| doko | hmm, libxau-dev and libxdmcp-dev are in the archives. strange, that they cannot be found ... | 12:07 |
|---|---|---|
| doko | lamont: ^^^ | 12:09 |
| lamont | missing buildd-deps, I bet | 12:10 |
| lamont | actually libxdmcp-dev is installed | 12:11 |
| lamont | no -I/usr/X11R6/include though... | 12:11 |
| lamont | gcc -c -fsigned-char -I../.. -I../../exports/include -Dlinux -D__powerpc__ | 12:12 |
| lamont | +-D_POSIX_C_SOURCE=199309L -D_POSIX_SOURCE | 12:12 |
| lamont | +-D_XOPEN_SOURCE -D_BSD_SOURCE -D_SVID_SOURCE | 12:12 |
| lamont | +-D_GNU_SOURCE -DFUNCPROTO=15 -DNARROWPROTO | 12:12 |
| lamont | +-DXTHREADS -D_REENTRANT -DXUSE_MTSAFE_API -DMALLOC_0_RETURNS_NULL | 12:12 |
| lamont | +-DHAS_SNPRINTF -DLIBX11 -DPOSTLOCALELIBDIR=\"lib\" | 12:12 |
| lamont | +-g -O2 -fno-strict-aliasing -g -DUNIXCONN -DTCPCONN -DHAS_STICKY_DIR_BIT | 12:12 |
| lamont | +-DHAS_FCHOWN -DIPv6 -DX11_t -DTRANS_CLIENT -DFAIL_HARD ConnDis.c -o | 12:12 |
| lamont | +unshared/ConnDis.o | 12:12 |
| lamont | ConnDis.c:38:23: error: X11/Xauth.h: No such file or directory | 12:12 |
| === lamont wonders if maybe they used to come from ../../exports/include | ||
| doko | yes, maybe ... | 12:13 |
| lamont | gst-plugins0.8 is ftbfs on ppc, btw | 12:13 |
| doko | wuhuhuhu | 12:14 |
| doko | lamont: lying, take a penalty card | 12:14 |
| lamont | on which? | 12:14 |
| doko | http://people.ubuntu.com/~lamont/buildLogs/g/gst-plugins0.8/0.8.8-3ubuntu3/gst-plugins0.8_0.8.8-3ubuntu3_20050523-2138-powerpc-successful.gz | 12:15 |
| lamont | ah, was ubuntu2 that I looked at. my bad | 12:15 |
| lamont | I am taking a penalty card | 12:15 |
| doko | talking ;) | 12:16 |
| doko | BOOTSTRAPCFLAGS=-I/usr/include/X11 | 12:17 |
| doko | hmm, but that isn't on the command line either | 12:17 |
| doko | elmo: libestools1.2 (speech-tools) needs new/main love | 12:36 |
| elmo | doko: no, it needs to build | 12:38 |
| elmo | come on dude, these are things you can check yourself | 12:38 |
| doko | ok, time to go to bed ... | 12:38 |
| doko | libao is a *_GNU_TYPE victim ... | 02:12 |
| === rtcm [~jman@217.129.142.72] has joined #ubuntu-toolchain | ||
| === lamont gets dragged off | ||
| doko | please try to get pike7.6 and then swig1.3 in. | 02:40 |
| daniels | huh | 03:29 |
| daniels | so, for what it's worth, -I/usr/X11R6/include won't gain you a thing | 03:34 |
| daniels | ah, pkg-config and x11proto-core-dev | 03:38 |
| daniels | yeah, I remembered pkg-config while I was half-asleep, decided it wasn't worth getting back up for | 03:38 |
| doko | heh, and the package is native | 03:41 |
| daniels | b-d-i vs b-d? | 03:47 |
| daniels | ahr, stupid libGLw | 03:48 |
| doko | you did have Build-Depends-Indep, which is wrong, you build for architecture any, not all | 03:48 |
| daniels | yeah | 03:49 |
| doko | would you mind writing down an email, what kind of things changed, which build-deps are likely to be replaced? | 03:50 |
| daniels | hm? it's only for two packages that I fucked up on | 04:00 |
| doko | no, in general: xinerama-dev needs to be added, xlibmesa-glu-dev -> libglu-dev-xorg, and so on ... | 04:01 |
| daniels | uhm, the xinerama-dev thing has been taken care of long ago, when we broke out xlibs-static-dev before hoary | 04:02 |
| daniels | the only thing I've changed recently in terms of build-deps is xlibmesa-glu-dev -> libglu-dev-xorg | 04:02 |
| doko | and libxau-dev was needed for libao ..., and a replacement for xlibmesa-dev, and probably unknown things. *please* write it down, not every MOTU knows about it | 04:04 |
| daniels | um, again, this has been around since before hoary | 04:08 |
| daniels | and there's no direct mapping | 04:08 |
