**** BEGIN LOGGING AT Wed Sep 15 02:59:57 2010 Sep 15 06:08:29 gm everyone! Sep 15 06:16:33 khem, not sure what is going on with gcc/eglibc for nslue2be: with the testing branch it fails with a gcc internal error, it did work on last weeks' testing Sep 15 06:16:35 see http://tinderbox.openembedded.net/packages/754585/ Sep 15 06:16:46 ../include/time.h:43: internal compiler error: Segmentation fault Sep 15 06:19:23 nslu2le complies fine, will retest to see if it is reproducible Sep 15 06:58:48 hi, i have an issue building perl-5.10.1 I get: | LD_LIBRARY_PATH=/home/frans/oe/tmp_minimal/work/armv5te-oe-linux-gnueabi/perl-5.10.1-r10/perl-5.10.1${LD_LIBRARY_PATH:+:}$LD_LIBRARY_PATH ./miniperl -Ilib make_patchnum.pl Sep 15 06:58:48 | /bin/sh: ./miniperl: No such file or directory Sep 15 06:59:14 dont really understand the LD_LIBRARYPATH part Sep 15 06:59:22 and can't find miniperl Sep 15 06:59:27 this is minimal/sheevaplug Sep 15 06:59:42 good morning Sep 15 07:06:17 hi mckoan Sep 15 07:23:30 eFfeM_work, ${LD_LIBRARY_PATH:+:} looks broken. As far as i can tell, it will just translate to a colon Sep 15 07:24:16 grg, yeah thought that would be wrong, but the root cause is that miniperl apparently is not found Sep 15 07:24:52 have build this at home and log didn't go to tinderbox (actually it may well be that bitbake -k & tinderbox do not go together well) Sep 15 07:25:20 on second thought.... ${LD_LIBRARY_PATH:+:}$LD_LIBRARY_PATH = "" if ld_library_path is empty, it will append ld_library_path with a colon delimiter otherwise Sep 15 07:25:33 will peek at it tonight or maybe during some dead minutes during the day Sep 15 07:25:57 miniperl is a bootstrap perl, right? i don't know perl at all Sep 15 07:26:26 yes Sep 15 07:26:56 don't know too much about it but some of the libperl recipes have deps on perl modules that are not in 5.8.8 Sep 15 07:27:37 btw is Roman in this channel? don't know his nick (if any) he seems to be the perl wiz Sep 15 07:28:59 not that i know of Sep 15 07:30:05 i had a peek at the p4 stuff you mentioned, that is really old, not too happy on binary-only things either Sep 15 07:30:23 yeah binary only stuff scares me for embedded systems Sep 15 07:30:47 i only saw it because it halted by bitbake -k world Sep 15 07:31:04 lots of failures seem to halt bitbake -k Sep 15 07:31:25 (md5sum mismatch in this case) Sep 15 07:31:50 i'd say it would move on anyway Sep 15 07:34:25 if base_do_fetch fails, bitbake exits regardless of -k flag. Sep 15 07:34:36 anyway, i'm off now... Sep 15 07:34:40 bye Sep 15 08:20:54 Hello Sep 15 08:21:44 hi Sep 15 08:22:47 do we have any expert of CMake over here Sep 15 08:23:01 can anybody tell me how link,txt is generated when we use cmake Sep 15 08:31:58 khem, gcc/eglibc/nlsu2be issue did not reappear after removing tmpdir guess there was something strange around due to the nslu2le build I did before in the same tree Sep 15 12:41:10 hi eFfeM_work, thanks for the tip yesterday, regarding nkf target recipe Sep 15 12:41:26 you're welcome :-) Sep 15 12:41:36 it worked for me and now its building fine Sep 15 12:42:00 i added TARGET_CC_ARCH += "${LDFLAGS}" in the recipe Sep 15 12:42:07 should i push the change? Sep 15 12:42:14 yes, please Sep 15 12:42:26 hi effem Sep 15 12:42:39 hi woglinde (and everyone else ofc) Sep 15 12:42:46 okay Sep 15 13:14:02 hi what's work/iwmmxt-oe-linux-gnueabi ? Sep 15 13:14:38 angelox_123, I bet it doesn't find mplayer because