| 00:12.11 | kergoth | hm, the update-alternatives bbclass should be using prerm, not postrm |
| 00:12.12 | kergoth | heh |
| 00:15.31 | *** join/#oe marcansoft (1337@80.26.222.222) |
| 00:21.11 | pb_ | kergoth: mm, yeah |
| 00:22.47 | kergoth | otherwise there's a period of time where your alternatives link isnt usable, as the binaries have been removed |
| 00:22.51 | kergoth | heh |
| 00:23.34 | pb_ | right |
| 00:26.33 | CIA-8 | 03kergoth 07 * r1.3063.2.1 10openembedded/ (3 files in 3 dirs): Fix the findutils build. |
| 00:27.24 | kergoth | [g2]: there. |
| 00:27.36 | [g2] | thx |
| 00:28.16 | *** join/#oe concept10 (~chatzilla@c-67-166-167-125.client.comcast.net) |
| 00:28.27 | kergoth | [g2]: no offense, but you guys need to learn to fix compile problems. |
| 00:28.40 | kergoth | i really cant remain your only source of auto* fixes. |
| 00:28.48 | [g2] | we do |
| 00:28.50 | emte | aww cmon |
| 00:28.54 | emte | :P |
| 00:28.58 | [g2] | we're slowly getting better |
| 00:29.05 | emte | cant you just complie it for me kergoth ? |
| 00:29.08 | emte | :P |
| 00:29.16 | [g2] | maybe you could have an autotools remedial workshop for us |
| 00:29.16 | kergoth | lazy fuckers! |
| 00:29.18 | kergoth | :P |
| 00:29.19 | emte | how goes it btw? |
| 00:29.46 | kergoth | job sitaution is still all screwy, havent made a decision |
| 00:29.49 | kergoth | other than that, pretty good |
| 00:30.42 | emte | ah |
| 00:30.44 | emte | hmm |
| 00:30.58 | emte | dont know of any off hand |
| 00:31.10 | kergoth | hehe. i just need to decide what i want to do |
| 00:31.12 | kergoth | the jobs are out there |
| 00:31.28 | kergoth | until i do, i'm in this odd state not really pursuing positions 100% |
| 00:31.48 | emte | yeah .. |
| 00:32.04 | kergoth | heh |
| 00:32.14 | kergoth | debating going back to minnesota and working for a friend of mine while i figure shit out |
| 00:32.24 | kergoth | or just do some contract or consulting work for a bit |
| 00:33.04 | treke | going back to minnesota is probably a wise idea |
| 00:33.06 | pb_ | heh, sounds like you need your boss to fire you so you can make up your mind. |
| 00:33.14 | treke | cut down on expensive while you try and figure things out |
| 00:33.18 | treke | expenses |
| 00:33.50 | kergoth | yeah. i mean, i still own the house there |
| 00:33.54 | kergoth | can stay there with my dad |
| 00:33.56 | kergoth | split the costs |
| 00:34.01 | treke | yeah |
| 00:34.08 | emte | well if you grab a couple of those military contarcts it'll carry you for a few mo |
| 00:34.20 | emte | contracts* |
| 00:34.31 | pb_ | working as a mercenary? |
| 00:34.35 | kergoth | hehe |
| 00:35.23 | emte | deve |
| 00:35.27 | emte | devel* |
| 00:35.28 | *** join/#oe nofpu (~matthew@82-45-183-10.cable.ubr06.dals.blueyonder.co.uk) |
| 00:35.40 | emte | especially if you have the clearance |
| 00:36.04 | emte | do a months work for 50-60 grand |
| 00:40.42 | [g2] | kergoth build completed |
| 00:40.45 | *** join/#oe bluelightning (~bluelight@203-173-244-15.akl.ihugultra.co.nz) |
| 00:40.54 | kergoth | good |
| 00:43.07 | kergoth | what the hell |
| 00:43.15 | kergoth | i lost my h320 usb cable and charager |
| 00:43.16 | kergoth | charger |
| 00:43.17 | kergoth | hmm |
| 00:56.38 | kergoth | damnit |
| 00:56.39 | kergoth | not there |
| 00:56.41 | kergoth | and i couldnt find it at work |
| 00:56.47 | kergoth | and i cant find it here |
| 00:56.47 | kergoth | wtf |
| 00:57.27 | emte | its in your car |
| 00:57.38 | kergoth | no, it really isnt |
| 00:57.50 | kergoth | gr |
| 01:00.21 | ken_ | I need a new compiling machine |
| 01:00.27 | ken_ | this laptop is way to slow |
| 01:02.59 | *** join/#oe HuntsMan (~hunts@pc-148-118-120-200.cm.vtr.net) |
| 01:10.49 | [g2] | kergoth, ever consider spliting ppl between OE users and OE devs at some point ? |
| 01:11.29 | kergoth | have thought about it, but people have enough trouble figuring out whats a distro issue and what they should ask in here |
| 01:11.36 | kergoth | much less breaking it down more than that |
| 01:11.37 | kergoth | heh |
| 01:13.02 | pb_ | yah, and the "devs" channel would have hardly anyone in it |
| 01:13.08 | kergoth | heh, indeed |
| 01:13.16 | kergoth | itd be like me, pb, and mickey, and a few others |
| 01:13.21 | kergoth | we can talk in here just as well |
| 01:14.49 | [g2] | well I'd suggest you do 1 of 2 things |
| 01:15.07 | [g2] | lower the expectations of the users all being devs |
| 01:15.16 | [g2] | start a dev mentoring program |
| 01:15.36 | [g2] | or continue to do what you do now which is dev by fire |
| 01:15.42 | kergoth | i've been trying to get more users to be package maintainers for a long time now. |
| 01:15.49 | kergoth | becoming a maintainer doesnt require development knowledge |
| 01:17.27 | [g2] | you are right and I thing a howto be a maintainer guide would go a long way to helping ppl get over the initial hurdle |
| 01:17.48 | [g2] | and work as a guide to bring ppl up to speed on the deeper issues |
| 01:18.13 | kergoth | an OE users manual is under way by some of hte new oe users. |
| 01:18.20 | kergoth | which will cover all sorts of aspects of hte system |
| 01:18.23 | kergoth | and class behavior |
| 01:18.25 | kergoth | see the wiki |
| 01:19.41 | [g2] | hey the wiki look bigger |
| 01:24.08 | *** join/#oe Cwiiis (~cwiiis@user-214-207-151-83.e7even.com) |
| 01:28.02 | [g2] | documentation scaring the devs away :) |
| 01:30.42 | kergoth | heh |
| 01:35.12 | *** join/#oe hufnus (~slonsiki@mbd8036d0.tmodns.net) |
| 01:37.05 | [g2] | kergoth, are many ppl doing distcc builds ? |
| 01:37.10 | kergoth | not many |
| 01:37.15 | kergoth | iirc florian was playing with it |
| 01:37.30 | kergoth | really, compile time is a small portion of the total build time nowadays |
| 01:37.34 | [g2] | has it worked out OK or are there issues ? |
| 01:37.38 | kergoth | so its questionable how much time it saves. |
| 01:37.46 | kergoth | more time is spent unpacking, patching, and configuring than compiling. |
| 01:37.50 | [g2] | so what is the big nut |
| 01:38.09 | kergoth | cpu isnt hte major bottleneck anymore. |
| 01:38.22 | kergoth | i'd be willing to bet hard disk performance is more of a bottleneck than cpu is, today |
| 01:38.27 | kergoth | (in oe) |
| 01:38.56 | [g2] | I've got the 36G |
| 01:39.29 | [g2] | why don't we cache the unpacked, patches builds ? |
| 01:39.54 | kergoth | ? |
| 01:40.07 | kergoth | first, some packages build inside the source tree |
| 01:40.11 | kergoth | second, heh |
| 01:40.17 | kergoth | no one has wanted it |
| 01:40.21 | kergoth | hasnt been requested. |
| 01:40.45 | [g2] | I'd image it would be pretty straight forward to do |
| 01:40.51 | [g2] | imagine |
| 01:40.56 | kergoth | not likely. |
| 01:41.03 | kergoth | not all buildsystems even _support_ building outside the source tree. |
| 01:41.30 | kergoth | plus, people bitch enough about the files in tmp/work/. people create classes to remove the content after build completion to save hard disk space |
| 01:41.40 | kergoth | heh |
| 01:42.44 | [g2] | maybe I'll try an experiment in a couple weeks |
| 01:43.35 | [g2] | I've also thought about make the tmp directory a tmpfs mount point |
| 01:43.49 | kergoth | i hope you mean /tmp |
| 01:43.51 | kergoth | not TMPDIR |
| 01:43.54 | kergoth | heh |
| 01:44.06 | [g2] | build/tmp |
| 01:44.12 | kergoth | are you on crack? |
| 01:44.16 | kergoth | a world build is over 20 gigs |
| 01:44.21 | kergoth | a normal one is 2-3 |
| 01:44.42 | [g2] | so I've got 1G |
| 01:44.52 | kergoth | which isnt nearly enough to do much of anything |
| 01:45.00 | kergoth | plus 1 gig is needed for bitbake to run itself |
| 01:45.01 | [g2] | it'd be easy to have a 5G swap space |
| 01:45.06 | kergoth | .. |
| 01:45.10 | kergoth | what? |
| 01:45.16 | kergoth | so you're running off of hd anyway |
| 01:45.20 | kergoth | but through additional layers of indirection |
| 01:45.22 | kergoth | making it even slower |
| 01:45.24 | kergoth | that makes sense. |
| 01:45.44 | [g2] | no the copy and unpack is in memory |
| 01:45.49 | [g2] | no I/O |
| 01:45.53 | [g2] | == fast |
| 01:45.58 | kergoth | ? |
| 01:46.05 | kergoth | if tmp/work is 3 gigs, that isnt going to be in memory. |
| 01:46.21 | kergoth | if you dont mean put build/tmp in tmpfs, dont say put build/tmp in tmpfs |
| 01:46.52 | kergoth | what you want, is to _only_ have the WORKDIR for the package currently being built to reside there |
| 01:46.58 | kergoth | which isnt what putting build/tmp in tmpfs would do. |
| 01:47.16 | kergoth | and yes, that might be worth playing with |
| 01:47.47 | [g2] | right that's what I want and I thought the kernel is smart enough to swap out the stuff that's not active at the time |
| 01:47.55 | [g2] | because it's such a serial process |
| 01:48.07 | kergoth | and tmp/cross would be active nonstop. |
| 01:48.09 | kergoth | 100% of hte time |
| 01:48.17 | kergoth | you really didnt give much thought to this yet, did you? :P |
| 01:48.51 | [g2] | the kernels is already caching lots of files |
| 01:49.02 | kergoth | ? |
| 01:49.49 | *** join/#oe Timelord0 (~TL@64-147-65-129.access-4.intertex.net) |
| 01:50.10 | [g2] | do you mean tmp/staging ? |
| 01:50.13 | kergoth | what you're talking about would result in huge amounts of churn, as the kernel swaps shit in and back out again, based on what in tmp is being accessed. putting the whole thing in tmpfs wouldnt really be very efficient |
