IRC Logs for #crux-devel Friday, 2012-10-12

*** mike_k has joined #crux-devel02:16
*** mike_k has quit IRC04:41
*** horrorStruck has quit IRC05:27
*** frinnst has quit IRC05:41
*** pedja has quit IRC05:41
*** __mavrick61 has quit IRC05:41
*** jue has quit IRC05:41
*** Romster has quit IRC05:41
*** __mavrick61 has joined #crux-devel05:44
*** jue has joined #crux-devel05:44
*** pedja has joined #crux-devel05:44
*** Romster has joined #crux-devel05:44
*** frinnst has joined #crux-devel05:44
juecreated an initial version of our handbook for 2.8, did some s/2.7/2.8/ but not much more, would be nice if someone else could volunteer to fix/improve the handbook06:03
juesections 3.4 and 7.2 needs some fixes, probably06:08
juehmm, seems like we should remove 3.4 entirely?06:13
jaegerboth methods can probably be made to work with tweaking but I admit I don't have a need for either, so I've not tried06:15
jueok, I'd say it's less work if we remove it, if someone else comes up with working solutions it's sufficient to have that on a wiki page?06:21
jaegerThat's fine with me. :)06:24
jueok, 3.4 removed :)06:28
jaegerThe 2 bootstraps that I ran yesterday are done, I will make ISOs and upload them for testing today. Right now I'm getting ready for work but I will test the grub2 installation and update 7.206:31
juegreat06:32
jaegerI've already tested it with rc2 but didn't update the documentation, obviously :)06:43
pitillolet's see06:45
pitillosorry, wrong window :)06:45
jaeger50 DKP MINUS!06:47
pitilloDKP?06:49
*** horrorStruck has joined #crux-devel07:00
pitillois it related to Paragon levels? xD07:20
jaegerno :)07:25
jaegerDKP is an MMORPG system for item distribution07:25
jaegerIt was created back when everquest was the biggest MMORPG and 2 dragons were the big bosses07:26
jaegerDragon Kill Points (DKP)07:26
pitillojajaja07:29
pitillothat was the first thing I saw with a fast search :)07:29
jaegerah, ok :)07:29
pitilloI can rest DKP but no Paragon experience xD07:30
jaegerok, uploaded -rc3 to the pre-releases folder08:16
frinnsthah, you played everquest jaeger?08:17
frinnsti lost a couple of friends to that monstrosity :)08:18
horrorStruckteK__: just FTR, you'll need this for 2.8 for cyrus-sasl port: https://projects.archlinux.org/svntogit/packages.git/plain/trunk/0027_db5_support.patch?h=packages/cyrus-sasl08:24
jaegerfrinnst: yeah, I did... it was awesome way back in the day :)08:25
frinnstI have a waaay too addictive personality to even attempt to play a mmorpg08:26
frinnstit would consume me :)08:26
jaegerI'm able to step away from them so it's not as bad08:27
jaegerman, I love that mkfs.ext4 is smart enough to discard blocks when creating a new filesystem08:44
frinnstyet still insists on creating a lost+found09:00
frinnststupid asdfg screws up my rsync setyp09:00
horrorStrucksepen, not sure if you'll be reading logs but imlib2 1.4.4 fails to build with 2.8, 1.4.5 builds OK09:01
jaegerI have a question about the locale generation bit in the handbook09:02
jaegerin the second step, localedef -i sv_SE -f ISO-8859-1 sv_SE.ISO-8859-109:03
jaegerthe name (final arg) matches up with the charmap (-f)09:03
jaegerthe UTF-8 example does not09:03
jaegerlocaledef -i sv_SE -f UTF-8 sv_SE.utf809:03
jaegerutf8 vs UTF-809:03
jaegerwhy is that?09:03
frinnstlocales are witchcraft09:04
jaegerthe output from 'locale -a' is utf8 either way09:04
jaegeragreed on that, definitely09:04
jaegerfrinnst: core-x86_64 util-linux needs an update at some point09:08
jaeger(2.8)09:09
frinnstaye. Will look over it this weekend09:09
jaegerok :)09:09
frinnsti need to start exercise. too tired when i get home :)09:09
jaegerit definitely helps, I can say that from experience :)09:10
jaegerok, I've updated the GRUB instructions in the 2.8 appendix, if anyone sees a mistake or something please let me know09:53
juejaeger: thx, downloading rc3 now10:18
juejaeger: no idea about the locale stuff, but obviously (locale -a) it doesn't matter10:20
*** mike_k has joined #crux-devel10:22
juejaeger: but I'd change our handbook to use UTF-8, the debian man-page for localedef has an example for it -> http://man.he.net/man1/localedef10:35
jaegerjue: sounds good, let's change it10:42
pitillothe apendix seems fine, simple and easy with good references to grub's manual10:42
juedone10:44
jaegercool10:45
jaegerpitillo: thanks for looking10:45
