**** BEGIN LOGGING AT Tue Jun 04 02:59:59 2013 Jun 04 07:34:57 hi. Jun 04 07:36:28 i have a quick question about excluding packages from an image. It must be possible, but cannot find any documentation on it other than BBMASK. Jun 04 08:07:42 morning all Jun 04 08:09:33 morning bluelightning Jun 04 08:09:40 * zenlinux waves from Tokyo Jun 04 08:09:56 zenlinux: hey... I was thinking it was late for you :) Jun 04 08:10:18 zenlinux: good job on the minnowboard.org intro video btw, just watched it yesterday Jun 04 08:10:38 heh, I have another video to post once I get back Jun 04 08:10:52 in that one I smile a few times and don't act quite so serious :) Jun 04 08:11:22 headed to the Tokyo Hackerspace for a meeting tonight where I'll tell them about Minnow (and a quick Yocto introduction, too) Jun 04 08:12:03 zenlinux: this whole hackerspace thing is still a bit new to me... I've been to one and it was in portland, I think maybe you took us there :) Jun 04 08:12:23 did I? Or was it maybe Beth? Jun 04 08:12:34 can't recall, it was a couple of trips ago Jun 04 08:12:51 was it brainsilo - which is in a converted warehouse space in an industrial district Jun 04 08:13:13 nah it was something in town... they had bits of PC hanging on the walls :) Jun 04 08:14:45 hope you enjoy your time in Japan anyway, I'd love to go there someday Jun 04 08:15:35 this is the best trip ever for me. I can't wait to come back Jun 04 08:17:40 "best trip ever" "can't wait to come back" don't go together! Jun 04 08:18:20 can't wait to come back to Tokyo, not home :) Jun 04 08:18:31 ah :) Jun 04 08:19:12 hi bluelightning Jun 04 08:19:45 hi Net147 Jun 04 08:20:43 bluelightning: I got mesa llvmpipe to run properly. for some reason it's really slow unless I compile with -fno-omit-frame-pointer. it's now around 160 fps instead of like 2 fps. Jun 04 08:21:14 hmm, some kind of weird interaction I guess... Jun 04 08:22:31 bluelightning: I isolated it to swrast_dri.so. if I swap it out with one from Arch Linux system it works fine. I don't think they compile with -fno-omit-frame-pointer. strange. Jun 04 08:22:57 bluelightning: so something in the compiler, some strangeness in Mesa driver or some strange interaction with LLVM library Jun 04 08:23:20 bluelightning: I may do some profiling later to try to isolate the cause Jun 04 08:35:52 good morning Jun 04 08:51:32 morning all Jun 04 08:52:06 some imx6-freescale user here? I have the same issue like this: http://www.mail-archive.com/meta-freescale@yoctoproject.org/msg00033.html Give it a solution for it? Jun 04 09:11:29 Is it possible to use a generated toolchain in Eclipse by setting it up manually instead of ADT? I have no understanding of the automake tools it is using and want it to be clean and simple Jun 04 09:13:08 silviof: only imx5 here ;-) but Otavio will wake up later Jun 04 09:14:07 RagBal: yes it is possible Jun 04 09:14:37 mckoan: thx - than i will wait :-) Jun 04 09:14:38 Is there any guide for that? Because I can't seem to get it functional Jun 04 09:15:34 RagBal: no clue, have you tried with google? Jun 04 09:16:15 Ofc, but I now have found a guide for NetBeans, trying that out Jun 04 11:23:55 khem: ping Jun 04 11:27:55 What is the difference between include and require in build/conf/local.conf? Jun 04 11:38:16 rburton: how's the Mesa 9.1.3 update going? Jun 04 11:42:55 JaMa: I got Mesa 9.1.3 llvmpipe working. had to add -fno-omit-frame-pointer to compile flags though otherwise slow performance. Jun 04 11:48:29 Net147: good work, thanks! Jun 04 11:48:56 JaMa: all the work is in a .bbappend in my private layer