WebHosting Paid by #1Payday.Loans
[00:00] <th> ahh - i've it on the PATA
[00:00] <th> but i've the sata on for some lvm
[00:00] <fake> the sata stuff is a pita bios-wise
[00:00] <th> "pita"?
[00:00] <fake> netrunner: sasl has been moved from /usr/include/sasl to /usr/include/sasl1 and sasl2 from /usr/include/sasl2 to /usr/include/sasl
[00:01] <fake> th: pain in the ass
[00:01] <th> i can't really confirm that
[00:01] <fake> ... on this board.
[00:01] <th> but our use case could much differ
[00:01] <th> what board is that?
[00:01] <th> nforce4?
[00:01] <fake> i have an ASUS K8N4-E Deluxe
[00:02] <fake> ack.
[00:02] <th> sounds like exactly my board
[00:02] <fake> with that crappy sata sil additional "raid" thing on it?
[00:02] <th> ack
[00:02] <th> never used this
[00:02] <fake> me neither
[00:03] <th> sk98lin
[00:03] <th> *sigh*
[00:03] <fake> skge ;P
[00:03] <fake> th: skge is new
[00:03] <fake> th: doesn't it work for you?
[00:03] <netrunner> fake: why has the minor version not been increased?
[00:03] <th> yes it does
[00:03] <netrunner> fake: now I have users whom I need to tell to install a package they already have.
[00:03] <fake> netrunner: because we are l4m0rz
[00:04] <fake> netrunner: send a patch, please
[00:04] <th> uhhhhh
[00:04] <th> using_dma = 0 (off)
[00:04] <th> IO_support = 0 (default 16-bit)
[00:04] <th> why is that not on dma+32bit?
[00:04] <fake> o_O
[00:04] <th> and i was wondering why this machine feels so slow
[00:04] <th> god damn
[00:05] <fake> oh, btw, now i don't have any /dev/sda anymore
[00:05] <fake> the /dev/sda -> x______________yz____blabla things are probably a side effect of starting udevd from /etc/initscript
[00:06] <th> are some basic entries in /dev missing?
[00:06] <fake> th: no
[00:06] <th> i mean
[00:06] <th> BEFORE starting udevd in initscript
[00:06] <fake> how would i know?
[00:07] <th> i dunno
[00:07] <fake> mount --move /dev /mnt/cdrom wuahah
[00:07] <th> hm?
[00:07] <fake> so i can see what's inside
[00:07] <th> ah
[00:07] <th> ;)
[00:07] <th> why /mnt/cdrom?
[00:07] <th> you're sick
[00:07] <fake> th: because that's always unused ;)
[00:08] <fake> almost...
[00:08] <th> whimp
[00:08] <fake> hm
[00:08] <th> s/h//
[00:09] <fake> oh great. the whole ~1M lines of output from udevd start didn't get logged anywhere.
[00:09] <fake> great!
[00:09] <th> record a vmware video ;)
[00:12] <fake> hrmpf.
[00:13] <th> perhaps you should pee at the console
[00:13] <th> s/pee/tee/
[00:14] <fake> pervert...
[00:14] <th> TEE!
[00:14] <fake> i'll add a redirect to /etc/initscript
[00:14] <fake> into some /mnt/cdrom/* file ;-))
[00:15] <th> you could use `tee` for this ;)(
[00:15] <fake> i don't want the output printed to /dev/console - it takes 5 minutes
[00:16] <th> HDIO_SET_DMA failed: Operation not permitted
[00:17] <th> did i became unable to build a proper kernel?
[00:21] <th> NFORCE-CK804: BIOS didn't set cable bits correctly. Enabling workaround.
[00:21] <th> NFORCE-CK804: port 0x01f0 already claimed by ide0
[00:21] <th> NFORCE-CK804: neither IDE port enabled (BIOS)
[00:24] <fake> oh damn, i don't get it
[00:25] <fake> th: does the initramfs thing work?
[00:25] <th> in general?
[00:25] <th> yes.
