[12:51] <zul> BenC: ping got a couple of minutes
[12:51] <zul> ?
[12:52] <BenC> zul: Just a little
[12:52] <zul> sure so what should we do for the xen kernel?
[12:52] <zul> since there isnt going to be paravirt-ops
[12:53] <BenC> zul: I'm thinking unless something amazing happens, you should probably keep linux-source-2.6.17 around to build xen with
[12:53] <zul> yeah thats going to be a problem since udev doesnt like 2.6.17 under fiesty
[12:53] <BenC> we want to keep it in sync with one of our tarballs, but it looks like that wont be 2.6.20
[12:54] <BenC> maybe we can talk to Keybuk about some edgy compatibility in udev
[12:54] <zul> sure..
[12:54] <zul> i can try with 2.6.20 as well
[12:54] <Keybuk> BenC: ?
[12:56] <zul> BenC, well let me try with 2.6.20 before we pull the trigger 
[12:56] <BenC> Keybuk: Some people are saying that feisty udev + edgy kernel == badness
[12:56] <Keybuk> BenC: should work fine, unless they need ide-generic
[12:56] <zul> more like udev initramfs-tools badness
[12:56] <BenC> edgy initramfs-tools?
[12:57] <BenC> maybe explains the problem
[12:57] <zul> yeah spits out a warning
[12:57] <Keybuk> ah, hmm
[12:57] <Keybuk> actually udev requires 2.6.19
[12:57] <Keybuk> in feisty
[12:57] <Keybuk> so initramfs-tools will refuse to make the initramfs
[12:58] <zul> meh..
[12:58] <zul> ok ill try to head to 2.6.20 then
[12:59] <ajmitch> Keybuk: yeah, I've already had issues with trying to test xen updates, a bit of a pain
[12:59] <zul> bald is what ill be then ;)
[12:59] <Keybuk> ajmitch: the pain is outweighed against the pain of trying to guess anything that could break in 2.6.17
[12:59] <ajmitch> kids will do that to you :)
[12:59] <kylem> pfft, just buy more hardwar.
[12:59] <kylem> to hell with this virtualization stuff.
[12:59] <Keybuk> kylem: ENOMONEY
[01:00] <zul> kylem: ditto ENOMONEY
[01:00] <kylem> just stop buying food
[01:00] <zul> well redhat looks like its doing 2.6.19 maybe that could be a happy medium
[01:00] <kylem> GAH
[01:00] <kylem> 6391 N Jeannette Mcgra Jeannette check this.
[01:00] <zul> kylem: yeah im fat enough
[01:01] <kylem> everytime i add to my .procmailrc, they change the damn thing
[01:01] <Keybuk> tbh, I'd be really unhappy supporting divergent kernels
[01:01] <Keybuk> e.g. if we go ahead and do the user-space binding in feisty
[01:01] <Keybuk> then udev will be relying on that
[01:01] <Keybuk> which won't work if you ran 2.6.19+xen
[01:01] <ajmitch> I've already had enough issues with mdadm in feisty
[01:01] <Keybuk> ajmitch: don't run feisty
[01:02] <Keybuk> your system will be failing to boot for most of the development process
[01:02] <ajmitch> I just don't reboot it very often
[01:02] <zul> i wouldnt mind kvm but that eleminates alot of people from using virtualization
[01:02] <zul> including me
[01:02] <ajmitch> and I sort of need to test some of these things on hardware, which means running feisty
[01:02] <ajmitch> xen in vmware doesn't go so fast
[01:03] <zul> BenC: what about backporting paravirt?
[01:36] <zul> breezy kernel uploaded and accepted
[01:38] <kylem> zul, cool, thanks.
[02:46] <zul> sweet...a twisted sister christmas
[03:10] <xhaker> current feisty-alternate image presented odd behavior. I point my finger not only to debian-installer and the kernel until proof of the contrary.
[03:11] <xhaker> If someone is still there and willing to hear say so
[03:21] <BenC> describe it
[03:55] <lifeless> xhaker: I think BenC wants to know what was odd
[03:55] <xhaker> BenC, sorry I was filling a bug report about it. You can read it here. I assigned it to Colin Watson. But you should see it. since it is kernel related also
[03:55] <xhaker> https://bugs.launchpad.net/distros/ubuntu/+bug/74592
[03:55] <xhaker> lifeless, ;)
[03:56] <BenC> you shouldn't assign it till it's confirmed
[03:56] <xhaker> BenC, I didn't realize. I was trying to get colin's attention though.
