IRC Logs for #crux-devel Wednesday, 2012-11-21

Romsteri could add that part myself jaeger01:01
Romsteri'll just partition it for now... after seeing this bug report
*** Romster has quit IRC01:57
*** Romster has joined #crux-devel02:17
*** Romster has joined #crux-devel02:17
RomsterMISSING   -rwxr-xr-x      root/root       usr/lib/purple-2/ssl-gnutls.so04:42
Romster=======> ERROR: Building '/var/ports/packages/pidgin#2.10.6-1.pkg.tar.gz' failed04:42
*** Romster has quit IRC06:20
*** Romster has joined #crux-devel06:29
*** Romster has joined #crux-devel06:29
horrorStruckI know it's highly-non-critical but the --with-x option we have in gcc is java specific06:50
Romsterjue, gkrellm doens't seem to build /usr/bin/ld: plugins.o: undefined reference to symbol 'g_module_error' misisng LDFLAGS link06:54
*** Romster has quit IRC07:04
*** Romster has joined #crux-devel07:19
*** Romster has joined #crux-devel07:19
*** rmull_ has joined #crux-devel07:29
*** pitillo_ has joined #crux-devel07:29
*** pitillo has quit IRC07:34
*** rmull has quit IRC07:34
prologicpkg-get and pkgsync are unmaintained these days right?07:35
prologicpkg-get seems to still work okay though07:35
jaegerI believe so07:35
prologicanyone actively use pkg-get?07:36
pitillo_prologic: we were using it on arm devices and I think (if I'm right) that sepen is maintaining it07:38
prologicI guess there's been no updates/releases sinces then?07:40
frinnstnothing major it seems, but still07:53
juejaeger: thanks08:54
jueRomster: well, it's on my ports-to-drop list ;)08:55
jueRomster: you are using gkrellm?08:56
jaegerjue: np08:56
frinnsttry conky instead09:00
jueyep, conky is nice09:03
jaegerAny of you use tiling WMs primarily?09:03
jueyeah, spectrwm here (was scrotwm formerly)09:04
jaegerever used i3? it looks kinda cool09:05
jueno, but indeed, it looks promising09:09
jueFYI, just started a TODO30 wiki-page ;)09:17
juemore or less empty, though09:18
frinnstoh, and that should be the letter "o"09:21
jaegerjue: making an i3 port if you want to try it when I get it done09:23
frinnstheh, i3 screencast narrated by a thick german accent insures you that you are in good hands :)09:26
frinnstanybody looked at the libpcre, curl updates?09:35
frinnstah, libpcre was only testing asdf09:36
jaegernot I09:36
frinnsti'll give the curl update a go09:37
jueyeah, go for it, works fine and passes all test for me09:38
juewrt 3.0: the only important port update I see currently is binutils (2.23.1), running it here since some days without issues09:57
jaegerany ABI or whatever breaking packages we deferred for 2.8 that we want to do in 3.0?09:57
juedon't know of any09:58
juejaeger: is compat-32 still the prefered name for the additional repo?10:02
jaegerI think so, matches the port names10:04
jueok, will try to create the repo ...10:04
jaegertry? :)10:05
jaegerjue: if you're interested, and i3ports.tar.xz.md510:10
juejaeger: yeah, thanks10:10
jaegerthere's a yajl port in my repo, I think the rest were already available10:11
jaegerjue: I haven't used it extensively yet but I installed it on one of my machines here and poked around a bit10:16
jaegerjue: oops, already see a mistake, I forgot to put "dmenu" in the i3 deps, so install dmenu :)10:16
jaegerIt's not required but it's used by i3 to launch other stuff10:17
jaegers/not required/not required to build/10:17
*** horrorSt1uck has joined #crux-devel10:18
jaegerjue: cloning works, I propose the master branch contain nothing but a README10:19
*** horrorStruck has quit IRC10:21
jueyep, I've added it to test10:24
juejaeger: writing should work for you10:26
*** horrorStruck has joined #crux-devel10:31
*** horrorSt1uck has quit IRC10:34
jueyep, also visible in our gitweb now :)10:49
jaegerspeaking of gitweb, maybe we should remove some old repos10:49
jaegeror at least remove them from the visibility list10:49
jaegermisc/udev-rules, ports/attic, etc.10:50
jaegeris tools/ports used?10:50
jueguess not10:50
jaegerWe could use it, certainly, though I guess ports doesn't need to change very often10:51
jueit's invisible for now ;)10:52
jaegerok :)10:52
jaegerjue: when do you think we should start setting up the 3.0 core/opt/xorg/contrib trees?10:59
jaegerI've got some family stuff going on in the next few days due to thanksgiving but can probably start this week11:00
juewell, let's do it very soon, in a first step core and opt should be sufficient, xorg is the same for x86_6411:06
jaegerI'm inclined to put glibc-32 into core since it should be installed even if nothing from compat-32 is11:07
juehmm, opt-x86_64 has nothing we need on our ISO11:08
juejaeger: no objections11:08
jaegerwon't matter in 3.0 anyway11:08
jaegeronce the repos are merged11:08
jaegerI can go ahead and push an initial set to compat-32 right now if that's cool11:10
juelooks like we need only a 3.0 branch for core in the first step11:10
jueyeah, that would be cool :)11:11
jaegeradded compat-32's 3.0 to the default_heads in gitweb11:15