| daniels | there's just 'if it uses this, then it needs this, et al' | 04:09 |
| doko | yes, *just*. please document it. | 04:13 |
| doko | good night | 04:13 |
| fabbione | morning | 06:11 |
| lamont | morning fabbione | 06:12 |
| fabbione | doko: gcc-3.3 did built fine :) | 06:41 |
| fabbione | 3.4 starting now | 06:41 |
| === lamont sleeps | ||
| daniels | elmo: when you wake up, could I please have all of xorg's build-deps (in the newest versions) in concordia's chroot? | 07:05 |
| fabbione | hey daniels | 07:06 |
| fabbione | daniels: how difficult it is to switch a crappy Makefile to automake? | 07:06 |
| daniels | fabbione: not too hard ... program or library? | 07:07 |
| fabbione | a mix | 07:07 |
| fabbione | there is a bit of everything :) | 07:07 |
| daniels | shouldn't be too hard | 07:07 |
| daniels | heh | 07:07 |
| daniels | what is it? | 07:07 |
| fabbione | red hat cluster suite | 07:08 |
| daniels | ah | 07:08 |
| daniels | just look at the packaging in xorg/lib/Xau | 07:08 |
| fabbione | their makefile system is totally retarded | 07:08 |
| daniels | and add bin_PROGRAMS = foo, foo_SOURCES = bar.c baz.c | 07:08 |
| fabbione | meh dude... i add that to what? | 07:11 |
| fabbione | i know >< this much about autocrap | 07:11 |
| fabbione | i guess i will need to learn it | 07:13 |
| daniels | add that to Makefile.am | 07:18 |
| fabbione | i am still puzzled by all the crap :) | 07:19 |
| === infinity was considering diving into autotools this week too.. | ||
| infinity | But then I thought better of it. | 07:31 |
| fabbione | infinity: let's switch the kernel to autotools :) | 07:33 |
| infinity | NOOOOOOO! | 07:53 |
| === mvo [~egon@ip181.135.1511I-CUD12K-01.ish.de] has joined #ubuntu-toolchain | ||
| doko | morning | 09:15 |
| fabbione | hey doko | 09:15 |
| doko | NOOOOOOOOOOOOOOOOOOOOOO | 09:15 |
| fabbione | ? | 09:15 |
| doko | Package: x11proto-gl-dev | 09:15 |
| doko | Conflicts: xlibmesa-gl-dev (<< 6.8.2-19), xlibmesa-glu-dev (<< 6.8.2-19), libglu-dev-xorg (<< 6.8.2-19) | 09:15 |
| doko | we did just change dozens of packages, now again? | 09:16 |
| fabbione | doko: and you will keep changing them until the transition is completed | 09:17 |
| doko | no, I don't change them anymore, he can do it himself, | 09:19 |
| doko | and he doesn't want to document/tell the needed changes | 09:20 |
| daniels | sigh | 09:22 |
| daniels | i'm trying not to cause another ftbfs by waiting and testing | 09:23 |
| daniels | i can dump yet another version into the archive and see how that turns out, but I'd rather not | 09:23 |
| daniels | no-one needs to change anything | 09:23 |
| daniels | libglu-dev-xorg in -19 depends on x11proto-gl-dev | 09:23 |
| daniels | it's just an internal reorganisation | 09:23 |
| doko | I'll only believe it, when I see it. does -19 fix xbase-clients? | 09:26 |
| daniels | http://people.ubuntu.com/~daniels/xlibs-static-dev.txt | 09:31 |
| daniels | and no, it won't, as I explained to you before | 09:31 |
| daniels | one version needs to get built and accepted, and then the next one will fix xbase-clients | 09:32 |
| === chmj [~d3vic3@dumbledore.hbd.com] has joined #ubuntu-toolchain | ||
| daniels | doko: -19 source is uploading to chinstrap now, if you want to upload it then do that, but for the love of god test it in an up-to-date chroot first | 09:46 |
| daniels | i can't because my mirrors' mid-pulse, and I'm stuck in the middle of main -- can't debootstrap | 09:46 |
| daniels | i'm leaving soon and will be out for a few hours | 09:46 |
| doko | ok, I'll ask elmo for build-deps on concordia, if that builds ok, then I'll upload it. can I start the build with -jN ? | 09:50 |
| daniels | don't | 09:50 |
