[12:09] elmo, any chance to move some more libs to main? [12:09] doko: I've been doing them? [12:09] as they come up [12:09] any ones in particular? [12:10] id3lib3.8.3, needed for fltk [12:10] did that a while ago [12:10] ok, thanks [12:10] back in a few hours. === ..[topic/#ubuntu-toolchain:doko] : GNU Compiler Collection, Glibc, Binutils, Linux-kernel-headers | GLIBC Todo: hppa, sparc NPTL, i386 biarch, C++ ABI change: 33/55 library packages in the archives [12:45] ok, going to sleep [12:47] elmo, Kamion: could you move libgcj-dev to main, it's a (not yet) build-dep to db4.2. as an alternative, java-gcj-compat would be ok. at least we do need the Java support for OOo2. [12:47] good night [12:48] night doko [12:54] one more upload .... === svenl_ [~luther@AStrasbourg-251-1-51-64.w82-126.abo.wanadoo.fr] has joined #ubuntu-toolchain === fabbione heads to bed too [01:18] doko: I think I probably technically have the ability to do it at the moment, but I'd rather leave universe->main promotion to elmo [02:06] moo [03:47] doko: libx11-dev is depended upon by xlibs-dev [03:48] jbailey: fonts *were* split out a while back [03:49] doko: the main thing we gain is that eventually we'll end up with lots of tiny packages, almost all of which never, ever change [03:49] so uploads to X packages are really small and easy to do [03:50] lamont: that's not a missing Depends on SDL, that's a missing -I/usr/X11R6/include, BTW === minghua is now known as minghua-away === minghua [~minghua@ppp-69-153-138-159.dsl.hstntx.swbell.net] has joined #ubuntu-toolchain [04:48] Hi, a c++ transition question: If a source package foo build libfoo1 and libfoo1-dev, I know I should rename libfoo1 to libfoo1c2, but do I need to rename libfoo1-dev to libfoo1c2-dev as well? [04:49] I think not, since the API is not changed [04:49] but just to make sure [04:49] nope [04:51] daniels: good to have confirmation, thanks [05:12] daniels: woot [05:24] r128_driver.c:98:29: extensions/dpms.h: No such file or directory [05:24] make[8] : *** [r128_driver.o] Error 1 === lamont grumbles at daniels [05:24] make[8] : Entering directory `/build/buildd/xorg-6.8.2/build-tree/xc/programs/Xserver/hw/xfree86/drivers/ati' [05:25] daniels: thoughts on making xorg compile on hppa? === daniels grunts; that should be fixed already. [05:26] that's in -156 [05:26] -15 [05:26] -16 will fix it [05:32] thank you. when will that be? [05:32] or can I get a -15.1 from you? [05:32] (doesn't need to be uploaded immediately, mind you...) [05:33] -16 will be by the end of today [05:33] more modularisation, merging from gravity [05:33] the easy fix is to: [05:33] fakeroot debian/rules clean setup [05:34] cd build-tree/xc/programs/Xserver/hw/xfree86/drivers/ati [05:34] cp r128_driver.c{,.orig} [05:34] vim r128_driver.c [05:34] %s/extensions\/dpms\.h/X11\/&/ [05:34] :wq [05:34] cd ../../../../../../ [05:34] diff -urN xc/programs/Xserver/hw/xfree86/drivers/ati/r128_driver.c{.orig,} > ../debian/patches/999_lamont_special.diff [05:34] cd .. [05:34] debuild -us -uc -b [05:42] woot [06:01] heh.. missed one ../ :-) [06:01] oops [06:09] np [06:17] morning [06:17] /usr/include/linux/pci.h:453: error: parse error before 'pci_power_t' [06:17] crap [06:18] Hi, I've got a problem for c++ transition of arpack++ [06:18] In configure I get: [06:18] checking how to run the C preprocessor... /lib/cpp [06:18] configure: error: C preprocessor "/lib/cpp" fails sanity check [06:18] See `config.log' for more details. [06:18] make: *** [build] Error 1 [06:18] and /lib/cpp is a symlink to /usr/bin/cpp-4.0 [06:18] nothing seems related in config.log, the last lines: [06:18] #ifdef __cplusplus [06:18] extern "C" void std::exit (int) throw (); using std::exit; [06:18] configure: exit 1 [06:19] I asked in #ubuntu-motu but got no reply. Any insights here? [06:19] ops [06:19] ECHAN [06:19] ops [06:19] no === lamont finds 1400x1050 to be a _strange_ resolution [06:20] jbailey: i really really need a good l-k-h [06:20] nice, but strange [06:21] xorg