| 01:50.20 | kergoth | i mean what i say. |
| 01:50.28 | kergoth | yes, staging is another area thats accessed continuously |
| 01:50.34 | kergoth | just like tmp/cross, that i just named 3 minutes ago |
| 01:50.59 | [g2] | well I wiped tmp and I'm rebuilding and cross isn't there yet |
| 01:51.03 | kergoth | ? |
| 01:51.05 | kergoth | yse, and? |
| 01:51.11 | kergoth | your toolchain hasnt built yet. |
| 01:51.17 | [g2] | right |
| 01:51.26 | kergoth | itd make much more sense to just: |
| 01:51.31 | [g2] | but toolchains are tiny after hey are built |
| 01:51.53 | kergoth | WORKDIR = "/tmp/work/${PN}-${PV}-${PR}" |
| 01:51.56 | kergoth | with /tmp as a tmpfs |
| 01:52.03 | kergoth | and inherit the bbclass that wipes work after build complete |
| 01:52.07 | kergoth | and call it done. |
| 01:53.09 | gb2 | bitbake sure does give a confusing error if you forget to have BBPATH set |
| 01:53.16 | kergoth | heh |
| 01:53.25 | kergoth | bitbake locates all configuration files and things through BBPATH |
| 01:53.30 | kergoth | the first thing it wants is bitbake.conf |
| 01:53.40 | kergoth | could add a check earlier on for it explicitly |
| 01:56.12 | [g2] | kergoth, how about this idea for fetching |
| 01:56.32 | [g2] | split off the fetching and feed it to a torrent client |
| 01:56.52 | kergoth | just add torrent:// support to the bb fetcher |
| 01:57.00 | gb2 | haha |
| 01:57.00 | kergoth | thats all you'd have to do. |
| 01:57.12 | kergoth | then you could use PREMIRRORS to point everything at a torrent source if need be |
| 01:57.14 | kergoth | heh |
| 01:57.33 | gb2 | torrents are generally the wrong answer for anything small |
| 01:57.37 | [g2] | I'm gonna setup a slug with local PREMIRRORS |
| 01:58.18 | gb2 | i was envisioning it torrenting a distro dvd image to get one source file out of it |
| 01:58.26 | kergoth | haha |
| 01:58.35 | [g2] | dual-layer |
| 02:00.35 | *** join/#oe hufnus (~slonsiki@md68636d0.tmodns.net) |
| 02:00.36 | *** join/#oe treke|home (~ggilbert@69-175-32-221.ventca.adelphia.net) |
| 02:00.48 | kergoth | hey treke |
| 02:00.54 | [g2] | THX kergoth |
| 02:01.05 | kergoth | whatd i do? |
| 02:01.12 | kergoth | heh |
| 02:01.40 | [g2] | what you often do, help make the community better |
| 02:01.40 | gb2 | does the .bb syntax have a += that doesn't insert a space? |
| 02:01.50 | [g2] | help get things done |
| 02:01.51 | kergoth | gb2: _append |
| 02:01.55 | kergoth | VAR_append = "value" |
| 02:01.56 | gb2 | ah, right |
| 02:02.09 | kergoth | i'd still like to see that die in favor of like VAR .= "blah" |
| 02:02.12 | kergoth | heh |
| 02:02.13 | gb2 | yeah |
| 02:02.29 | gb2 | i was remembering that, but it'd been so long and i wasn't sure if it was added so i was reading the docs :) |
| 02:02.34 | gb2 | and noticed nothing obvious |
| 02:02.36 | kergoth | its in the bitbake users manual |
| 02:02.43 | gb2 | .= isn't |
| 02:02.45 | kergoth | http://bitbake.berlios.de/manual/ |
| 02:02.51 | kergoth | .= doesnt exist. _append is in there |
| 02:02.53 | kergoth | heh |
| 02:03.29 | kergoth | i should do some oe toolchain cleanup |
| 02:03.29 | kergoth | pull in some updated patchsets |
| 02:03.29 | gb2 | i'd forgotten about the _append syntax which didn't stick out at me since there's a whole section on "Appending (+=) and Prepending (=+)" |
| 02:09.45 | *** join/#oe raster (~raster@YahooBB219198104088.bbtec.net) |
| 02:14.48 | [g2] | kergoth is there a way that I can get bb to stop after/before a certain stage ? |
| 02:15.02 | kergoth | ? |
| 02:15.06 | [g2] | like stop after configure |
| 02:15.11 | kergoth | ? |
| 02:15.14 | kergoth | bitbake -c configure blah |
| 02:15.16 | kergoth | ta da |
| 02:15.25 | kergoth | what are you trying to do? |
| 02:15.39 | [g2] | If I rm tmp |
| 02:15.51 | [g2] | and do a time bb -c configure openslug-image |
| 02:16.35 | [g2] | that would show me the time elapsed and system time to do the unpacking/pathcing/configuing |
| 02:16.46 | kergoth | eh? |
| 02:16.52 | kergoth | you cant configure if your dependencies arent staged. |
| 02:16.54 | kergoth | configure will fail. |
| 02:17.20 | [g2] | Ok so - patch |
| 02:17.32 | kergoth | yes, that would work fine |
| 02:17.38 | [g2] | that'll show the unpacking/patching |
| 02:17.40 | kergoth | people use -c all the time to prefetch sources at work and bring them home, and the like |
| 02:17.47 | kergoth | why dont you just do what i do? |
| 02:17.54 | kergoth | instrument oe |
| 02:18.04 | kergoth | oe has event handlers that fire at each task start and completion and failure |
| 02:18.07 | [g2] | that's what I'm talking about |
| 02:18.11 | kergoth | you can dump time metrics at those to a log |
| 02:18.23 | kergoth | i have a bbclass or .inc around here somewhere |
| 02:18.36 | kergoth | just a modification of the current event handler to add the time/date to each message |
| 02:18.43 | kergoth | the current event handlers already support logging to a file |
| 02:18.45 | kergoth | if you enable it |
| 02:18.50 | kergoth | read base.bbclass |
| 02:19.17 | [g2] | I'd like to dump sql records to a db |
| 02:19.20 | [g2] | for import |
| 02:19.34 | kergoth | feel free. use pysqlite in an eventhandler. |
| 02:19.36 | kergoth | heh |
| 02:20.45 | [g2] | that'd be interesting |
| 02:21.27 | [g2] | the 340 bb files and 50 packages used to build openslug broken down by time and stage |
| 02:22.14 | *** join/#oe _drak0_ (~rob@ip68-98-148-202.dc.dc.cox.net) |
| 02:41.39 | gb2 | kergoth: did you break gcc-cross building with uclibc? |
| 02:43.00 | gb2 | oh, i may have been building against a very old tree when it worked yesterday, hmm. wonder how long this has been broken |
| 02:43.43 | kergoth | no, i build with uclibc every day |
| 02:43.59 | gb2 | .../codecvt_specializations.h:52: error: `iconv_t' does not name a type |
| 02:44.07 | kergoth | never seen that. |
| 02:44.21 | kergoth | oh, are you using uclibc's iconv? |
| 02:44.27 | kergoth | i've not done that in a while |
| 02:44.33 | gb2 | don't know :) |
| 02:44.35 | kergoth | i generally do USE_NLS_uclibc = "no" |
| 02:44.48 | kergoth | give that a shot |
| 02:44.58 | kergoth | give that a shot |
| 02:45.01 | kergoth | bah |
| 02:45.05 | gb2 | ok |
| 02:45.07 | gb2 | :) |
| 02:45.15 | kergoth | :) |
| 02:45.35 | treke|home | hmmm. joshua's dad died :( |
| 02:46.13 | kergoth | doh :( |
| 02:47.19 | treke|home | yeah |
| 02:49.58 | CIA-8 | 03kergoth 07 * r1.3074 10openembedded/ (53 files in 3 dirs): (log message trimmed) |
| 02:49.58 | CIA-8 | Gcc 3.4.0, 3.4.1, 3.4.2: |
| 02:49.58 | CIA-8 | <PROTECTED> |
| 02:49.58 | CIA-8 | Gcc 3.4.3: |
| 02:49.58 | CIA-8 | * Adjust the gcc build to patch away the STMP_FIXINC/STMP_FIXPERM screwup when |
| 02:49.58 | CIA-8 | not setting --with-sysroot, and stop passing --with-sysroot .. setting it was |
| 02:50.01 | CIA-8 | resulting in our g++ binary adding -L/lib and -L /usr/lib to its link lines, |
| 02:51.21 | kergoth | ~botmail for pb_: killed off the old gcc 3.4.x versions.. however, there was no gcc-cross-s3.4.x at all. it'd be good if you'd create one for 3.4.3, using the .inc's, which shoudl ease maintainance. |
| 02:51.31 | kergoth | <PROTECTED> |
| 02:51.37 | kergoth | wtf |
| 02:51.45 | kergoth | my windows box is possessed or something |
| 02:51.53 | gb2 | that looks rather garbled |
| 02:52.03 | emte | its windows ... its possesed by billy |
| 02:52.15 | kergoth | yeah, it swallowed some of my words, and douplicated others |
| 02:52.15 | emte | sorry bad pun |
| 02:52.24 | kergoth | i think i overran windows keyboard buffer |
| 02:52.25 | kergoth | heh |
| 02:52.46 | emte | printed fine for me |
| 02:52.58 | kergoth | ~botmail for pb_: that is, that last botmail was referring to gcc-cross-sdk. |
| 02:53.03 | kergoth | heh |
| 02:53.16 | kergoth | it shoudlve been 'there was no gcc-cross-sdk 3.4.x' |
| 02:53.22 | kergoth | close enough i guess |
| 02:53.25 | kergoth | stupid windows |
| 02:53.39 | emte | needs a better frame :P |
| 02:55.17 | gb2 | kergoth: still keeping two 3.3.x versions though? |
| 02:55.54 | kergoth | gb2: need to look at those again. oe 3.3.x hasnt been touched in quite a while |
| 02:55.59 | kergoth | patchset needs updating |
| 02:56.04 | gb2 | ah |
| 02:56.11 | kergoth | will get it down to one version there too |
| 02:56.14 | gb2 | what's the -sdk package for? |
| 02:56.20 | kergoth | i think one .bb per major branch is sufficient |
| 02:56.41 | kergoth | creates an sdk |
| 02:56.56 | kergoth | packages up the cross toolchain, libs & headers from other packages, etc |
| 02:57.04 | gb2 | ah, ok. |
| 02:57.04 | kergoth | (the -sdk stuff in general does, that is) |
| 02:57.14 | kergoth | pb_'s work, i never messed with it |
| 02:57.27 | gb2 | dev from cross rather than dev from target build? |
| 02:57.42 | kergoth | yeah, sdk is cross stuff |
| 02:57.52 | kergoth | native dev on the target works nowadays btw |
| 02:57.55 | kergoth | out of hte box |
| 02:58.02 | kergoth | including autoconf/automake/libtool |
| 02:58.09 | gb2 | cool. |
| 02:59.02 | kergoth | gb2: so haventseen you around here in a while. how you been? |
| 02:59.09 | gb2 | i've been good |
| 02:59.15 | gb2 | really busy though |
| 03:00.15 | gb2 | getting some new basestation RF hardware working mostly which is kinda draining work since it's pretty repetative debugging |
