T2 IRC Log: 2004-11-09

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 Tue Nov 09 00:00:19 2004
00:51 < rxr> re
00:51 < rxr> hehe - minit is quite i18n documented:
00:51 < rxr> /* Schritt 1: argv[1] als Service nehmen und starten */
00:51 < rxr> ;-)
00:53 < rxr> da da da, ba ba ba, da ...
01:15 < mnemoc> rehi
01:16 < rxr> hi mnemoc
01:16 * rxr should be in bed
01:17 < rxr> currently fighting mol to load my emulate my kernel in a running linux
01:17 < mnemoc> and? :)
01:17 < mnemoc> mol?
01:18 < rxr> mac on linux
01:18 < rxr> GPL vmware alike for PowerPC ...
01:19 < mnemoc> aha
01:19 < rxr> way faster then qemu - aside from qemu only crashing for me for PowerPC system emulation ...
01:19 < rxr> only x86 system emulation working here on the PowerPC box ...
01:19 < rxr> da da, da da da du - yeah ...
01:20 < rxr> http://www.maconlinux.org/
01:20 < mnemoc> *click
01:20 < mnemoc> *
01:25 < mnemoc> is it a ppc emulator only or s.th like wine?
01:25 < rxr> it is like vmware
01:25 < rxr> whole system virtual machine
01:26 < mnemoc> ic
01:27 < mnemoc> how much overhead?
01:27 < rxr> quite fast, like vmware - maybe 90% cpu speed or so
01:27 < rxr> way faster then qemu ...
02:09 < rxr> crazy what old sun hardware did at ancient times ...
02:11 < rxr> ok, sparc build going along:
02:11 < rxr> 952 builds total, 213 completed fine, 3 with errors.
02:12 < rxr> and me in bed (have been to long wake anyway) - cu
02:12 < rxr> n8 all
03:45 -!- rxr_ [~rene@p213.54.244.19.tisdip.tiscali.de] has joined #t2
03:45 -!- Topic for #t2: T2 | the system development environment | http://www.exactcode.de/t2
03:45 -!- Topic set by valentin [] [Sun Nov 7 16:41:27 2004]
03:45 [Users #t2]
03:45 [ _Ragnar_] [ daja77] [ martin_] [ rxr ] [ sparc-kly]
03:45 [ CIA-8 ] [ jsaw ] [ mnemoc ] [ rxr_] [ valentin ]
03:45 -!- Irssi: #t2: Total of 10 nicks [0 ops, 0 halfops, 0 voices, 10 normal]
03:45 -!- Channel #t2 created Sun Aug 8 21:15:33 2004
03:45 -!- Irssi: Join to #t2 was synced in 11 secs
03:59 -!- rxr [~rene@p213.54.205.215.tisdip.tiscali.de] has quit [Read error: 110 (Connection timed out)]
04:00 -!- martin_ [~martin@brln-d9ba0340.pool.mediaWays.net] has quit [Read error: 110 (Connection timed out)]
04:31 < jsaw> re
05:31 -!- martin_ [~martin@brln-d9ba25e7.pool.mediaWays.net] has joined #t2
08:39 -!- mtr [~michael@Hb8c3.h.pppool.de] has joined #t2
10:32 -!- nzg [~tschmidt@p508EA3ED.dip.t-dialin.net] has joined #t2
13:14 -!- You're now known as rxr
13:14 < rxr> moin
13:15 < rxr> ouhm - qt fails on the sparc ....
13:15 < rxr> QThreadStorage: thread 4000 exited after QThreadStorage destroyed
13:15 < rxr> make[4]: *** [qmake_image_collection.cpp] Bus error (core dumped)
13:15 < rxr> make[4]: *** Deleting file `qmake_image_collection.cpp'
13:15 < rxr> [1] base/linux26 [1] base/linux24
13:15 < rxr> [1] sparc/silo [5] x11/glut
13:15 < rxr> [5] base/mtools [5] printing/cups
13:15 < rxr> [5] kde/qt
13:15 < rxr> 952 builds total, 287 completed fine, 7 with errors.