pitillono no jaeger, thanks for all the impressive job all people is doing10:47
jaeger:)10:49
jaegerHrmm, need to figure out how to fix gl-select for 2.810:53
*** joe9 has quit IRC10:58
jueFTR, I had a subtile problem with it after the update to xorg-server 1.13:10:59
jueI did the update while I had selected nvidia, meaning libdri.so from xorg-server 1.12 was renamed to libdri_so at that moment11:00
jaegerI always 'gl-select use xorg' before xorg-server or nvidia updates, for what that's worth11:01
juebecause of that libdri_so was still installed after the update, which normaly would have removed libdri.so11:01
jueafter that I've selected xorg to test nouveau and got an old libdr.so back ...11:02
jaegerheh11:03
juejaeger: yeah, that wouldn't happened if I did that as well11:03
*** Romster has quit IRC11:06
juebut the biggest problem was that xorg-server 1.13 happily loads the old lib which causes issues with nouveau11:06
jaegercharming =/11:06
jue:)11:07
*** Romster has joined #crux-devel11:07
juelaptop updated to rc3, no problems11:25
jaegerexcellent11:25
*** mike_k has quit IRC11:25
jaegerRomster: http://jaeger.morpheus.net/linux/crux/files/wine-update.diff <-- if you're interested, a patch that updates wine and fixes the gecko/mono installs15:05
jaegerI'll try to finally fix python-32 today or tonight, too, shouldn't be too difficult15:47
*** frinnst has quit IRC17:41
*** frinnst has joined #crux-devel17:43
*** frinnst has joined #crux-devel17:43
*** Nomius has joined #crux-devel17:54
Romsterjaeger, i'm not sure why your looking at that old version i got wine at 1.5.14 and wine gecko at 1.7 granted it needs a version bump bit it's not as old as your diff is.19:47
Romsters/bit/but19:47
jaegerRomster: 2.819:48
Romsteryou haven't pulled in 2.7 ports to 2.8?19:48
jaegerI have pulled some but I don't merge them regularly19:48
Romsterk19:48
jaegerI basically haven't done any 2.7 development in the multilib trees since 2.8 is nearly done19:49
Romsteri have bigger problems atm like boost and stuff wont compile on 2.8 i686 not sure what i'm gonna do about all this yet looking though the easy ones to fix first.19:49
jaegerok19:49
jaegerI can fix wine if you want to work on other stuff19:49
Romsteri can bump that shortly19:50
jaegerI'm gonna try to fix python-32 now as well19:50
Romsterbut i'll need python-32 working19:50
jaegeraside from the version bump my diff added the mono msi to wine19:50
*** pedja has quit IRC19:50
Romsteruh didn't spot that before how did i miss that19:50
jaegerpython-32 will build with the --with-dbmliborder specified but I want to fix the weird include junk19:50
Romsteralso your mime type should be plan text not binary on diff files on your host.19:51
jaegerprobably so19:52
Romsterthere is a mono 0.0.8 i see, i'll look into it off the wine site wiki19:52
jaegerThere is but it wants 0.0.419:53
Romsteri guess there implementing the entire dot net stuff now.19:53
jaegerI checked the strings in wine for that19:53
Romsteryeah need to be careful what version needs what.19:53
jaegerat least the gecko page on their wiki is clear19:53
jaegerthe mono page is not, but 0.0.4 is what it wants19:54
Romsterhttp://wiki.winehq.org/Mono well that says 0.0.419:54
jaegertrue, but it lists wine "1.5.6 and later" so I checked to see if a later version was listed in the source19:55
jaegerer, 1.5.3 and later19:56
Romsterah i'll talk to a wine dev about clearing that up19:56
Romsterlooks unclear19:56
jaegerwith all that said, I tested the port with a fresh wineprefix and it worked, didn't prompt me for either gecko or mono19:56
Romsterseems odd if mono i ahead of wine dev19:57
Romsters/i/is19:57
Romsteri'll have a talk and see why that is so, but first i'm going out for lunch for a few mates then i'll at least bump wine and see what else i can get fixed for 2.819:58
Romsteri don't feel confident in letting 2.8 out until these other ports are not broken, or at least a warning wiki page on known broken ports on 2.819:58
jaegerWe've already delayed it quite a bit, so I'd like to get it released soon19:59
jaegerthough of course even after it's released users will take their time to migrate to it19:59
*** pedja has joined #crux-devel20:01
jaegeron a side note, changed the diff mime type20:07
Romsterhrm it's got a ton of broken ports though20:08
Romsterhow about a topic entry on known broken ports in #crux perhaps?20:21
