T2 IRC Log: 2008-12-22

This is the log as captured by an IRC bot in the channel. The statements are those of the individual people and might not neccessarily reflect the policy and legal rules as set forth by the T2 SDE Project.

« prev | next »

--- Log opened Mon Dec 22 00:00:42 2008
00:23 -!- wtracy [n=wtracy@adsl-68-122-7-13.dsl.pltn13.pacbell.net] has joined #t2
02:56 -!- wtracy [n=wtracy@adsl-68-122-7-13.dsl.pltn13.pacbell.net] has quit [Remote closed the connection]
08:09 -!- tfl0pz_ [n=tfl0pz@mail.constalant.com] has quit [Remote closed the connection]
09:17 < rxr> yes, it used to build
09:17 [Users #t2]
09:17 [@ChanServ] [ CIA-24] [ dsoul] [ LMJ ] [ mtr] [ Stealth]
09:17 [ axionix ] [ Cyda ] [ koan ] [ mqueiros_] [ rxr] [ TobiX ]
09:17 -!- Irssi: #t2: Total of 12 nicks [1 ops, 0 halfops, 0 voices, 11 normal]
10:01 < CIA-24> rene * r31590 /trunk/package/x11/numlockx/ (. numlockx.cache numlockx.desc):
10:01 < CIA-24> Koen Vervloesem :
10:01 < CIA-24> * added numlockx (1.1) - A small tool to start X with NumLock turned on
10:02 < koan> I'm just packaging some random small tools I regularly use :-)
10:03 < rxr> yeah - thats great
10:03 -!- mpp [n=chatzill@alpha695.server4you.de] has joined #t2
10:04 < mpp> moinsen
10:04 < rxr> moin
10:04 < koan> hi mpp
10:04 < rxr> does anyknow know of a rhel (4) download URL ?
10:04 < rxr> or do I have to use centos to try something out for a client's problem on RHEL 4 ?
10:04 < mpp> probably you do
10:06 < koan> I'd try CentOS, yes
10:06 < mpp> rxr: does kvm+qemu trunk build on your reference machine ?
10:06 < rxr> qemu usually yes
10:06 < mpp> i could use at least some kind of virt. for testing t2
10:06 < mpp> on my side i breaks
10:06 < rxr> kvm also, but it has a shared file (kvm.h) that I ignored for the last weeks .-(
10:07 < mpp> meaning ?
10:07 < rxr> hm - ok - indeed qemu failed in my nightly tester
10:07 < mpp> told ya
10:07 < rxr> make: *** [subdir-m68k-linux-user] Error 2
10:07 < mpp> btw
10:07 < rxr> kvm "fails" also - due to the mentioned shared file (linux/kvm.h or so)
10:07 < mpp> i dealt with OO - what a trip
10:07 < mpp> are these guys nuts ?
10:07 < rxr> yes, they are ...
10:08 < mpp> i'll stick with abiword for the time beeing
10:08 < rxr> I also have to get the current 3 version running in trunk for an project
10:08 < mpp> OO sucks
10:08 < rxr> abiword is not so reliable
10:08 < mpp> what is the problem on your side ?
10:08 < rxr> we tried to use it as wordprocessor for our office
10:08 < mpp> i don't get passed the bootstrap stage
10:08 < rxr> but it is way to limited and unstabled
10:08 < rxr> we had to mirgrate to openoffice
10:08 < rxr> despite it's "amazing" code base ...
10:09 < mpp> yes but since OO doesn't build on trunk
10:09 < rxr> I got openoffice-org (the no go- variant) to build thru in i386 in my nightly tester
10:09 < mpp> i need some wordprocessor at least
10:09 < rxr> fails on the x86-64 one
10:09 < mpp> ah
10:09 < mpp> yes i'm trying on the x86-64
10:09 < mpp> but it fails
10:09 < rxr> but on a "slightly older / mixed" t2 branch of a client I still have a thing to hunt
10:09 < mpp> k
10:09 < mpp> be happy to use it on trunk
10:09 < rxr> on x86-64 I think it failed on my side (my laptop) with some nss mixup
10:10 < mpp> but i couldn't figure it out
10:10 < rxr> I'll track the two errors I'm aware of today
10:10 < rxr> for kvm it's a little sad that now that I got the Mac OS X virtualization patches in our trunk we can not update kvm anymore as they rewrote some block handling logic and thos Mac OS X virtualization patches need an major rediff due to that ...