is FTBFS on sparc [06:21] r128_driver.c? [06:21] nope.. the above [06:21] l-k-h related [06:21] not sure if it will fail later [06:27] hrm... I got 5.5MB into the logfile [06:27] 3.3 here [06:27] i think [06:28] 5656 -rw-r--r-- 1 sparcbuildd sparcbuildd 5776501 May 19 01:50 xorg_6.8.2-15_20050518-2215 [06:28] but what gets builded changes quite a lot between arches [06:28] true [06:28] sparc doesn't build r128 anyway [06:29] i'm wondering why it wasn't ftbfs on i386, amd64 or powerpc, which all built r128 [06:39] where is jbailey? [06:39] given that it's about midnight his time.... [06:40] and he is not here? *BAD* === fabbione ponders hijacking l-k-h [06:43] hijacking is bad, mmkay? :P === lamont throws xorg back into the chompers [06:47] xorg_6.8.2-15hppa1 building [06:47] daniels: fix is already in -16-to-be, yes? [06:47] mmm-hmm [06:51] hm [06:53] well, the last failure was 2 hours in... I don't think I'm going to wait up for it [07:09] sure [07:09] night dude [07:09] night lamont [07:13] didn't say I was going to bed... [07:13] anything I can do to speed up graphics on a: 0000:01:00.0 VGA compatible controller: ATI Technologies Inc RV350 [Mobility Radeon 9600 M10] [07:13] ? [07:13] you can try the stuff from r300.sf.net [07:13] binary drivers? [07:13] or yeah, fglrx [07:14] daniels: I'll bug you more tomorrow on setting that up then [07:14] 0000:02:04.0 Ethernet controller: Atheros Communications, Inc. AR5212 802.11abg NIC (rev 01) === lamont wonders which restricted-package he needs for firmware [07:15] none for firmware [07:15] just l-r-m-$(uname -r) [07:15] right [07:15] where did you get this new toy? :) [07:15] where/when [07:16] liberated an nc6000 from my new bos [07:16] s [07:16] nice [07:16] so I can have it installed before I start work [07:17] oh i had one too [07:17] it was doing pretty well [07:19] daniels: on the first boot, it had lots of tearing on the screen, as well as some echos. restarting X made it all better === lamont found that somewhat concerning [07:20] um, cool [07:20] odes this happen reproducibly? [07:20] not since [07:20] but changing the resolution from 1400x1050 to 1280x1024 didn't make it go away either. [07:21] if you can reproduce it, let me know [07:21] will do [07:21] you *are* using hoary/breezy, right? [07:21] hoary+security [07:21] righto [07:21] warty is like, OLD [07:22] indeed [07:22] sid is like, older [07:22] (in terms of X) [07:22] yeah [07:27] ath0 doesn't find the accesspoint that my palm does. [07:27] grumble [07:27] you have brought the interface up, right? [07:29] uhuuhuh [07:29] configfs on /config type configfs (rw) [07:29] now i want you to guess what is that :) [07:30] daniels: how absolutely strange. I much prefer aironet cards... :) [07:34] hrm.. /me has gmail invites [07:37] is gmail any good? [07:38] dunno - don't use it. [07:38] but I do have an account [07:41] I think just about everyone (except fabbione, apparently) has an account and a mess of invites. [07:42] yeah [07:42] I check mine once every 2 or 3 months when someone reminds me about it. === lamont just logged in for the first time in a few months [07:42] feb 17 was last login [07:43] You are currently using 0 MB (0%) of your 2207 MB. [07:43] infinity: well it's like i don't really need it? :) [07:43] Where did you find the last login info? [07:44] from my one and only mail message (the intro one) [07:44] Ahh. :) [07:44] haven't logged in since I created it, until tonight. [07:44] at which point I deleted about 6 bounce messages [07:44] My last read message is Feb 17th, so I'm assuming that's my last login. [07:44] Creepy. [07:44] all for spam that I didn't send [07:44] And I have 50 invites. === lamont considers inviting his cat [07:46] lamont: so your cat has a email address already? ;-) [07:46] Why wouldn't it? [07:46] Cats need spam too. [07:50] 'zactly [07:54] lamont : So, when did/do you start back at HP? [07:54] June 1 [07:54] Looking forward to it? [07:54] yes [07:54] You've been institutionalised, haven't