--- Log opened Sat Oct 11 00:00:05 2008
00:36 < CIA-8> aldas * r30746 /trunk/package/audio/pulseaudio/pulseaudio.desc: * updated pulseaudio (0.9.12 -> 0.9.13)
00:36 < CIA-8> aldas * r30747 /trunk/package/gnome2/gparted/gparted.desc: * updated gparted (0.3.8 -> 0.3.9)
00:37 < CIA-8> aldas * r30748 /trunk/package/office/gnucash/gnucash.desc: * updated gnucash (2.2.5 -> 2.2.7)
00:37 < CIA-8> aldas * r30749 /trunk/package/mail/dovecot/dovecot.desc: * updated dovecot (1.1.3 -> 1.1.4)
00:37 < CIA-8> aldas * r30750 /trunk/package/multimedia/vlc/vlc.desc: * updated vlc (0.9.3 -> 0.9.4)
00:43 < CIA-8> aldas * r30751 /trunk/package/develop/scons/scons.desc: * updated scons (1.0.1 -> 1.1.0)
00:49 < CIA-8> aldas * r30752 /trunk/package/network/bacula/bacula.desc: * updated bacula (2.4.2 -> 2.4.3)
01:13 < CIA-8> aldas * r30754 /trunk/package/news/slrn/slrn.desc: * updated slrn (0.9.8.1 -> 0.9.9p1)
01:39 -!- mpp [n=user@i53876F85.versanet.de] has quit ["good night - good fight"]
02:18 < CIA-8> aldas * r30753 /trunk/package/develop/cmake/ (cmake.desc uname.patch): * updated cmake (2.4.8 -> 2.6.2)
02:41 -!- hikenboot [n=vmuser@c-75-67-11-49.hsd1.ma.comcast.net] has joined #T2
10:43 [Users #t2]
10:43 [@ChanServ] [ dsoul ] [ hikenboot] [ mtr ] [ Stealth]
10:43 [ axionix ] [ felanha] [ koan ] [ Ragnarin] [ TobiX ]
10:43 [ CIA-8 ] [ gw ] [ mqueiros ] [ rxr ]
10:43 -!- Irssi: #t2: Total of 14 nicks [1 ops, 0 halfops, 0 voices, 13 normal]
11:02 -!- mpp [n=user@i53874979.versanet.de] has joined #t2
11:33 -!- mpp_ [n=user@i5387665F.versanet.de] has joined #t2
11:40 -!- mpp [n=user@i53874979.versanet.de] has quit [Read error: 60 (Operation timed out)]
11:57 -!- mpp_ is now known as mpp
12:01 < CIA-8> rene * r30755 /trunk/package/lua/lpeg/ (lpeg.cache lpeg.desc): * updated lpeg (0.7 -> 0.8)
12:04 -!- PiP [n=pip@unaffiliated/pip] has joined #T2
12:04 < PiP> Nice
12:05 < CIA-8> rene * r30757 /trunk/package/base/screen/screen.desc:
12:05 < CIA-8> * updated screen Download tag, former uni server no longer
12:05 < CIA-8> available, use long GNU location instead
13:50 -!- PiP [n=pip@unaffiliated/pip] has left #T2 ["It's not your problem."]
13:57 < CIA-8> rene * r30756 /trunk/package/lua/lpeg/lpeg.desc: * updated lpeg (0.8 -> 0.9)
14:27 < CIA-8> rene * r30758 /trunk/package/develop/subversion/subversion.desc: * updated subversion (1.5.2 -> 1.5.3)
15:24 < CIA-8> rene * r30759 /trunk/package/gnome2/gtkmm/gtkmm.desc: * updated gtkmm (2.14.0 -> 2.14.1)
16:51 < mpp> hey
16:51 < mpp> x-compile hangs on openssl
16:51 < mpp> stage 1
16:51 < mpp> i686-t2-linux-gnu-as: unrecognized option `-PIC'
16:51 < mpp> make[1]: *** [x86cpuid-elf.o] Error 1
16:51 < mpp> any ideas ?
16:52 < mpp> has anyone successfully x-compile t2 lately ?
16:52 < mpp> i mean with a fresh system ? no previous ccache or build dirs ....
17:00 < rxr> x in stage ? guess you cross compile ?