[00:25] <fake> by using mkinitrd
[00:25] <th> i setup an infrastructure for creating it already
[00:25] <th> using gen_cpio_init
[00:26] <th> no never used mkinitrd for it. it wont work
[00:26] <th> i want a fast IDE disk on this board _now_
[00:27] <fake> th: there are some drivers on the nvidia page...
[00:27] <th> ahh right
[00:27] <th> i forgot about thos/
[00:27] <th> this really doesn't work with open source?
[00:27] <fake> th: btw, if i re-start udevd in the booted system and re-emit the uevent stuff, the links are correct
[00:27] <fake> th: no idead - works here (2.6.16.1)
[00:28] <fake> IO_support = 1, using_dame = 1
[00:28] <fake> dma
[00:28] <fake> on the ide bus
[00:28] <fake> pata
[00:28] <th> hmm
[00:28] <fake> on sata, IO_support = 0
[00:29] <th> anything special in kernel?
[00:29] <fake> th: default rock config, just added sata_nv as builtin
[00:30] <th> i can get IO to 32 bit.
[00:30] <th> but no dma
[00:30] <fake> (and removed the support for wlan and atm)
[00:30] <th> sata_nv should have nothing todo with the pata
[00:30] <fake> ack
[00:31] <fake> th: wait, what is your bios revision?
[00:31] <th> how can i tell?
[00:31] <th> it's one of the very first boards
[00:31] <fake> lshw
[00:31] <fake> lshw | grep Revision
[00:32] <th> lshw where is that from?
[00:32] <th> did not knew
[00:32] <th> version: ASUS A8N-SLI DELUXE ACPI BIOS Revision 1002 (11/26/2004)
[00:32] <fake> version: ASUS K8N4-E Deluxe ACPI BIOS Revision 1011-003 (02/08/2006)
[00:32] <th> uhh
[00:32] <th> so we might have different boards
[00:32] <fake> it's a beta version, but it fiex a _lot_
[00:32] <th> we do
[00:33] <netrunner> mine shows nothing at revision
[00:33] <fake> netrunner: that whole string after "version:" is vendor-specific
[00:33] <netrunner> version: 1UETA5WW (1.55 ) (01/07/2005)
[00:33] <netrunner> :)
[00:34] <netrunner> no revision. was just curious what you do there :)
[00:34] Action: netrunner going to get another beer.
[00:34] <th> Please select packages for installation:
[00:34] <th> [*] NVIDIA audio driver for Linux-x86 (1.0-7)
[00:34] <th> [*] NVIDIA network driver for Linux-x86 (1.0-13)
[00:34] <th> EOF
[00:34] <fake> ...
[00:35] <th> perhaps audio drivers enable my dma
[00:35] <fake> th: i would check on asus.com wether there's a newer bios
[00:37] <th> 1016 would be the latest stable
[00:37] Action: fake rebooting with syslogd / klogd started in /etc/initscript...
[00:37] <fake> th: that's a whole lot more
[00:37] <fake> beb
[00:37] <fake> brb
[00:41] <th> damn do i need a floppy drive for this?
[00:45] <th> wow there is some windows application which downloads new bios from internet and flashs it
[00:45] <th> what a luck that this is the gf's gaming machine
[00:51] <fake> i used my shiny new usb stick together with a win98 boot cd and usb disk emulation ;)
[00:52] <fake> okay, got it in the logs now
[00:53] <th> and i got a new bios splash screen
[00:53] <th> the bios spoke to me
[00:54] <fake> hm. i creates the device node /dev/vcc/a1
[00:54] <fake> *it
[00:54] <th> vcc?
[00:55] <fake> over and over again. four million six hundred and three thousand and nine hundred and eighty-nine times.
[00:55] <fake> exactly.
[00:56] <fake> the only hint is in the very beginning of the log:
[00:57] <th> what's vcc?
