[12:30] <BenC> mjg59: it's part of a patchset I sent kyle for inclusion
[12:33] <kylem> yeah, clearly it's all my fault.
[12:34] <pkl_> kylem: makes a change from being my fault :)
[12:39] <mjg59> BenC: Is there any way to determine which patches were in feisty and aren't currently in gutsy?
[12:40] <BenC> kylem: wasn't pointing fingers, just letting him know that the patch wasn't lost
[12:40] <BenC> mjg59: yeah, it's basically the file I sent to kyle
[12:41] <BenC> mjg59: forwarded to you
[12:41] <mjg59> Thanks
[12:44] <BenC> mjg59: when is toshiba_acpi stuff going to get into mainline? :)
[12:44] <mjg59> 23
[12:44] <Nafallo> kylem: I've bought the laptop earlier this evening. will let you know how the Intel-stuff works out :-)
[12:45] <kylem> Nafallo, if it doesn't work, it's my job to make it, so feel free to ping me
[12:45] <Nafallo> kylem: I know, and I will :-)
[12:45] <BenC> mjg59: cool, thanks
[12:54] <zul_> BenC: ping
[12:54] <BenC> zul_: yo
[12:55] <zul_> BenC: there is problem in the i386 config PAE is not enabled the hypervisor needs that or it will a kernel mismatch when booting
[12:55] <BenC> pkl_: ^^
[12:56] <zul_> freaking nickserv
[12:59] <mjg59> Haven't we been over the PAE on i386 thing many, many times before?
[12:59] <BenC> mjg59: this is a xen specific config/flavour
[12:59] <BenC> at least I hope that's what zul meant
[12:59] <zul_> yep it is
[01:00] <zul_> besides its the generic flavour renamed i386
[01:00] <BenC> sure you wouldn't rather use -server redone for -xen?
[01:00] <BenC> -server already has PAE
[01:01] <zul_> arent there drivers missing from -server?
[01:01] <BenC> no, right now it's basically -generic with HZ=100 and no preempt
[01:01] <BenC> plus a little bump is cpu target (686 as opposed to 586)
[01:02] <zul_> doesnt matter to me then
[01:04] <zul_> oh when are those policies going to be on the wiki?
[01:04] <Nafallo> zul_: are you doing openvz btw? :-)
[01:05] <zul_> Nafallo: no
[01:05] <Nafallo> zul_: you got any idea if someone is? :-)
[01:06] <zul_> Nafallo: probably the openvz guys, i really dont have time for it 2 virtualizations are enough for me
[01:07] <Nafallo> zul_: sounds sensible enough :-)
[01:07] <Nafallo> I might need it for work, but I hope I can avoid maintaining a patchset :-P
[01:10] <pkl_> zul_ bummer
[01:11] <pkl_> zul_: the config.generic flavour has HIGHMEM4G and not CONFIG_X86_PAE.  I made sure the config matched that.  Why isn't that appropriate?
[01:12] <kylem> xen only supports pae, afaik.
[01:13] <pkl_> kylem/zul_: so the xen flavour with PAE will only run on the server kernel?
[01:13] <zul_> pkl_: you CONFIG_HIGHMEM64G enabled
[01:13] <zul_> pae is default you can build it without pae but it would suck
[01:14] <pkl_> zul_: sorry, just checked, I had CONFIG_HIGHMEM4G
[01:15] <zul_> pkl_: yep no probs
[01:15] <BenC> pkl_: xen guest kernels only run under a xen dom0 kernel, and I think it's dom0 that needs pae
[01:15] <zul_> yep
[01:16] <kylem> anyway, bob dylan. l8rs.
[01:16] <zul_> oh yes bluesfest have fun
[01:16] <BenC> but since the xen/config.i386 is both dom0 and domU, just make that config PAE and we're good
[01:16] <BenC> kylem: later
[01:17] <pkl_> zul_: adding HIGHMEM64G/X86_PAE is no problem.
[01:17] <zul_> pkl_: thanks
[01:18] <pkl_> zul_: in your patch you had both -pae and non pae i386.configs, and so I figured no-pae configs worked on xen.
[01:19] <zul_> they do if the dom0 is built without pae 
[01:19] <zul_> we had both in feisty 
[01:20] <pkl_> zul_: the xen/config.i386 config is both dom0 and domU?  And so it would work with no pae?
[01:21] <zul_> its bit dom0 and domU but dom0 needs PAE in order to boot
[02:10] <pkl_> Seems like non PAE support in xen was scheduled to go away in 3.1.
[02:15] <zul> pkl_: correct but distros such as fedora and opensuse still supports it
[02:17] <pkl_> zul: yeah, but I couldn't remember having to use a PAE enabled dom0.   You were saying dom0 had to be PAE enabled. 