10:11 < rxr> so updating kvm is not so easy if we do not want to loose the ability to run Mac OS X virtualized ... .-(
10:13 < rxr> but I first have to get openoffice-org building before I can think about anything else
10:13 < rxr> (and install centos in a VM to track some strange thing :-()
10:13 < mpp> btw i have patched grub to work with smartarray controllers
10:13 < rxr> :-)
10:13 < mpp> if anyone is interested
10:14 < rxr> if it can be applied cleanly to trunk :_)
10:14 < mpp> they have a diff partition naming scheme
10:14 < mpp> i need to fix the installing stage of grub
10:14 < mpp> should be easy
10:14 < mpp> i'll submit off course
10:14 < mpp> another thing that **** me off :
10:15 < mpp> gimp build is broken on trunk x86_64
10:15 < koan> mpp: also here on my ppc64
10:15 < rxr> koan also noticed some segfaulting last night
10:15 < rxr> on ppc64
10:15 < rxr> which reminds me I could boot the G5 ...
10:16 < rxr> but I only have today left to fix some client's problems before I'm abroad ...
10:16 < koan> mpp: which error do you get?
10:16 < mpp> for the virtualization part:
10:16 < mpp> ill try to setup a package for virtualbox
10:16 < koan> here it's something about pygtk.h which is not found
10:16 < mpp> on gimp ?
10:16 < koan> yes
10:16 < mpp> pygtk/pygtk.h not found
10:16 < mpp> wich is true btw.
10:17 < koan> yes I also didn't find it manually
10:17 < rxr> mpp: I tried virtualbox - their code is as ugly as openoffice
10:17 < mpp> damnit
10:17 < rxr> mpp: I already added some dependency packages to t2 some months (or year?) ago
10:17 < mpp> regarding ?
10:17 < rxr> mpp: but good luck to get the main bundle built ...
10:17 < mpp> ah k
10:17 < mpp> yes they have tons of deps listed
10:18 < rxr> that much time ago it was mostly really crappy code not buildable with "decent GCCs"
10:18 < rxr> like only with gcc-3 ...
10:18 < mpp> OSS - the new madcow for geeks
10:18 < rxr> I had no time to try again - maybe it's already fixed upstream ...
10:18 < mpp> i figured that one
10:18 < rxr> (or some other's distribution package)
10:18 < mpp> ill do so
10:18 < rxr> but I find KVM performance quite good ...
10:18 < rxr> but more packages welcome anytime, anyways
10:18 < mpp> yes - if it was working on trunk ill be happy
10:19 < mpp> i also prefer qemu+kvm+kqemu
10:19 < mpp> no doubt about that
10:19 < rxr> kvm should only have a shared header file that I have not yet decided how it's best to handle
10:19 < rxr> but please try virtualbox - a t2 package of that would be great - choice is always good (tm)
10:23 < CIA-24> rene * r31591 /trunk/package/gnome2/pango/pango.desc: * updated pango (1.22.3 -> 1.22.4)
10:23 < CIA-24> rene * r31593 /trunk/package/xorg/xf86-video-trident/xf86-video-trident.desc: * updated xf86-video-trident (1.3.0 -> 1.3.1)
10:23 < CIA-24> rene * r31592 /trunk/package/xorg/xf86-video-vesa/xf86-video-vesa.desc: * updated xf86-video-vesa (2.0.0 -> 2.1.0)
10:23 < CIA-24> rene * r31594 /trunk/package/perl/perl-mime-types/perl-mime-types.desc: * updated perl-mime-types (1.25 -> 1.26)
10:23 < CIA-24> rene * r31595 /trunk/package/gnome2/pycairo/pycairo.desc: * updated pycairo (1.6.4 -> 1.8.0)
10:24 < CIA-24> rene * r31596 /trunk/package/python/pyslsk/pyslsk.desc: * updated pyslsk (1.2.7b -> 1.2.7c)
10:50 < rxr> my is centos^Wredhat installing slowly ...