| daniels | falls over badly when you try to parallelise it | 09:50 |
| doko | ok | 09:50 |
| fabbione | is anybody using concordia? | 09:56 |
| fabbione | or can i spin its load up? | 09:57 |
| doko | fabbione: no, please don't, trying to build xorg, when it's uploaded and build-deps are installed | 09:57 |
| fabbione | doko: sure.. no problem | 09:57 |
| doko | daniels: you have to fix xorg, to compete with fabbione for CPU time ;-) | 09:58 |
| fabbione | doko: tsk :) | 09:59 |
| fabbione | don't make me run -j400 | 09:59 |
| fabbione | doko: btw.. 80% of my load is disk I/O bounded | 10:00 |
| fabbione | it's not all real CPU | 10:00 |
| fabbione | i have ccache configured on concordia | 10:00 |
| fabbione | doko: but you are not even logged in on concordia! | 10:02 |
| fabbione | oh great | 10:03 |
| fabbione | FUCK | 10:03 |
| doko | ? | 10:03 |
| doko | _when_ it's uploaded | 10:03 |
| fabbione | i hate DPATCH | 10:03 |
| fabbione | 1.2 is FTBFS | 10:04 |
| fabbione | FUCKING HELL | 10:05 |
| fabbione | elmo: can you please install a sane version of vim in halley/breezy chroot? kthxbye | 10:23 |
| === _mvo_ [~egon@ip181.135.1511I-CUD12K-01.ish.de] has joined #ubuntu-toolchain | ||
| elmo | fabbione: done | 11:11 |
| fabbione | elmo: thanks mucho | 11:11 |
| doko | elmo: please could you install xorg deps on concordia/breezy? | 11:35 |
| fabbione | elmo: don't :) i am building the kernel :P | 11:35 |
| === doko slaps fabbione ;) | ||
| elmo | fabbione: err, too late, sorry | 11:37 |
| elmo | (sorry, I really did start, like 20 mins ago) | 11:38 |
| elmo | I should have realised that's why it was going so slowly | 11:38 |
| fabbione | ahha :) | 11:38 |
| fabbione | don't worry... i can stop if you need speed | 11:38 |
| elmo | nah, it's cool, as long as it doesn't break your build | 11:39 |
| fabbione | it shouldn't | 11:39 |
| fabbione | just don't switch compiler version :) | 11:39 |
| fabbione | that's enough | 11:39 |
| elmo | doko/daniels: done | 11:47 |
| doko | elmo: please could you move libestools1.2 to main | 12:01 |
| elmo | doko: done, 30 mins ago | 12:01 |
| doko | thanks | 12:03 |
| doko | fabbione, daniels: xorg -19 did cleanly build on amd64. I'm uploading the sources now | 01:06 |
| fabbione | doko: hold on | 01:06 |
| fabbione | did you change anything in the MANIFEST ? | 01:06 |
| fabbione | doko: or can i see the diff please? | 01:07 |
| doko | nothing, see /home/daniels on chinstrap | 01:08 |
| fabbione | doko: ok... seems alright.... | 01:10 |
| === Seveaz [~seveas@seveas.demon.nl] has joined #ubuntu-toolchain | ||
| === doko_ [~doko___@dsl-084-059-053-209.arcor-ip.net] has joined #ubuntu-toolchain | ||
| mvo | it looks like the code in AM_GNU_GETTEXT is no longer working with 64bit/gcc-4.0. this caused aptitude to fail | 02:00 |
| mvo | has anyone seen that before? | 02:01 |
| mvo | ^--- doko | 02:03 |
| doko | no | 02:04 |
| doko | hmm, that should be fixed upstream? | 02:04 |
| mvo | yes | 02:06 |
| mvo | I'll have a look | 02:06 |
| jbailey | mvo: I've seen it with g++, I think. | 02:08 |
| mvo | the bad code is: | 02:10 |
| mvo | return (int) gettext ("") + _nl_msg_cat_cntr + *_nl_domain_bindings | 02:10 |
| mvo | conftest.cc:100: error: cast from 'char*' to 'int' loses precision | 02:10 |
| mvo | and it comes from gettext.m4 | 02:10 |
| mvo | its just a matter to change it into long | 02:12 |
| jbailey | Yup, that's the beast. | 02:13 |
| jbailey | You need to run the gettext -f command or some such like that. | 02:13 |
| mvo | -f ? | 02:13 |
| jbailey | For some reason autoreconf -f -i didn't seem to be enough, I think because it calls autopoint rather than gettext (which seems a bit uselesS) | 02:13 |
