IRC Logs for #crux-devel Thursday, 2010-07-08

*** deus_ex has quit IRC00:06
*** sepen has joined #crux-devel03:51
sepenfrinnst, ping03:51
sepenhi guy, you should put more attention with opt changesets, so you --removed-- libnotify without asking other people for confirmation, and libnofity is required for more ports (in xfce and contrib collections), I would have taken if you had asked before doing anything04:18
sepenI think you removed it due to xulrunner changes, any other reason for dropping the port?04:20
frinnstno. that was the only reason it was added as well. i didnt find any other ports in opt/contrib that depended on it. my bad i guess04:21
sepenjust use $ prt-get dependent --all libnotify04:21
sepenxfce/xfce4-settings xfce/exo contrib/popper04:22
sepenanyways if we're a TEAM we should take some decissions as a TEAM, and more when we're in opt04:23
frinnstagreed. mea culpa04:24
sepenfrinnst, I want to adopt the port again, agree?04:29
pitillohello good morning, can someone confirm that 256Mb isn't enought to boot 2.7test1?04:30
sepenjust I think that opt is the best place to maintain this port (due to contrib and xfce requeriments on it)04:30
sepenpitillo, ram?04:30
pitillosepen: yes, I'm testing under vbox, started with 256MB(ram) and fb640, now making other tests04:32
sepenpitillo, I reported similar issues with 2.6 in the past. but can't remember how many ram it was04:32
pitilloyes, that's why I started with 256, trying to be sure (your report was 196MB if I remember well)04:32
sepenpitillo, try to remove from the initram's rc (maybe linuxrc?) some modules, it would save some memory04:34
pitillowell, seems to boot now (with 256MB and fb640) sorry for the noise04:34
pitillonah, si a mi me la suda bastante el resto04:35
sepenwell I'll adopt libnotify in opt if nobody is disagree04:38
pitillowin 404:47
pitillopfff sorry again04:47
jaegerhrmm... anyone else see a segfault on the -test1 iso boot process during the crux media search?08:30
jaegertime to build a debug iso08:35
jaegerah, it's the block device check08:56
jaegergrep -E '[sh]d' /proc/partitions | sed -e 's/.*\([sh]\)d/\1d/'08:56
jaegerthe grep is fine, the sed segfaults08:56
*** sepen has quit IRC10:13
*** deus_ex has joined #crux-devel10:40
jaegerhrmm... I don't have any idea how to figure out why busybox's sed would segfault. not much by way of debugging tools there11:10
jaegermight be this, testing:
jaeger-fno-tree-pta by itself doesn't fix the problem11:49
jaeger-fno-tree-pta -O0 does, testing -O0 by itself as well11:49
jaeger-O0 seems sufficient11:52
jaeger-fno-tree-pta causes sed to output garbage11:53
*** jue has joined #crux-devel11:53
jaegerdifferent gargabe every time it's called11:53
juejaeger: hi, thanks for -test1, will try it tomorrow11:54
jaegerheyo :) no problem11:54
jaegerI've already found a problem with it11:54
jaegerI've got a fix ready if you haven't downloaded it yet, though11:54
jaegerit's a small issue, though, won't break an install from CD, just other media11:55
juejust read the logs, you mean the busybox/sed issue?11:55
tilmancool, sepen created contrib.git/2.711:55
jaegertilman: nice11:55
jaegerI'll go ahead and overwrite the current test1 ISO if you guys have no objections11:56
tilmando it11:56
jaegerok, building it now, will be a quick upload11:57
jaegerok, new -test1 is uploaded with sed fix12:02
jaegerIf you guys have no problems with -test1 I'll push my setup, setup-helper, and Makefile patches12:03
jaegerusing as a temporary dump for the patches12:05
juedownloading now ...12:16
jaegerthe gcc bug seems fixed upstream but since 4.5.1 isn't released yet I've worked around it in busybox12:24
*** jue has quit IRC15:22
*** mavrick61 has quit IRC21:39
*** mavrick61 has joined #crux-devel21:40

Generated by 2.11.0 by Marius Gedminas - find it at!