| 03:00.26 | gb2 | so i haven't had the will to do my independant hacking projects much |
| 03:00.34 | kergoth | know how that is |
| 03:00.51 | kergoth | my job is repetative driver coding. churning them out day after day. sam problem |
| 03:00.54 | kergoth | same, even |
| 03:01.17 | gb2 | yeah, i've got to do a new driver in the next two weeks or so |
| 03:01.29 | gb2 | for poorly documented hardware ofcourse |
| 03:02.33 | gb2 | the RF hardware is kinda fun once it's finally near working, since it's satisfying to run ./start_bss.sh, and try to make a test phone call from two cell phones at my desk :) |
| 03:05.59 | gb2 | I did use some of my time to learn python well, though that was doing a gui, so it got old pretty quick |
| 03:13.17 | gb2 | kergoth: you done much with gcc4 yet? |
| 03:14.24 | joshua_ | darncool. the new nvidia drivers support XRandR. |
| 03:14.54 | gb2 | joshua: rotation acceleration even? |
| 03:14.55 | joshua_ | of course, neither my mouse nor my stylus manage to rotate with the display... |
| 03:14.59 | joshua_ | I think accelerated. |
| 03:15.40 | *** join/#oe raster (~raster@YahooBB219198104088.bbtec.net) |
| 03:16.24 | joshua_ | framerate takes a dump, but not software rendering level dump. |
| 03:17.06 | joshua_ | from 1100fps to 230fps |
| 03:17.39 | gb2 | that's a huge drop |
| 03:17.47 | gb2 | 2D seems fast though? |
| 03:17.53 | gb2 | try dragging a window around |
| 03:18.07 | joshua_ | trying |
| 03:18.47 | joshua_ | hmm |
| 03:18.52 | joshua_ | it becomes rectally slow when I do so. |
| 03:19.06 | emte | been told the pre-catalyst driver set works bets for most cards |
| 03:19.12 | emte | best* |
| 03:19.20 | joshua_ | heh, isn't catalyst ATI? |
| 03:19.22 | emte | er no catalyst is ati |
| 03:19.53 | emte | the set before nVidia started their all-in-one set |
| 03:20.18 | joshua_ | that has to be really ancient |
| 03:20.24 | emte | yeah it is |
| 03:20.42 | emte | sounds like the fps is almost double though |
| 03:20.58 | joshua_ | I wonder if it even supports the gf4go |
| 03:21.01 | emte | not sure why, havent tried it myself |
| 03:21.20 | emte | i dont really like my nVidia card |
| 03:21.44 | emte | so its just has the normal nVidia driver installed |
| 03:21.59 | joshua_ | heh |
| 03:22.13 | emte | just doesnt perform very well |
| 03:22.34 | emte | wish my new radeon would show up |
| 03:22.40 | [g2] | joshua_, which nvidia drivers ? |
| 03:22.46 | joshua_ | 71xx |
| 03:22.55 | emte | using this trident as a temp replacement bites |
| 03:22.59 | joshua_ | heh |
| 03:23.12 | emte | cooked my other radeon |
| 03:23.31 | emte | 5min of dri with a locked up fan sorta does that i guess |
| 03:24.10 | joshua_ | heh. |
| 03:24.43 | emte | did buy a zalman dual HP for the new card though |
| 03:24.50 | emte | no more bad fans |
| 03:24.56 | joshua_ | heh. |
| 03:25.36 | [g2] | no but those new DP via's look nice |
| 03:25.38 | joshua_ | I haev a gf3ti200 that's been running for awhile now with no fan... it was dirty, I cleaned it, it popped off, the card hasn't compmlained... |
| 03:25.50 | joshua_ | (complete with 3d accel and all..) |
| 03:26.50 | joshua_ | when I went dualhead, I was going to replace that with a radeon something or other... DRI didn't support my card, and I coudln't wrestle fglrx into actually working for more than a few minutes. |
| 03:29.10 | emte | yeah my replacement is a dual 9600XT |
| 03:32.41 | *** join/#oe hufnus (~slonsiki@m288536d0.tmodns.net) |
| 04:22.08 | gb2 | ER renewed till 2008 |
| 04:27.16 | treke|home | good god how long has it been running? |
| 04:41.44 | *** join/#oe micropal (ident@139.80-202-82.nextgentel.com) |
| 04:48.35 | gb2 | oh, i meant to say that on another network, heh |
| 04:51.10 | kergoth | heh |
| 04:51.36 | kergoth | hmm, speaking of networks, i need to getm y proxy connecting to all the networks i'm on |
| 04:51.39 | kergoth | heh |
| 04:52.06 | gb2 | kergoth: played with gcc4 much? |
| 04:52.29 | *** join/#oe offroadgeek (~michael@c-24-118-57-32.mn.client2.attbi.com) |
| 04:52.37 | kergoth | not a whole lot. tried a couple snaps, but havnet had time to really play with it extensively |
| 04:52.41 | kergoth | seems promising |
| 04:53.25 | gb2 | yeah. |
| 04:54.04 | gb2 | that's kinda what this cheap dell box i bought is for |
| 04:54.23 | gb2 | i'm planning to put fedora-devel on it, which is now built all with a gcc4 snapshot |
| 04:55.06 | kergoth | ah, neat |
| 04:55.17 | chouimat|Zzz | night |
| 04:55.22 | kergoth | should give the gcc-snapshot build in oe some TLC and play with it |
| 04:55.44 | chouimat|Zzz | anyone played with xen? |
| 04:55.52 | gb2 | chouimat: yeah |
| 04:56.03 | gb2 | fedora-devel also has that |
| 04:56.13 | chouimat|Zzz | gb2: worth the installation? |
| 04:56.22 | gb2 | kergoth: their plan is to hopefully have gcc4 out in april, fc4 is targetted for a may release |
| 04:56.27 | kergoth | ah, nice |
| 04:56.43 | gb2 | chouimat: it works pretty well, there are a few quirks still, but it's a great way to have parallel installs of multiple distros |
| 04:56.54 | chouimat|Zzz | gb2: ok |
| 04:58.00 | gb2 | chouimat: it's got great performance, NPTL requires some fixups because of how thread local storage works, so that runs with a performance degradation. they're working on fixes for that though. |
| 04:58.12 | kergoth | guess it depends what your goals are. to a certain extent you can get a number of distros running happily just with uml, as long as you have disk images handy |
| 04:58.14 | gb2 | it's otherwise pretty close to native speed |
| 04:58.40 | chouimat|Zzz | ok thanks I was planning to give a try and install netbsd and freebsd too =0 |
| 04:58.56 | gb2 | kergoth: yeah, xen is a good way of doing things at higher performance though. |
| 04:59.33 | gb2 | chouimat: yeah, that too. UML isn't very good for running netbsd ;) |
| 04:59.55 | kergoth | hehe |
| 05:00.28 | chouimat|Zzz | gb2: yup |
| 05:00.50 | gb2 | kergoth: I'm hoping to split my machine into two xen VMs for general use, granting the PCI network devices into one of them, everything else in the other, and using one of the xens for all my network firewalling in a single machine. |
| 05:01.01 | gb2 | doing some security separation that way. |
| 05:01.09 | kergoth | ah, thats a cool use |
| 05:02.28 | chouimat|Zzz | but I will have to rebuild this box to install xen :( |
| 05:02.49 | gb2 | kergoth: it also has some interesting applications for doing hardware driver development. |
| 05:03.16 | kergoth | work on them without risk to the stability of your main system, i suppose |
| 05:03.27 | gb2 | chouimat: that's why I bought this $229 dell box :) |
| 05:03.36 | kergoth | to a certain extent, anyway |
| 05:03.37 | kergoth | heh |
| 05:03.40 | kergoth | $229? nice |
| 05:03.44 | chouimat|Zzz | gb2: this one is using Reiser4 |
| 05:04.10 | gb2 | kergoth: you can do some cross-VM debugging and crash dump analysis |
| 05:04.33 | gb2 | kergoth: if you tell your device to lock up the pci bus, you're still screwed though. |
| 05:04.35 | kergoth | nod. the same sorts of thing you end up doing for driver development with 2 achines. |
| 05:04.46 | kergoth | right |
| 05:04.59 | kergoth | hence 'to a certain extent' |
| 05:05.00 | kergoth | :) |
| 05:05.06 | kergoth | this world build is taking ages |
| 05:05.21 | kergoth | ~/code/openembedded$ ls ../build-x86/tmp/stamps/*.do_package|wc -l |
| 05:05.22 | kergoth | 902 |
| 05:05.24 | kergoth | ~/code/openembedded$ ls ../build-x86/tmp/deploy/ipk/|wc -l |
| 05:05.25 | kergoth | 5035 |
| 05:06.07 | gb2 | kergoth: i've done a bunch of serial console debugging and more recently netconsole stuff, but it would still be nicer to have more access to what's going on in the system. I should try kdb again. and perhaps the netconsole based kgdb |
| 05:06.31 | kergoth | how is netconsole? i've never gotten around to playing with it |
| 05:06.47 | gb2 | i've successfully captured good oops stack traces using it |
| 05:06.55 | kergoth | nifty |
| 05:07.05 | gb2 | i have machines like my laptop which have no serial port |
| 05:07.15 | gb2 | since that's the more normal way of capturing them |
| 05:20.36 | *** join/#oe hufnus (~slonsiki@m243c36d0.tmodns.net) |
| 05:47.06 | *** join/#oe hufnus (~slonsiki@mc18c36d0.tmodns.net) |
| 05:58.24 | emte | alright, after three days i give up ... how do i override the version of automake it tries to build? |
| 05:59.04 | emte | i've tried putting PREFERRED_VERSION_automake-native = "1.9.3" in every spot i can think of |
| 05:59.23 | emte | and it still builds 1.8.4 |
| 05:59.54 | kergoth | heh |
| 06:00.01 | kergoth | 1.9.3 is default in upstream oe today |
| 06:00.11 | kergoth | but yes, that preferred versin line looks fine |
| 06:00.37 | emte | but i cant get it to build |
| 06:00.51 | emte | maybe i should try an update |
| 06:00.59 | emte | cant hurt i guess |
| 06:01.38 | emte | i am hoping the best place for it is in the distro file? |
| 06:01.40 | kergoth | the automake-native 1.9.3 bb exists, right? |
| 06:01.44 | emte | yup |
| 06:01.46 | kergoth | distro or local |
| 06:01.49 | kergoth | yeah |
| 06:03.01 | emte | its strange since my autoconf preferred works fine |
| 06:03.41 | kergoth | see why it thinks it cant build it, perhaps?e why it thinks it cant build it, perhaps? |
| 06:03.46 | kergoth | goddamnit windows |
| 06:04.36 | emte | its not taht it cant build it ... its the fact its not using tehright version |