| jbailey | force it to overwrite anything it had there before? | 02:14 |
| mvo | ok | 02:14 |
| jbailey | If that doesn't work, ping me. My brain is split a few other ways atm so I can't check it out. | 02:16 |
| jbailey | mvo: You will need to do an autoreconf after, though, to make sure that configure picks up the changes. | 02:16 |
| mvo | yes, I think it will work fine | 02:17 |
| \sh | mvo: which package was it, where u had this error? | 02:39 |
| mvo | \sh: aptitude, it was fixed by using the latest gettext.m4 from the gettext package | 02:45 |
| fabbione | jbailey: impressive.. benc is on irc :) | 02:54 |
| jbailey | Wow. | 02:56 |
| jbailey | Should we pounce and scare him away? =) | 02:56 |
| \sh | mvo: was it something like this? check the last output of the build logs (http://hwdb.ubuntu.com/buildlogs/?show=http://people.ubuntu.com/~lamont/buildLogs/d/dar/2.2.1-1ubuntu1/dar_2.2.1-1ubuntu1_20050524-1145-ia64-failed.gz) | 02:56 |
| jbailey | Last I heard from benc on IRC, he was doing fiarly well in a poker tournament. | 02:56 |
| fabbione | jbailey: i am already working on it.. but he has 23 hours of idle | 02:57 |
| jbailey | He's apparently not that far of a drive from here. | 02:58 |
| fabbione | jbailey: what are you waiting to knock on his door? | 02:58 |
| jbailey | But I think the border guards might question me having a body bound and gagged in the trunk to bring home with me. | 02:58 |
| jbailey | Nothing. I've long ago accepted that I will get nothing useful out of him. | 02:58 |
| lamont | jbailey: that's why it's _very_important_ that he be quiet | 02:59 |
| fabbione | ehehhe | 02:59 |
| mvo | \sh: let me check | 03:00 |
| lamont | and covered with coffee grounds, to confuse the dogs. | 03:00 |
| jbailey | Better that than poppy seeds... | 03:01 |
| lamont | heh | 03:02 |
| === lamont figures border guarddogs are trained to alert on coffee as well... | ||
| jbailey | I'll do it at the Vancouver crossing. | 03:06 |
| jbailey | Between Vcr and Seattle, the guards would go nuts if they looked at coffee. | 03:06 |
| jbailey | The hard part is getting the trunk full of grinds up to body temperature so that he doesn't show up on a thermograph/ | 03:06 |
| mvo | \sh: yep, just copy the gettext.m4 from /usr/share/aclocal/gettext.m4 to m4/ and call autoreconf (seems to work for me at least) | 03:06 |
| jbailey | mvo: That will probably work. I usually prefer to update all of gettext so that at least I'm in a somewhat supported config. | 03:07 |
| mvo | is there a wiki page about common problems with g++-4.0 and how to fix them? could be interessting for the MOTUs to share experience | 03:07 |
| mvo | jbailey: runining "gettexize -f" ? | 03:07 |
| mvo | jbailey: runining "gettexize -f -c" | 03:08 |
| mvo | ? | 03:08 |
| jbailey | ADd -c | 03:08 |
| lamont | Group.cpp: In member function 'std::string libfwbuilder::Group::_ZTv0_n16_NK12libfwbuilder5Group11getTypeNameEv() const': | 03:09 |
| lamont | Group.cpp:55: internal compiler error: in cp_expr_size, at cp/cp-objcp-common.c:101 | 03:09 |
| lamont | Please submit a full bug report, | 03:09 |
| \sh | mvo: ah .. u r my rescue :) | 03:09 |
| lamont | woot. same as arts | 03:09 |
| \sh | mvo: i will try this workaround | 03:12 |
| mvo | \sh: jbailey advice is probably better, use "gettextize -f -c; autoreconf" instead of just copying the gettext.m4 file | 03:19 |
| jbailey | lamont: Still on hppa? | 03:20 |
| mvo | \sh: you can check the result by searching in the generated configure file. what should happen is: "return (int)gettext(""); " -> "return * gettext("")" | 03:20 |
| lamont | jbailey: yes | 03:20 |
| lamont | it would appear that we have two instances of the same ICE | 03:21 |