13:24 < rxr> and no, I did not got up that late, but as early and sleepy that I had to rush out into university immediately ...
13:24 < rxr> did some linuxrc work in the lecture ...
13:59 < CIA-8> rene * r4626 /trunk/package/graphic/ghostscript/ (ghostscript.conf ghostscript.desc gs-fix-gc.patch):
13:59 < CIA-8> * re-updated ghostscript (7.05.6 -> 7.07.1), including the garbage-
13:59 < CIA-8> collector patch needed for most architectures (tested on powerpc)
13:59 < CIA-8> (thanks to Mandrake and Gentoo - real source quite unclear)
13:59 < CIA-8> * fixed ghostscript status output on the way
14:04 < rxr> sparc: 292 completed fine
14:07 < jsaw> [5] emulators/hercules [5] kde/kdebindings
14:07 < jsaw> [5] contrib/directfb [5] network/gsmlib
14:07 < jsaw> [5] e17/efsd
14:07 < jsaw> 1316 builds total, 488 completed fine, 5 with errors.
14:08 < jsaw> looks nice so far :)
14:09 < rxr> yep, ut qt is aboring with bus error :-((((
14:09 < rxr> this will cuase quite a lot of build errors until fixed :-(
14:09 < jsaw> haven't seen "bus error" since the netscape days
14:14 < jsaw> rxr: besides the "imperfect indentation", the "*" is also superfluos...
14:19 < CIA-8> jsaw * r4627 /trunk/package/develop/gnupth/ ( gnupth.desc head-1-usage.patch): * add GNU Pth 2.0.2
14:36 < CIA-8> rene * r4628 /trunk/package/network/samba/samba.desc: * updated samba (3.0.7 -> 3.0.8) - SECURITY
14:37 < CIA-8> rene * r4629 /trunk/package/www/firefox/ (firefox.conf firefox.desc): * updated firefox (1.0rc2 -> 1.0) - untested
14:41 -!- Sparc-kly|G4 [~sparc@66-50-123-220.prtc.net] has joined #t2
14:42 < rxr> hi Sparc-kly|G4
14:42 < Sparc-kly|G4> hi :D
14:51 < rxr> hm, qt internal memory corruption or so?:
14:51 < rxr> 3602 munmap(0x70740000, 8192) = 0
14:51 < rxr> 3602 munmap(0x70f38000, 80592) = 0
14:51 < rxr> 3602 munmap(0x70f54000, 455520) = 0
14:51 < rxr> 3602 munmap(0x70ee8000, 87552) = 0
14:51 < rxr> 3602 munmap(0x70f08000, 182704) = 0
14:51 < rxr> 3602 write(2, "QThreadStorage: thread 4000 exit"..., 66) = 66
14:51 < rxr> 3602 --- SIGBUS (Bus error) @ 0 (0) ---
14:51 < rxr> 3602 +++ killed by SIGBUS (core dumped) +++
15:02 -!- tschmidt_ [~tschmidt@p508EB337.dip.t-dialin.net] has joined #t2
15:04 < jsaw> leaving, cu later
15:08 -!- nzg [~tschmidt@p508EA3ED.dip.t-dialin.net] has quit [Read error: 60 (Operation timed out)]
15:21 -!- Sparc-kly|G4 [~sparc@66-50-123-220.prtc.net] has quit ["Leaving"]
15:25 < rxr> cu jsaw
15:44 < rxr> amazing, even quality 0 ogg vorbis files still should very good, yielding ~56kBit/s files ...
15:44 < rxr> I now try qulity -1
15:53 < mnemoc> moin
15:58 < rxr> moin mnemoc
15:58 < rxr> wow - even quality -1 is still really good for those jazz recordings ....
16:20 < mtr> hi rxr, mnemoc
16:20 < mtr> you remember the uclibc thing we talked about a few days ago?