| 06:05.16 | kergoth | bbread|grep PREFERRED |
| 06:05.25 | kergoth | confirm things are what you think they are |
| 06:05.27 | kergoth | sanity check |
| 06:05.42 | emte | k |
| 06:07.34 | emte | hmm looks liek its not getting it from my distro file ... |
| 06:07.42 | emte | time to figure out why |
| 06:07.47 | kergoth | :) |
| 06:08.02 | kergoth | when things arent doing what you expect, always check your assumptions |
| 06:13.42 | *** join/#oe zap (~zap@217.170.93.196) |
| 06:19.50 | emte | hmm |
| 06:20.06 | emte | its not getting the info from my distro file ... |
| 06:21.10 | kergoth | woo! |
| 06:21.15 | kergoth | my world build finished |
| 06:21.17 | kergoth | finally |
| 06:21.17 | kergoth | <PROTECTED> |
| 06:21.27 | kergoth | heh |
| 06:21.51 | emte | hmm |
| 06:21.55 | emte | odd ... |
| 06:22.23 | emte | it behaves if i put it in openembedded's tree |
| 06:22.48 | kergoth | sounds like an incorrectly set BBPATH |
| 06:22.57 | kergoth | or you're putting it in th ewrong place |
| 06:23.09 | kergoth | mkdir build/conf/distro; cp blah.conf build/conf/distro/ |
| 06:23.09 | kergoth | heh |
| 06:23.17 | emte | hard to say, my igortree is primary not openembedded |
| 06:23.39 | emte | so it should hev been taking it from igortree |
| 06:30.07 | gb2 | kergoth: you run into widechar issues in various c++ packages building with uclibc? |
| 06:30.10 | emte | lets see if i can hit my A1 target now |
| 06:30.49 | kergoth | gb2: dont think so, no |
| 06:30.54 | kergoth | gb2: got an example? |
| 06:31.26 | gb2 | opensp |
| 06:32.47 | kergoth | thats a new gcc thing i think. just being more picky |
| 06:32.50 | kergoth | fails here too |
| 06:32.57 | kergoth | with glibc |
| 06:33.00 | gb2 | ah |
| 06:37.13 | gb2 | huh. looks kinda strange. |
| 06:37.29 | gb2 | like it needs an additional include in that file |
| 06:38.50 | kergoth | robtaylor_: why does the linux-atm .bb install _target_ binaries into staging in do_stage? |
| 06:40.52 | gb2 | yeah, that one line fix did it |
| 06:41.47 | *** join/#oe raster (~raster@YahooBB219198104088.bbtec.net) |
| 06:44.51 | emte | heya raster |
| 06:45.32 | raster | emte: yo yo yo. pop |
| 06:46.50 | emte | lol |
| 06:46.58 | emte | how goes it? |
| 06:48.08 | raster | mucho busy |
| 06:48.09 | raster | as usual |
| 06:48.19 | raster | trying to nuke this mini freeze todo list |
| 06:51.26 | raster | hehe |
| 06:51.30 | raster | no can do |
| 06:51.34 | raster | need to fix the bugs/issues ON it |
| 06:51.35 | raster | :) |
| 06:54.44 | gb2 | grrr, stupid iputils |
| 06:55.43 | emte | aw cmon you dont need the internet do ya :P |
| 06:56.05 | gb2 | it does some "glibc fix" which doesn't do the right thing for uclibc |
| 06:56.29 | emte | aha there we go... now we're fixing chikens |
| 06:56.52 | emte | gb2, not supprising ... probably a gettext thing |
| 06:56.59 | gb2 | socket bits |
| 06:57.02 | gb2 | actually |
| 07:08.56 | *** join/#oe micropal_ (ident@139.80-202-82.nextgentel.com) |
| 07:21.10 | kergoth | gb2: heh, yeah, that happens. since uclibc defines __GLIBC__. |
| 07:21.19 | kergoth | gb2: thankfully it also defines __UCLIBC__, so can ifndef on that |
| 07:31.18 | *** join/#oe hufnus (~slonsiki@mab8e36d0.tmodns.net) |
| 07:34.18 | *** join/#oe |dkey| (~dkey@L0009P16.dipool.highway.telekom.at) |
| 09:01.30 | *** join/#oe nofpu (~matthew@82-45-183-10.cable.ubr06.dals.blueyonder.co.uk) |
| 09:10.28 | *** join/#oe koen (~koen@cl-148.ams-05.nl.sixxs.net) |
| 09:10.48 | koen | good morning all |
| 09:21.52 | *** join/#oe [cc]smart (~smart@gw.ptr-62-65-149-158.customer.ch.netstream.com) |
| 09:40.19 | *** join/#oe hanjo (~hanjo@pD95D0BC8.dip.t-dialin.net) |
| 09:59.44 | *** join/#oe _alwin_ (alwin@cable-195-14-198-78.netcologne.de) |
| 10:02.48 | *** join/#oe gremlin[it] (~gremlin@194-185-220-119.f4.ngi.it) |
| 10:02.53 | gremlin[it] | hi to all |
| 10:03.09 | gremlin[it] | i'm looking hard about my trouble with perl-native |
| 10:03.20 | gremlin[it] | seem a problem in config file |
| 10:03.22 | *** join/#oe darkschneider (~gab@213-140-22-64.fastres.net) |
| 10:03.47 | gremlin[it] | meybe someone more expert than me can give me an hint |
| 10:04.12 | gremlin[it] | is really sethostent_r function undefined in linux's gnu libc ? |
| 10:11.15 | *** join/#oe ar_ (~ar@port-ip-213-211-231-91.reverse.mdcc-fun.de) |
| 10:45.37 | CIA-8 | 03koen 07 * r1.3070.1.1 10openembedded/packages/libmatchbox/libmatchbox_svn.bb: |
| 10:45.37 | CIA-8 | libmatchbox_svn.bb: |
| 10:45.37 | CIA-8 | <PROTECTED> |
| 10:45.44 | gremlin[it] | hi koen ! |
| 10:45.51 | koen | hey gremlin[it] |
| 10:46.08 | gremlin[it] | i problebly resolve problem with perl-native |
| 10:46.31 | gremlin[it] | but have some problem instructing bitbake to compile as i need ;( |
| 10:47.03 | koen | kergoth pushed a cset last night to fix some gcc problems, maybe that will solve it |
| 10:48.38 | gremlin[it] | basically in configuring perl-native it recognise the exixtence of function sethostent_r thta really don't exist :( ... |
| 10:48.42 | gremlin[it] | and some others ... |
| 10:53.58 | gremlin[it] | i suppose perl-native don't have dependencies in gcc (arm-gcc) |
| 10:55.26 | gremlin[it] | grrr everithing i do in ./Configure in perl are discarged ;( :( |
| 10:55.39 | gremlin[it] | >>> |
| 10:55.42 | gremlin[it] | *** WHOA THERE!!! *** |
| 10:55.42 | gremlin[it] | <PROTECTED> |
| 10:55.42 | gremlin[it] | <PROTECTED> |
| 10:55.43 | gremlin[it] | <<< |
| 11:03.10 | gremlin[it] | if i do 'bitbake -c configure -b somewhere/package.bb' i'll stop to configure step ? |
| 11:04.31 | *** join/#oe pb_ (~pb@2002:5168:d330:1:a00:1fff:fe06:93c) |
| 11:04.37 | gremlin[it] | hi pb_ |
| 11:04.47 | pb_ | hi gremlin[it] |
| 11:06.45 | gremlin[it] | grrr how can 'overwrite' perl define/undefine status !!! grrr |
| 11:08.19 | *** join/#oe univac (~univac@ann11.neoplus.adsl.tpnet.pl) |
| 11:08.31 | univac | hi |
| 11:09.07 | univac | when i run bitbake program i have |
| 11:09.12 | univac | OE Build Configuration: |
| 11:09.12 | univac | TARGET_ARCH = "INVALID" |
| 11:09.12 | univac | TARGET_OS = "INVALID" |
| 11:09.30 | univac | when i unhash this lines in local.conf i have python errors |
| 11:09.56 | gremlin[it] | have u defien something in MACHINE and DISTRO in local.conf ? |
| 11:10.54 | gremlin[it] | go to lunch .. bye |
| 11:10.57 | gremlin[it] | cu later |
| 11:11.06 | univac | when i unhash this lines i have in arch arm and targer_os linux |
| 11:11.15 | univac | <PROTECTED> |
| 11:11.16 | univac | bb.parse.ParseError: conf/local.conf:67: unparsed line: ' TARGET_OS = "linux"' |
| 11:22.31 | *** join/#oe Rupert-Giles (~willow@neilmathers.co.uk) |
| 11:23.23 | koen | univac: remove the leading space |
| 11:23.47 | univac | where? |
| 11:24.05 | koen | ' TARGET_OS = "linux"' |
| 11:24.22 | koen | lines in local.conf can't start with a space |
| 11:24.35 | univac | i have commented this line |
| 11:25.15 | univac | # TARGET_OS = "linux" |
| 11:25.19 | univac | # DISTRO = "familiar" |
| 11:25.55 | univac | remove this lines? |
| 11:26.16 | koen | remove '# ' from those lines |
| 11:26.41 | univac | when i remove # from this 2 lines i have python syntax error |
| 11:27.13 | pb_ | indeed. that's why you need to remove "# ", not "#". |
| 11:27.43 | RP | morning all |
| 11:27.47 | koen | hey RP |
| 11:27.47 | pb_ | hi rp |
| 11:28.07 | CIA-8 | 03koen 07 * r1.3076 10openembedded/conf/local.conf.sample: |
| 11:28.07 | CIA-8 | local.conf.sample: |
| 11:28.07 | CIA-8 | <PROTECTED> |
| 11:28.28 | univac | look |
| 11:28.29 | univac | http://rafb.net/paste/results/UqctVR18.html |
| 11:28.47 | koen | univac: we know |
| 11:29.01 | univac | and what can i do? |
| 11:29.03 | univac | to repiar it |
| 11:29.08 | koen | univac: lines in local.conf can't start with ' ' |
| 11:29.27 | univac | i must remove # and '? |
| 11:29.34 | pb_ | good grief |
| 11:29.51 | koen | remove "# ", not just "#" |
| 11:31.41 | koen | the line should read 'TARGET_ARCH = "arm"' instead of ' TARGET_ARCH = "arm"' |
| 11:31.51 | *** join/#oe aboeglin (~aboeglin@213.154.79.4) |
| 11:32.52 | RP | remove the space before TARGET_ARCH |
| 11:33.06 | univac | yes i know |
| 11:33.16 | univac | now i make handling |
| 11:33.28 | univac | bitbake files |
| 11:33.34 | pb_ | congratulations |
| 11:33.39 | univac | ;p |
| 11:33.54 | univac | and where is save the packages whn is the buildest? |
| 11:34.13 | koen | $TMPDIR/deploy |
| 11:35.02 | koen | *sigh* |
| 11:35.17 | koen | evo2 crashes when trying to save the vlc shar |
| 11:36.11 | koen | pb_: when minimo works, I'll push the new mozconfig dougt sent me |
| 11:36.17 | univac | its like portage from gentoo |
| 11:36.23 | pb_ | koen: very good |
| 11:36.40 | koen | univac: bitbake is based on portage :) |
| 11:37.21 | univac | :) |
| 11:37.33 | univac | ^^ |
| 11:38.23 | koen | luckily we eradicated any portage code in bitbake/oe |
| 11:42.25 | RP | Problem 1, the screen is upsidedown :) |
| 11:42.39 | koen | RP: bootsplash? |
| 11:43.07 | RP | The "Powered by openembedded, GPE-2.6" screen |
| 11:43.49 | koen | yeah, that's been upside down for ages on corgi |
| 11:44.16 | *** join/#oe perlguru (~maurice@cc444495-a.hnglo1.ov.home.nl) |
| 11:44.39 | RP | It should also probably say something like "please wait" as I was thinking this had crashed! :) |
| 11:46.38 | koen | RP: 3d rotating status bar! |
| 11:47.26 | RP | koen: Are you saying it needs on or that I missed it? |