| lamont | couple more, and we can make a drink. :=) | 03:21 |
| jbailey | lamont: Do you have time to reduce it to a testcase? Sounds like it might be time to punt it to upstream bugzilla. | 03:21 |
| lamont | I'll make time in the next day or two | 03:21 |
| jbailey | I'm trying to remember is hppa-hpux is a secondary platform. | 03:22 |
| \sh | well | 03:26 |
| \sh | mvo: first of all I get this after autoreconf: | 03:26 |
| \sh | configure.ac:17: error: possibly undefined macro: AC_PROG_LIBTOOL | 03:26 |
| \sh | autoreconf: /usr/bin/autoconf failed with exit status: 1 | 03:26 |
| \sh | apt-get install autobook for more brain autotools magic | 03:27 |
| mvo | \sh: dar? | 03:27 |
| jbailey | \sh: apt-get install libtool? | 03:28 |
| doko | mvo: not explicitely, only references to the upstream release notes. Feel free to start one ;-) | 03:28 |
| \sh | mvo: yepp | 03:29 |
| \sh | jbailey: hmm...strange.. | 03:29 |
| mvo | doko: http://www.ubuntulinux.org/wiki/GCC4CommonProblems | 03:39 |
| \sh | hmmm | 03:40 |
| doko | mvo: ok, we need to link it ... | 03:41 |
| \sh | gettextize -f -c is asking me to hit return ;) | 03:41 |
| \sh | i should go home | 03:42 |
| lamont | \sh: that'd be bad for the autobuilder. :-( | 03:42 |
| doko | daily aspell orgy ... | 03:44 |
| mvo | doko: updated and linked | 03:47 |
| doko | mvo: thanks! | 03:49 |
| \sh | lamont: yeah | 03:49 |
| \sh | mvo: did u see the same behaviour? | 03:49 |
| \sh | i have to check again | 03:50 |
| mvo | \sh: about the return thing? yes. that's ok. just follow the instructions there and press ok. it needs to be done only once (and not on the buildd) | 03:51 |
| \sh | mvo: so on the buildd this behaviour won't happen? | 03:52 |
| mvo | \sh no | 03:52 |
| \sh | good :) | 03:53 |
| mvo | \sh: :) the really interessting bit is "aclocal -I m4; autoconf" | 03:53 |
| mvo | \sh: feel free to add you experience to the wiki page! | 03:54 |
| jbailey | mvo: I think that's usually set through ACLOCAL_FLAGS or some such. | 03:58 |
| jbailey | Upstream shouldn't trust that the user remembers to add -I m4 when running autoreconf. | 03:58 |
| mvo | jbailey: ok, thanks | 03:59 |
| \sh | mvo: autoreconf should call them all | 04:08 |
| \sh | hmmm... | 04:10 |
| \sh | i need a faster package mirror | 04:12 |
| \sh | de.archive.u.c is not fast enough with syncing ;) | 04:12 |
| elmo | use de.archive.u.uc as a fall back | 04:13 |
| jbailey | \sh: autoreconf seems to call autopoint now, which is a bit on the useless side. | 04:14 |
| \sh | jbailey: soi u mean calling aclocal and friends "by hand" is better then to call autoreconf? | 04:14 |
| jbailey | No, autoreconf is always the right thing. | 04:14 |
| jbailey | Just that in this case getting the gettext stuff updated seems a bit broken. | 04:15 |
| jbailey | Probably because of the risk of screwing up C apis or something. | 04:15 |
| doko | elmo: please could you install mysql-dfsg build-deps on halley/breezy? | 04:15 |
| elmo | doko: done | 04:18 |
| doko | thanks | 04:27 |
| daniels | just uploaded xorg -20 and I'm going to bed | 04:39 |
| daniels | this means it's sure to FTBFS | 04:39 |
| lamont | daniels: lol | 04:40 |
| doko | xorg built while daniels was sleeping ;) | 06:28 |
| === cartman [foobar@cartman.developer.konversation] has joined #ubuntu-toolchain | ||
| === cartman [foobar@cartman.developer.konversation] has joined #ubuntu-toolchain | ||
| lamont | IDLAny.cc:372: internal compiler error: in cp_expr_size, at cp/cp-objcp-common.c:101 | 08:51 |
| lamont | that's 3. I guess I'd best start working on that testcase soon... | 08:51 |
| jbailey | lamont: Three instances at least makes it sound like it shouldn't be that hard to reduce. | 09:01 |