of that: Sep 15 13:14:40 work/iwmmxt-oe-linux-gnueabi Sep 15 13:14:55 apparently you've build for a machine iwmmxt Sep 15 13:15:07 yes Sep 15 13:15:07 yes for a1200 uses it Sep 15 13:15:27 a1200/deploy/ipk$ ls Sep 15 13:15:27 a1200 all armv5te iwmmxt Packages Packages.filelist Packages.gz Packages.stamps x86_64 Sep 15 13:15:38 mplayer is not found during do_rootfs Sep 15 13:15:47 try copy... Sep 15 13:15:53 and it's the only package in iwmmxt Sep 15 13:15:55 bad idea Sep 15 13:16:03 yes;srry Sep 15 13:16:04 that's a bad workarround Sep 15 13:17:09 strange.. Sep 15 13:17:19 why? Sep 15 13:17:33 here on my builds the most of packages go on armv5te ... Sep 15 13:17:38 how many iwmmxt machines exists Sep 15 13:17:47 because else there is a dead simple workarround Sep 15 13:17:54 put that to machine arch Sep 15 13:18:07 the news motorola devices uses too.. Sep 15 13:20:04 ah ok all zaruses Sep 15 13:21:05 yes... Sep 15 13:21:14 waiting some help here... Sep 15 13:24:01 angelox_123, I've really no time for that sorry Sep 15 13:24:06 try to build it yourself Sep 15 13:24:18 and then report the error on the openembedded mailing list Sep 15 13:24:45 because a1200 has iwmmxt as extra arch Sep 15 13:24:51 so I don't know why it doesn't work Sep 15 13:26:42 run.do_rootfs.2224: for arch in all any noarch arm armv4 armv4t armv5te a1200 Sep 15 13:26:47 no idea why it does that Sep 15 13:44:45 gm Sep 15 13:47:07 ? Sep 15 13:47:20 hi likewise Sep 15 13:52:52 hi stefan Sep 15 13:52:58 hi woglinde Sep 15 13:57:00 03Tom Rini 07org.openembedded.dev * radd861a395 10openembedded.git/recipes/apr/ (4 files): Sep 15 13:57:00 apr, apr-util: Remove 1.2.* versions Sep 15 13:57:00 These don't patch and aren't pinned by anyone. Sep 15 13:57:00 Signed-off-by: Tom Rini Sep 15 13:57:11 03Tom Rini 07org.openembedded.dev * r3fb4bb5b92 10openembedded.git/conf/machine/p2020ds.conf: Sep 15 13:57:11 p2020ds: Pull in tune-ppce500v2.conf Sep 15 13:57:11 Signed-off-by: Tom Rini Sep 15 13:57:12 03Tom Rini 07org.openembedded.dev * r0278c78046 10openembedded.git/conf/distro/include/angstrom-eglibc.inc: Sep 15 13:57:13 angstrom-eglibc.inc: Update SPE check Sep 15 13:57:13 Use BASE_PACKAGE_ARCH to determine if we should be SPE, or not Sep 15 13:57:13 which is what sane-toolchain.inc does. This saves us having to Sep 15 13:57:13 update this file every time a new e500 machine is added. Sep 15 13:57:13 Acked-by: Koen Kooi Sep 15 13:57:13 Signed-off-by: Tom Rini Sep 15 13:57:14 03Tom Rini 07org.openembedded.dev * r8e2d76c156 10openembedded.git/ (5 files in 4 dirs): Sep 15 13:57:14 mpc8544ds: Add this machine Sep 15 13:57:15 Signed-off-by: Tom Rini Sep 15 14:05:23 hello. i am getting these errors: NOTE: Couldn't find shared library provider for libgcc_s.so.1 and NOTE: Couldn't find shared library provider for libstdc++.so.6 Sep 15 14:05:28 which packages provide these? Sep 15 14:06:51 or to put it another way: how can i track which package provides a certain file? Sep 15 14:35:50 khem, eFfeM_work: Had some fetch troubles with my first bitbake -k world steps. Going better now. 7400 from 71000. Sep 15 14:42:25 stefan_schmidt: cool Sep 15 14:42:28 http://oreilly.com/store/ddccc.html hmm Sep 15 14:42:49 stefan_schmidt: I could only do 51000 before I ran out of disk even with rm_work Sep 15 14:43:07 khem: wow Sep 15 14:43:08 I think if I disable pstage I could get some more Sep 15 14:43:27 khem: how much space was it? Sep 15 14:43:34 90G Sep 15 