10:50 < CIA-24> rene * r31597 /trunk/package/develop/xulrunner/xulrunner.conf: * added a libxul-nss compatibility sym-link for openoffice (3.0.0)
10:53 < rxr> (how to stress your MacBook Pro to know how good your defective NVidia chips are before the warranty is over: install CentOS in KVM, Emerge "update" -system your system and Emerge openoffice-org: accumulated load: 5.09)
10:53 < rxr> grep . /sys/devices/platform/applesmc.768/*temp*
10:53 < rxr> /sys/devices/platform/applesmc.768/temp10_input:53500
10:53 < rxr> /sys/devices/platform/applesmc.768/temp11_input:40000
10:53 < rxr> /sys/devices/platform/applesmc.768/temp2_input:35000
10:53 < rxr> /sys/devices/platform/applesmc.768/temp3_input:87500
10:53 < rxr> /sys/devices/platform/applesmc.768/temp4_input:73500
10:53 < rxr> /sys/devices/platform/applesmc.768/temp5_input:66000
10:53 < rxr> /sys/devices/platform/applesmc.768/temp9_input:49750
10:54 < koan> it's getting hot in your office? :-)
10:54 < rxr> /sys/devices/platform/applesmc.768/fan1_input:6009
10:54 < rxr> /sys/devices/platform/applesmc.768/fan2_input:6013
10:54 < rxr> koan: yes, the multi core servers are contributing to it ...
10:54 < rxr> we do not neat to turn the heating on :-)
10:55 < koan> :-)
10:55 < rxr> (how can an installer be as slow .... still waiting for the centos install)
10:55 < koan> I had that too when I was a student: my G5 heated the whole room :-)
10:55 < rxr> ok - what else is on the whiteboard :-)
10:56 < rxr> heh :-)
10:56 < rxr> at home my Cube is not heating my flat too much :-(
10:56 < rxr> maybe we should relocate some office servers to my flat ?
10:57 < koan> my iMac G3 was also cosy at winter
10:57 < CIA-24> rene * r31598 /trunk/package/network/plptools/plptools.desc: * updated plptools (1.0.5 -> 1.0.6)
10:57 < CIA-24> rene * r31600 /trunk/package/network/whois/whois.desc: * updated whois (4.7.29 -> 4.7.30)
10:57 < CIA-24> rene * r31599 /trunk/package/x11/worker/worker.desc: * updated worker (2.17.0 -> 2.17.1)
10:59 < rxr> mpp: was your openoffice-org failure this one:
10:59 < rxr> !> checking for PK11_GetCertFromPrivateKey in -lnss3... no
10:59 < rxr> !> configure: error: PK11_GetCertFromPrivateKey missing but needed.
10:59 < rxr> !> See https://bugzilla.mozilla.org/show_bug.cgi?id=262274.
10:59 < rxr> !> Fixed since nss 3.9.3 (contained by e.g. mozilla >= 1.7.5)
11:03 < mpp> hm ... can't remember exactly
11:03 < mpp> ill have to get back to you later
11:03 < mpp> im busy atm
11:03 < rxr> sure
11:11 < rxr> load average: 7.66
11:11 < rxr> maybe with 2 VMs and some t2 tests I should get more RAM ... for my laptop
11:22 < rxr> if your's is the above error you just need to rebuild nspr from trunk:HEAD
11:52 < CIA-24> rene * r31601 /trunk/package/emulators/kvm/90-lsi-update.patch: * imported a lsi scsi emulation update to kvm for improved compatibility
12:11 < CIA-24> rene * r31602 /trunk/package/security/nss/nss.conf: * fixed libnss.so symlink for libdir != lib (x86-64, ppc64 et. al)
12:50 < mpp> re
12:50 < mpp> grub patch done
12:50 < mpp> works now with cciss and /dev/cciss/c%dd%dp%p naming schema for smartarray controller
12:51 < mpp> anyone with ibm serveraid ?
12:51 < mpp> could be that the ips needs patching too
12:52 < rxr> not on my side
12:54 < mpp> cause i have no mail right here
12:54 < mpp> can i pastebin the patch and rxr pushes it to the svn ?
