T2 IRC Log: 2009-11-09

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 Nov 09 00:00:17 2009
07:39 -!- capey [n=mp@telakka.wapice.com] has joined #t2
08:02 -!- CIA-28 [n=CIA@208.69.182.149] has quit [Read error: 60 (Operation timed out)]
08:03 -!- CIA-28 [n=CIA@208.69.182.149] has joined #t2
08:27 -!- mjungwirth [n=mjungwir@213.174.234.68] has joined #t2
08:27 < mjungwirth> moin
08:39 < capey> morning
09:31 < rxr> moin
10:00 < CIA-28> rene * r34844 /trunk/package/graphic/epdfview/epdfview.desc:
10:00 < CIA-28> Gerardo Di Iorio :
10:00 < CIA-28> * updated epdfview (0.1.6 -> 0.1.7)
10:26 < rxr> http://rene.rebe.name/photos/?p=/Berlin/2009/20%20Years%20Fall%20of%20the%20Berlin%20Wall/
10:53 < CIA-28> rene * r34845 /trunk/package/gnome2/f-spot/f-spot.desc: * updated f-spot (0.6.1.3 -> 0.6.1.5)
10:58 < rxr> http://rene.rebe.name/2009-11-09/20-years-since-the-fall-of-the-berlin-walls/
11:12 -!- csiadmin [n=csiadmin@81.144.152.52] has joined #t2
11:13 < CIA-28> rene * r34846 /trunk/package/gnome2/gmime/gmime.desc: * updated gmime (2.4.10 -> 2.4.11)
11:36 < rxr> 9http://t2-project.org/packages/virtualbox.html
11:38 < CIA-28> rene * r34847 /trunk/package/x86/flashrom/flashrom.desc: * updated flashrom (0.9.0 -> 0.9.1)
11:38 < CIA-28> rene * r34848 /trunk/package/develop/vala/vala.desc: * updated vala (0.7.7 -> 0.7.8)
11:38 < CIA-28> rene * r34849 /trunk/package/graphic/libsdl_mixer/libsdl_mixer.desc: * updated libsdl_mixer (1.2.9 -> 1.2.10)
12:03 -!- Xor [n=oscar@11.Red-80-38-62.staticIP.rima-tde.net] has joined #t2
12:04 < Xor> hello all
12:04 -!- Xor is now known as Guest69410
12:07 < Guest69410> im getting a lot of "too few arguments" errors in some "xf86-input-xxx" packages
12:12 -!- Guest69410 is now known as Xor_
12:13 < rxr> some are deprecated upstream
12:13 < rxr> we might have to remove those, soon, if noone updates them to the new X.org ABI/API
12:19 < Xor_> can this errors affect to the final build? I am building a livecd generic target for second time. First time, the script Create-ISO returned an error.
12:19 < rxr> the ISO should create without those packages
12:25 < Xor_> Then i don't understand why create-iso returned this error: Not a valid: isofs.txt must exists
12:26 < rxr> because there is no isofs.txt in the build/.../TOOLCHAIN (or so) directory
12:26 < rxr> should be trival to review what went wrong at the end of the build
12:26 < rxr> which architecture do you build for?
12:28 < Xor_> x86-i486
12:28 < rxr> you have selected an installer or live image as output?
12:29 < Xor_> Do you refer to Target Distribution ?
12:30 < rxr> I think so, would need to lookup the exact menu name
12:33 < Xor_> Target Distribution: Generic System / Minimal-Xorg / livecd
12:33 < rxr> no
12:34 < rxr> (install) Install CD / disks
12:34 < rxr> 3rd row or so
12:35 < Xor_> I have selected "Live CD", "Install CD/disks" is unchecked
12:46 < rxr> ok, livecd should be good, too
13:31 < csiadmin> hi everyone
13:31 < csiadmin> anybody know why I'm not getting a initrd.img file even though the rescue build completed successfully?