| 11:47.38 | RP | I can't log in over the serial lead (login: no valid shadow password) |
| 11:48.28 | koen | RP: it's needs a status bar |
| 11:48.37 | koen | and gpe zaps the rootpasswd |
| 11:48.49 | koen | (as will opie in familiar-0.8.2) |
| 11:49.06 | RP | koen: So how am I supposed to get into this |
| 11:49.08 | RP | ? |
| 11:49.25 | koen | set a root passwd using the gui |
| 11:49.49 | RP | gpe is now showing me a black screen with a cursor on it. How do I get the gui? |
| 11:49.57 | koen | heh |
| 11:50.21 | koen | it should be on your screen |
| 11:50.31 | RP | It isn't... |
| 11:50.40 | koen | looks like you need to build an image without zap_root_passwd |
| 11:58.07 | RP | Looks like there's no way around it :-/ |
| 11:58.25 | _alwin_ | the mail-led on C8x0/C7x0 don't work with 2.6.x kernel? |
| 11:58.44 | RP | _alwin_: correct |
| 11:58.48 | _alwin_ | hm. |
| 11:59.08 | RP | I can't decide on the interface. Its easy enough to add when I decide on one... |
| 11:59.10 | _alwin_ | and I'm searching the error in opiemail-applet like a fool |
| 11:59.25 | RP | It does say that on the 2.6 kernel status page... |
| 12:00.13 | _alwin_ | may be ;) but 1. the were the next question - may I get the link again (lost it) 2. there are lot of statuspages - sometimes I forget some details ;) |
| 12:00.35 | RP | http://www.rpsys.net/openzaurus/ |
| 12:00.38 | _alwin_ | RP: eg, the problem is you don't know the hardware interface |
| 12:01.25 | RP | _alwin_:I know how to turn the LED on and off. I'm just not happy about the right way to expose it to userspace... |
| 12:01.33 | _alwin_ | ah! |
| 12:02.08 | _alwin_ | I would prefer sysfs/arm ;) |
| 12:02.42 | RP | _alwin_: so would I but it isn't that easy |
| 12:03.28 | _alwin_ | Yes, of course, and I think 2.6.x is real working fine mostly on that devices. |
| 12:04.23 | RP | 2.6 is very nearly there. Just needs sound and power support sorting out proplerly. leds are a minor issue... |
| 12:04.31 | koen | RP: I noticed mq1100fb uses 'generic software acceleration', would w100fb benefit from that too? |
| 12:04.40 | _alwin_ | and that OPIE can not make suspend when the pda is closed |
| 12:04.50 | RP | I think w100fb already uses that |
| 12:04.54 | _alwin_ | is this a kernel problem - or may I take a look into opie |
| 12:04.59 | koen | ah, cool |
| 12:06.20 | RP | What exactly do you mean by "generic software acceleration"? |
| 12:06.31 | koen | the cfb* modules |
| 12:06.38 | RP | Yes, it uses them |
| 12:06.59 | RP | I have hardware accel planned for the w100fb. I think it can be done |
| 12:07.07 | koen | in the fb layer? |
| 12:07.17 | RP | yes |
| 12:07.23 | koen | nice |
| 12:07.34 | RP | although once code it there, opie/xserver can probably be hooked in |
| 12:07.51 | RP | _alwin_: suspending when closing the unit is an opie issue |
| 12:08.08 | RP | We report hinge status to opie... |
| 12:11.00 | _alwin_ | hm. |
| 12:13.28 | *** join/#oe [g2] (~g2@g2.nslu2-linux) |
| 12:13.52 | _alwin_ | ai, I see. three states, hinge1, hinge2, hinge3.... I'should check how them comes from system to opie and whats going wrong |
| 12:14.19 | RP | It gets there via the input system |
| 12:14.56 | RP | The input maintainer is unhappy with how I've done that and its going to have to change slightly although nothing that will invalidate what you're doing... |
| 12:15.13 | _alwin_ | yes, found it. |
| 12:26.03 | *** join/#oe reenoo_ (~r@p5489E114.dip.t-dialin.net) |
| 12:26.48 | reenoo_ | afternoon |
| 12:27.22 | RP | hi reenoo_ |
| 12:27.37 | reenoo_ | hey RP |
| 12:34.10 | pb_ | hi reenoo_ |
| 12:35.36 | reenoo_ | hey pb_ |
| 12:36.15 | RP | Where should I be asking for help in debugging this gpe image? Here or #gpe? |
| 12:37.12 | *** join/#oe aboeglin (~aboeglin@213.154.79.4) |
| 12:38.46 | koen | bleh |
| 12:39.01 | koen | minimo fails in wince/winEmbed.cpp |
| 12:42.11 | koen | goody |
| 12:42.22 | koen | and no way to disable a wince build either |
| 12:42.50 | reenoo_ | heh |
| 12:44.04 | koen | reenoo_: I got mplayer to work |
| 12:44.31 | reenoo_ | koen: interesting. what have you changed? |
| 12:44.34 | CoreDump|home | morning |
| 12:44.41 | koen | cmdline options |
| 12:44.50 | reenoo_ | heh |
| 12:44.57 | koen | <PROTECTED> |
| 12:45.02 | koen | that works |
| 12:45.51 | koen | will test -vo x11 after my new gpe-image is finished |
| 12:46.12 | reenoo_ | that bypasses X though, doesn't it? |
| 12:48.07 | koen | it does |
| 12:50.12 | koen | reenoo_: there is a patch on the oe-ml to make vlc-gpe a bit more pda friendly |
| 12:51.46 | reenoo_ | oh, really? all I've seen was a patch that makes the pda gui compile for vlc 0.8.1 |
| 12:52.06 | koen | that same patch removes some fp stuff |
| 12:52.43 | univac | :> |
| 12:57.36 | reenoo_ | koen: I don't see anything like that in the "fix-pda.patch". but I haven't extracted the .shar either. that's like zip .exe's in windows |
| 12:58.34 | koen | I though I saw something like "fix slider to use non-floating point math" somewhere |
| 12:58.57 | koen | I read the shar as text without extracting it |
| 12:59.58 | reenoo_ | that's part of the pda gui by default |
| 12:59.58 | reenoo_ | the patch doesn't touch it |
| 13:00.21 | koen | aha |
| 13:00.27 | koen | I misread that part |
| 13:00.40 | koen | reenoo_: http://handhelds.org/scap/port.17735.png |
| 13:02.42 | koen | that's with -vo sdl |
| 13:08.21 | reenoo_ | ah, nice |
| 13:10.47 | reenoo_ | have you read that vlc-devel posting btw? it's kind of funny how people always blame autoconf (or even OE) first |
| 13:11.19 | reenoo_ | instead of _reading error/warning messages_ and fixing their own mistakes |
| 13:11.19 | koen | via.ecp.fr timesout over here |
| 13:11.27 | reenoo_ | heh |
| 13:13.26 | reenoo_ | http://pastebin.ca/7313 |
| 13:15.10 | koen | heh |
| 13:15.46 | *** join/#oe hanjo (~hanjo@pD95D0BC8.dip.t-dialin.net) |
| 13:20.14 | _alwin_ | RP: the hingesensor is /dev/event0 - right? |
| 13:20.45 | koen | depending when the modules are loaded |
| 13:21.08 | koen | when the ts get's loaded first that will be event0 |
| 13:21.39 | _alwin_ | how may I checked that? |
| 13:21.46 | koen | sysfs or ioctls |
| 13:22.09 | koen | don't about the details, zecke or pb_ know more about that |
| 13:30.02 | RP | _alwin_: Look at the output from dmesg |
| 13:30.18 | RP | but it normally is /dev/event0... |
| 13:30.25 | *** join/#oe [cc]smart (~smart@gw.ptr-62-65-149-158.customer.ch.netstream.com) |
| 13:30.29 | _alwin_ | RP: hm. it seems that there is a strong failure inside button-handling of opie... |
| 13:31.42 | RP | I know mickeyl did update opie to handle the new keyboard handling. I've yet to test it though. (an opie image is on my list after I've got gpe-image working) |
| 13:41.04 | univac | ok i build some pakcages and how i cand install it on familiar? |
| 13:44.57 | koen | univac: use ipkg |
| 13:45.23 | univac | i know, but where i can do it? |
| 13:45.28 | univac | copy packages on ipaq? |
| 13:47.45 | pb_ | ipkg install <url> |
| 13:47.55 | univac | all packages is in openembedded/package? |
| 13:51.12 | *** join/#oe oe_newbie (~kvirc@195.234.128.72) |
| 13:51.15 | univac | what is the good light browser for gpe? |
| 13:51.46 | chris144|home | hi folks |
| 13:52.54 | pb_ | univac: minimo |
| 14:03.48 | *** join/#oe gremlin[it] (~gremlin@194-185-170-139.f4.ngi.it) |
| 14:03.59 | gremlin[it] | have a question |
| 14:04.16 | gremlin[it] | .. wait.. first of all ... "good afternoon" !!! |
| 14:05.26 | gremlin[it] | well ... if in local.conf i define CVSDATE_pack1=20050201 CVSDATE_pack2=20050202 and CVSDATE=20050220 |
| 14:05.57 | gremlin[it] | all cvs based package except pack1 and pack2 will be no nore recent than 2005-02-20 ... right ? |
| 14:06.31 | gremlin[it] | univac, explain me ... maybe i can help |
| 14:07.01 | *** join/#oe raster (~raster@YahooBB219198104088.bbtec.net) |
| 14:07.09 | univac | i build packages with bitbake |
| 14:07.25 | univac | but i don;t know what i can install this and upgrade old packages |
| 14:08.29 | gremlin[it] | mhhh i have just clean all ... but if i remember correct in stuff/build/tmp/deploy/ipkg u have all packeges :) |
| 14:08.46 | univac | yes i have |
| 14:08.55 | chris144|home | gremlin[it]: I think what you will see is that pack1 and pack2 will be doenloaded with your given CVSdate, all other packages will be downloaded with 20050220. Should be fairly easy to try out. |
| 14:09.00 | univac | but what i have build hmm new gpe and upgrade |
| 14:09.12 | univac | or install new with denepds |
| 14:10.11 | gremlin[it] | chris144|home thanks :) ... is what i suppose but i would have another answer ;) |
| 14:10.32 | chris144|home | gremlin[it]: what would you like to see? |
| 14:10.42 | gremlin[it] | univac is such dir would be also present Package.something right |
| 14:10.50 | chris144|home | maybe it can be done... |
| 14:11.29 | gremlin[it] | chris144|home exactly what u said ... i'm working with familiar-0.8.1 snapshot so i have to limit CVSDATE to no more recent than release date ... |
| 14:13.26 | univac | hmm maybe i use standart packages from ipkg ;[ |
| 14:13.49 | chris144|home | gremlin[it]: but familar 0.8.1 already has 20050222 in its conf file. it should work also for other packages i think |
| 14:14.22 | chris144|home | gremlin[it]: what exactly is your problem? |
| 14:14.56 | gremlin[it] | chris144|home mhhh no ... |