14:43:44 ouch Sep 15 14:43:47 thats *with* rm_work? Sep 15 14:43:50 yes Sep 15 14:43:52 you could try using the completion scheduler Sep 15 14:43:58 khem: ok, I should have ~160GB over here for it Sep 15 14:43:59 so it completes each recipe before moving on to the next Sep 15 14:44:02 might slow the build down though Sep 15 14:44:22 BB_SCHEDULER = "completion" rather than BB_SCHEDULER = "speed" (default) Sep 15 14:44:37 kergoth: eventually it will hit the disk full isnt it Sep 15 14:45:12 I will give it a shot with pstage disabled then I can save some space Sep 15 14:45:31 before I get a bigger disk Sep 15 14:46:35 yeah, scheduler just makes it get to the rm_work faster for each recipe, the end result will be the same Sep 15 14:56:06 kergoth_: if it merges with this: http://oreilly.com/oreilly/cs/b2g1f.html then that's an extremely good deal Sep 15 15:02:03 likewise: heh, if only -- they don't let their promo codes combine, it makes me sad Sep 15 15:02:48 kergoth_: thanks anyhow, I ordered 6 Sep 15 15:03:06 the commission should be yours :-) Sep 15 15:03:36 likewise windows7? Sep 15 15:04:01 woglinde: ?? Sep 15 15:04:34 the link Sep 15 15:04:44 there is a windows7 book shown Sep 15 15:05:08 woglinde: oh, it's unrelated to the discount though :-) Sep 15 15:05:14 likewise: 6? sheesh :) nice Sep 15 15:05:22 i love their ebook deal of hte day, makes them worth buying Sep 15 15:05:25 $10 is nice and reasonable Sep 15 15:05:38 woglinde: trust me, no windows 7 on my bookshelf (not yet anyway) Sep 15 15:06:04 hi kergoth Sep 15 15:06:05 kergoth_: depends on the book, but the cookbooks seem nice enough for learning stuff you "need on the side". Sep 15 15:08:02 likewise: now if only i'd get around to actually *reading* all the ebooks and physical books I've bought... :) Sep 15 15:08:06 hey pb_ Sep 15 15:25:22 kergoth_: I have a shitload (read:box) full of paper o'reilly python books (for diving in bitbake): never got to it. Sep 15 15:25:56 hehe Sep 15 15:26:48 i started using goodreads -- my 'to-read' shelf has 200 books in it at the moment Sep 15 15:26:54 (course, i don't own all those, but still) Sep 15 15:29:14 likewise: the git version of efikamx kernel does not clone git form me Sep 15 15:29:27 likewise: I have set DEFAULT_PREFERENCE to -1 for now Sep 15 15:41:27 ~heil debian's apt-get dist-upgrade Sep 15 15:41:38 ~hail debian's apt-get dist-upgrade Sep 15 15:41:39 * ibot bows down to debian's apt-get dist-upgrade and chants, "I'M NOT WORTHY!!" Sep 15 15:41:44 mckoan? Sep 15 15:41:48 aptitude Sep 15 15:41:54 makes the job slight easier Sep 15 15:42:08 woglinde: I'm a rude guy ;-) Sep 15 15:42:23 rude dude Sep 15 15:42:23 mckoan he I used dselect for a long time Sep 15 16:03:24 khem: indeed, I can reproduce, there is something strange happening indeed. if I run the clone from the shell it works, though. Sep 15 17:27:22 likewise: I will push the fix and let u figure out issues with svn Sep 15 17:27:27 errr git Sep 15 17:31:45 khem: yes, I'll try to. Sep 15 17:35:48 hmmm someone should let CIA know of master branch Sep 15 17:35:57 if you push into master CIA is not catching it Sep 15 17:40:03 hi Sep 15 17:45:45 hi Sep 15 17:46:00 could anyone review/merge this patch? http://patchwork.openembedded.org/patch/2943/ Sep 15 17:48:56 anarsoul: I have it staged Sep 15 17:49:04 khem: ok Sep 15 17:49:18 anarsoul: as I said before it takes a little while Sep 15 17:49:33 khem: ok, just wanted to make sure