[00:58] <fake> no db file to read /dev/.udev/db/class@vc@vcs1: no such file or directory
[00:59] <fake> i dont' know. anyhow, it gets add- and remove- events for the device
[00:59] <th> version: ASUS A8N-SLI DELUXE ACPI BIOS Revision 1016 (12/01/2005)
[00:59] <th> ok so this worked
[00:59] <fake> n1c3
[00:59] <th> wow
[00:59] <th> and now hwscan proposes nvnet for the other NIC
[00:59] <fake> how often is /etc/initscript executed? for each vc?
[01:00] <th> yes
[01:00] <fake> th: forcedeth here, btw, not skge, that's my laptops' marvell yukon, i mixed that up
[01:00] <th> still no dma
[01:00] <fake> :(
[01:00] <fake> starting udevd from there sucks...
[01:01] Action: fake moving udevd start to the beginning of /etc/init.d/system for a try...
[01:04] <th> why not start it before init?
[01:04] <fake> th: how?
[01:04] <fake> th: no initrd working here
[01:05] <th> ah
[01:05] <th> ok
[01:05] <th> so this is no option
[01:05] <th> you could use 2.6.12 with devfs
[01:06] <fake> any _useful_ suggestions?
[01:06] <fake> ;)
[01:06] <th> ohh
[01:06] <th> static /dev ?
[01:06] <th> dont use computers?
[01:06] <th> use windows?
[01:07] <th> well
[01:07] <th> try init.d/system
[01:07] <th> when do you mount your /dev?
[01:07] <fake> th: if the $devtype is udev, i don't do anything in /etc/initscript and mount it from system.init
[01:10] Action: fake rebooting...
[01:10] <th> The IDE class driver will not be able to use DMA, which results in lower IDE performance. This can be worked around by patching the driver with the appropriate device IDs.
[01:28] <fake> ha!
[01:28] <fake> found it
[01:29] <th> who should get the whale?
[01:29] <fake> the problem is that our /dev on the root filesystem is *empty* by default
[01:29] <th> that was my initial question
[01:29] <fake> whereas the initrd's /dev is pre-populated with a few devices
[01:29] <fake> th: i booted from the bootcd to check
[01:29] <netrunner> th: save the whale
[01:30] <fake> we need to add exactly those devices to the $root/dev
[01:30] <fake> then it should even work from /etc/initscript
[01:31] Action: fake trying his thesis
[01:35] <th> using_dma = 1 (on)
[01:35] <th> i compiled it into the kernel instead of loading it as module
[01:40] <fake> th: the ide driver?
[01:40] <th> the amd/nvidia ide thing
[01:40] <fake> btw: it works. i'm gonna patch stone_mod_install.sh to create those device nodes in /dev - or should they even be part of the sysfiles package? (for all you non-installer users)
[01:41] <th> and the via chipset stuff
[01:41] <th> how about postinstall
[01:42] <fake> device nodes in gem files suck, indeed
[01:43] <fake> (and /dev/null is part of the party)
[01:43] <fake> even /dev/console, now that would be fun
[01:44] <fake> but what would be the hook? creation of /dev ?
[01:44] <fake> then it's part of 00-dirtree
[01:45] <th> hmm
[01:45] <th> perhaps kernel
[01:46] <fake> hm... i think it should indeed be part of 00-dirtree. those devices are the bare basics...
[01:46] <th> but not part of the gem, i hope
[01:46] <fake> ram0, null, zero, tty, console
[01:46] <fake> where ram0 isn't needed
[01:47] <fake> so null, zero, tty and console
[01:47] <th> and the links
[01:47] Action: fake trying to create a gem with those
[01:47] <fake> no
[01:47] <fake> the links aren't needed, neither are the pts/shm directories.
[01:48] <fake> once the ramfs is mounted on /dev, they're copied over
[01:48] <th> the links are needed. just not for what you're currently doing
[01:48] <th> so if we add anything - why not add the links as well?
[01:48] <fake> th: they are either added by MAKEDEV.sh, devfs or udev
[01:48] <th> or are they never needed before /dev mounting?
[01:49] <fake> they need not be there in all cases. that disqualifies them from being part of the package, imho...