juewow, that's a lot, to get wine running everything is needed?11:23
jaegermany of them are romster's, not mine. with that said, wine does have quite a few deps11:25
horrorStruckjaeger: i know you just created the branch but libpthread-stubs-0.2 should really die for 3.011:26
jaegeryes, it probably should11:26
jaegerthat was one of the things we talked about but deferred in 2.811:26
jaegerI'll set up a 3.0 build environment and work on purging that abomination this week/weekend11:27
jaegerI'll add that to TODO30 as well11:28
jaegerIs libpthread-stubs going to continue to be needed or is the upstream intention to get rid of it entirely?11:29
juenot needed anymore, version 0.3 is just a empty *.pc file11:30
jaegerright, I'm just wondering if the port itself can eventually be removed, rather than just updating it to 0.3 and fixing all the link issues11:30
juewe should remove the port, even if that throws some work to our users11:31
jaegerok. I'll work on that assumption, then, and see how it goes11:32
jaegerI wonder how many upstream things still check for it via pkgconfig11:32
horrorStruckjaeger: just curious, will transforming 3.0 into pure64 be as simple as removing glibc-32 and rebuilding the toolchain?11:35
jaegerpossibly, though I wouldn't do it in that order11:35
jaegerAll things considered I don't think it'd be worth the effort, honestly11:35
jaegerI'd just leave glibc-32 installed and never install anything else from compat-3211:35
horrorStruckok thanks11:36
jaegerIf you decide you really want to do it anyway, it'd be best to rebuild the whole toolchain to be sure11:37
horrorStrucki think i'll do it, i dont need multilib and this OCD thing will force me to do that :P11:38
horrorStrucktotally unrelated but your choice not build graphite for 2.8 was wise:
jaegergraphite is such a mess :(11:40
jaegerI think it could be cool if they would sort it out11:40
jaegerthat is one of the documents that made me avoid graphite, for reference11:40
jaegerAFK for a while, lunch and have to run some work errands11:41
horrorStruckah ok, i just discovered recently11:41
horrorStruckbon appetit11:41
horrorStruckjue: just FTR, i dont actually have OCD :P11:42
juehorrorStruck: :)11:43
horrorStruckjust curious, why do we have objc support in gcc?11:43
juePer wants it, and it doesn't hurt/costs something11:46
horrorStruckah ok. are you still in contact if i may ask?11:46
juenot really, our last mail was around 2 years ago11:47
horrorStruckok thanks, sorry for my curiousness :)11:48
juenp :)11:49
juehorrorStruck: btw, saw your bug-report at eudev11:50
juebut it's not usable currently11:51
jueat least not the version from yesterday11:51
horrorStrucki tried it in a vm 2 or 3 days ago, udevd died on me11:54
*** Romster has quit IRC11:54
juewell, was able to boot with it, even if I got a segfault of udevd; but second had a linker error while building xorg-server and lvm against it11:56
juefor reference I tried the LFS build stuff, works without issues11:58
horrorStruckyeah i got the same segfault and i stopped there11:59
horrorStruckudev-lfs you tried was 195?12:02
juehorrorStruck: port is here ->
horrorStruckthanks, i was doing one, i really need some sort of notification in my irc client12:24
horrorStruckare you using it ATM?12:26
juetested it on my laptop, but not with my desktop12:27
horrorStruckah they just released 19612:28
jueone point: udev-lfs doesn't build a libudev.a which we need for libdevmapper and lvm212:28
horrorStrucki think i'll stay we good ol' 182 for now, i'll try it in a vm however12:31
jueyeah, 196 made {usb,pci}utils obsolete, there's a own thing called 'hwdb'12:32
horrorStrucksystemd will soon provide its own kernel12:33
juenext they will integrate dbus and avahi ;)12:34
jueeven if it sounds so, at all it's not funny12:36
horrorStrucki joined #systemd briefly a few days ago, that's exactly what kay was talking about (well, i guess, in context, as i missed the beginning of the convo), for dbus at least.12:37
horrorStruckCRUX/kFreedBSD will be our salvation12:39
frinnstthey are looking to get dbus into the kernel12:47
frinnst(i am serious)12:47
horrorStruck(we know you are if the sentence doesn't end with *.gif)12:54
horrorStruckfor now, 1:54 am is the right time to wish you gentlemen a good night12:55
*** c0x has quit IRC13:04
*** c0x has joined #crux-devel13:07
*** horrorStruck has quit IRC14:46
*** horrorStruck has joined #crux-devel14:52
*** horrorStruck has quit IRC15:20
*** horrorStruck has joined #crux-devel15:26
*** Amnesia has quit IRC16:55
*** joe9 has quit IRC18:38
horrorStruckfrinnst: it looks like you could replace your lib64 -> lib sed lines in glibc (x86_64) with: echo "slibdir=/lib" >> configparms , before configure. I get a 100% matching .footprint20:09
horrorStruck(the ldd one is still needed however)20:12
*** _mavrick61 has quit IRC21:40
*** __mavrick61 has joined #crux-devel21:41

Generated by 2.11.0 by Marius Gedminas - find it at!