it is not lost Sep 15 17:49:42 should I change status of patch? Sep 15 17:50:43 anarsoul: I will do that Sep 15 17:50:55 ok Sep 15 18:58:24 re Sep 15 19:04:28 does PHP need all the statically linked .a libraries in the /usr/lib folder, or are the library methods available within the php executable? Sep 15 19:13:48 khem wrt your remark on ML "do you need these m4 recipes " I hope you mean p4 not m4 (as the thread was on p4) Sep 15 19:14:00 yes I saw it after sending Sep 15 19:14:36 even though m and p are far away on keyboard Sep 15 19:16:23 well m4 is something we all know Sep 15 19:16:56 this is the p4 thing: http://www.perforce.com/ Sep 15 19:17:01 :) yes Sep 15 19:17:08 I just wanted to keep focus on patch Sep 15 19:17:13 commercial sw with a binary trial license Sep 15 19:17:19 the amd note suggest CosmicPenguin involvement Sep 15 19:17:52 CosmicPenguin: care to comment ? Sep 15 19:18:04 trial is time bound ? Sep 15 19:18:06 or what Sep 15 19:18:29 who in the what now? Sep 15 19:18:44 gm CosmicPenguin Sep 15 19:18:53 CosmicPenguin: p4 recipes in OE need some love Sep 15 19:19:18 CosmicPenguin: if we still need them Sep 15 19:19:20 without love they are bound to die :-) Sep 15 19:19:21 you mean recipes that use P4 or that build P4? Sep 15 19:19:33 I think p4 itself Sep 15 19:19:38 yes Sep 15 19:19:39 I'm not on the mailing list, so i am lacking a lot of context Sep 15 19:19:46 and it would be nice to know what all use p4 Sep 15 19:19:48 ah ok Sep 15 19:20:38 CosmicPenguin: http://thread.gmane.org/gmane.comp.handhelds.openembedded/36962 Sep 15 19:21:14 Ah, okay Sep 15 19:21:36 You can probably zap those unless anybody is seriously using a p4 fetcher Sep 15 19:22:05 no one complained yet Sep 15 19:22:20 but we could move to obsolete for a while Sep 15 19:22:25 though I disagree about the adversion to binary blob downloads - sometimes you gotta do what you gotta do Sep 15 19:22:25 if that is more desired Sep 15 19:22:56 i know, sometimes you have no choice (e.g. wireless firmware) Sep 15 19:23:22 Citrix comes to mind too for thin clients Sep 15 19:23:41 But anyway, all that stuff was added by AMD for our internal P4 stuff - probably not interesting for the community Sep 15 19:23:55 I doubt anybody is running a public P4 repo.. :) Sep 15 19:24:32 not that I am aware of, actually never heard of it before the post showed up. Sep 15 19:24:41 hm planet.linuxtogo.org is no more? Sep 15 19:24:56 ='( Sep 15 19:24:57 it should be Sep 15 19:25:00 * florian checks Sep 15 19:25:16 oops Sep 15 19:29:42 while building an image I get: libc6-dev: unsatisfied recommendation for localedef-dev Sep 15 19:29:55 is there an easy way to find out who drags in libc6-dev in the first place ? Sep 15 19:33:41 bitbake -g ? Sep 15 19:34:03 good idea Sep 15 19:34:09 thanks Sep 15 19:39:58 bitbake -g did not help, i use eglibc, libc6 as string does not appear in the .dot file eglibc-dev does in package-depends.dot (not in task-depends.dot) and only at the left side Sep 15 19:41:12 libc6-dev is the renamed version Sep 15 19:41:51 yeah, that's why i searched for eglibc-dev Sep 15 19:42:15 but that does not appear at the right hand side of a -> line in the .dot file Sep 15 19:43:06 i did a full rebuild last night (after the update for libgcc and libstdc++) , didn't seem to have this before Sep 15 19:43:52 actually there were other things as well, seemed it build ncurses but there was no ipk for it, (but the work dir