12:54 < mpp> would be nice
12:54 < rxr> sure
12:55 < mpp> http://pastebin.com/m5d124e01
12:55 < mpp> grub segfault on non known commands
12:56 < mpp> but thats not important atm
12:57 < rxr> I think this is a bug in grubs "no ncurses" codepath
12:57 < rxr> as on x86-64 we have a 64bit ncurses and the 32bit grub can not link against it ...
12:57 < rxr> have you written the patch of "found it somewhere" ?
12:59 < mpp> i have ...
12:59 < mpp> i searched some other patches to get a hint
12:59 < mpp> but there are million of grub patches
12:59 < mpp> some weired stuff very distro-ugly
12:59 < mpp> so
12:59 < mpp> i kept it a minimum to do the job
13:00 < rxr> I think I buy some 4G of RAM for my notebook in the luncch break ...
13:00 < mpp> thats cool
13:00 < rxr> load average: 3.57, 6.42, 7.90
13:00 < rxr> -ETOO_MANY_VMS
13:00 < mpp> i have a server with l5420 xeon
13:00 < mpp> 8 cores
13:00 < rxr> yeah - /me too
13:00 < mpp> it builds fast !!!!
13:00 < rxr> yeah - 8way 8G server here as VM nexus, likewise
13:00 < mpp> thats why i needed to patch grub
13:00 < mpp> what is vm nexus ?
13:01 < rxr> but I mostly put this load to my notebook because a) the server is likewise busy, and b) I will probably need those VMs for support in the holiday days ...
13:01 < rxr> mpp: my name for "running all the VMs on" :-9
13:01 < mpp> ah k
13:01 < mpp> allright bbl
13:02 < mpp> enjoy your break/lunch rxr :-)
13:02 < rxr> still some minutes away
13:02 < rxr> another coffee first :-)
13:02 < rxr> n'guten, likewise
13:02 < mpp> yeah kaffeine good idea !
13:02 < mpp> i'll grab some too asap
13:09 < rxr> mpp: what name did you used for the cciss smartarray .patch ?`
13:16 < mpp> mpp@0x3381e.de
13:16 < mpp> i guess
13:17 < mpp> or do you mean the filename ?
13:27 < rxr> filename :_)
13:36 < mpp> ahm i had it in my target actually
13:36 < mpp> since there is already a grub patch i wasn't sure how to name it
13:37 < rxr> what's the name?
13:37 < rxr> hp-cciss ?
13:37 < rxr> or hp-smartarray ?
13:37 < mpp> cciss
13:37 < rxr> something like this .patch ?
13:37 < mpp> grub-0.97-cciss.patch
13:37 < mpp> would do fine
13:38 < rxr> urgs
13:38 < rxr> the patch has DOS line endings ?
13:38 < mpp> sorry
13:38 < mpp> its probably due to pastebin i guess
13:38 < rxr> I just stripped them :-)
13:38 < rxr> commit in a second
13:39 < mpp> thanx alot
13:39 < mpp> so now im on to the virtualization stuff
13:39 < mpp> gee t2 keeps me busy
13:39 < mpp> :-)
13:42 < CIA-24> rene * r31603 /trunk/package/x86/grub/ (grub-0.97-cciss.patch grub.conf):
13:42 < CIA-24> mpp@0x3381e.de:
13:42 < CIA-24> * implemented HP CCISS support for grub
13:44 < koan> if I want to run a minimal t2 installation in a virtual environment, which target should I choose?
13:44 < mpp> generic-10
13:44 < mpp> ;-)
13:45 < rxr> depends what you want to use it for
13:45 < rxr> maybe start with a generic + minimal selection
13:48 < koan> rxr: I'm investigating some virtual appliance platforms, such as Ubuntu JeOS, Red Hat Appliance Operating System, rPath Linux, etc. and I was thinking T2 could also be perfect for this purpose
13:51 < rxr> we already quite use it for this purposse here internally already
13:51 < rxr> (without surrounding colorful products, yet)
13:52 < koan> what's the footprint of such a T2 appliance then?