Romsterseems i got a bit of time before i go out, starts fetching 2.820:22
Romsteroh yougot rc2 or rc3 on multilib jaeger  i got the rc2 is that good enough for me to use in a chroot20:22
jaegerIt's good enough. There are updates between rc2 and rc3 but nothing you can't build20:22
jaegerThere is an rc3, though, if you have to fetch it anyway20:24
Romsterk20:24
Romsteruse rc2 do a sysup i'd imagine be ok20:24
jaegerShould be, yeah20:24
jaegerok, testing my fixed python-3220:28
Romsteryou running 2.8 on your desktop yet?20:31
jaegerI'm running it in several different places but not my desktop20:31
jaegerVMs, laptop, etc.20:32
jaegerinteresting. looks like talloc-32 was missing python support before20:35
jaegerbecause with the fixed python-32 port it builds some new files20:35
Romsterhmm20:38
Romsterstill waiting on my chroot i's nearly done20:38
Romsterit's*20:39
Romsteralso i think need a new keyboard this is very old now.20:39
Romsteri just pulled the ports and opt-x86_64-multilib/wine is at 1.5.1420:41
Romsteri guess you did a merge since that wine diff.20:41
jaegernope20:41
jaegerI'm guessing you pulled 2.7, though, not 2.820:41
Romsteroh wait my prt-get.conf is missing opt-x86_64-multilib20:42
jaegerah20:42
Romsterthat one is in opt20:42
Romstershouldn't that be fixed on the prt-get.conf on the multilib iso?20:42
jaegerYeah, just haven't done that yet. It's been on the list for a while20:43
Romsteralso how do we deal with prefer higher since your not keen on that merged opt core contrib thing.20:43
jaegerIn my opinion we don't. Once 3.0 hits it won't matter so I'm disinclined to worry about it overmuch20:43
Romsternot have it set but if someone enables that i expect breakage at some point.20:43
Romsterhmm oh it wont probably matter then... as the fight is between i686 and x86_64 priority.20:44
Romsteryeah it'll fix itself.20:44
Romsteri jsut hate hidden known landmines.20:44
jaegerBesides, theoretically CRUX users should be advanced enough to see what's wrong and fix it :)20:44
Romsteryou cna use this but... watch out for x y z that can break it.20:45
jaegerOr at least report it so I can fix it20:45
Romstercan*20:45
Romstertrue20:45
jaegerpersonally I feel preferhigher is a landmine on its own20:45
Romsteroh actaully it probably is if oyu got personal repos of same name ports.20:45
jaegerMost likely the port maintainer a) intended the version that's available, or b) hasn't updated yet20:45
jaegerYeah, that's a factor as well20:45
Romsterbut then it can be done intentionally too.20:46
jaegerYes20:46
jaegerSo my personal opinion is that the port maintainer should be trusted over the version numbers20:46
Romstersafer option is the /usr/ports/foo:bar above other repos.20:46
jaegerI never use preferhigher, for what that's worth20:46
jaegeryeah20:46
Romsteri was but then i disabled that.20:47
Romsterbefore i ran into problems.20:47
Romsteryeah definitely fix that prt-get.conf for multilib, also i did manually edit finddeps and stuff to look at them extra multilib stuff paths and revdep for lib32 but i never made a patch for that.20:48
jaegerrevdep shouldn't need any changes20:49
Romsterhmm strange prt-get.conf does have core and opt multilib20:49
Romstermight of just been findeps it was a while ago i messed in them20:49
Romsterxorg-multilib is missing from that list though20:50
jaegerthat's probably a change you've made locally, then... prt-get isn't even in the multilib repos20:50
Romsteryeah i did it manually on my system20:51
Romsteri'm always bumping the reported wine version in #winehq i guess the others expect i'll do it all the time now.20:55
jaegerothers?20:55
Romsteroh obvious mistake i see opt-x86_64 but no opt-multilib that's why i'm seeing the wrong wine version20:57
Romsterthere is wine devs and a few moderators like man_in_shack and a few others that help out in the channel sometimes they used to set it before i even had a chance too. now one time it was like 4 days later that no one even changed the topic.20:58
Romsterso i've been doing it every second saturday whenever a new release is out.20:58
jaegerAh. Didn't know you were an op in there, didn't make sense without the context20:59
Romsterthere core-multilib opt-multilib and xorg-multilib added now i see the right wine.21:00
jaeger:)21:00
Romsteroh sorry i've been a chanop for a year or two now.21:00