was there) Sep 15 19:44:34 and (for likewise :-) ) there was a day wrap in it, started the build around 23.00 pm CET Sep 15 19:45:31 will try bitbake -vvvv -D -D -D or so Sep 15 19:45:41 to find out about the libc6-dev Sep 15 19:56:00 btw anyone an idea on how to override SRC_URI for a specific arch? see http://thread.gmane.org/gmane.comp.handhelds.openembedded/36893 Sep 15 19:57:39 eFfeM im sure there is a way to do it correctly; but if you have control over the download locations, you could just add ${ARCH} in the bb? Sep 15 19:58:56 eFfeM: that should work. it doesn't. therefore its a bug. Sep 15 19:59:08 * kergoth_ hasn't had a chance to look into it yet Sep 15 19:59:55 warflyr: what do you mean with adding${ARCH} in the bb, (see the link on what I did) Sep 15 20:00:25 kergoth, ah thanks, for the update, was somewhat afraid it was lost Sep 15 20:00:37 khem Sep 15 20:00:39 ping Sep 15 20:00:56 ka6sox-work: hellpo Sep 15 20:01:11 hiya, how goes the ld issue? Sep 15 20:01:32 guess the cached data needs to be cleaned, but didn't know enough of python and the logic to diagnose this further Sep 15 20:02:06 ka6sox-work: does not happen with native install Sep 15 20:02:22 eFfeM i remember seeing that before SRC_URI_arch; i think youve got that right Sep 15 20:02:24 I have local installation of 10.04 Sep 15 20:02:51 64bit or 32bit? Sep 15 20:04:22 03Roman I Khimov 07org.openembedded.dev * r7a9ad0037c 10openembedded.git/recipes/dnsmasq/ (5 files): Sep 15 20:04:22 dnsmasq: remove versions prior to 2.52 Sep 15 20:04:22 These have known unpatched security issues, 2.52 is in the repo for quite some Sep 15 20:04:22 time, anyone who cares should be using that already. Sep 15 20:04:22 Signed-off-by: Roman I Khimov Sep 15 20:04:22 Acked-by: Frans Meulenbroeks Sep 15 20:04:24 03Henry von Tresckow 07org.openembedded.dev * ra5418685b3 10openembedded.git/recipes/dnsmasq/ (dnsmasq-dbus_2.55.bb dnsmasq_2.55.bb): Sep 15 20:04:24 dnsmasq: add version 2.55 from upstream Sep 15 20:04:24 Signed-off-by: Henry von Tresckow Sep 15 20:04:24 Acked-by: Roman I Khimov Sep 15 20:04:25 Signed-off-by: Roman I Khimov Sep 15 20:04:39 eFfeM hows OE coming on the nios2? Sep 15 20:05:23 ka6sox-work: both 32 bit and 64 bit Sep 15 20:05:25 well the toolchain, kernel and busybox builds, haven't tried much other stuff yet Sep 15 20:05:46 need to update it but need to finish another project first Sep 15 20:06:03 khem, is your 64bit version of glibc and the one on garnet different? Sep 15 20:06:12 ~seen ant Sep 15 20:06:27 ant was last seen on IRC in channel #oe, 561d 19h 54m 40s ago, saying: 'nite all'. Sep 15 20:06:32 ka6sox-work: no Sep 15 20:06:32 ~seen ant_ Sep 15 20:06:34 ant_ <~andrea@host84-105-dynamic.50-82-r.retail.telecomitalia.it> was last seen on IRC in channel #oe, 1d 20h 56m 29s ago, saying: ''nite'. Sep 15 20:06:59 khem, version or md5sumj? Sep 15 20:07:26 eFfeM is there some module/vhd in quartus or nios2 that gives you an arm core or something? Sep 15 20:07:58 eFfeM or do you make that module with the compile kernel and/or rootfs? Sep 15 20:10:14 ka6sox-work: version Sep 15 20:10:31 okay what version do you need? Sep 15 20:10:34 ka6sox-work: but there are different packages on my system Sep 15 20:10:38 let me see if its available Sep 15 20:10:50 stefan__, you here? Sep 15 20:11:13 warflyr: i'm unaware of an arm core although there is talk about it Sep 15 20:11:32 warflyr: nios2 itself is a different cpu arch with