| 14:15.03 | gremlin[it] | CVSDATE_genext2fs-native=20050127 |
| 14:15.04 | gremlin[it] | CVSDATE_genext2fs=20050127 |
| 14:15.04 | gremlin[it] | #CVSDATE_cairo=20050204 |
| 14:15.04 | gremlin[it] | #CVSDATE_xserver-kdrive=20050207 |
| 14:15.04 | gremlin[it] | CVSDATE_minimo=20050210 |
| 14:15.25 | *** join/#oe zecke_ (fakeuser@pD9E7DCF3.dip0.t-ipconnect.de) |
| 14:15.29 | gremlin[it] | i have only this CVSDATE defile |
| 14:15.33 | gremlin[it] | defines |
| 14:17.15 | chris144|home | gremlin[it]: sorry, I still do not understand what you want to do. Where are these CVSDATe entries? in your local.conf file? |
| 14:17.44 | gremlin[it] | yes ... but not a "globa;" CVSDATE .... so i have to add |
| 14:19.26 | univac | hmm whats is the good terminal for gpe? |
| 14:19.43 | gremlin[it] | rxvt-aa |
| 14:21.26 | univac | gremlin[it]: can you help me with install this pacgaes from my box on pda? |
| 14:21.27 | univac | :> |
| 14:21.50 | chris144|home | gremlin[it]: I think the familar distro conf files are a good example: familiar-0.8.1.conf include a global CVSDATE. The included preferred-gpe-versions.inc includes special CVSDATES for cairo and xserver-kdrive. |
| 14:21.57 | koen | univac: this is not a distro support channel, see #familiar for that |
| 14:22.27 | chris144|home | hi koen |
| 14:22.27 | *** join/#oe qfh (~qfh@ppp-82-135-12-140.mnet-online.de) |
| 14:22.30 | gremlin[it] | where is the familiar-0.8.1.conf ??? |
| 14:22.36 | koen | hey chris144|home |
| 14:22.50 | gremlin[it] | DHO .. find !!! |
| 14:23.00 | chris144|home | gremlin[it]: in openembedded/conf/distro |
| 14:23.25 | gremlin[it] | but so what some packspecific_CsVDATE are defined in local.conf and others in familiar-0.8.1.conf ?!?!?! |
| 14:23.40 | gremlin[it] | but why some ... |
| 14:24.41 | chris144|home | gremlin[it]: I think the stuff in local.conf is global, distribution independant stuff. |
| 14:25.26 | gremlin[it] | yes ..but in the familiar snapshot existe conf/local.conf ... so i try to use it ... |
| 14:30.21 | Pendalar | whoa.. flipped on disney for my son, and after Wiggles, there was a TMBG song about finding letters |
| 14:30.35 | Pendalar | expose TMBG early on heh.. rock ^_^ |
| 14:31.48 | Pendalar | of course probably only kergoth could appreciate that =-s |
| 14:43.52 | *** join/#oe Ken|JL (~Ken@h72n2fls34o1105.telia.com) |
| 14:44.37 | *** join/#oe Timelord (~TL@64-147-65-107.access-3.intertex.net) |
| 14:47.33 | *** join/#oe loial (~loial@cp10860-a.tilbu1.nb.home.nl) |
| 14:47.44 | loial | lo |
| 14:48.52 | *** join/#oe Ken|JLimeX (~Ken@h72n2fls34o1105.telia.com) |
| 14:51.05 | *** join/#oe hanjo (~hanjo@pD9E77B6B.dip.t-dialin.net) |
| 14:51.59 | cbrake | any opinions what is the best filesystem for OE development? I'm thinking reiser as it is supposed to be good w/ small files. |
| 14:52.29 | koen | xfs performs well over here |
| 14:52.51 | *** join/#oe Ken|JL (~Ken@h72n2fls34o1105.telia.com) |
| 15:35.14 | *** join/#oe dkey| (~dkey@L0004P09.dipool.highway.telekom.at) |
| 15:47.51 | gremlin[it] | question: all 'native' packages are build with gcc and glibc already present on distro riht ? |
| 15:47.51 | gremlin[it] | riht? |
| 15:47.55 | gremlin[it] | right? |
| 15:48.37 | koen | no |
| 15:49.09 | koen | they should be built against glibc in staging |
| 15:49.30 | koen | since not everyone builds on linux |
| 15:50.02 | gremlin[it] | koen ok ... i'm still trying to figure out why seem i'm the only one who experience the perl-native issue :( ... |
| 15:50.27 | koen | I had a perl-native issue too, but I hacked around it |
| 15:50.35 | koen | (too lazy to rebuild from scratch) |
| 15:50.39 | gremlin[it] | what about u ? |
| 15:51.14 | koen | I edited the makefile and copy the missing perl modules from my host system (yes, that's bad, I know) |
| 15:51.26 | koen | s/copy/copied/ |
| 15:51.30 | gremlin[it] | sethostent_r, setprotoent_r, setservent_r, endhostent_r, endprotoent_r, endservent_r undefined :( |
| 15:51.43 | gremlin[it] | ahhh no seem different from mine ... |
| 15:51.46 | koen | I didn't see that |
| 15:52.28 | gremlin[it] | i'm working with familiat snapshot ... so aquite old oe repo :) |
| 15:52.34 | koen | hmm |
| 15:52.45 | koen | the current repo should build too |
| 15:53.03 | koen | with a small alteration to xext and xserver-kdrive |
| 15:53.58 | gremlin[it] | last time i try build updated repo (begin of month) got perl-native issue in both :( ... |
| 15:55.56 | gremlin[it] | koen, monday i'll got athlon64 with dual chanel 1GByte RAM ... i hope do be more fast building ;) |
| 15:56.11 | gremlin[it] | on building |
| 15:56.19 | koen | heh |
| 15:56.32 | koen | still waiting for a network connection for the uni buildbox |
| 15:56.50 | koen | dual opteron 244/2gb ram and u320 scsi disks |
| 15:56.59 | koen | too bad it's not mine |
| 15:57.07 | gremlin[it] | conenction will be my next issue ... but i have to change home for that :) |
| 15:57.41 | gremlin[it] | asap i'll buy a WD raptor harddrive ... not a U320 but really good ;) |
| 15:57.59 | *** join/#oe sc (~sc@p50925D66.dip0.t-ipconnect.de) |
| 15:59.28 | chouimat | morning |
| 15:59.53 | gremlin[it] | chouimat, late afternoon here ;) |
| 16:01.07 | *** join/#oe marcansoft (1337@80.26.222.222) |
| 16:01.51 | gremlin[it] | koen, can i query u ? |
| 16:06.45 | *** join/#oe pb_ (~pb@2002:5168:d330:1:a00:1fff:fe06:93c) |
| 16:23.32 | RP | _alwin_: Could you send a message to the opie people about the hinge problem please? |
| 16:42.04 | *** join/#oe alwin_ (alwin@cable-195-14-254-244.netcologne.de) |
| 16:42.36 | alwin_ | RP: i had fixed it in cvs |
| 16:43.29 | *** join/#oe drw (~drw@c-24-0-157-195.client.comcast.net) |
| 16:44.00 | RP | alwin_: Even better :) |
| 16:58.00 | *** join/#oe hufnus (~slonsiki@m961336d0.tmodns.net) |
| 17:06.57 | *** join/#oe CRIMANY (~XEN@d198-166-22-94.abhsia.telus.net) |
| 17:10.20 | *** join/#oe hufnus (~slonsiki@me31836d0.tmodns.net) |
| 17:30.52 | _alwin_ | RP: do you have a timeline about deciding how to integrate the LEDs into the userspace access? |
| 17:37.13 | *** join/#oe treke|home (~ggilbert@69-175-32-221.ventca.adelphia.net) |
| 17:43.22 | gb2 | hrm, what's a good way of convincing bitbake to roughly output something i could interpret as a dependency graph necessary for building a particular package |
| 17:43.47 | emte | odd |
| 17:43.57 | emte | now where did this come from? |
| 17:43.59 | emte | | checking for i686-linux-gcc... ccache i686-linux-gcc |
| 17:43.59 | emte | | checking for C compiler default output file name... configure: error: C compiler cannot create executables |
| 17:44.43 | emte | gb2, write a script to follow depes |
| 17:44.55 | emte | i was thinking about that earlier |
| 17:45.09 | koen | parse bitbake -v? |
| 17:45.10 | emte | sed seems the logical choice |
| 17:45.25 | koen | or hack up bitbake to display it |
| 17:45.51 | gb2 | there used to be something that would roughly do it |
| 17:47.25 | emte | actually ... the easiest way would be to add a bit to the do download |
| 17:47.48 | emte | which would put i'ts file name in another file and then append all other filenames after |
| 17:47.57 | reenoo_ | gb2: bitbake -nv in a clean build tree IIRC |
| 17:48.41 | RP | _alwin_: Not really. I should probably take John Lenz's led driver and use that for now... |
| 17:48.48 | emte | t:~/fabrication/build$ bitbake -nv |
| 17:48.48 | emte | Nothing to do. Use 'bitbake world' to build everything, or run 'bitbake --help' |
| 17:48.48 | emte | for usage information. |
| 17:48.57 | gb2 | reenoo: yeah, that's the best i've found sofar |
| 17:49.00 | emte | seems you have to do world first |
| 17:49.02 | gb2 | emte: give it a package name |
| 17:49.05 | emte | aha |
| 17:51.00 | gb2 | checking host system type... Invalid configuration `i386-linux-uclibc': system `uclibc' not recognized |
| 17:51.01 | gb2 | boo |
| 17:52.05 | emte | bad quotes btw |
| 17:52.28 | emte | odd it does that |
| 17:52.44 | gb2 | emte: that's how unicode quotes should be actually |
| 17:53.04 | emte | as opposed to both single? |
| 17:53.29 | gb2 | you're supposed to do: ``foo'' for double quoting too |
| 17:53.33 | gb2 | kinda like in TeX |
| 17:53.39 | emte | hmm |
| 17:54.04 | emte | tex uses propper quotes |
| 17:54.14 | emte | better alignment on them tho |
| 17:54.42 | gb2 | huh? |
| 17:54.54 | emte | now i am curious how many programming languages that would mess up |
| 17:55.08 | gb2 | when you write TeX, you're supposed to do: `` for an open double quote, and a '' for a closing double quote. they typeset differently |
| 17:55.15 | emte | i suppose if you called a unicode library it shouldnt ... |
| 17:55.29 | gb2 | newer gcc outputs stuff like that too iirc |
| 17:55.34 | *** join/#oe hrw|gone (szczepan@blue.stone.pl) |
| 17:55.44 | emte | my paste was std |
| 17:56.11 | gb2 | depends what standard you're following |
| 17:56.23 | emte | as for tex i guess i should read more since i use the standard typewriter method singel and double quotes |
| 17:56.36 | gb2 | int main(void) { return bob; } |
| 17:56.41 | gb2 | annabel% gcc -c foo.c |
| 17:56.41 | gb2 | foo.c: In function `main': |
| 17:56.41 | gb2 | foo.c:1: error: `bob' undeclared (first use in this function) |
| 17:57.29 | gb2 | quoting is supposed to be locale specific |
| 17:57.30 | emte | hmm |
| 17:57.33 | emte | locale is utf8 |