at the moment Jun 04 11:48:58 I'm building updated mesa+wayland atm Jun 04 11:49:35 JaMa: do the LLVM patches look ok? Jun 04 11:50:12 did you had to specify which llvm-config should be used? because here it was missing 2.9 suffix to include dir when llvm-config --cflags were used (llvm-config2.9 works fine) Jun 04 11:50:31 Net147: yes, I've only squashed 2 of them (you can see it in contrib/jansa/in-test) Jun 04 11:50:51 Net147: I plan to merge them after next jenkins builds Jun 04 11:51:51 JaMa: I just add export WANT_LLVM_RELEASE = "3.2" Jun 04 11:52:13 JaMa: I am testing with LLVM 3.2 Jun 04 11:53:19 JaMa: I had problems with LLVM 2.9 due to missing symbol. I think may have to do with LLVM 2.9 being built with cmake. Jun 04 11:55:14 JaMa: I observed building with autotools creates an all in one libLLVM-*.so with static .a for each component. building with cmake creates *.so for each component but doesn't create an all in one libLLVM-*.so Jun 04 11:55:44 JaMa: it has been recommended to link LLVM was shared library for Mesa due to various bugs using static LLVM Jun 04 11:56:55 JaMa: my .bbappend so far - http://pastebin.com/3YwJZ8m1 Jun 04 12:00:28 thanks I'll convert them to PACKAGECONFIG options for mesa in master Jun 04 12:00:43 so that we can enable galium and llvmpipe with much smaller bbappends later Jun 04 12:01:09 JaMa: I will submit the LLVM shared library cross compile fix for Mesa upstream Jun 04 13:05:43 JaMa: note that some configure options are not recognized anymore in Mesa: --disable-glu, --disable-glw, --disable-glut. build will still continue but it will just show warning. Jun 04 13:27:59 gonzzor: "include" will include the file if it exists and not error if it doesn't; require will include the file and error if it doesn't exist Jun 04 13:28:29 gonzzor: so we tend to use require unless we expect the build to continue if the file isn't there Jun 04 13:42:56 Hi, how to "fill" one file into a static ubi-volume? Jun 04 13:44:29 rburton, Net147: please check last 9 commits http://git.openembedded.org/openembedded-core-contrib/log/?h=jansa/mesa it's based on our and Net147's .bbappends, still WIP Jun 04 13:44:45 Crofton: Crofton|work Jun 04 13:44:53 gm Jun 04 13:44:57 gm. Jun 04 13:45:23 gnuradio hackfest this weel Jun 04 13:45:27 Net147: can I merge your llvm-cross.patch to it? I would wait for you to submit it upstream, so that we can have correct Upstream-Status Jun 04 13:45:31 Crofton: cool Jun 04 13:45:50 JaMa, we see the source of the gnuradio issue you reported Jun 04 13:46:50 Crofton: great Jun 04 13:49:56 JaMa: it's submitted upstream. check on mesa-dev mailing list Jun 04 13:50:10 JaMa: albeit, I added a bit more to it Jun 04 13:51:09 ok, thanks Jun 04 13:51:19 JaMa: http://lists.freedesktop.org/archives/mesa-dev/2013-June/040167.html and http://lists.freedesktop.org/archives/mesa-dev/2013-June/040168.html Jun 04 13:51:35 JaMa: rburton added the first one to his 9.1.3 update Jun 04 13:52:44 yup my branch is based on his mesa/wayland/gtk upgrade Jun 04 13:54:29 JaMa: there are more gallium drivers than swrast. r300,r600,radeonsi,nouveau,svga,swrast for example. r600 needs LLVM 3.3 Jun 04 13:54:55 JaMa: LLVM 3.3 needs valgrind-native recipe I think which doesn't exist yet Jun 04 13:56:25 JaMa: LLVM 3.3 work in progress based on RC2 - http://pastebin.com/Y48B3YAC Jun 04 14:00:57 Net147: people can extend GALLIUMDRIVERS list in .bbappend like we do for DRIDRIVERS Jun 04 14:01:22 JaMa: I see Jun 04 14:01:29 JaMa: good work Jun 04 14:03:47 JaMa: I don't see -fno-omit-frame-pointer flag in your changes Jun 04 