| lamont | jbailey: exactly | 09:05 |
| lamont | #3 was libfwbuilder | 09:06 |
| doko | Failed to fetch http://ports.ubuntu.com/ubuntu-ports/dists/hoary/Release Unable to find expected entry main/binary-hppa/Packages in Meta-index file (malformed Release file?) | 09:25 |
| doko | Reading package lists... Done | 09:25 |
| doko | lamont: ^^^ | 09:25 |
| lamont | yeah. no hoary/hppa | 09:26 |
| lamont | hoary/hppa is at http://people.ubuntu.com/~lamont/ubuntu-hppa/tree iirc | 09:26 |
| doko | ok, ok, now to the breezy side of life ... | 09:26 |
| lamont | that's at ports.ubuntu.com/ubuntu-ports where it belongs | 09:28 |
| lamont | and is slowly populating | 09:28 |
| lamont | 37 packages queued up to sneak through the upload-straw | 09:28 |
| fabbione | bag | 09:49 |
| fabbione | bah | 09:49 |
| fabbione | i just trashed the gcc-3.4 build 3/4 of the way | 09:50 |
| fabbione | suckage | 09:50 |
| fabbione | doko: is it safe to start the build again with ./debian/rules build ? | 09:50 |
| fabbione | it was running the testsuite when i did stop it | 09:50 |
| doko | fabbione: yes, if stamps/*build-stamp exists. do you want to restart the check? | 09:51 |
| fabbione | doko: i basically did a killall make :) | 09:51 |
| fabbione | i didn't touch the source or the build dir | 09:51 |
| fabbione | so i am just re-running ./debian/rules build | 09:51 |
| doko | if you do not want to have tstresults, run WITHOUT_CHECK=yes debian/rules binary-arch | 09:52 |
| lamont | May 24 13:34:08 buildd-uploader: Setting to Uploaded(breezy): eel2 | 09:54 |
| lamont | progress! | 09:54 |
| lamont | fabbione: you aren't planning to upload that abused child, are you? | 09:54 |
| lamont | evil evil man. :-) | 09:54 |
| fabbione | lamont: i plan to do a debdiff before uploading :) | 09:55 |
| lamont | hehe | 09:55 |
| fabbione | 3 minutes to the meeting | 09:56 |
| jbailey | Thanks for the reminder. =) | 09:57 |
| jbailey | Hmm. This is just the maintainer candidates review, right? | 09:57 |
| doko | it was an "urgent" meeting ... | 09:58 |
| jbailey | Oh? Hmm. | 09:59 |
| doko | fabbione, lamont: please add speech-tools, festival and firefox to the list of libraries to build first (sparc and hppa) | 10:00 |
| fabbione | firefox? | 10:01 |
| fabbione | i only have mozilla-firefox banned as application | 10:02 |
| lamont | doko: actually the state of both architectures is that it's banned all cxxapps building... the rest just build as they come | 10:03 |
| === fabbione restarts gcc-3.4 build | ||
| fabbione | i didn't like some scary messages while building the debs :) | 10:17 |
| doko | which ones? | 10:19 |
| fabbione | doko: nothing to be worried about.. it was the interrupted build | 10:19 |
| jbailey | Doh. | 10:31 |
| jbailey | Helps if I remember that it's a c++ compiler. | 10:31 |
| jbailey | doko: Should be an easy fix for mysql. Removing variable called 'new' from asm-ia64/atomic.h =) | 10:31 |
| doko | heh, fine | 10:32 |
| === lamont wonders if anyone in particular has been babysitting the cxxlibs builds to get them fixed for xorg/dpkg pain | ||
| doko | hey man, doing that for the last 5 days ... | 10:48 |
| jbailey | lamont: I see that there's a parisc CVS update for the kernel. Is there a ChangeLog maintained for those patches? It would be nice to know when it's time to see if some glibc stuf is happier. | 10:49 |
| doko | jabiley: ask on parisc-linux.org | 10:51 |
| jbailey | doko: They're now on oftc, but yeah. | 10:52 |
| fabbione | doko: now i am 100% sure... gcc on sparc is more strict than on other arches | 10:52 |
| fabbione | and this sucks | 10:52 |
| fabbione | big times | 10:52 |