13:53 < rxr> this depends
13:53 < rxr> the really cool thing about t2 that is scales from very tiny
13:53 < rxr> to full desktop systems
13:53 < koan> indeed
13:53 < rxr> if you want you can use t2 to create < 2MB linux solutions (we did for environmental monitors) etc.
13:54 -!- |beowulf| [i=1014@sourcemage/mage/beowulf] has joined #t2
13:54 < rxr> hand selected kernel options, dietlibc + app ...
13:54 < rxr> a normal t2 minimal install already is some 100 MB (due gcc etc.)
13:54 < rxr> but for pure virtual user-space without devel environment some multiple of 10MB are possible
13:54 < koan> nice
13:54 < rxr> (e.g. glibc apache php mysql)
13:55 < rxr> with all normal Unix strings detached
13:55 < koan> the platforms I mentioned are all above 100MB
13:59 < rxr> I think some hard-core vserver devels also routinely use dietlibc for their virtual-machine user-space
13:59 < rxr> (at least the used to contribute quite some fixes to dietlibc over the last years)
14:00 < CIA-24> rene * r31604 /trunk/package/xorg/xf86-input-vmmouse/xf86-input-vmmouse.desc: * updated xf86-input-vmmouse (12.6.2 -> 12.6.3)
14:00 < CIA-24> rene * r31605 /trunk/package/xorg/xf86-video-apm/xf86-video-apm.desc: * updated xf86-video-apm (1.2.0 -> 1.2.1)
14:00 < CIA-24> rene * r31606 /trunk/package/xorg/xf86-video-ark/xf86-video-ark.desc: * updated xf86-video-ark (0.7.0 -> 0.7.1)
14:00 < CIA-24> rene * r31607 /trunk/package/xorg/xf86-video-ast/xf86-video-ast.desc: * updated xf86-video-ast (0.85.0 -> 0.87.0)
14:00 < CIA-24> rene * r31608 /trunk/package/xorg/xf86-video-chips/xf86-video-chips.desc: * updated xf86-video-chips (1.2.0 -> 1.2.1)
14:04 < CIA-24> rene * r31609 /trunk/package/xorg/xf86-video-s3/xf86-video-s3.desc: * updated xf86-video-s3 (0.6.0 -> 0.6.1)
14:04 < CIA-24> rene * r31610 /trunk/package/xorg/xf86-video-s3virge/xf86-video-s3virge.desc: * updated xf86-video-s3virge (1.10.1 -> 1.10.2)
14:04 < CIA-24> rene * r31611 /trunk/package/xorg/xf86-video-glint/xf86-video-glint.desc: * updated xf86-video-glint (1.2.1 -> 1.2.2)
14:04 < CIA-24> rene * r31612 /trunk/package/xorg/xf86-video-sis/xf86-video-sis.desc: * updated xf86-video-sis (0.10.0 -> 0.10.1)
14:05 < CIA-24> rene * r31613 /trunk/package/xorg/xf86-video-neomagic/xf86-video-neomagic.desc: * updated xf86-video-neomagic (1.2.1 -> 1.2.2)
14:06 < CIA-24> rene * r31614 /trunk/package/xorg/xf86-video-tdfx/xf86-video-tdfx.desc: * updated xf86-video-tdfx (1.4.0 -> 1.4.1)
14:06 < CIA-24> rene * r31615 /trunk/package/xorg/xf86-video-tseng/xf86-video-tseng.desc: * updated xf86-video-tseng (1.2.0 -> 1.2.1)
15:15 < CIA-24> rene * r31616 /trunk/package/office/openoffice-org/openoffice-org.conf: * fixed openoffice-org to disable ldap support if openldap is not installed
15:55 < mpp> hey rxr:
15:55 < mpp> i pasted the incomplete grub patch
15:55 < mpp> sorry -
15:55 < mpp> iam fixing it now
15:55 < rxr> :-/
15:56 < mpp> no problem
15:56 < rxr> at least I compile tested it :-)
15:56 < mpp> blame it on me
15:56 < mpp> the part with the partition renaming is missing
15:56 < mpp> thank god i have my local repo
15:56 < mpp> puh
15:57 < rxr> heh
16:08 < rxr> well, I have 3 instances of OOo building for some hours, now ...