13:43 < Xor_> build finish with this message: "Creating isofs_arch.txt file .." but Create-Iso script says: isofs.txt must exist. What i'm doing wrong ?
13:58 < Xor_> I have make a soft link isofs.txt -> isofs_arch.txt and CreateISO seems to works fine
14:20 < rxr> did you modify the target build, or is this an unmodified build ?
14:20 < Xor_> this is a unmodified build
14:24 < Xor_> Why in the Config_cd1.iso does no exists the squash root file system if this is a cdlive iso?
14:27 < rxr> I can not magically answer why files are missing on your side, it just works here since years
14:27 < rxr> what linux do you build on?
14:34 < Xor_> I have T2 Linux as the base system
14:35 < Xor_> and working with t2-trunk
14:37 -!- capey [n=mp@telakka.wapice.com] has quit ["home"]
15:18 < csiadmin> Could an build failure in stage 5 for mc-mp, be responsible for getting no initrd.img?
15:24 < csiadmin> ...hmm not sure this error has anything to do with it.
15:32 < rxr> no mc-mp should not affect this
15:34 -!- Baldzius [n=aldas@87-198-133-94.ptr.magnet.ie] has joined #t2
15:34 < Baldzius> moin
15:39 < rxr> hi Baldzius
15:39 < rxr> http://rene.rebe.name/2009-11-09/20-years-since-the-fall-of-the-berlin-walls/
15:40 < Baldzius> hey rxr
15:43 < Baldzius> was trying to compile t2 on Ubuntu 9.10 64 bits yesterday
15:43 < Baldzius> everything went fine until it reached 2-gcc
15:43 < Baldzius> I see someone also pasted this error on 16th of October
15:43 < Baldzius> 22:48 < dpc> configure:7657: result: no
15:43 < Baldzius> 22:48 < dpc> configure:7821: checking dynamic linker characteristics
15:43 < Baldzius> 22:49 < dpc> configure:8239: error: Link tests are not allowed after GCC_NO_EXECUTABLES.
15:48 < rxr> hm
15:58 < Baldzius> was googling but couldn't find anything useful
15:59 < Baldzius> it doesn't look to be trigered by bintutils update
15:59 < Baldzius> and not sure if it has to do something with gcc as ubuntu runa on 4.4.1
15:59 < Baldzius> runs*
16:34 < Baldzius> rxr: btw is it possible to get installable 64 bits iso image on 32 bit host system with cross compile enabled with x86_64 cpu?
16:34 < Baldzius> never cross compiled before
16:35 < rxr> sure
16:35 < rxr> since 8.0-rc one can cross build bootable images
16:35 < rxr> all -rc images where fully cross built
16:36 < Baldzius> even powerpc or sparc?
16:38 < rxr> yepp
16:38 < rxr> all isos built on xeon 8-core
16:39 < Baldzius> would it create iso for bootstrap target?
16:40 < rxr> hm, no idea if bootstrap has enough for iso creation
16:41 < rxr> i built the minimal
16:41 < Baldzius> ok, will try myself
16:41 < Baldzius> thanks
16:41 < Baldzius> btw seems like mtr is preparing gcc-4.4 update
16:42 < Baldzius> not sure if he have mentioned something here
16:42 < rxr> yes, we discussed it yesterday or the night before :-)
16:43 < Baldzius> any dates mentioned?
