:49] Anyway, I am suggesting a PAE kernel not because of >4G but because of NX. [23:49] well the wrapper can handle flash [23:49] but you still lack java, no offical java plugin for 64 bit exists [23:49] maybe you can use icetea or so [23:50] and how do you use win32 codecs [23:50] yuhong: but do people really need NX, that's the question [23:50] XP SP2 enables PAE with a compatiblity hack on NX capable processors. [23:50] Kano: why do you need illegal codecs? ;) [23:51] JanC: to view my legal wmv files [23:51] with audio [23:51] yuhong: I can so it might be useful in a security-enhanced kernel [23:52] Kano: well most wmv-files play fine in totem with gstreamer-ffmpeg or such now [23:52] well there is a project to make it possible to use win32 codecs over a 32 bit server tool with 64 bit apps [23:52] Movie codec patents are a big problem today, BTW. [23:52] JanC: not the video is the problem the AUDIO [23:52] Audio codec too. [23:52] just not the latest one [23:52] That is why there is Ogg Vorbis. [23:53] yuhong: are you on the ubuntu-hardened project? [23:53] No. [23:53] maybe they are interested in NX [23:53] Correct. [23:53] and there might be a use case for a special extra-secure kernel [23:54] yuhong: did you ever buy a wmv hd dvd? [23:54] No [23:54] But I am aware of the problem. [23:54] Kano: why would we buy crap ;) [23:54] I am interested in a PAE desktop kernel because of NX> [23:54] JanC: because it looks better than mpeg2 standard res [23:54] Exactly [23:55] what's the use of it if they forbid you to play it? [23:55] Anyway, I am interested in a PAE desktop kernel because of NX. [23:55] yuhong: what cpu do you have got? [23:55] yuhong: I suggest you talk to the ubuntu-hardened people [23:55] Why. [23:56] and there might be a use case for a special extra-secure kernel [23:56] yuhong: to determine how long it would take to get a kernel ;) [23:56] I didn't intend this issue to be related to ubuntu-hardened. [23:56] "ubuntu-hardened wants this" is a bit stronger than "yuhong wants this" ;-) [23:57] You are right that ubuntu-hardened would want this, though? [23:57] yuhong: using NX is is about hardening your system, no? [23:57] Yes. [23:57] yuhong: how about booting a 64 bit system + 32 bit chroot for special purpose? [23:57] so... ツ [23:57] Fedora have a PAE desktop kernel, BTW. [23:59] do you have your kernel config you want to use ready [23:59] generic with PAE enabled. [23:59] Same as generic with PAE enabled. [23:59] did you enable it in your config