=== lamont grumbles at gcj-4.0. or maybe it's launchpad... [12:15] lamont: will be demoted to universe, if that helps [12:18] Demoting lamont? Say it's not so! [12:21] jbailey: sigh [12:21] "if that helps" =) [12:21] doko: It would mean less work for him to do. =) [12:22] doko: it's more that once gcj-4.0 builds on i386, then libgcj6 becomes uninstallable on any architecture that's not done building it yet [12:22] at least, every time you bump the versioned depend on libgcj-common [12:22] jbailey: just send you email about glibc on powerpc [12:22] that's actually more of an archive issue than a doko issue, but it's more fun to blame doko [12:22] lamont: good hint, will fix that [12:22] doko: Cool, thanks. [12:23] doko: the issue is that libgcj-common is arch: all [12:23] lamont: can you get some hppa guys to fix the java thing? [12:23] doko: Just on the phone with a customer. I haven't looked at your email yet, sorry. =( [12:23] and the archive stuff drops earlier versions of it, even though arch: foo still needs it, since there's a newer package of that name. [12:23] doko: I think someone is working on it, but dunno with what speed. [12:23] what exactly is the failure? [12:23] or issue [12:24] http://lists.parisc-linux.org/pipermail/parisc-linux/2006-February/028428.html [12:25] http://lists.parisc-linux.org/pipermail/parisc-linux/2006-February/028439.html [12:25] http://lists.parisc-linux.org/pipermail/parisc-linux/2006-March/028470.html [12:26] -Wl,-soname -Wl,liba1.so.0 === lamont giggles [12:27] it had email yesterday, so that's a good sign [12:29] doko: So it *doesn't* do it on Ubuntu now? [12:34] no, don't ask me why ... I did repeat both test-suites on the same machine (Power5) [12:35] lamont: what exactly is the problem with libgcj-common? there is no versioned dependency on it [12:43] doko: It can still be discrepancies between the glibc on Debian and the one in Ubuntu. [12:43] They're using gcc-4 in Debian, they have some extra patches and such. [12:43] Because of the Dapper lifecycle, I've tried to be *really* conservative. [12:43] The excitement starts for us the day after Dapper releases. [12:43] MWAHAHAHAHAHA [12:43] yes, seems to be a good decision ... [12:44] Well, ideally a week before so that all new things in 6.10 get build with gcc-4.1, glibc-2.4. [12:44] I argued with drow a bit about this, I didn't think it was a good decision for Debian either. [01:00] gcj-4.0-base Depends: libgcj-common (>= 4.0.2-8) bu t4.2.0-7ubuntu3 is to be installed [01:00] or something like that === lamont bootstraps gcj-4.0 again, to make life easier for infinity once the hppa buildd is existant [01:14] lamont: that's an unversioned dependency now, but honestly I didn't expect hppa that far behind ... [06:26] doko: hppa has been offline (and remains so) ever since the switch to soyuz. I'm hoping that we get buildd's online in the DC sometime in the next week or so. === fabbione [i=fabbione@gordian.fabbione.net] has joined #ubuntu-toolchain === doko [n=doko@dslb-084-059-115-131.pools.arcor-ip.net] has joined #ubuntu-toolchain [11:57] fabione, infinity: cmov is everywhere, even in hoary in sarge, looking further ... [12:50] doko: .. You'd think we would have had complaints before now if it was actually causing a problem... [12:59] infinity: we do have the kernel emulation for it turned on? [01:02] I have no idea if there even is kernel emulation for it. [01:02] But Sarge users (many of whom are custom-kernel-compiling types) would be screaming if Sarge didn't work on the old C3.. [01:02] So maybe we're not seeing what we think we're seeing here. === jbailey [n=jbailey@modemcable139.249-203-24.mc.videotron.ca] has joined #ubuntu-toolchain === doko [n=doko@dslb-084-059-069-105.pools.arcor-ip.net] has joined #ubuntu-toolchain