16:48 < rxr> nope, just that he wanted to try to update :-)
16:48 < rxr> maybe it would be best to branch / tag the current stable state as 8.0 before the update, though
16:49 < rxr> http://t2-project.org/irclogs/t2-2009-11-07.html
16:49 < rxr> ^- mentioning gcc 4.4
16:49 < rxr> btw. I know that virtualbox does not install yet, just wanted to share the finally compiling package base for other to have a chance to contribute / finalize, too
16:50 < rxr> http://t2-project.org/packages/virtualbox.html
16:53 < Baldzius> I'd say we shoudn't rush for branch
16:53 < Baldzius> merging involves huge work
16:53 < Baldzius> I remember when we were releasing 6.0
16:53 < Baldzius> that was 1 _busy_ month of merging
16:56 < Baldzius> but in general there's nothing much left to do : udev, orbit2 and some other small stuff
16:56 < Baldzius> I would really like to get my patch in for 8.0
16:57 < Baldzius> as I test instaltions from usb key
17:02 < rxr> ah yes, install right
17:02 < rxr> well, I'd like to finally have a nice stable major release (series)
17:02 < rxr> I do not want gcc-4.4 instability getting in beween that
17:03 < rxr> I guess the new library dependenceies of gcc-4.4 will certainly cause cross compile regressions
17:03 < CIA-28> rene * r34850 /trunk/package/develop/autogen/autogen.desc: * updated autogen (5.9.9 -> 5.10)
17:04 < Baldzius> hm
17:09 < csiadmin> guys, I'm using trunk to try to build a bootable initramfs, I'm getting no errors but don't get an initrd.img output - can anyone help?
17:13 < rxr> csiadmin: initramfs as target build output ?
17:13 < csiadmin> rxr: yeah
17:13 < rxr> hm - been some months, year?, since I last used it
17:14 < rxr> which target and package selection? maybe too huge?
17:15 < csiadmin> rxr: it's a custom build based on rescue (but with a few files in rootfs)
17:15 < rxr> did it built before?
17:15 < rxr> built and worked ?
17:16 < csiadmin> rxr: yes I built it recently without issue, it's very small ~100 packages
17:16 < rxr> and worked?
17:16 < rxr> booted and all?
17:17 < csiadmin> rxr: yep it was fine, I used it to install a bigger t2 build on some machines
17:17 < rxr> Xor_ also gets a build without isofs.txt stuff
17:17 < rxr> ok, I try to test myself ASAP
17:17 < rxr> though maybe a trivial typo?
17:17 < rxr> have you tried to trace where somethign goes wrong, aborts?
17:18 < csiadmin> rxr: Not yet - it says 'Creating output files...' then unmounts as normal
17:18 < rxr> maybe something as "stupid" as a bash-4.0 regression, ...
17:19 < csiadmin> rxr: I appreciate your help, don't have enough experience to know where to look (yet)
17:20 < rxr> csiadmin: you could start adding some echo "I'm here" things on the build.sh way to the final initrd creation
17:20 < rxr> maybe first try enabling this XTRACE config option, if you are lucky it gives a hitn
17:20 < rxr> hint
17:21 < rxr> maybe, thoug, the XTRACE has no effect in the "not building a package" scripts, though
17:21 < csiadmin> rxr: I had a look in target/share/build.sh - is this the build file you mean?
17:22 < rxr> and the initrd specific one
17:22 < rxr> target/share/initrd/build.sh
17:22 < csiadmin> rxr: sorry that's the file I meant
17:23 < rxr> try enable XTRACE in the CONfig
17:23 < rxr> Config
17:23 < rxr> maybe you seee something new where it fails, stops
17:23 < rxr> if not start adding some "poor man's echo /tracepoints/"
17:24 < csiadmin> rxr: just ran a build with xtrace enabled...http://pastebin.ca/1663562
17:27 < rxr> we shoudl fix mc-mp :-)
17:27 < rxr> ok - not so intersting
17:28 < rxr> maybe something generic is wrong with output image creation as Xor_ also has proglems
17:28 < rxr> as it appears to be generic maybe bash 4.x related
17:28 < rxr> as my build server has an older root
17:29 < csiadmin> rxr: would it matter that I'm build a 32bit image on a 64bit machine?
17:29 < rxr> doubt
17:29 < rxr> I suspect some generic issue
17:29 < rxr> either something like bash-4 or recently introduced typo
17:29 < rxr> or cross vs. non-cross build typo etc
17:30 < rxr> as this is post the build it should be really possible to trace it down with some echo's and trial'n error
17:30 < rxr> question is, do I leave and go look at domino wall toppling :-)
17:30 < Xor_> now I'm trying with version 7.0 !