14:06:27 When I create a ubi-static-volume with "ubimkvol", how to write a kernel-image into that new static-ubi-volume? Jun 04 14:21:59 Net147: it's there now, but I'm still waiting for test build to finish Jun 04 14:22:11 so consider this only as WIP to see what I plan to do Jun 04 14:25:26 is there a way to trace the files that set and modiry DISTRO_FEATURES? Jun 04 14:25:32 *modify Jun 04 14:25:52 JaMa: I wonder if we should only apply the flag if using llvmpipe Jun 04 14:27:24 Net147: it would be nice to understand why this flag slows whole thing down, then we could decide if it's good or bad to enable it for all or only llvmpipe builds Jun 04 14:27:58 JaMa: yes. I wonder if it's related to the GCC compiler version used. Jun 04 14:28:12 JaMa: since other distributions compile it without the flag and work fine Jun 04 14:29:28 JaMa: more specifically, it is something inside swrast_dri.so Jun 04 14:30:16 JaMa: omitting frame pointer is a compiler optimisation that is supposed to speed things up Jun 04 14:32:58 JaMa: --disable-glu --disable-glw --disable-glut are not recognized anymore in Mesa 9.1.3 so I wonder if we should drop those flags in mesa.inc Jun 04 14:35:43 yes we should, I'll drop them in next patch Jun 04 14:39:35 bluelightning: Thanks Jun 04 15:00:04 YPTM: Saul here Jun 04 15:00:07 YPTM: Mark is here Jun 04 15:00:09 YPTM: jzhang on the call Jun 04 15:00:29 YPTM: welcome to the meeting, please let me know who's on the bridge. Thanks! Jun 04 15:00:35 YPTM: Tom Z here Jun 04 15:00:36 YPTM: Beth just joined Jun 04 15:00:37 YPTM: Paul Eggleton is on the call Jun 04 15:00:41 Song_Liu: Richard is on the call Jun 04 15:00:52 YPTM: Kevin Strasser is dialing in Jun 04 15:00:56 YPTM: Corneliu joined Jun 04 15:01:59 YPTM: Denys is here Jun 04 15:02:24 YPTM: Laurentiu Palcu joined Jun 04 15:03:19 YPTM: Björn Stenberg joined the call Jun 04 15:03:44 YPTM: belen is on the call Jun 04 15:04:20 YPTM: Any opens? Jun 04 15:05:11 YPTM: nitin is on the call Jun 04 15:08:26 joined YPTM Jun 04 15:10:33 https://wiki.yoctoproject.org/wiki/Planning#Roadmaps_and_Schedules Jun 04 15:12:14 darren joined YPTM Jun 04 15:14:28 joined YPTM Jun 04 15:18:50 isn't gcc flagging a bit short-sighted? Jun 04 15:19:12 yeah it hurts to rebuild everything, but you *did* replace the compiler after all Jun 04 15:20:23 we have e500 builds, but haven't started testing 4.8 yet Jun 04 15:20:39 Zagor: its a question you can't win on Jun 04 15:20:49 its also configurable Jun 04 15:21:19 ya.. versioning gcc 4.8.1 vs 4.8.0 is likely a loss.. but 4.7 to 4.8 is likely a win.. Jun 04 15:21:32 that was why my suggestion on using GCCVERSION = ... Jun 04 15:21:39 it covers only the main version string Jun 04 15:21:49 it's the e500 and e500v2 tune BTW Jun 04 15:24:43 and I've had about 10 people tell me that gcc 4.8.1 is now out.. :P (I think they're hoping that will fix the ICE...) but I havn't even looked at that, I usually defer to Khem Jun 04 15:32:32 YPTM: thank you all for joining the meeting. You all have a nice day/evening Jun 04 15:32:42 thanks Song_Liu Jun 04 15:36:50 morning Jun 04 15:45:10 sgw1: did you have something more you wanted to talk about, or can I go make dinner? :) Jun 04 15:45:40 Zagor: I think we are good, if you can just send a list of packages you have pending that would be great. Jun 04 15:45:51 yep, I'll do that Jun 04 15:58:31 morn' Jefro Jun 04 15:59:17 mranostay good morning - is it grey up there? thick fog down here (coastal northern CA) Jun 04 16:01:11 oh i keep forgeting you are in SC not JF Jun 04 16:01:43 blue skies today Jun 04 16:02:04 