its own instruction set Sep 15 20:12:02 eFfeM aahhh, i was wondering that Sep 15 20:12:48 ive only put verilog on nios2s before Sep 15 20:13:32 warflyr: i didn't construct the fpga code, but think it is made with sopc builder or so Sep 15 20:13:37 still need to peek at that Sep 15 20:14:03 ohhh, i remember how it works now; nios2 is the instruction put on whatever fpga was on that dev board ( i think it was a spartan2 or 3) Sep 15 20:14:11 spartan III Sep 15 20:14:53 i've heard about cortex m3 and a8 being mentioned, not sure if that was for altera or xilinx, but afaik it is not available yet Sep 15 20:15:04 and m3 might be w/o mmu Sep 15 20:15:55 khem, what version do you need? Sep 15 20:17:44 ka6sox-work: no worries I dont have root cause yet Sep 15 20:21:50 khem, okay... Sep 15 20:23:38 ka6sox-work: back now. whats up? Sep 15 20:24:46 stefan__, is there anything fancy about the wiki? Sep 15 20:24:53 any fancy plugins? Sep 15 20:25:23 ka6sox-work: which wiki? Sep 15 20:25:33 nm Sep 15 20:25:39 sorry.. Sep 15 20:26:00 wrong guy? At least I got the feeling :) Sep 15 20:26:22 ya, sorry. Sep 15 20:27:08 np :) Sep 15 20:30:09 So, bitbake gurus (kergoth_?) why do we need: http://pastebin.com/t17DeKu6 Sep 15 20:30:18 bitbake 1.8.x, bitbake master, presumably bitbake 1.10.x Sep 15 20:31:48 kernel_ is added by bitbake Sep 15 20:32:00 rather whatever the name of bbclass Sep 15 20:32:41 but its confusing Sep 15 20:33:53 Er Sep 15 20:34:03 We do kernel_do_compile and EXPORT_FUNCTIONS it Sep 15 20:34:09 and it's no do_compile, for the fn Sep 15 20:34:19 Is it intended that to modify the depends part we have to call it do_compile Sep 15 20:34:27 (and it's just confusing) Sep 15 20:34:40 Or is it a bitbake bug that the fn is remapped, but not the depends/ Sep 15 20:40:49 can a file be put into two pacakges Sep 15 20:40:51 ? Sep 15 20:41:10 eFfeM: you mean two ipks Sep 15 20:41:10 looking at this in mythtv Sep 15 20:41:11 bb.data.setVar("FILES_lib%s%s" % (m, pv), "${libdir}/lib%s-%s.so.*" % (m, pv), d) Sep 15 20:41:11 bb.data.setVar("FILES_lib%s%s-dev" % (m, pv), "${libdir}/lib%s-%s.*" % (m, pv), d) Sep 15 20:41:17 khem, yes Sep 15 20:41:22 usually no Sep 15 20:41:33 this suggests the .so.* files end up in both Sep 15 20:41:42 bad Sep 15 20:41:46 but they only end up in -dev and no regular lib is created Sep 15 20:41:59 Tartarus: I wondered too Sep 15 20:42:10 khem: k. kergoth_ ? ;) Sep 15 20:43:17 Tartarus: it could be that function is remapped to usuall do_* stuff Sep 15 20:44:09 planet.linuxtogo.org is back Sep 15 20:44:10 about an hour and I need to do an rsync of the git server. Sep 15 20:44:32 guys keep full uptodate copies Sep 15 20:44:41 just in case something happens to melo Sep 15 20:44:55 I have two trees uptodate Sep 15 20:46:29 khem, I'm hoping nothing happens to melo...but I'm just making sure we have backups Sep 15 20:46:40 sure Sep 15 20:46:52 CIA does not catch commits to master Sep 15 20:47:00 who can help fixing that Sep 15 20:47:28 what script is used there? Sep 15 20:47:53 I'm sure branch is hardcoded inside hook script or cia script Sep 15 20:48:35 well other branches show up well Sep 15 20:48:40 stable too Sep 15 20:48:46 but master is recent addition Sep 15 20:48:53 may be CIA needs to update Sep 15 20:49:46 ah.. Sep 15 21:17:21 about 15mins from now I'm starting an upgrade of the git server Sep 15 21:21:20 ka6sox-work: there is a /README with some info about what is on the server Sep 15 21:22:24 cbrake, where? Sep 15 21:23:48 ka6sox-work: on the server Sep 15 21:24:03 looking Sep 15 21:24:58 ka6sox-work: it has info on what services are running, etc Sep 15 21:25:04 not sure if it is current Sep 15 21:25:29 looking Sep 15 21:26:18 ka6sox-work: the bandwidth to .nl has been recovered. 