17:01 < rxr> looked like a regression from my new AS=... default change
17:10 < mpp> its a 32-target on a 64-host
17:11 < mpp> i could need that fixed
17:11 < rxr> with "this is a cross build enabled="
17:11 < mpp> as i would like to use 64bit host only
17:11 < mpp> to build 32bit targets
17:11 < mpp> ill check
17:12 < mpp> yes it is enabled.. (off course :-)
17:12 < mpp> btw .
17:12 < rxr> well - you would not need to have it enabled, as your 64bit system can run the "old" 32bit binaries
17:12 < mpp> purchased a phen x4 with 8gb ram
17:12 < mpp> ?
17:13 < rxr> you only need to enable the "this is a cross build" if your build host can not run the target binaries
17:13 < mpp> can i build a 32bit target on 64bit host without x-compile ?
17:13 < rxr> yes
17:13 < mpp> ah...
17:13 < mpp> cool
17:13 < mpp> so phenom x4 really kicks ass compiling t2 :-)
17:13 < rxr> but of course you found a real but, new regression will fix it as soon as I hit it (test builds running)
17:13 < mpp> cool !
17:14 < mpp> i would be more than happy to have only 64bit build hosts
17:14 < rxr> probably fixable by overwriting the new T2 default AS= in the makeopt's of the package
17:14 < rxr> xorg-server it is failing you said ?
17:14 < mpp> xorg-server ?
17:14 < rxr> ah - sorry - openssl ...
17:14 < mpp> yes
17:14 < mpp> right
17:14 < rxr> should hit that sooner than later
17:14 < mpp> okay
17:15 < rxr> but patches welcome if you want to try it ...
17:15 < rxr> something like var_append makeopt ' ' "AS=$CC" # or so
17:16 < mpp> to the openssl package ..?
17:16 < rxr> yes
17:16 < mpp> ill try
17:17 < mpp> does the t2 minimal target runn 32bit binaries by default ß
17:17 < mpp> ?
17:17 < mpp> cause i had trouble in the past ifir correctly
17:17 < mpp> could be wrong though
17:17 < rxr> you mean the minimal package selection of the generic target? you can build it for x86-64 if you like
17:17 < mpp> yes i mean the running host
17:18 < mpp> a x86-64 host can run 32bit binaries ?
17:18 < rxr> normally yes (unless you disable CONFIG_COMPAT32 or what it was named in the linux kernel
17:18 < mpp> that step failed me when i was trying to build a 64bit target on a 32bit host
17:19 < mpp> it complained about stub-32 or something
17:19 < rxr> building stuff is another case
17:19 < rxr> you can cross build x on any host y
17:20 < mpp> however . if i do not need a 32bit host to build 32bit targets then i'll switch the remaining machine also to 64bit t2
17:20 < mpp> which would be the dl2.t2-project.org :-)
17:21 < rxr> yes, you can (except some DOS emulation fluff IIRC) any 32bit linux program on a x86-64 lilnux kernel
17:21 < mpp> k
17:21 < mpp> than dl2 will have some downtime later today or perhaps tommorow !
17:21 < rxr> no problem
17:22 < mpp> thanx rxr
17:22 < rxr> do you work at some ISP ?
17:22 < rxr> mpp: you're welcome!
17:22 < mpp> nope - why
17:22 < rxr> because of exchanging boxes at those fixed ip's?
17:22 < mpp> ah...
17:23 < mpp> i don't understand ..
17:23 < mpp> you mean the OS ?
17:23 < mpp> thats no probl with server4you
17:23 < rxr> ah - ok - I thought you would update the hardware, too
17:23 < mpp> they have a pxe-recovery which allows you to do whatever you like on your server
17:24 < mpp> no . not possible for those low budget servers
17:24 < mpp> but theyre all 64bit amd systems
17:24 < rxr> ah
17:24 < rxr> we used to had one a weperoni (or how it was spelled) they offered similar functionality
17:25 < mpp> i'll have some blade servers for t2 stuff in the near future at my disposal in my company
17:25 < mpp> for now i use a single with 2 xeons and 4gb ram to build stuff
17:25 < mpp> theyre pretty cool
17:26 < rxr> wich root server offering do you use at server4you? they apear to srtart at 38.99/m ?
17:26 < mpp> ahm i have an old one which is not avail anymore
17:26 < mpp> i recommend the eco server
17:27 < mpp> i have the one with 2gb ram 2x160hdd and amd2
17:27 < rxr> :-)
17:27 < mpp> its called ecoserver big i think
17:27 < rxr> we have a 8-way with 8GB RAM for automated t2 builds here in the office for a year, now
17:27 < mpp> for the money the service is decent and the cpu power is suffice
17:27 < mpp> what system is it ?