16:08 < rxr> i386 in a VM, x86-64 natively on my notebook, and ppc64 native on a dual-core G5 ...
16:09 < mpp> yea OOo takes hours to build
16:09 < rxr> yeah - even on multi core boxes ...
16:09 < rxr> with plenty of RAM
16:11 < mpp> http://pastebin.com/m3ff6dc86
16:11 < mpp> there you go
16:11 < mpp> i have nothing to strip whitespaces
16:11 < mpp> probably due to pastebin
16:11 < rxr> sed -i 's/\r//'
16:11 < mpp> ah k
16:11 < rxr> err
16:11 < rxr> sorry
16:12 < rxr> sed -i 's/\r$//'
16:12 < rxr> of course
16:12 < mpp> i admit - bashing ? never heard of it :-)
16:12 < rxr> bashing ?
16:12 < rxr> btw. which editor do you use?
16:12 < mpp> yea i lak some of the basic *n*x tools knowledge
16:12 < mpp> ahm nano
16:13 < rxr> that is advanced sed use anyway :-)
16:13 < mpp> k
16:13 < rxr> and IIRC the BSD sed does not handle that even
16:13 < rxr> at least not the Mac OS X one - but that may have other defects anyway
16:13 < mpp> if you say so
16:14 < mpp> .. mpp has no clue of various basic tools
16:14 < rxr> (wanted to use that on 10.5 recently and it mangled all 'r' instead of carriage-return ...)
16:14 < rxr> 10.5 as in Apple's Mac OS X 10.5 aka. Leopard ...
16:14 < mpp> only if stumble across occasionaly - then i read the *** man pages
16:14 < mpp> right - heard of it
16:15 < rxr> yeah - a broken clone of Unix - at least pretty :_)
16:15 < rxr> s/clone/fork of FreeBSD/ :-)
16:15 < mpp> well it at least has something similar to a microkernel - i heard ...
16:16 < mpp> anyways ...
16:16 < rxr> yeah - but they are not really using it
16:16 < rxr> it's there mostly for the fun of it ...
16:16 < rxr> just wasting some cpu cycles here and there ...
16:16 < rxr> all major stuff lives in the BSD user-space task
16:17 < rxr> if some driver breaks the whole thing panics, just like Linux :-9
16:17 < rxr> (been there, done that)
16:17 < mpp> that occured to me too
16:17 < mpp> a couple of people i know showed me those crashes
16:17 < rxr> and I even found some ways to kernel panic even the currently shipping flavor :-)
16:17 < mpp> fun
16:17 < rxr> and another to stall the UI completely ...
16:17 < mpp> welcome to the real world
16:18 < rxr> (and I did not even try hard, just noticed during normal development ...)
16:18 < koan> speaking of crashing drivers, how's the minix port of t2 going?
16:19 < mpp> wohaaaa !
16:19 < rxr> minix source code quality distracted me a little
16:19 < rxr> but there is nothing preventing that in theory
16:20 < koan> the quality is not good?
16:20 < rxr> little hacky "proof of concept" hackery
16:20 < rxr> not as clean code as I hoped
16:21 < rxr> but it would be straight forward packaing of their components for a t2 port
16:21 < rxr> I just do not have the time to do so anymore these days
16:22 < koan> I once tried minix3, I was just a little disappointed with the number of available packages and the limited hardware support
16:23 < koan> but if it can be ported to T2, the number of packages you can run is already a factor bigger
16:24 < rxr> yeah - modulo those one would have to adapt / patch to compile on minix
16:24 < rxr> but all the non-low-level stuff should mostly just build
16:24 < mpp> got my x target ready
16:25 < mpp> with fluxbox - firefox - abiword and stuff
16:25 < mpp> no it's more enjoyable to build on a t2-host
16:26 < koan> :-)
16:26 < mpp> one thing i observed on the dep-detection by t2 ....