[03:57] <BenC> ok, none of that is a kernel bug
[03:58] <BenC> the hda->sda thing is because we are switching from ide to ata modules..ata uses the scsi subsystem
[03:59] <BenC> xhaker: Some of those issues may be know, so you may want to search debian-installer reports
[03:59] <xhaker> BenC, my hard drive isn't scsi? :( joke
[04:01] <xhaker> I don't think there are reports of these.
[05:09] <AnAnt> BenC: ping
[05:10] <BenC> AnAnt: yo
[05:10] <AnAnt> BenC: I am using Edgy, and I found out that it doesn't support SD cards
[05:10] <BenC> AnAnt: That's correct
[05:10] <AnAnt> BenC: although in dapper it did
[05:10] <BenC> are you sure?
[05:11] <AnAnt> BenC: well, let me correct this
[05:11] <AnAnt> while Edgy was still beta, I used Dapper's kernel
[05:11] <AnAnt> but I applied a 2 line patch to support > 1GB cards
[05:12] <AnAnt> and some colleage lent me his SD and I used it with that kernel (which was only patched to support >1GB cards) and it worked
[05:12] <AnAnt> I will try it again later today to make sure though
[05:12] <AnAnt> as I still have that kernel here
[05:13] <lifeless> BenC: I have a SD card - or at least its ~ that form factor :) and it worked on dapper, edgy and fisty
[05:15] <AnAnt> what's ~ ?
[05:15] <BenC> around/about
[05:15] <lifeless> its a sdcard slot, but it may be a less-proprietary card is what I mean
[05:16] <AnAnt> oh
[05:16] <AnAnt> BenC: btw, the SD card was a Transcend SD card
[05:17] <BenC> AnAnt: Are you using it in the same type hardware slot?
[05:17] <BenC> the card isn't as important as the hardware you are reading it with
[05:17] <AnAnt> BenC: yeah, that TI slot
[05:17] <AnAnt> BenC: in my laptop
[05:17] <BenC> using the mmc driver?
[05:17] <lifeless> BenC: is mine relevant, I can give details
[05:17] <AnAnt> Texas Instruments PCI6411, PCI6421, PCI6611, PCI6621, PCI7411, PCI7421, PCI7611, PCI7621 Secure Digital (SD) Controller
[05:18] <AnAnt> BenC: well, dunno which driver I use, it auto-detected 
[05:18] <BenC> lifeless: Is it usb, or mmc?
[05:18] <AnAnt> BenC: I recall there was another module with mmc that is loaded
[05:19] <BenC> AnAnt: Yeah, check in /lib/modules/`uname -r`/kernel/drivers/mmc/
[05:19] <AnAnt> BenC: there's sdhci there (in the dapper's kernel)
[05:19] <lifeless> BenC: mmc
[05:20] <lifeless> 02:01.2 Generic system peripheral [0805] : Ricoh Co Ltd R5C822 SD/SDIO/MMC/MS/MSPro Host Adapter (rev 17)  0805: 1180:0822
[05:20] <AnAnt> lifeless: I think sdhci was loaded too, anyways, I'll try again when I go to work today
[05:21] <lifeless> AnAnt: it neds to be if its going to work ;)
[05:21] <AnAnt> BenC: btw, is there any simple article about the difference between CONFIG_PREEMPT, and CONFIG_PREEMPT_VOLUNTARY ?
[05:22] <AnAnt> BenC: also sometimes I get a message saying something about "underrun", what's that ?
[05:22] <BenC> AnAnt: voluntary is where preempt is done when the kernel calls yield()
[05:22] <BenC> PREEMPT is real preemptibility
[05:23] <AnAnt> BenC: well, I followed the thread about PREEMPT in devel, I don't care about sound, but I run simulations, and CONFIG_PREEMPT in dapper made them fast indeed
[05:23] <AnAnt> BenC: but I'm not sure if VOLUNTARY in edgy made it slower
[05:23] <BenC> AnAnt: Then the lowlatency kernel is what you'll want
[05:23] <AnAnt> seems I got to test it
[05:23] <AnAnt> BenC: ok, thanks
[05:24] <AnAnt> does the "underrun" have to do with the voluntary thing ?
[05:32] <AnAnt> ?
[05:32] <AnAnt> BenC: does the "underrun" have to do with the voluntary thing ?
[05:42] <BenC> probably
[05:42] <AnAnt> thanks
[05:42] <AnAnt> oh, what's the sky2 bug? is it skype related?
[05:43] <AnAnt> "sky2 hardware hung" ?