16:22 < mtr> now the result is: 197 builds total, 189 completed fine, 7 with errors
16:47 < rxr> mtr: good :-)
16:58 < mtr> rxr: most of the work was already done by mnemoc
16:59 < mtr> i just put together current patches for binutils and gcc
17:01 < mtr> and then the Build-Pkg problem, i guess this was the point where mnemoc stopped
17:02 < mtr> there is the LD_PRELOAD setting for fl_wrapper.so
17:03 < mtr> and if it is preloaded before chroot'ed, some dependencies are resolved against the host systems glibc
17:05 < mtr> i will post the patches when the target system is up and running
17:10 < rxr> mtr: yep, a complete mail with patches to the list would be great!
17:11 < mtr> rxr: i used an update for binutils, and found your remark: 2.15.92.0.2 is broken on powerpc
17:12 < mtr> so i assume you will not accept my patch until 2.15.93.x is out
17:13 < rxr> do you really really need the newer binutils for some reason?
17:14 < mtr> no, not for a special reason, i've used to get the most current version
17:15 < mtr> s/used to/used it to/
17:15 < rxr> on powerpc the version works, but it is stricter, so certain assembler fragments are not longer accepted ...
17:15 < rxr> resulting in gcc and the linux kernsl to not compile ...
17:15 < rxr> for gcc a patch exists, and it is included in the latest gcc release ...
17:15 < rxr> but the linux kernsl where the reason I hesitated to use the update ...
17:16 < rxr> i would like some powerpc kernel guru say whether binutils is defect, or the assembler code should be mofiied, and this it was the system init and such code I did not wanted to mess in it that day ...
17:16 * rxr needed a booting kernel ... :-)
17:17 < mtr> hmm, ok, i will put together the patches for 2.15.92.0.2
17:18 < mtr> as you mention gcc problems, i have one with 3.4.2 and large c++ apps, it's the .linkonce bug
17:18 < mtr> searching in bookmarks ...
17:19 < rxr> oops
17:20 < rxr> my ultrasparc 30 just froze, even using the debian kernel I use for reference test reasons right now ...
17:20 < rxr> so the only difference between my kernel and the debian one is: my kernel makes the box freeze ever day and the debian one every four days ... :-(
17:22 < rxr> but this time it happend exactly at the time my fridge went out and i heared some click in the loudspeakers ...
17:22 < rxr> maybe the U30 is quite touchy to the power supplied?
17:22 < rxr> although I do not have the problem with any other box ...
17:24 < mtr> yep, here it is: http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16276
17:24 < CIA-8> jsaw * r4630 /trunk/package/graphic/ghostscript/ghostscript.conf: * insert missing quote in ghostscript.conf
17:25 < mtr> i had the same failure when i tried to compile inkscape on my gcc-3.4.2 system
17:26 < daja77> O_o thought we fixed that ages ago
17:28 < rxr> mtr: I do not have ti
17:28 < rxr> this problem on x86 or powerpc so far
17:28 < rxr> hm - maybe I have, stlport might not compile here due to that, /me checking
17:29 < mtr> i never had this before, but with inkscape-0.40pre1
17:29 < rxr> ah
17:29 < rxr> mtr: so it is in 3.4.3 ?
17:30 < rxr> Additional Comment #28 From Mark Mitchell 2004-10-31 01:03 [reply] -------
17:30 < rxr> Postponed until GCC 3.4.4.
17:30 < rxr> ?!?!
17:30 < mtr> the last comment on gcc-bugzilla on the very buttom of the page: "Postponed until GCC 3.4.4."
17:30 < rxr> yeah - just read it ...
17:32 < rxr> hm in my stlport build I have those:
17:32 < rxr> `.L18779' referenced in section `.rodata' of ../lib/obj/GCC/ReleaseD/locale_cata
17:32 < rxr> log.o: defined in discarded section `.gnu.linkonce.t._ZNK4_STL9money_getIcNS_19i
17:32 < rxr> streambuf_iteratorIcNS_11char_traitsIcEEEEE6do_getES4_S4_bRNS_8ios_baseERiRNS_12
17:32 < rxr> basic_stringIcS3_NS_9allocatorIcEEEE' of ../lib/obj/GCC/ReleaseD/locale_catalog.
17:32 < rxr> o
17:33 < rxr> so it might be indeed that bug ...