16:26 < mpp> it is somewhat fuzzy at least
16:26 < CIA-24> rene * r31617 /trunk/package/x86/grub/grub-0.97-cciss.patch:
16:26 < CIA-24> mpp@0x3381e.de:
16:26 < CIA-24> * updated^W complete grub cciss patch
16:26 < koan> yesterday I finally managed to get my dual head setup working, now I also have finally an enjoyable t2 host :-)
16:26 < mpp> but respecting the crappy packages - no wonder
16:27 < rxr> :-)
16:27 < rxr> btw. before it's going downhill later
16:27 < mpp> have a vm host setup
16:27 < mpp> do the emerge
16:27 < mpp> do the revert
16:27 < mpp> do the -deps=none....
16:27 < mpp> fun story really
16:27 < rxr> I'll be abroad during the christmas days
16:27 < mpp> most packages don't even complain on the configure stage
16:27 < rxr> so happy christmas and happy new year if we do not read later
16:27 < mpp> what a oss clusterfuck !!!
16:28 < mpp> yea hohohooooo
16:28 < mpp> :-)
16:28 < mpp> marry x-mass to ya all
16:28 < rxr> and, if you want something in t2 there are 20+ people with write access ... e.g. mtr
16:28 < koan> happy holidays rxr
16:28 < rxr> so best send to the list and/or ping him
16:28 < mpp> yeah got it
16:28 < mpp> ill do so
16:28 [Users #t2]
16:28 [@ChanServ] [ Cyda ] [ LMJ ] [ mtr ] [ TobiX ]
16:28 [ axionix ] [ dsoul] [ mpp ] [ rxr ] [ |beowulf|]
16:28 [ CIA-24 ] [ koan ] [ mqueiros_] [ Stealth]
16:28 -!- Irssi: #t2: Total of 14 nicks [1 ops, 0 halfops, 0 voices, 13 normal]
16:28 < mpp> thanx for the coop
16:29 < mpp> will keep in touch
16:29 < rxr> but I'll be in the office some more hours before I back together in a hurry .-) as usual .
16:29 < mpp> mpp is too lazy to get his *** to the ccc at berlin
16:29 < mpp> sorry :-)
16:39 < rxr> I'll not be there neither
16:39 < rxr> have been in the past, but with my and my girls parents at different ends of germany we "have to" travel to two other points in germany ...
16:40 < koan> I'm lucky Belgium is such a small country :-)
16:41 < rxr> ah - thus fosdem :-9
16:41 < rxr> :-) even
16:41 < koan> yes I'm lucky FOSDEM is 20 km from my home :-)
16:42 < rxr> yeah - thats indeed great
16:42 < rxr> never have been there :-)(
16:42 < rxr> ( ...
16:42 < mpp> been there alot
16:42 < mpp> my relatives lived there of a couple of years
16:43 < mpp> rxr: many packages do parallel jobs - but the cpu utilization is somewhat poor
16:44 < rxr> been at ccc some years ago, likewise
16:44 < rxr> mpp: yes, I know
16:44 < rxr> the start and end of the builds are seriell ...
16:44 < rxr> unpack; patch; configure; post-install checks; and package compression
16:45 < mpp> thought so
16:45 < rxr> so only the "bulk build in the middle" is subject to parallel build
16:45 < mpp> i meant belgium above
16:45 < rxr> in the past the ROCK code base had "cluster build", but it required NFS and was otherwise crappy
16:45 < mpp> yeah
16:45 < mpp> distcc comes to mind
16:45 < rxr> anyway, in theory with evaluation of the real dependency graph we should be able to parallel build multiple packages
16:45 < mpp> have to give it some try
16:46 < rxr> mpp: distcc only helps distributiing the same build-in-the-middle-build phase
16:46 < rxr> I mean an in-T2 parallel build of whole packages
16:46 < mpp> k
16:46 < rxr> like building firefox and thunderbird in parallel etc.