[02:18] <zul> which kernel did you use?
[02:18] <zul> sorry crabby baby
[02:20] <zul> dom0 is pae enabled by default under 3.1
[02:20] <pkl_> zul: I haven't played about with xen for many years.  2004/2005?  When was it first released?  It seemed something interesting when it got first released, but I never got time to play around with it afterwards.
[02:21] <zul> ah well you probably used the defaults but anyways..
[02:24] <pkl_> zul: maybe, but a google brings up many instances of non PAE dom0/domU usage.
[02:26] <zul> pkl_: try it out and see :)
[02:28] <zul> brb
[02:35] <zul> pkl_: ie: https://bugs.launchpad.net/ubuntu/+source/xen-source-2.6.17/+bug/65096
[02:38] <pkl_> zul: yeah, but that's because there was a mismatch between dom0 and domU, dom0 was PAE, and domU wasn't.  He was using the server flavour.
[02:40] <zul> you can use the dom0 kernel as a domU kernel, thats the way we do it and debian does the same way now
[02:40] <zul> thats the way weve done it since edgy as well
[02:43] <pkl_> zul: I can't see where this pissing contest is going to be honest.  I said I'd update the config to use PAE.  My comments afterwards was just to try and prove I'n not a complete idiot.  Idiot maybe, but not a complete one.
[05:45] <calc> BenC: here still?
[05:45] <calc> BenC: i have the oops
[05:50] <calc> BenC: i emailed you the oops output
[05:50] <calc> BenC: going to fsck my disk now
[06:19] <calc> BenC: showed as clean and oops the same after reboot and apt-get dist-upgrade attempt again
[07:53] <yulin> Question: http://kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.20/    differences between 2.6.20-mm1 and 2.6.20-mm2, maybe the difference between -mm1 and -mm2?
[07:54] <crimsun> come again?
[07:54] <yulin> 2.6.20-mm1 and 2.6.20-mm2 are directory in the http: directory
[07:56] <crimsun> what are you attempting?
[07:57] <crimsun> if you simply want to see a diff between the two, just grab http://kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.20/2.6.20-mm1/2.6.20-mm1.bz2 and http://kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.20/2.6.20-mm2/2.6.20-mm2.bz2 and use interdiff(1)
[07:57] <yulin> sorry, my xchat has some problem in show text, wait for a moment, i change to weechat.
[07:58] <yulin> i am back
[07:58] <yulin> can you resend you msg?
[07:58] <crimsun> if you simply want to see a diff between the two, just grab http://kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.20/2.6.20-mm1/2.6.20-mm1.bz2 and http://kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.20/2.6.20-mm2/2.6.20-mm2.bz2 and use interdiff(1)
[07:59] <yulin> ok
[08:01] <yulin> the patches in the two packages organized by what? time? effect? branch?
[08:05] <yulin> our product will use 2.6.20 kernel for several monthes, but we often faced new hardwares such as southbridge, chipset, should i use the patches in -mm?
[08:07] <yulin> Now i used the kernel source from feisty, where i can get the patch of ubuntu-kernel?
[08:14] <crimsun> we don't carry broken-out patches.  Please see http://kernel.ubuntu.com/git/
[08:19] <yulin> ok, i will git.
[10:18] <yulin> q
[10:18] <yulin> quit
[06:14] <Kano> hi
[06:14] <Kano> how about adding aufs, thats more stable then unionfs
[06:16] <Kano> also 2 little kernel packages that only add new pci ids would be nice to have, now i always need to patch the source
[06:16] <Kano> http://kanotix.com/files/kernel/kernel-update-pack/source/2.6.20-via-quirks.patch
[06:17] <Kano> http://kanotix.com/files/kernel/kernel-update-pack/source/t-sinus_111card-2.6.16.diff
[06:17] <mjg59> Kano: What does the first patch do?
[06:17] <Kano> well thats the intel pci id of the via southboard fix
[06:18] <Kano> in the kernel right now are only the amd boards, but the problem happens on the intel version too.
[06:19] <Kano> the fix works fine, just needs to be enabled
[06:19] <mjg59> Kano: Do you have a reference for the requirement?
[06:20] <Kano> a similar thing for the 2nd patch. it only adds another pcmcia card
[06:20] <Kano> ask locsmif
[06:20] <Kano> on oftc.net
[06:20] <Kano> he needed it
[06:21] <mjg59> Kano: It would be helpful to have a complete description of the issue
[06:21] <Kano> it is just the same issue like the well known via southbrigde bug
[06:22] <Kano> just for the via chipset for intel cpus
[06:23] <Kano> it is needed just for those board without bios update
[06:29] <Kano> the 2nd patch is confirmed too by Adapter45 (online on this server)