17:30 < rxr> http://rene.rebe.name/2009-11-09/20-years-since-the-fall-of-the-berlin-walls/
17:31 < csiadmin> rxr: I'll have a go at debugging the script - you should definitely check out the domino wall :-)
17:32 < rxr> well, it will be very^3 crowded there,
17:32 < rxr> doubt I'll see anything :-) and I got pretty pictures yesterday :-)
17:32 < csiadmin> rxr: I had a look through the pictures already - looks great
17:33 < rxr> berlin feels great, too :-) though I only know it united
17:33 < rxr> was 9 when the wall fell, never visited east germany, or berlin before
17:33 < rxr> now living in berlin for nearly a decade, though
17:33 < rxr> time goes by pretty fast
17:34 < rxr> wow, I just read the domino bricks weight 20kg!
17:34 < csiadmin> rxr: I visited Berlin in May and had a great time, I thought the city had a great party atmosphere!
17:35 < rxr> definetly
17:35 < rxr> very multi-cultural and with a lot of students
17:35 < rxr> new people moving to berlin from everywherre
17:36 < rxr> it's like 3 universities, 2 unversities of applid science, and the additional medical schools etc.
17:36 < csiadmin> rxr: I thought everyone would have the day off work!
17:37 < rxr> today?
17:37 < csiadmin> rxr: yeah I thought it'd be a public holiday
17:37 < rxr> well, public holiday celebrating the reunation is the 2nd october
17:38 < csiadmin> rxr: ah I see - I remember watching the wall come down on the news, I was 14 at the time
17:40 < rxr> then you're some years old than I'm .-)
17:40 < rxr> as I mentioned in my blog post this year(s) quite some thing come together
17:41 < rxr> 60 years federal republic of germany (post WWII) 20 years since fall of the wall, and next year it will be 20 years since reunification
17:42 < csiadmin> rxr: plenty to celebrate, the celebrations are getting lots of coverage here on tv
17:42 < rxr> nice
17:43 < csiadmin> http://news.bbc.co.uk/1/hi/in_depth/world/2009/walls_around_the_world/default.stm
17:43 < csiadmin> ...plenty more walls to come down
17:43 < rxr> yeah
17:44 < rxr> with the german wall 20 years ago a way larger icon curtain fell
17:44 < rxr> making europe, the world a way opener place
17:45 < csiadmin> rxr: absolutely
17:45 < rxr> http://www.google.de/logos/berlinwall09-hp.gif
17:45 < rxr> ^- german dodle (google.de) image of the day
17:46 < csiadmin> rxr: very cool
17:49 < csiadmin> rxr: will try to debug and let you know how I get on tomorrow. As always thanks for your assistance :-)
17:51 -!- csiadmin [n=csiadmin@81.144.152.52] has left #t2 []
22:01 < mtr> hi
22:02 < mtr> gcc-4.4 build minimal+x is ready ;-)
22:03 < mtr> the ppl and cloog libraries even do cross build, an embedded target works as well
22:07 < mtr> rxr: from what you wrote this afternoon about the branching strategy I'll withhold the gcc update commit
22:08 < mtr> rxr: maybe I should commit the ppl and cloog-ppl packages, prepare the 2-line patch for gcc.conf, and you could give it a try on your reference builder
22:09 < mtr> ah, there is an open issue i had: as ppl needs the host g++ i got "undefined references to __sync_fetch_and_add_4"
22:10 < mtr> usually setting -march=i686 solves this for me
22:10 < mtr> is it in general safe to use "-march=$HOSTTYPE" (for sparc, ppc, et. al.) ?
23:26 -!- Baldzius [n=aldas@87-198-133-94.ptr.magnet.ie] has quit ["Leaving"]
--- Log closed Tue Nov 10 00:00:17 2009