suppose to be good weather for the whole weeke Jun 04 16:02:07 *week Jun 04 16:09:16 Darren: I like to know if you have decided on any format of configuring disk images Jun 04 16:10:00 Ramana_, do you mean partitioning or some other aspect? Jun 04 16:10:49 yes partitioning Jun 04 16:11:01 Ramana_, tomz1 is looking into that via https://bugzilla.yoctoproject.org/show_bug.cgi?id=3847 Jun 04 16:11:01 as part of bug 3847 Jun 04 16:11:02 Bug 3847: enhancement, Medium, 1.5 M2, tom.zanussi, IN PROGRESS DESIGN , New partitioning description and tooling Jun 04 16:11:04 Bug https://bugzilla.yoctoproject.org/show_bug.cgi?id=3847 enhancement, Medium, 1.5 M2, tom.zanussi, IN PROGRESS DESIGN , New partitioning description and tooling Jun 04 16:11:08 right :-) Jun 04 16:11:26 Ramana_, we haven't made a lot of progress on that yet Jun 04 16:11:43 Ramana_, tomz1 will be picking that up in the next few weeks I believe Jun 04 16:12:23 Darren, ok, I am also working on making a common configuration file format. Jun 04 16:12:42 Ramana_, please collaborate with tomz1 on that bug Jun 04 16:12:55 Darren, I have come up with a format and a parser, so wanted to check with you Jun 04 16:13:07 Darren, OK Jun 04 16:14:18 Darren, thanks I will check with tomz1, what is full name of tomz1? Jun 04 16:14:36 Ramana_, see the bug owner Jun 04 16:15:08 Darren, got it thanks Jun 04 16:16:57 Ramana_, thanks for letting us know, great to work together and not duplicate effort Jun 04 16:18:35 tomz1, Hi ru there? **** BEGIN LOGGING AT Tue Jun 04 17:53:05 2013 Jun 04 17:58:02 otavio: hi there Jun 04 18:14:39 otavio: Eric has done the 4.0.0 release Nitrogen6X Jun 04 18:14:41 otavio: it seems to work. Jun 04 18:24:46 wmat: hey canadian Jun 04 18:25:19 mranostay: hey portlandian Jun 04 18:26:10 haha **** BEGIN LOGGING AT Tue Jun 04 18:36:00 2013 Jun 04 19:04:46 _darknighte_: welcome to the jungle Jun 04 19:05:10 <_darknighte_> mranostay: what's up? Jun 04 19:05:23 <_darknighte_> my znc bouncer just freaked out a bit there... Jun 04 19:08:18 darknighte_znc: make up your mind! :P Jun 04 22:31:25 I look forward to bitbake requiring python 2.7 Jun 04 22:45:21 RP: we should think about adding a new event for operation against a finalized config datastore, which bitbake then uses for config, so we can avoid manual finalizing in ConfigParsed handlers for particular cases Jun 04 22:45:58 RP: though, the event would have to carry both finalized and non-finalized datastores with it, since only non-finalized would affect recipes Jun 04 22:46:01 * kergoth ponders Jun 04 22:48:29 kergoth: that is the dilemma. The finalised store isn't useful to bitbake and could end up confusing users even more since writes to it won't have any affect :/ Jun 04 22:50:25 kergoth: I wish we could move to 2.7 too... Jun 04 22:56:23 the finalized store is extremely useful to bitbake. without it, we can't combine overrides with PREFERRED_PROVIDER & the like Jun 04 22:56:53 but yeah, confusion is a definite concern Jun 04 22:56:55 * kergoth shrugs Jun 04 23:07:50 hi kergoth **** BEGIN LOGGING AT Tue Jun 04 23:42:28 2013 Jun 05 00:02:11 hello Jun 05 00:02:24 how can i clean the sstate for a recipe? Jun 05 00:03:33 rm -rf ? :) Jun 05 00:03:45 bitbake -c cleansstate Jun 05 00:04:02 clean -- clean the working directory Jun 05 00:04:09 cleansstate -- perform clean & remove sstate Jun 05 00:04:17 cleanall -- preform cleansstate & remove download(s) Jun 05 00:27:13 thank you very much Jun 05 00:44:41 hey fray did I sayd thank you ? Jun 05 00:44:53 oh I did Jun 05 00:57:02 np **** ENDING LOGGING AT Wed Jun 05 02:59:57 2013