600kB/s now. Thanks for the time put into this! Sep 15 21:26:59 likewise, ya, I think I might have found something here too...so that should be better. Sep 15 21:27:20 when you were being hammered I realized that there were 15 users doing pulls Sep 15 21:34:48 cbrake, its probably not current...but I see what ports are open Sep 15 21:57:32 my home and end keys dont work when im SSHd into a angstrom build... what typically controls this? **** BEGIN LOGGING AT Wed Sep 15 23:16:14 2010 Sep 15 23:28:22 03Florian Boor 07org.openembedded.dev * r7d909ba426 10openembedded.git/conf/machine/babbage.conf: babbage.conf: Add framebuffer video driver to XSERVER Sep 15 23:28:27 03Florian Boor 07org.openembedded.dev * r21ab39d55f 10openembedded.git/recipes/gpe-conf/gpe-conf_0.2.9.bb: gpe-conf: add 0.2.9 Sep 15 23:34:31 03Tom Rini 07org.openembedded.dev * r25e515ba2f 10openembedded.git/ (3 files in 3 dirs): (log message trimmed) Sep 15 23:34:31 kernel.bbclass: Fix INITRAMFS_IMAGE logic Sep 15 23:34:31 Due to a bitbake bug to be fixed in 1.10.1, EXPORT_FUNCTIONS isn't Sep 15 23:34:31 remapping kernel_do_compile[depends] to do_compile[depends], so Sep 15 23:34:31 make a comment about cleaning this up once we depend on a working Sep 15 23:34:32 bitbake. Next, it needs to add to do_configure (like the users Sep 15 23:34:33 have manually set a dependancy on) rather than do_compile so that Sep 15 23:37:05 you guys might need to re-commit those...I hope not but Sep 15 23:39:19 oops :) k Sep 15 23:39:56 is it easy to make the repo r/o? just thinking ahead for the next maint window Sep 15 23:40:13 not to my knowledge Sep 15 23:40:46 k Sep 15 23:46:25 install an update hook that blocks everything Sep 15 23:46:35 then chmod +x / -x the hook when enabling/disabling the maintenance window Sep 15 23:46:42 you coudl even have it display a message about why you can't push right now Sep 15 23:47:22 k Sep 15 23:49:43 I just shut things down for now... Sep 16 00:54:16 ka6sox-work: is the maintenance done Sep 16 00:56:47 khem, I wish Sep 16 01:44:25 kergoth: strange issue with svn fetching Sep 16 01:44:39 svn ls http://svn.openmoko.org/trunk/src/target/OM-2007.2/libraries Sep 16 01:44:45 works Sep 16 01:44:51 but when put in a recipe Sep 16 01:45:16 NOTE: Fetch svn://svn.openmoko.org/trunk/src/target/OM-2007.2/;module=libraries;proto=http Sep 16 01:45:25 svn: URL 'http://svn.openmoko.org/trunk/src/target/OM-2007.2/libraries' doesn't exist Sep 16 01:48:44 hmmm missing SRCREV Sep 16 01:48:53 so SRCREV is a must ? Sep 16 01:54:59 seems to be usually Sep 16 01:55:14 if you need head you gotta specify SRCREV=HEAD Sep 16 01:55:24 er Sep 16 01:55:35 proto=http;rev=HEAD Sep 16 01:56:08 And that fills in SRCREV=${AUTOREV} Sep 16 01:56:35 wow. I'm tired. I mean, add SRCREV=${AUTOREV} -- and I think thats how it works Sep 16 01:56:42 time to go home for dinner Sep 16 01:59:07 git is up....bugs and tinderbox are offline for a little while longer Sep 16 02:14:38 i'm getting xtscal: xcb_io.c:506: _XReply: Assertion `!dpy->xcb->reply_data' failed. when running xtscal using xserver-kdrive-fbdev with top of tree OE Sep 16 02:33:53