[06:39] <BenC> Subject: More ARM binutils fuckage
[06:40] <BenC> I love it
[02:29] <AnAnt> BenC: ok, I just tested it
[02:30] <AnAnt> BenC: ping
[02:30] <AnAnt> BenC: 2.6.15 kernel does support SD card
[02:30] <AnAnt> BenC: I got this from dmesg:  mmcblk0: mmc2:b368 SDC   975360KiB <NULL>
[02:36] <AnAnt> BenC: I'm using 2.6.17 kernel, now, it did detect the SD card too
[02:37] <AnAnt> BenC: but it gave some errors in dmes
[02:37] <AnAnt> BenC: but it gave some errors in dmesg
[02:50] <zul_> heylo
[02:52] <gnomefreak> is there no more ppc kernels?
[03:38] <zul> BenC: ping when you get a second can i bounce an idea off you?
[03:41] <BenC> zul: Sure
[03:42] <zul> its xen related so it might drive you crazy
[03:42] <ajmitch> run now
[03:43] <BenC> someone needs to rename xen, it's pronunciation is a misnomer
[03:44] <kylem> lol.
[03:44] <zul> like you-are-going-to-make-me-pull-my-hair-out-and-shoot-myself-with-a-gun?
[03:45] <Mithrandir> BenC: you clearly don't understand it, then. :-P
[03:45] <BenC> zul: We can call it crap for short
[03:45] <zul> heh..
[03:45] <BenC> Mithrandir: I must not be in the correct state of mind then :)
[03:45] <zul> or yagtmepmyhoasmwag
[04:07] <zul> kylem: ping
[04:08] <kylem> yo
[04:09] <kylem> just making some coffee, be back in a sec
[04:11] <kylem> k
[04:12] <zul> do you have the patch for CVE-2006-4572 for dapper?
[04:12] <kylem> no, not without an abi change, i'm working on a better fix now.
[04:12] <zul> ok can you pass it along when you are done for breezy?
[04:12] <kylem> sure.
[04:13] <zul> thanks, you are the best!
[04:14] <zul> ill owe you a beer
[04:15] <BenC> mmm...black angus
[04:15] <zul> yeah more than enough to get my tipsey
[04:16] <BenC> glad I checked that...I didn't hit the 'g' key well enough and almost typed it as "mmm...black anus"
[04:16] <zul> lol
[04:47] <BenC> zul: Oh, i meant to tell you, akpm said Andi still plans a 2.6.20 merge for paravirt-ops
[04:48] <zul> thank god
[04:48] <zul> so we should be ok?
[05:16] <AnAnt> BenC: ping
[05:23] <BenC> AnAnt: pong
[05:54] <AnAnt> BenC: yeah, you got what I said about SD ?
[05:55] <BenC> AnAnt: yeah
[05:55] <AnAnt> BenC: I forgot to tell you that I tried a 2 GB SD but it didn't work
[05:55] <AnAnt> will try again some day I hope
[05:55] <AnAnt> one day I'll mail you the details (the errors from dmesg, and the 2GB SD & so)
[06:33] <kylem> sigh. having a testcase for some of these CVEs would really be nice.
[06:34] <BenC> kylem: Sucks when most of them are "in theory"
[06:34] <BenC> but that one you are working seems like it could be tested with proper packets
[06:35] <BenC> I don't know enough about ipv6 packets to write one up for you though :)
[06:36] <kylem> me either
[08:22] <Mithrandir> kylem: take a look at packit?
[08:55] <zul> kylem: ping got the patch yet?
[08:55] <kylem> sort of, i've sent it to patrick for reviwew.
[08:56] <zul> ok cool
[11:33] <BenC> bcollins@huffy:~$ uptime
[11:33] <BenC>  17:32:22 up 1 min,  1 user,  load average: 0.36, 0.17, 0.06
[11:33] <BenC> bcollins@huffy:~$ grep ata /proc/modules 
[11:33] <BenC> bcollins@huffy:~$ grep ata /proc/ioports 
[11:33] <BenC>   01f0-01f7 : libata
[11:33] <BenC> bcollins@huffy:~$ dmesg | grep libata
[11:33] <BenC> bcollins@huffy:~$
[11:33] <BenC> I'd like to know how the fuck that can happen
[11:50] <BenC> Hmm...the problem seems to come from a PCI quirk, but I'm not sure why it is triggering
[11:50] <BenC> still doesn't make sense because that quirk should grab 0x170 too, and it doesn't
[11:51] <BenC> oh wait, I see it