16:46 < mpp> but packages without interdependency could be built on another host
16:46 < rxr> and thus also parallelizing the start and end (prologue and epliogue) of the packages
16:46 < mpp> k
16:46 < rxr> that additinoally
16:46 < rxr> but also on a single host in times of "massive" multi-codes
16:46 < rxr> multi-cores
16:47 < mpp> i deal alot with bladeservers / centers
16:47 < mpp> so having a cluster capable t2 build would kick some serious ***
16:48 < mpp> we need a kind of virtualize build farm for t2
16:48 < mpp> spread over to all users involved
16:49 < mpp> could speed up things somewhat :-)
16:49 < mpp> we could initiate a comunity build of a target
16:49 < koan> mpp has big plans :-)
16:49 < mpp> just saying
16:49 < koan> t2 as a cover for world domination :-)
16:50 < mpp> t2 is cool - some usability improvements would give it a popularity boost
16:50 < mpp> yeah !!!
16:50 < koan> I agree
16:50 < mpp> all your t2 are belong to us !
16:50 < mpp> :-) LOL
16:51 < mpp> ill up my x target tonight so a fresh trunk build will be available for the newbie devs / pros / amateurs
16:51 < mpp> ;-)
16:53 < mpp> i wonder if the t2 build process would be portable to the GPU http://gpu.sourceforge.net/
16:53 < mpp> could save us some build time
16:53 < koan> :-)
16:54 < mpp> having the standard targets built automagically
16:56 < mpp> k i'm off
16:56 < mpp> so have fun :-)
16:56 -!- mpp [n=chatzill@alpha695.server4you.de] has quit ["ChatZilla 0.9.84 [Firefox 3.0.4/2008111317]"]
17:12 < CIA-24> rene * r31618 /trunk/package/emulators/kvm/ (14 files):
17:12 < CIA-24> * updated kvm (78 -> 81), rediffed most of our patches, some major
17:12 < CIA-24> Mac OS X, multiboot, virtualization patches are still missing
17:23 -!- axionix [n=axion@cpe-67-242-94-6.nycap.res.rr.com] has quit [Read error: 60 (Operation timed out)]
17:25 < CIA-24> rene * r31619 /trunk/package/emulators/kvm/ (07-qemu-macmodel.patch 07-qemu-macmodel.patch.disabled): * rediffed kvm/07-qemu-macmodel.patch
17:35 < CIA-24> rene * r31620 /trunk/package/emulators/kvm/ (14-maccpuid.patch 14-maccpuid.patch.disabled): * rediffed kvm/14-maccpuid.patch
18:21 < rxr> ieek: Unknown savevm section or instance 'rtl8139' 0
18:22 < rxr> got some "just blinking sleep indicator light on the MacBook Pro" after exchaning th 2x1G for 2X2G RAMs, this helped: http://conradc.wordpress.com/2006/06/01/macbook-wont-start-flashing-sleep-light/ (re-inserting the modules)
18:24 < rxr> just some 50EUR for 4GB of laptop RAM, insane times ...
18:40 -!- mpp [n=mpp@i53874270.versanet.de] has joined #t2
18:41 < mpp> re
19:57 < CIA-24> rene * r31621 /trunk/package/emulators/kvm/ (09-qemu-multiboot.patch 09-qemu-multiboot.patch.disabled): * initial kvm/09-qemu-multiboot.patch port to new option_rom handling (does not yet work quite right)
20:10 < mpp> hm... neither kvm nor qemu build
20:12 < rxr> mpp: what is the kvm error?
20:12 < rxr> maybe it is "just" the shared file (kvm.h) I talked about?
20:35 < mpp> nope
20:35 < mpp> some rom stuff
20:35 < mpp> i got virtualbox closed source running
20:35 < mpp> it's a start
20:35 < mpp> on t2
20:36 < mpp> only thing missing is the modules.symvers
20:48 -!- wtracy [n=wtracy@adsl-68-122-7-13.dsl.pltn13.pacbell.net] has joined #t2
21:20 -!- mpp [n=mpp@i53874270.versanet.de] has quit ["ChatZilla 0.9.84 [Firefox 3.0.3/2008092510]"]
23:22 -!- wtracy [n=wtracy@adsl-68-122-7-13.dsl.pltn13.pacbell.net] has quit [Remote closed the connection]
23:31 < rxr> checking out T2 via 3 is astonishing fast, consumed 10M
23:32 < rxr> RX bytes:9567158 (9.1 Mb) TX bytes:245811 (240.0 Kb)
23:32 < rxr> via 3G even
--- Log closed Tue Dec 23 00:00:48 2008