| fabbione | http://people.ubuntu.com/~lamont/buildLogs/g/gstreamer0.8/0.8.10-0ubuntu1/ | 10:52 |
| fabbione | cothreads.c: In function 'cothread_switch': | 10:53 |
| fabbione | cothreads.c:654: error: invalid lvalue in assignment | 10:53 |
| fabbione | make[5] : *** [libcothreads_la-cothreads.lo] Error 1 | 10:53 |
| fabbione | on sparc is FTBFS | 10:53 |
| fabbione | and that error is usually a macro on lvalu | 10:53 |
| fabbione | lvalue | 10:53 |
| doko | that's 3.4? | 10:53 |
| lamont | fabbione: nah - it's that strict on others too | 10:53 |
| fabbione | that's 4.0 | 10:54 |
| lamont | I have lots of those | 10:54 |
| fabbione | i have seen a few.. | 10:54 |
| fabbione | but why? | 10:54 |
| fabbione | given that is the same gcc and same code | 10:54 |
| lamont | because gcc-4.0 doesn't let you cast lvalues randomly | 10:54 |
| Kamion | that line of code is #ifndef HAVE_MAKECONTEXT | 10:55 |
| fabbione | hmmm | 10:55 |
| Kamion | I'm guessing sparc doesn't have that | 10:55 |
| Kamion | (or else the test is wrong) | 10:56 |
| lamont | doko: figured you were. was wondering if it was helpful for me to announce them as I trip over them, or if you're already just doing it and it's only channel-noise | 10:56 |
| fabbione | GST_ARCH_SETUP_STACK ((char *) cothread->sp); | 10:57 |
| jbailey | daniels: Hmm, the default path is now to /usr/bin/X11, but xbase-clients seems to still have things in /usr/X11R6/bin on ppc. =) | 10:57 |
| fabbione | that's what gcc doesn't like :/ | 10:57 |
| === fabbione hits seb128 with a sparc64 bat | ||
| jbailey | fabbione: Oo, did I miss something good? =) | 11:01 |
| jbailey | Oh, I see. This is gstreamer. | 11:01 |
| fabbione | jbailey: it's all l-k-h fault! | 11:02 |
| fabbione | i know that! | 11:02 |
| fabbione | you are hiding it very well :P | 11:02 |
| jbailey | fabbione: I try. =) | 11:03 |
| === jbailey waits for :03 to watch the lkh upload. | ||
| fabbione | it's already :04 :) | 11:04 |
| jbailey | Hmm. /me reruns ntpdate. | 11:04 |
| jbailey | Wow, 49 seconds off. | 11:04 |
| jbailey | Off an uptime of 3 days. =( | 11:05 |
| fabbione | Kamion: wow.. you did really a lot of magic in debootstrap :) | 11:05 |
| fabbione | Kamion: but really.. don't get too crazy about sparc, even if i really appreciate the extra work | 11:05 |
| Kamion | fabbione: that wasn't extra work | 11:06 |
| Kamion | fabbione: that was automatic :-) | 11:06 |
| fabbione | ahhhh | 11:06 |
| fabbione | ehheh | 11:06 |
| === Kamion hugs ./breezy-update | ||
| fabbione | i guess breezy will be a porting hell if Debian doesn't switch to gcc-4 soon | 11:11 |
| jbailey | Or if they change the transition plan somehow. | 11:12 |
| === fabbione ponders the idea of making only sparc server | ||
| fabbione | jbailey: can they? ;) | 11:12 |
| fabbione | let see... | 11:12 |
| fabbione | Kamion for sure will not vote for a different plan... | 11:13 |
| jbailey | fabbione: Depends if the release manager and ftp master decide that they hate Ubuntu that day... =) | 11:13 |
| fabbione | i know vorlon's wife.. that's like having vorlon's testicle in my hand | 11:13 |
| Kamion | it's been run past -release, no objections | 11:13 |
| doko | heh, the plan was discussed on -release ... | 11:13 |
| fabbione | jvw owns me a few tons of beer | 11:13 |
| fabbione | we can buy elmo | 11:13 |
| fabbione | do we need anybody else? :) | 11:14 |
| jbailey | Just doko, I guess. =) | 11:14 |
| === fabbione makes doko an offer he can't refuse | ||
| jbailey | I wonder how many packages in Debian will silently grow -#ubuntu# versions as lazy maintainers just upload the Ubuntu version. =) | 11:15 |
| doko | we should switch to HEAD in unstable, it can compile KDE ;-) | 11:15 |
| fabbione | jbailey: almost nobody.. | 11:15 |