17:27 < rxr> yes, looks good
17:28 < mpp> they also offer flatrate at 100mbit
17:28 < rxr> # cat /proc/cpuinfo | grep name
17:28 < rxr> model name : Intel(R) Xeon(R) CPU X5365 @ 3.00GHz
17:28 < rxr> ...x8
17:28 < mpp> nice !
17:29 < mpp> but the amd phenom x4 does kick ass for a low budget
17:29 < rxr> yeah - though I need to change the virtualization setup on the box, openvz loocks it up sooner than later, somewhen between some days and months, highest uptime was just 76days or so ...
17:29 < rxr> probably will use KVM domains in the future
17:29 < mpp> yes virtual is a matter id like to talk to you
17:30 < mpp> vmware fails with all kinds of errors
17:30 < rxr> I can not recomment any of the linux "in-kernel" partitionzer (openvz/vserver)
17:30 < mpp> kvm and qemu are broken
17:30 < rxr> the "patching every syscal with domain conditionals" really hurts stability
17:30 < mpp> i can imagine
17:30 < mpp> what virt do you prupose
17:30 < mpp> i really need one on my t2 boxes
17:31 < mpp> asap actually
17:31 < rxr> well - I'm not yet in the VM business ... so ... and I historically avoided vmware ...)
17:31 < mpp> is the openvz usable ?
17:31 < rxr> for embedded and system development I use qemu for years, and kvm performance looks pretty good
17:31 < mpp> but kvm and qemu do not build with t2 trunk on my side
17:32 < rxr> mpp: openvz is only soft partitioning linux containers, and that's what I used on the 8way which is not too stable for this kind of heavy duty work)
17:32 < mpp> hmm..
17:32 < mpp> mpp needs a vm
17:32 < rxr> ok - actually I only added kvm at a client in a hurry, and use HEAD checkouts of KVM and Qemu ...
17:32 < mpp> do they build on trunk ?
17:33 < rxr> well - Qemu SVN yes
17:33 < rxr> though our normal package should normally also buil
17:33 < rxr> d
17:33 < rxr> I can only imagine that the dyngen process fails in rare corner cases due our new gcc-4.x
17:33 < mpp> ill give it a try on one of my hosts
17:33 < rxr> but the upcomming qemu will use another codegen "technology" and no longer depend on a old GCC
17:34 < mpp> what packages do i need to select in generic target ?
17:34 < mpp> X qemu
17:34 < mpp> X kvm
17:34 < rxr> our trunk qemu is quite patched to build despite the newer gcc, which usually worked for me
17:35 < rxr> maybe kvm alone is enough for the hardware based virtualization I have to double check what kind of kvm package I actually comitted when I was abroad at a client some months ago ...
17:35 < mpp> if you find time to check if they build on trunk - i would be more than happy
17:35 < rxr> let me just finish some other C++ code I was just working on :-)
17:35 < mpp> sure
17:35 < mpp> thanx alot for the work
17:35 < rxr> oh - that reminds me - need to test the ppc64 cd from koan
17:36 < mpp> btw . i had to stop syncing the binaries for dl2 + dl2
17:36 < mpp> dl3
17:36 < mpp> the amount of hdd did go beyond the 34gb you mentioned
17:36 < rxr> hm - strange
17:36 < mpp> so for now i left only the 7.0 and trunk binaries
17:37 < mpp> and i will add binaries as requested by people
17:37 < rxr> hm - according to my "du" stable should be 36G and unstable 2G
17:37 < mpp> cause to be honest havin 2.0.0 and 6.x binaries makes rarely sence
17:37 < rxr> maybe you skip the 2.x ones and only sync 6.x and 7.x ...
17:38 < mpp> ill do so
17:38 < mpp> ..
17:38 < mpp> okay ill be afk for a few hours
17:38 < mpp> greetz to berlin.
17:38 < rxr> cu then - have some other C++ code to finish, likewise :-)
17:38 < rxr> cu then - have fun
17:39 < mpp> bye
17:47 -!- mpp [n=user@i5387665F.versanet.de] has quit ["good night - good fight"]
18:25 -!- axionix [n=axion@cpe-67-242-94-6.nycap.res.rr.com] has quit [Read error: 110 (Connection timed out)]
--- Log closed Sun Oct 12 00:00:10 2008