| 17:57.42 | _alwin_ | RP: as the interface would not much change later it would be a temporary solution I think. Ok, it isn't that important. But you know: Users have this LED so them will see it blink. |
| 17:57.45 | gb2 | even if i set my locale to C, it produces: `' |
| 17:59.47 | emte | so what do you use for apostrophes then? |
| 18:00.23 | gb2 | ' is an apostrophe |
| 18:00.26 | gb2 | it's also a closing single quote |
| 18:00.47 | emte | hmm |
| 18:02.05 | RP | _alwin_: The trouble is if I implement it and then change later, I'll get complaints anyway. I can't win... |
| 18:03.37 | RP | _alwin_: I'll think about it :) |
| 18:06.03 | *** join/#oe micropal_ (ident@139.80-202-82.nextgentel.com) |
| 18:09.36 | kergoth | morning |
| 18:09.43 | emte | lol |
| 18:09.48 | emte | heya Ken_Laptop |
| 18:09.52 | emte | bah |
| 18:09.55 | emte | kergoth, |
| 18:11.07 | gb2 | kergoth: should base_do_package be python? |
| 18:11.23 | pb_ | hey kergoth |
| 18:11.45 | kergoth | gb2: heh, you get bit by the bug where not inheriting package breaks the build? |
| 18:11.53 | gb2 | yeah |
| 18:12.03 | gb2 | turning that into python seems to fix it |
| 18:12.20 | gb2 | all it does is : right now |
| 18:12.25 | kergoth | yeah, we need to do something about that. not sure what, exactly. few possible solutions. changing base_do_package would probably be the quickest fix |
| 18:12.39 | kergoth | no need for :, can empty it, bitbake will realize that it doesnt need to run an empty function |
| 18:12.42 | kergoth | heh |
| 18:13.00 | gb2 | i'm also finding a bunch of: checking host system type... Invalid configuration `i386-linux-uclibc': system `uclibc' not recognized, this morning |
| 18:13.15 | kergoth | gb2: those are packages that are trying to bypass our autoreconf. |
| 18:13.26 | kergoth | gb2: if autoreconf isnt run, the config.guess/config.sub dont get updated |
| 18:13.34 | gb2 | what's the normal way that they bypass it? |
| 18:13.38 | kergoth | change the do_configure to call gnu-configize before oe_runconf |
| 18:13.42 | kergoth | they override do_configure |
| 18:13.51 | kergoth | bypass the autotools bbclass version |
| 18:13.51 | gb2 | do_configure () { |
| 18:13.54 | gb2 | <PROTECTED> |
| 18:13.54 | gb2 | } |
| 18:13.55 | kergoth | yeah |
| 18:13.58 | kergoth | put gnu-configize before that |
| 18:14.03 | kergoth | to update the config. files |
| 18:14.04 | *** join/#oe rn-hlmt (~rn-hlmt@p5493A465.dip0.t-ipconnect.de) |
| 18:14.06 | kergoth | or.. fix the damn thing |
| 18:14.09 | gb2 | heh |
| 18:14.49 | kergoth | doing that is a cheap way to get it building if you have neither the time nor inclination to fix whatever horrid auto* errors are preventing the autoreconf from working |
| 18:14.52 | kergoth | heh |
| 18:15.23 | *** join/#oe andersee (~andersee@codepoet.org) |
| 18:15.30 | gb2 | first i'll try just using the default do_configure and see how bad it is |
| 18:18.31 | gb2 | ah, configure.in uses stuff in aclocal.m4, which gets blown away in the reconf |
| 18:19.18 | gb2 | so it's improperly packaged |
| 18:21.46 | kergoth | thats really common |
| 18:21.58 | kergoth | in a lot of our do_configure's, we just mv aclocal.m4 acinclude.m4 |
| 18:22.02 | kergoth | then call autotools_do_configure |
| 18:22.03 | kergoth | heh |
| 18:22.09 | gb2 | ick :P |
| 18:22.15 | kergoth | yep |
| 18:22.17 | gb2 | you can endup with old macros that way |
| 18:22.28 | kergoth | or you patch it, and break every time they change a macro |
| 18:22.43 | gb2 | true |
| 18:23.03 | kergoth | thats a situation where you really need to fix upstream |
| 18:23.09 | kergoth | give the maintainers a good cluebatting |
| 18:23.13 | gb2 | yep |
| 18:23.52 | kergoth | heh, thats on my list |
| 18:23.55 | kergoth | item #6 on http://www.openembedded.org/cgi-bin/moin.cgi/Buildsystems_2fRulesOfThumb |
| 18:24.33 | gb2 | heh |
| 18:26.55 | gb2 | i should really find a psyco package for fc |
| 18:26.56 | gb2 | 3 |
| 18:28.41 | emte | just pull it in |
| 18:28.57 | emte | pretty sure you can get it through python's svn |
| 18:30.34 | gb2 | psyco is a sourceforge project |
| 18:31.00 | emte | i know |
| 18:31.41 | gb2 | then why should it come from python's svn? |
| 18:31.42 | emte | i belive that i have pulled and installed it through python's svn |
| 18:32.57 | emte | suppose i could be wrong but i dont think so |
| 18:33.06 | gb2 | well, bdist_rpm seemed to work on the tarball that comes from sf |
| 18:37.29 | gb2 | hrm, net-snmp needs more than just trivial fixups |
| 18:37.52 | kergoth | i did a world build last night.. there are like 180 current fialures |
| 18:37.53 | kergoth | :\ |
| 18:37.57 | kergoth | heh |
| 18:39.07 | gb2 | it needed a bunch of cross-build fixes to its autoconf i think |
| 18:39.12 | pb_ | kergoth: doh |
| 18:39.53 | kergoth | pb_: note, now that bb doesnt emit messages for things it doesnt build, bitbake -n world is useful |
| 18:40.09 | kergoth | after build completion, -n world shows what its going to try to build.. in other words, only the failed packages |
| 18:40.10 | kergoth | heh |
| 18:41.17 | pb_ | ah, cool |
| 18:45.01 | gb2 | there are a bunch of packages that depend on the kernel for headers still too |
| 18:45.20 | gb2 | they ought to just be using the headers from linux-libc-headers |
| 18:45.26 | emte | and they are not supposed to? |
| 18:45.43 | gb2 | emte: generally they shouldn't |
| 18:45.45 | kergoth | userspace apps requiring the headers they use to match the kernel they're going to run on, are just broken. |
| 18:45.50 | kergoth | with very few exceptions |
| 18:47.12 | emte | so any kernel patches with new headers are supposed to be added to linux-libc-headers then? |
| 18:47.27 | kergoth | ? |
| 18:47.36 | emte | non-mainstreme patches |
| 18:47.41 | emte | mainstream* |
| 18:47.49 | kergoth | i dont understand the question. why would an _application_ need those headers at all? |
| 18:48.08 | gb2 | usually if you're a non-mainstream patch with non-mainstream tools, your tools should carry along with them the non-mainstream headers necessary to build |
| 18:48.11 | kergoth | right |
| 18:48.43 | emte | hmm |
| 18:49.06 | emte | so i'll have to patch the apps then |
| 18:49.09 | emte | okay |
| 18:49.15 | gb2 | what's your non-mainstream patch? |
| 18:49.39 | emte | looking at a few delete, crytpo, and loop patches |
| 18:50.50 | gb2 | usually it's just a matter of carrying along a dir include/linux/*.h in your tool package, and buliding against that |
| 18:51.11 | kergoth | it gets more ugly when there are differetn versions of the headers |
| 18:51.12 | gb2 | and ofcourse your APIs should be robust enough so that the binaries don't lose too hard if you run them on an inappropriate system. |
| 18:51.16 | kergoth | like in the wireleess-tools case |
| 18:51.17 | gb2 | kergoth: wireless-tools |
| 18:51.19 | gb2 | heh |
| 18:51.20 | kergoth | :) |
| 18:51.24 | kergoth | they carry every version around |
| 18:51.26 | gb2 | that whole api should just be discarded |
| 18:51.30 | kergoth | yeah |
| 18:51.30 | gb2 | and redone from scratch |
| 18:51.31 | kergoth | its a mess |
| 18:51.46 | gb2 | jgarzik and a bunch of other network people really want that done |
| 18:52.09 | gb2 | but nobody wants the effort |
| 18:54.22 | emte | urg ... time to start the global patch against the whole oe tree ... |
| 18:54.39 | emte | or i could not use the oe tee ... |
| 18:54.44 | emte | decisions decisions |
| 18:57.10 | emte | html emails are amusing when you have html disabled |
| 18:57.37 | emte | the html part tries to sell you software or perscriptons |
| 18:57.53 | emte | and the non-html is some non-sensical story |
| 18:58.57 | emte | "In 1934 Jenny Jones beached a whale on the Alps in Chicago while laundry ate Nike" |
| 19:22.07 | gremlin[it] | seem teleport don't have libgpepimc ad RDEPEND or DEPEND (had to build by hand) |
| 19:23.35 | pb_ | it shouldn't. |
| 19:24.28 | gremlin[it] | mistake ... miss libtododb ... not libgpepimc (that is need by libtododb) |
| 19:26.03 | gremlin[it] | here teleport fail on linking case don't find libtododb ... building it with "bitbake -b" resolve |
| 19:27.15 | kergoth | there, did a quick genext2fs patch to add a cmdline opt to change the 'bytes per inode' ratio used in the automatic inode count. thats often much more useful than the max inode option. |
| 19:27.23 | kergoth | heh |
| 19:27.25 | kergoth | hrm |
| 19:29.38 | pb_ | gremlin[it]: the bug is in teleport's makefile. It shouldn't be linking with those libraries. |
| 19:29.46 | pb_ | kergoth: very good |
| 19:29.54 | gremlin[it] | ok :) |
| 19:30.27 | gremlin[it] | bp_ ... i'm working with familiat snapshot of oe ... so maybe is already patched ... |
| 19:30.38 | gremlin[it] | familiat --> familiar |
| 19:31.17 | gremlin[it] | kergoth still here ? |
| 19:32.02 | gremlin[it] | have u scare of me ?!?!?! a small black gremlin !?!?! :) :) :) |
| 19:32.08 | kergoth | hehe |
| 19:32.39 | gremlin[it] | kergoth i'm abe to build perl-native i found the trouble |
| 19:33.05 | chris144|home | folks, one question, since I am not sure I understand it right: I have built a full rootfs, which works. Now I will start writing an app. I have to use the .h files from staging for compilation and link the app to .o and .so files in staging right? |
| 19:33.41 | kergoth | gremlin[it]: ah, what was it? |
| 19:34.03 | gremlin[it] | <PROTECTED> |
| 19:34.04 | kergoth | chris144|home: yep. and of course, add the tmp/cross/bin to your PATH for the crosscompiler |