| doko | heh, pike7.6 was'nt prepared for an "ubuntu" in the release part of the version | 11:15 |
| fabbione | i am pretty sure they are too proud of their pure versions | 11:15 |
| doko | hmm, but xorg are the same packages, aren't they? | 11:16 |
| fabbione | nope | 11:16 |
| fabbione | not that i know off at least | 11:16 |
| fabbione | had 0 time to look at xorg in debian | 11:16 |
| doko | oh no ... | 11:16 |
| Kamion | they're pretty similar | 11:17 |
| Kamion | they've been merging changes back and forward AFAIK? | 11:17 |
| fabbione | Kamion: yes | 11:17 |
| fabbione | but they are not the same | 11:17 |
| Kamion | and I was under the impression that they more or less imported Ubuntu packaging | 11:17 |
| Kamion | I think doko meant "same structure" | 11:17 |
| jbailey | fabbione: I thought of doing it with cdbs. =) | 11:17 |
| fabbione | Branden is correctly over paranoid about maintainer scripts and code cleanup | 11:17 |
| doko | gcc-4.0 (4.0.0-8) experimental; urgency=low | 11:18 |
| doko | * Synchronize with Ubuntu. | 11:18 |
| fabbione | Kamion: we are going modular... so there will be not much to share and i doubt daniels and Overfiend will agree on the same way of splitting or pkg names | 11:18 |
| doko | Kamion: same "-dev" packages would be very nice | 11:18 |
| fabbione | doko: well that's because you upload both of them? :P | 11:18 |
| fabbione | doko: i will try to play some mind jedi tricks to make that happen | 11:19 |
| fabbione | but i am not sure i can manage | 11:19 |
| fabbione | the 3 sides of the Force are all really strong | 11:19 |
| doko | yup | 11:20 |
| fabbione | Kamion: let's take over ubuntu and make it only a minimal server install! at least i will manage breezy with sparc :P | 11:21 |
| Kamion | fabbione: gravity seems to be the guy actually doing most of the xorg work in Debian? | 11:22 |
| doko | jbailey, the arch you wanted fix lkh for, did FTBFS | 11:24 |
| fabbione | Kamion: probably... i didn't even have the time to check out the tree | 11:25 |
| fabbione | i planned some debian work during the next weekend :) | 11:25 |
| fabbione | wife is away with scouts :) | 11:26 |
| jbailey | doko: And how do you manage to see these so quickly after they happen? =) | 11:27 |
| doko | well, I'm looking when I can safely upload next packages or poke lamont retrying some others, the read noise just hurd^Dts ;) | 11:29 |
| doko | Kamion: you did have a page with the status of installable/not installable packages. where? | 11:30 |
| Kamion | doko: http://people.ubuntu.com/~cjwatson/testing/breezy_probs.html | 11:30 |
| doko | thanks | 11:31 |
| doko | oops, the list got longer ... | 11:31 |
| lamont | Kamion: would it be terribly painful to do another such report containing the scc architectures? | 11:31 |
| Kamion | lamont: it's on my list :-) | 11:31 |
| lamont | or maybe one-per report - hppa will look really ugly for a while.. :-) | 11:31 |
| lamont | ENOGLIBC | 11:31 |
| Kamion | need to knock something up on rookery | 11:31 |
| lamont | Kamion: why is xorg 6.8.2-10 being used? -20 should be in the archive... | 11:32 |
| lamont | firecall | 11:32 |
| doko | lamont: plese retry swig1.3 when the fire is out | 11:33 |
| Kamion | lamont: being used for what? | 11:33 |
| Kamion | lamont: oh, stale binaries I imagine | 11:34 |
| doko | hmm, that's just KDE | 11:34 |
| Kamion | those are the latest versions of xlibmesa-glu-{dbg,dev} in breezy | 11:34 |
| === fabbione heads to bed | ||
| fabbione | good night | 11:41 |
| jbailey | g'n Fabio | 11:45 |
| === ajmitch [~ajmitch@port162-41.ubs.maxnet.co.nz] has joined #ubuntu-toolchain | ||
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!