Romsterusually only patch submitters get chanop privileges and very rarely at that. i got in by chance and a need at the time of trolls and that i was around alot more than i am now so i could remove offenders.21:01
Romsteralso seemed ot have ganed chanops in lvm and device-mapepr for being so helpful on data recovery.21:02
Romsteri spread myself too thinly :/21:02
jaegernice to be appreciated, eh?21:02
Romsteri guess so21:02
Romsterbut i really do far too much stuff.21:03
Romsterfull time work cramps my style too.21:03
Romsternot as creative, might be a good thing not coming up with new ideas21:03
jaegerhrmm... that sed in freeglut could be shortened21:08
jaegersed -i -e '/^SUBDIRS =/s/progs //' Makefile.in21:08
jaegerI've got a few updates to push for 2.8 now - glu-32, freeglut-32, mesa3d-32, python, python-32, talloc-3221:09
jaegerwine is building now, if it works properly I'll push them all21:09
Romstereh i've edited wine in my chroot already and downloading the sources21:10
Romsterbut if oyu done it already just push it i'm heading out to lunch21:10
jaegerI hadn't planned to push it but I will if you want to save the work21:10
jaegerI was going to push all the rest, though21:10
Romsteroh do that for freeglut-32 i'll shorten the one in opt later.21:11
jaegerok21:11
jaegerto be clear I was referring to wine rather than freeglut-32 with my push statement21:11
Romsteri was more concentrating on making stuff work than being optimized.21:12
Romsterah ok21:12
Romsterpush wine i'll fix freeglut-32 freeglut21:12
jaegerI will push all the ports I listed above but was going to avoid pushing wine since you were working on it; with that said, if you want me to push wine as well, I can. :)21:12
Romstersince you've already done the work.21:12
jaegerhope that clarifies21:12
Romsterpush wine :)21:13
jaegerI'll do freeglut-32 as well since it's already ready, if you don't mind21:13
jaegerok21:13
Romsterk21:13
Romsterit's pretty much a free for all atm in multilib do we need to define arch maintainers or happy with current arrangements, quite honestly it's silly having one maintainer per a port than a group.21:14
Romsterexcept for special ports that need alot of attention.21:14
jaegerMy assumption is that we'll move to the group method when it goes official so I haven't put much thought into it21:14
jaegersince we already do groups with the official stuff21:14
jaegerto some extent, at least21:15
Romstermultilib group?21:15
Romsteri'm pretty sure i could coax nogagplz intot hat group too since he is running multilib as well21:15
Romstersince he does all the emulator stuff.21:15
jaegerWill have to ask jue as well but I have no objections to that as long as everyone agrees on some standardized guidelines21:16
jaegerThe one I'm thinking of specifically right now is no duplication of includes to include32, for example21:16
Romsterfor evidence he is quite smart. i wouldn't recommend him if he wasn't clued up.21:16
Romsterthat's if he was keen on joining.21:16
jaegerright. no point if he's not interested21:16
Romsterthat was a hack it wouldn't work any other way for me.21:17
jaegerYou'll see how I fixed it soon, when I push these updates21:17
jaegerthe same can be done for libdsl-32, I'm sure21:17
Romstersome type def or something, if you can fix that to not need include32 go for it.21:17
jaegerseems to work fine for python-3221:17
jaegerfixed talloc-32, at least, and I've seen no build errors since21:17
Romsteri've been neglecting multilib a little on version bumps21:18
Romsterexcept that nvidia-32 i did.21:18
jaegerI've done a lot of them recently21:18
Romsterhave alot on my plate lately21:19
Romsteralso i got my work cut out for me in contrib21:19
Romstergstreamer ffmpeg newer boost ifi  can get that to work nice with delgue etc due to a long standing bug21:19
Romsterthen there is this compile issue on 4.7.221:20
jaeger2.7 won't disappear when 2.8 is released, anyway, so there's plenty of time to fix compile issues21:20
Romsterprobably not many that has as many dependencies as i have probably though21:21
Romsteryeah i'l be on 2.7.1 for a while yet21:21
Romsterok bbl21:22
jaegerok21:22
*** __mavrick61 has quit IRC21:52
*** __mavrick61 has joined #crux-devel21:53
jaegeropt/multilib changes are pushed to 2.822:41
jaegermesa3d-32 as well now22:57
*** prologic has quit IRC23:37

Generated by irclog2html.py 2.11.0 by Marius Gedminas - find it at mg.pov.lt!