[01:51] BoS|afk (n=BoS@dslb-088-072-035-247.pools.arcor-ip.net) left irc: Remote closed the connection
[01:51] BoS|afk (n=BoS@dslb-088-072-033-150.pools.arcor-ip.net) joined #rocklinux.
[01:52] <th> fake: ok
[02:14] <fake> test build running...
[02:14] <fake> (need to reach stage 1)
[02:16] <netrunner> cool. sparda pays 0,05eur per online money transfer
[02:16] <netrunner> max 0,55eur per month.
[02:16] <netrunner> if I create 1000 accounts, and add 0,01 eur to the first, and then transferit all around I can gain a basic income of 550eur :)
[02:17] <owl> O_O
[02:19] kasc_ (n=kasc@dslb-084-060-109-148.pools.arcor-ip.net) joined #rocklinux.
[02:25] <fake> th: patch coming up
[02:26] kasc (n=kasc@dslb-084-060-111-118.pools.arcor-ip.net) left irc: Read error: 113 (No route to host)
[02:26] Nick change: kasc_ -> kasc
[02:29] Action: fake happy that the damn -i aliases are gone
[02:32] <fake> th: it's in
[02:33] <fake> th: not severely tested though, i'll start a testbuild against vanilla trunk (r7398)
[02:36] <fake> argh!!
[02:37] <fake> th: kde-i18n-{es,it,de,fr,ru} and koffice-l10n-{the same} are now (trying to be) built in the bootdisk target
[02:37] <fake> th: besides that, with my journal i linked earlier today, no regressions in bootdisk, and only kaffeine failing due to missing cdparanoia in crystal
[02:39] <fake> th: patch for cdparanoia coring is now in sm
[02:42] Action: fake afk for tonight
[02:52] <owl> gn8 fake
[03:59] mnemoc (n=amery@200.73.88.2) left irc: Read error: 110 (Connection timed out)
[09:18] blindcoder (n=blindcod@tor/session/x-66a5f14f3f457a95) joined #rocklinux.
[09:21] <blindcoder> fake: not my problem if messagebus and haldaemon are bad :P
[09:31] <blindcoder> 3227 builds total, 1394 completed fine, 101 with errors.
[10:26] daja77 (n=daja77@dslb-088-072-042-170.pools.arcor-ip.net) joined #rocklinux.
[10:38] daja77_ (n=daja77@dslb-088-072-044-048.pools.arcor-ip.net) left irc: Read error: 110 (Connection timed out)
[10:42] ija_ (n=ija@84.19.222.84) joined #rocklinux.
[10:55] ija (n=ija@116.242.80.212.versanetonline.de) left irc: Read error: 113 (No route to host)
[10:55] Nick change: ija_ -> ija
[11:09] <th> 02:32:23 * fake happy that the damn -i aliases are gone
[11:09] <th> i did this in a silent night
[11:52] <th> usuallt i'm the xterm guy. but for testing purposes now i built aterm from our package...
[11:52] <th> and the flist only contains the man page
[14:01] <owl> moin
[17:14] <fake> moin
[17:15] <owl> hi fake
[17:23] [raphael] (n=raphael@raphael.netpark.at) joined #rocklinux.
[17:54] [raphael] (n=raphael@raphael.netpark.at) left irc: "using sirc version 2.211+KSIRC/1.3.12"
[18:04] icelbox (i=simon@roedelkiste.net) joined #rocklinux.
[18:05] <icelbox> hi
[18:58] <fake> hi icelbox
[19:06] <blindcoder> moi
[19:06] <blindcoder> n
[19:06] <blindcoder> finally back home
[19:07] <fake> home = berlin?
[19:12] <blindcoder> yeah
[19:21] Action: fake going to be in IN from 5th through 11th
[19:22] <blindcoder> I'll be the weekend after that
[19:34] <fake> th: noatun works here
[19:34] <th> fake: this was all realted to my dma issues :-/
[19:34] <fake> th: ah, ok
[19:57] owl (n=owl@193.93.28.218) left #rocklinux.
[23:34] <fake> sleepy sunday
[00:00] --- Mon Apr 3 2006