| 19:34.23 | chris144|home | kergoth: all I needed, thanks! |
| 19:34.30 | kergoth | np |
| 19:34.52 | kergoth | gremlin[it]: hmm, odd |
| 19:35.00 | kergoth | gremlin[it]: i'll check into it, i think i know the cause |
| 19:35.32 | gremlin[it] | ok |
| 19:36.20 | gremlin[it] | and seem -Uaaa options to ./Configure are ignored (unable to change default behaviour) :(* |
| 19:36.27 | kergoth | odd |
| 19:37.00 | gremlin[it] | i had to change the setup.sh script by hand after the fail and then rebuild with bitbake |
| 19:37.43 | gremlin[it] | seem commandline cannot change behaviour of already defined options loaded from "standard" setup ;( |
| 19:38.38 | gremlin[it] | well guy ... i got a stable familiar-0.8.1 gpe-image build myself ... |
| 19:39.07 | gremlin[it] | now i try do my work to reduce optimally for a h3600 without jackets :) |
| 19:40.31 | gremlin[it] | kergoth, list of functions are : sethostent_r, setprotoent_r, setservent_r, endhostent_r, endprotoent_r, endservent_r. |
| 20:00.57 | *** join/#oe Ken|JLimeX (~Ken@h72n2fls34o1105.telia.com) |
| 20:02.50 | *** join/#oe Ken|JL (~Ken@h72n2fls34o1105.telia.com) |
| 20:04.00 | *** join/#oe Ken|JLimeX (~Ken@h72n2fls34o1105.telia.com) |
| 20:05.24 | *** join/#oe dkey (~dkey@L0004P09.dipool.highway.telekom.at) |
| 20:09.09 | *** join/#oe Ken|JL (~Ken@h72n2fls34o1105.telia.com) |
| 20:12.49 | *** join/#oe Ken|JLimeX (~Ken@h72n2fls34o1105.telia.com) |
| 20:16.37 | pb_ | ah, that was dumb |
| 20:16.44 | kergoth | hm? |
| 20:16.51 | pb_ | I just loaded a firewall policy onto our router that blocked my ssh connection. |
| 20:16.55 | kergoth | hah |
| 20:16.57 | kergoth | i hate it when i do that |
| 20:17.00 | pb_ | yeah |
| 20:17.18 | pb_ | oh well. I can't be bothered to drive over to the office and fix it. |
| 20:17.24 | pb_ | guess it can wait until Monday |
| 20:17.33 | *** join/#oe cyn (~cyn@adsl-068-209-224-046.sip.mem.bellsouth.net) |
| 20:22.03 | *** join/#oe Ken|JL (~Ken@h72n2fls34o1105.telia.com) |
| 20:25.40 | CIA-8 | 03kergoth 07 * r1.3074.1.1 10openembedded/packages/genext2fs/ (3 files in 2 dirs): |
| 20:25.40 | CIA-8 | genext2fs 1.3+cvs: |
| 20:25.40 | CIA-8 | <PROTECTED> |
| 20:25.40 | CIA-8 | <PROTECTED> |
| 20:25.40 | CIA-8 | <PROTECTED> |
| 20:27.03 | *** join/#oe chris144|test (~chris@195.234.128.72) |
| 20:33.09 | *** join/#oe tigrux (~tigrux@charybdis-ext.suse.de) |
| 20:33.20 | *** join/#oe HuntsMan (~hunts@pc-148-118-120-200.cm.vtr.net) |
| 20:36.26 | kergoth | switched the wiki over to the stylesheet punkass did |
| 20:38.03 | pb_ | cool |
| 20:38.55 | kergoth | slowly but surely working through some of the small items on my TODO |
| 20:38.55 | kergoth | heh. |
| 20:38.59 | Ken_Laptop | eeh, im getting this while compiling gcc - "The following requested languages were not found : f77" |
| 20:47.48 | *** join/#oe hufnus (~slonsiki@me31836d0.tmodns.net) |
| 20:55.27 | gremlin[it] | gpe-session-scripts don't need gpe-bluetooth ad RDEPEND ... |
| 21:02.22 | gremlin[it] | bye to alll ... |
| 21:17.41 | kergoth | hah, with punkass's stylesheet, the first line of text in the page overlaps the line below the navbar |
| 21:21.47 | koen | only 5k? |
| 21:21.56 | kergoth | thats what a world build emits |
| 21:22.03 | kergoth | 180 packages failed to build. |
| 21:22.03 | koen | a world build should get you over 8k |
| 21:22.09 | koen | ah, I see |
| 21:22.19 | koen | wait |
| 21:22.32 | *** join/#oe micropal (ident@139.80-202-82.nextgentel.com) |
| 21:22.41 | koen | a world build for 6 machines emits 8.1k ipkgs |
| 21:22.46 | kergoth | hehe |
| 21:22.54 | koen | lot's of kernel modules |
| 21:23.14 | kergoth | we've been at around 5,000 for a single machine for a while now |
| 21:23.21 | koen | that's with 79 failed packages and familair 0.8.1 |
| 21:23.33 | kergoth | i cant believe 180 failed, that hurts |
| 21:23.35 | kergoth | heh |
| 21:23.55 | koen | X broke a lot with the -D_GNU_SOURCE thing |
| 21:24.09 | kergoth | some broke from automake 1.9.x'ism |
| 21:24.18 | kergoth | probably a few missing upstream sources |
| 21:24.20 | kergoth | as usual |
| 21:41.13 | chris144|home | nite folks |
| 21:49.37 | kergoth | ~seen mickeyl |
| 21:49.53 | ibot | mickeyl <~LPA@141.2.14.64> was last seen on IRC in channel #oe, 7d 23h 17m 25s ago, saying: 'c'ya with hopefully all bones'. |
| 21:49.53 | kergoth | .. |
| 21:50.26 | CIA-8 | 03kergoth 07 * r1.3078 10openembedded/packages/gnu-config/ (gnu-config_cvs.bb gnu-config/config-guess-uclibc.patch): Fix the autom4te_perllibdir bits in gnu-configize, to make it usable natively on the target. |
| 21:51.00 | [g2] | that looks like a cool one ! :) |
| 21:51.50 | kergoth | heh, forgot to push it |
| 21:51.56 | kergoth | been sitting here since i got things happy |
| 21:55.32 | CIA-8 | 03kergoth 07 * r1.3080 10openembedded/conf/bitbake.conf: |
| 21:55.32 | CIA-8 | Per the recommendation of Erik Andersen, kill off the setting of -rpath in our |
| 21:55.32 | CIA-8 | TARGET_LDFLAGS. We now rely on ldconfig more than ever. This was necessary, |
| 21:55.32 | CIA-8 | because the linker looks in both rpath _and_ rpath-link to find things |
| 21:55.32 | CIA-8 | during the build. |
| 21:55.35 | CIA-8 | 03kergoth 07 * r1.3079 10openembedded/conf/bitbake.conf: Fix the naming of the .ext2 image inside the .ext2.gz for that image type. |
| 21:57.40 | pb_ | mm. It seems kind of like a bug in ld that it searches rpath directories during a cross link. I think it should treat it like LD_LIBRARY_PATH and only search there during a native link. |
| 21:57.57 | kergoth | yeah, that does seem strange. |
| 22:04.03 | *** join/#oe chrta (~cht@dsl-082-082-232-140.arcor-ip.net) |
| 22:04.36 | chrta | hello |
| 22:05.01 | pb_ | might be worth sending a note to the binutils list to see if anyone knows why it's that way |
| 22:06.34 | CIA-8 | 03kergoth 07 * r1.3081 10openembedded/packages/slugimage/slugimage.bb: Pass the password in the cvs:// uri in the slugimage .bb |
| 22:13.26 | chrta | i'm trying to 'make' the bitbake user manul, but it fails: make: *** [pdf] Error 1 |
| 22:13.53 | kergoth | which tells us nothing |
| 22:13.56 | kergoth | if you have an error message, spit it out |
| 22:14.02 | kergoth | heh |
| 22:14.04 | chrta | where do i find informations how to build a small hello word using bitbake/openembedde ? |
| 22:14.14 | chrta | XMLLINT ./usermanual.xml |
| 22:14.14 | chrta | XMLTO pdf ./usermanual.xml |
| 22:14.14 | chrta | Making portrait pages on a4 paper (210 mm x 297 mm) |
| 22:14.14 | chrta | make: *** [pdf] Fehler 1 |
| 22:14.22 | kergoth | try html. |
| 22:14.28 | chrta | This is the whole output of make |
| 22:14.39 | chrta | thx |
| 22:14.44 | chrta | html works |
| 22:14.47 | kergoth | not sure whats up with pdf |
| 22:14.51 | kergoth | i'll look into it |
| 22:30.50 | CIA-8 | 03kergoth 07 * r1.3083 10openembedded/packages/php/ (7 files in 2 dirs): Add php 4.3.10, and move common bits into a .inc. |
| 22:30.53 | CIA-8 | 03kergoth 07 * r1.3082 10openembedded/site/ (6 files): Add php test results to the i686 linux* site files. |
| 22:38.36 | pb_ | kergoth: ah, cool, you wrote back to Tim. |
| 22:38.48 | pb_ | I meant to do that yesterday but, er, forgot. |
| 22:40.11 | kergoth | yeah, going through my email backlog finaly |
| 22:40.16 | kergoth | ~praise saturdays |
| 22:40.18 | ibot | All hail saturdays! |
| 22:40.34 | pb_ | heh |
| 22:41.26 | pb_ | hrm, there are no trains to the damn airport on Sundays. |
| 22:41.30 | pb_ | guess I have to drive there. |
| 22:42.04 | pb_ | well, no worthwhile trains anyway |
| 22:57.13 | *** join/#oe zecke (~ich@pD9512DF6.dip.t-dialin.net) |
| 22:57.25 | zecke | pb_: have a pleasant flight |
| 23:07.05 | pb_ | zecke: heh, thanks |
| 23:07.11 | pb_ | I gather you were checking up on our stand the other day. |
| 23:09.46 | zecke | pb_: I've seen it ;) |
| 23:10.12 | zecke | pb_: Currently I'm tired and should better get some sleep :) |
| 23:12.07 | pb_ | zecke: very good :-) |
| 23:12.50 | CIA-8 | 03kergoth 07 * r1.3084 10openembedded/packages/mysql/ (4 files in 2 dirs): Upgrade mysql to 4.1.10a. |
| 23:13.19 | pb_ | zecke: reviewing my itinerary for Tuesday, I see that I will be arriving at Berlin Zoo at about 19:00 hours. |
| 23:13.50 | zecke | pb_: and when is your flight scheduled? |
| 23:14.45 | pb_ | zecke: I think the flight is at 22:30, so I guess I should plan to be at Schonefeld by around 21:30. |
| 23:15.09 | koen | 'night all |
| 23:20.15 | CIA-8 | 03kergoth 07 * r1.3085 10openembedded/packages/grep/grep_2.5.1.bb: Add grep 2.5.1. |
| 23:21.20 | *** join/#oe bluelightning (~bluelight@203-173-244-152.akl.ihugultra.co.nz) |
| 23:22.36 | pb_ | zecke: anyway, call me on Tuesday if you want to get together for a coffee or something. My cellphone number is +44 7973 869202, if you don't already have it. |
| 23:25.04 | *** join/#oe marcansoft (1337@80.26.222.222) |
| 23:36.26 | *** join/#oe dkey (~dkey@L0004P09.dipool.highway.telekom.at) |
| 23:45.56 | *** join/#oe ar_ (~ar@port-ip-213-211-231-91.reverse.mdcc-fun.de) |
| 23:46.11 | CIA-8 | 03kergoth 07 * r1.3086 10openembedded/ (6 files in 3 dirs): Clean up the freetype builds.. use autotools_stage_includes. |
| 23:49.40 | zecke | emte: quite an expensive pizza that is |
| 23:52.59 | emte | lol |
| 23:53.18 | emte | pretty sure he would appreciate it |
| 23:53.29 | emte | the thought was mildly amusing |