you? :) [07:55] not just because I'm getting paid to keep hacking on linux [07:55] A year without an HP ID badge must have been hell. [07:55] ahaha [07:55] infinity: it was the lack of health insurance that made it hell... [07:55] but yeah. I've been institutionalized [07:55] Ahh, that too. [07:55] I keep forgetting about the US and its lack of public health system. [07:55] pisser is that someone stole lamont@hp.com from me [07:55] You're kidding? [07:56] I hope thye enjoy all the spam it gets. [07:56] no. there were 3 other lamont's at HP. [07:56] one of them took it... bastard. [07:56] That was a pretty widely publicised email address. [07:56] Results 1 - 10 of about 846 for lamont@hp.com. (0.32 seconds) [07:57] not bad for something a year stale [07:58] well. anyway, time for bed I think. [07:58] cfdisk [07:58] Segmentation fault [07:58] interesting :) [07:58] lamont : 'Night. [07:58] fabbione : I blame ncurses. [07:59] infinity: possibly [07:59] night lamont === Seveas [~seveas@seveas.demon.nl] has joined #ubuntu-toolchain === chmj [~d3vic3@dumbledore.hbd.com] has joined #ubuntu-toolchain === cartman [foobar@cartman.developer.konversation] has joined #ubuntu-toolchain [09:46] daniels: up for a small X-on-breezy question? [09:46] the answer is 'don't' ;) [09:46] what's up/ [09:46] ? [09:46] :) [09:47] daniels: as nvidia driver will look for /usr/X11R6 and X is going to /usr [09:47] will be there symlinks? [09:47] not for a while though [09:47] from X11R6 to /usr ? [09:47] yeah, when xorg all hits /usr, there will be symlinks [09:47] but the server, Mesa, and drivers, will all be the last to go [09:47] daniels: cool thanks [09:47] since that's Tricky [09:47] yeah :) [09:47] I was worried :) [09:49] it shouldn't be too painful, hopefluly [09:49] gotta run now, though [09:49] see you [10:07] <\sh> morning [10:12] <\sh> something new about dpkg? [10:29] what is wrong with dpkg? === cartman [foobar@cartman.developer.konversation] has joined #ubuntu-toolchain [10:31] hmm X.org still compiled with old gcc ? [10:31] libGLU still links to libstdc++.so.5 here [10:34] good morning all [10:35] hi doko [10:35] morning [10:36] so the buildd's have stopped? [10:36] doko: hmm not that i know off [10:36] why? [10:37] on i386 the failed packages don't get a rebuild [10:37] flac, qt-x11-free, libsdl1.2, gtkmm2.4, ... [10:38] failed != dep-wait [10:38] there's nothing to dep-wait [10:38] dep-wait is kicked automatically [10:38] failed no [10:39] http://people.ubuntu.com/~lamont/buildLogs/g/gtkmm2.4/2.6.2-0ubuntu2/ [10:39] ? [10:40] After installing, the following source dependencies are still unsatisfied: [10:40] libglibmm-2.4-dev(inst 2.6.1-0ubuntu1 ! >= wanted 2.6.1-0ubuntu2) [10:40] Source-dependencies not satisfied; skipping gtkmm2.4 [10:40] but the dependency is already built [10:40] probably the chroot hasn't been updated [10:46] lamont, elmo: please could you check for the chroots, so that flac, qt-x11-free, libsdl1.2, gtkmm2.4 can be rebuilt? === cartman [foobar@cartman.developer.konversation] has joined #ubuntu-toolchain === chmj [~d3vic3@dumbledore.hbd.com] has joined #ubuntu-toolchain [11:40] elmo: some time for C++ ABI stuff (as long as lamont is not awake)? [11:44] what stuff? [11:45] fail to see why db4.2 is failing to build, but all build-deps are in main ... [11:45] wvstreams (main) needs some stuff from the xplc package, which is still in universe [11:46] doko: can you confirm we can build the new dpkg? [11:46] and that it was in the list by miskate? [11:46] (at least i remember this way) [11:47] fabbione: I don't have any problems with it now. yes, I removed it from the list at chinstrap:~doko/cxxapps.txt [11:47] doko: thanks [11:47] elmo: we can push it :) [11:48] qt-x11-free isn't tried to build on two archs [11:49] i386 buildds are turned off [11:49] I don't know why - did lamont say anything in here? === fabbione checks the irclogs [11:49] no, nothing [11:50] nothing in the logs [11:50] nope [11:51] meh, timezones are just teh suck [11:52] humpf [11:52] i wonder if he just forgot to turn them on again [11:53] i guess we will have to wait a bit [11:54] I'm going to have a shower, but once I'm back I'll have a look around, and if there's nothing obvious, turn one on again [11:54] :) [11:54] ok [11:54] i need food and i will be off soon today [11:54] take a cold shower ... === cartman [foobar@cartman.developer.konversation] has joined #ubuntu-toolchain === cartman [foobar@cartman.developer.konversation] has joined #ubuntu-toolchain [01:11] fabbione: Heya fabio - I still haven't seen a failure that shows that lkh is bad. The failures you showed me were all other things... [01:15] jbailey: xorg :) [01:16] Got a build log? =) === cartman [foobar@cartman.developer.konversation] has joined #ubuntu-toolchain [01:16] usual place [01:16] needsreallove/xoef_6.8.2-15 [01:16] /usr/include/linux/pci.h:453: error: parse error before 'pci_power_t' [01:16] The imap server takes me almost 10 minutes to get connected to. I tried arguing with the evo developers and they claim that they need to walk the entire tree in order to know the folder structure. =( [01:17] am I correct that Xorg is still compiled with g++3.3 ? [01:17] cartman: no [01:17] it's builded with gcc-4 [01:17] jbailey: you can ssh to vultus5 [01:17] fabbione: -15 still links to libstdc++.so.5 here [01:17] fabbione: need -16 or something? [01:17] less logs/xorg_ [01:17] Tx. [01:18] jbailey: imap and logs are strictly binded [01:18] 6.8.2-15 here [01:18] i was working with elmo to get the logs on people [01:18] A failure in linux/ should affect all arch's. I'm curious. [01:18] cartman: what is still linked to libstdc++.so.5? [01:18] Oh, probably a definition of __bitwise. [01:18] [~] > ldd /usr/X11R6/lib/libGLU.so|grep stdc++ [01:18] libstdc++.so.5 => /usr/lib/libstdc++.so.5 (0x00002aaaab0ca000) [01:18] fabbione: ^^ [01:19] ssh: connect to host vultus5.fabbione.net port 22: Network is unreachable [01:19] jbailey: you need to bounce via trider-g7 [01:20] or get ipv6 :) [01:20] cartman: check with daniels [01:20] @ WARNING: REMOTE HOST IDENTIFICATION HAS CHANGED! @ [01:20] joy === jbailey eyes fabbione suspiciously... [01:20] cartman: also.. what version of libglu did you install? [01:20] jbailey: what ssh command are you trying to use? [01:21] ssh trider-g7.fabbione.net [01:21] I think that's what I did before... [01:21] hmm no [01:21] i did open a port for you to do that [01:21] that was bouncing to vultus5 in ipv4 [01:21] that's why [01:21] Ah, non-standard port? Gimme a sec. [01:21] Actually, FWIW I could probably just enable my ipv6 stuff again. [01:21] if you can tell me the port again, that would be nice [01:22] jbailey: well it needs to be routed too :) [01:22] Yes, dear. =) [01:23] jbailey: do you still have my mail with the port handy? [01:23] Somewhere. Enabling by freenet6 daemon will be faster, though. I just need to move the config from my belle's box to mine. [01:23] doesn't freenet6 provide a /64? [01:24] You can either get them to provide a network or a host endpoint. [01:24] jbailey: ok hold on a sec [01:25] jbailey: ssh trider-g7 -p 1000 [01:26] that should redirect you to vultus5 [01:26] modulo slowliness [01:39] And this isn't biting other arch's? Only sparc? =( [01:39] Nothing about this line should be arch specific. [01:40] yeah only sparc [01:41] Is my breezy chroot still there? [01:41] jbailey: yes. it might need to be updated tho [01:41] or do you want me to recreate one for you? [01:42] (that's probably faster) [01:42] Please. [01:42] ok [01:42] At this point I'm at 2 or 3 seconds lag to a keystroke, so the least I have to type on this the better. [01:43] jbailey: yes i know [01:43] upload is the sux from here [01:43] hopefully within the next year we will get fibers === fabbione is looking forward to 20Mb MBR [01:44] =) [01:44] it's almost there :) [01:44] My new appt will have better bandwidth too. Right now if I upload, it cuts out the phone connection. [01:51] ok... gcc-4 is building on sparc now [01:59] jabailey: you want to make a gcc-4 upload today or tomorrow? [01:59] I did hear some rumors [01:59] elmo: just as a status, is the i386 buildd running? [02:01] doko: Well, you said that I could just upload a biarch-enabled gcc, I said that I would do it after the transition was done. Originally that sounded like Friday. =) [02:03] jbailey, well, yes, provided we don't have dpkg, xorg, and other uploads. I'll put a new tarball on chinstrap tomorrow, so you can work from this one? [02:04] how come linux-kernel-headers got demoted from ship to supported in today's CD build? did something stop depending on it? [02:04] Right. Nothing to interfere, just to make it so that I'm maintaining fewer glibc trees here. =) [02:04] Kamion: Err... [02:05] Kamion: Is libc6-dev usually in ship? [02:05] yes [02:05] build-essential is in ship [02:06] libc6-dev doesn't depend on l-k-h any more though [02:06] apt-cache rdepends on ppc seems to show it still. [02:06] What arch are you looking at? [02:06] Version: 2.3.5-1ubuntu2 [02:06] Architecture: powerpc [02:06] Depends: libc6 (= 2.3.5-1ubuntu2) [02:07] same on i386 and amd64 [02:07] looking in the archive on rookery with dpkg -I [02:07] Ah, bugger, yeah I see it too. [02:07] Lemme troll the logs to see why the perl script didn't run. =( [02:08] Kamion: Thanks for the headsup [02:08] np [02:09] Oh, I know what it will be, probably. [02:09] More linux/linux-gnu mapping fallout from the dpkg change. [02:09] ah, yes [02:10] +if ($DEB_HOST_GNU_SYSTEM eq "linux") { [02:10] that? [02:10] Ayup [02:11] Well, no. [02:11] That's where I'm setting it to linux-gnu internally. [02:11] even if I get linux - older versions of dpkg will make us behave consistantly. [02:12] morning [02:12] Err, sorry that looked like the debian/rules snippet for a sec. === lamont curses , visits the buildd [02:12] Perhaps I'll eat breakfast before attempting to hack more. [02:12] lamont, action now! ;-) [02:15] now gcc changed target to amd64-linux on amd64 [02:15] will this stop? :) === lamont finds that 1 of the i386 buildd's was offline. that's a far cry from "all" [02:18] Hmm. m/^i386-linux$/ could just as easily be m/^i386-linux/ with probably no sideeffects, right? [02:19] jbailey: should imagine so ... === lamont grumbles at daniels [02:22] amd64-linux is final target for amd64 gcc? ( it was x86_64-linux then x86_64-linux-gnu and now amd64-linux ) [02:22] From lwn: The usual fallback is to identify all the stakeholders and get them to say "yes Andrew, this code is cool and we can use it", but I don't think the clustering teams have sufficent act-togetherness to be able to do that. [02:22] Does this mean the clustering people are poorly clustered? [02:22] jbailey: haha [02:22] jbailey: most certainly === lamont is going to take a few hours of downtime this morning [02:24] starting shortly [02:25] mostly because I'm going to fall over [02:26] lamont: Can I have a glibc build log from you for hppa? [02:27] Or could you make them generally appear with the others in ~lamont? =) === Seveas [~seveas@seveas.demon.nl] has joined #ubuntu-toolchain [02:28] dude - I gotta get X to build [02:28] then I was going to build glibc [02:28] Ah? okay. I thought for some reason you had fired up glibc last night. [02:28] and making sparc/hppa appear is on the list of things that we're trying to figure out [02:28] no - I fired up X, and it's still FTBFS [02:29] must flee - will be back on in about 75-90 min or so === cartman [foobar@cartman.developer.konversation] has left #ubuntu-toolchain ["Ich] [02:29] lamont: Travel safely. =) [02:46] jbailey: you want me to run glibc now, and X later? [02:46] eep. [02:46] had hoped it wasn't :45 yet [02:48] lamont: could we have a look at some packages? [02:48] I'll start glibc and then xorg [02:48] doko: ?? [02:51] db4.2 [02:51] lamont: It doesn't matter that much. I just want to tell Carlos what the final failure set is. We're going to hack on generic function descriptors after that to solve a couple more of them. [02:51]