17:33 < rxr> just for reference, might be nice (C++ template based image processing):
17:33 < rxr> http://kogs-www.informatik.uni-hamburg.de/~koethe/vigra/
17:35 < rxr> mtr: do you imported a patch for that bug in your build?
17:41 < mtr> rxr: do you mean a gcc-patch in my uclibc-build? No
17:43 < mtr> btw - ot - a patch for transcode and gcc-3.4.2 already exists
17:44 < mtr> it is in the other project, which name should not be spoken ;-)
17:46 < rxr> ;-)
17:46 < rxr> mtr: i meant the gcc patch ...
17:46 < rxr> do you test it, or should I review this, here?
17:47 < mtr> compiles and links, but no further tests done
17:48 < rxr> URL for the patch you used?
17:48 < rxr> mtr: do you want to work on / with t2 and want write access? or are you still in some orientatio process?
17:52 < CIA-8> rene * r4631 /trunk/package/multimedia/transcode/gcc34.patch: * imported transcode/gcc34.patch from ROCK Linux
17:54 < mtr> to your first question, i would answer yes, i will continue working on and using t2
17:55 < mtr> but i am not sure if get enough spare time in the future
17:57 < mtr> on the other hand i have some other packages for my settopbox: vdr, noad, dvb-driver
17:59 < mtr> i did not submit the patches until now cause of SM which has a patch naming schema i cannot work with
18:00 < mtr> i am really bad in linking and remembering numbers like 2004110916185800589 with a real object
18:01 < rxr> re
18:02 < rxr> I can understand this ;-)
18:03 < rxr> just notify me when you want arite access to commit s.th. - but just sending patches to me or the list is also very ok
18:08 < mtr> ok, i am thinking about the write access, and i will send the uclibc patches to the list
18:08 < rxr> == 11/09/04 17:54:42 =[9]=> Finished building package firefox.
18:15 < CIA-8> rene * r4632 /trunk/package/x11/glut/glut.conf: * fixed glut for non-x86 systems
18:22 < rxr> == 11/09/04 18:22:19 =[5]=> Finished building package transcode.
19:25 < rxr> do we want the checkiso5dm bload in our linuxrc?
19:48 * rxr testing stlport with patched gcc ...
19:49 < rxr> nope, the tiny patch is not enought:
19:49 < rxr> !> `.L18694' referenced in section `.rodata' of ../lib/obj/ ..
19:49 < rxr> !> `.L18767' referenced in section `.rodata' of ../lib/obj/ ..
19:49 < rxr> !> `.L18779' referenced in section `.rodata' of ../lib/obj/ ..
19:49 < rxr> !> `.L18885' referenced in section `.rodata' of ../lib/obj/ ..
19:54 < rxr> mtr: did you apply the single file patch , or the multiple extended single file ones ?
19:57 < rxr> I think I live with the bug until it is fixed in a real release - I do not want to pull that big patchset right now ...
19:57 < mtr> ehem, missunderstanding here: i did not test any of the gcc patches for .linkonce
20:00 * mtr did not manage to get it work :-(
20:04 < rxr> ah, I tought you patched it to get inkscape compiled ...
20:04 < rxr> in my desktop build only stlport failed due to that so far ...
20:04 < rxr> I do the more important stuff first ...
20:04 < rxr> if anyone wants to pull the real patches in feel free to do so ...
20:05 < rxr> the gcc-3.4-linkonce-1.patch alone does not help me ...
20:05 < rxr> so /me a bit off-line - cu
20:32 -!- Tchmnkyz [~jeremy@c-24-131-109-234.mw.client2.attbi.com] has joined #t2
22:25 < rxr> n8 - all
22:43 -!- Tchmnkyz [~jeremy@c-24-131-109-234.mw.client2.attbi.com] has quit [Read error: 113 (No route to host)]
22:46 < mtr> gn8 rxr
23:21 -!- mtr [~michael@Hb8c3.h.pppool.de] has quit ["using sirc version 2.211+KSIRC/1.3.11"]
--- Log closed Wed Nov 10 00:00:33 2004