| paulsherwood | http://fpaste.org/139312/ | 08:42 |
|---|---|---|
| paulsherwood | looks like it built ok, to me? but install failing... | 08:42 |
| paulsherwood | those install-commands are from our original integration of sdl 1 - presumably it's all different now | 08:44 |
| radiofree | these are the install commands from daniels chunk though | 08:44 |
| paulsherwood | weird | 08:45 |
| paulsherwood | i think he may have bodged things, not used full install | 08:45 |
| radiofree | well /usr/smartdevicelink-qt/bin has smartdevicelink core | 08:45 |
| radiofree | so i'm guessing make install does that | 08:46 |
| radiofree | so we don't have to do the copy | 08:46 |
| paulsherwood | ok so i'll scrap that command? | 08:46 |
| * radiofree removes the mkdir + copy stuff from smartdevicelink-qt + smartdevicelink-web | 08:46 | |
| radiofree | yeah, just do make install DESTDIR=$DESTDIR | 08:46 |
| radiofree | oh well | 08:47 |
| radiofree | leave the web one alone | 08:47 |
| radiofree | paulsherwood: this build on x86 though? | 08:47 |
| paulsherwood | yup | 08:49 |
| paulsherwood | still waiting for morph to make up its mind on jetson | 08:49 |
| paulsherwood | in other news, when i create a vm in virtualbox following the baserock instructions these days, it crashes with some network error. then i just go to network settings vi vbox ui, all seems fine, and machine works.... | 08:53 |
| paulsherwood | 2014-10-05 08:39:55 Stratum [lua-runtime] version is 9195292 | 08:53 |
| paulsherwood | 2014-10-05 08:53:20 [Build 286/351] [smartdevicelink-qt] Building chunk smartdevicelink-qt | 08:53 |
| paulsherwood | 13 1/4 minutes thinking time | 08:54 |
| paulsherwood | morph has got big and fat | 08:55 |
| radiofree | did you build with --verbose? | 08:55 |
| paulsherwood | no, i like tidy ouput. there'll be more in the log, though | 08:57 |
| paulsherwood | want to see? | 08:57 |
| radiofree | ok | 08:59 |
| radiofree | maybe it's actually building? | 08:59 |
| radiofree | or do you see "running build commands"? | 08:59 |
| radiofree | smartdevicelink-qt built and installed! | 08:59 |
| paulsherwood | it is building. i'm just saying morph shouldn't spend 13 mins debating what to do | 08:59 |
| paulsherwood | :) | 09:00 |
| radiofree | oh yeah that's bad | 09:00 |
| paulsherwood | w00t! does it work, though? | 09:00 |
| radiofree | i still have a lot to build before i can test | 09:01 |
| radiofree | [Build 1530/1848] | 09:01 |
| radiofree | but at least qt5 is built, so most of the big stuff is done now | 09:01 |
| paulsherwood | ah... you're using an older morph | 09:01 |
| radiofree | god damn it | 09:04 |
| radiofree | ERROR: In staging area /src/tmp/staging/tmpDLn8Qe: running command 'sh -c make' failed. | 09:04 |
| radiofree | smartdevicelink-web | 09:04 |
| radiofree | max jobs again i think | 09:04 |
| paulsherwood | heh | 09:05 |
| * paulsherwood wonders if there's a default splitting rule to apply that will reduce these systems by 50% | 09:12 | |
| paulsherwood | (to save on deploy times) | 09:12 |
| paulsherwood | (and space) | 09:12 |
| radiofree | hurrah smart device link is done | 09:16 |
| radiofree | now onto gtk by the looks of it | 09:16 |
| paulsherwood | yup | 09:18 |
| paulsherwood | but you also have llvm up ahead :) | 09:19 |
| radiofree | why? | 09:21 |
| radiofree | isn't llvm only needed by mesa? | 09:22 |
| radiofree | i'm not sure why we need gtk2? | 09:22 |
| radiofree | i'm removing this | 09:22 |
| paulsherwood | jjardon has been trying to tidy this up | 09:23 |
| radiofree | oh | 09:23 |
| radiofree | glibmm will need it | 09:23 |
| radiofree | i'm building stuff in x-generic | 09:30 |
| radiofree | so i guess i won't need to rebuild llvm | 09:30 |
| radiofree | ya | 09:30 |
| radiofree | y | 09:30 |
| radiofree | but at least i have the smartdevicelink stuff built now, if this build isn't finished by 12.30ish i can at least manually test smartdevicelink | 09:31 |
| paulsherwood | do you have a working bluetooth? | 09:33 |
| radiofree | hmm not sure, easy fix though | 09:36 |
| paulsherwood | http://wiki.baserock.org/guides/baserock-jetson/ i've updated it to use the 14.40 image, crossing my fingers that works | 09:47 |
| radiofree | i don't understand why we need glibmm, gtk | 09:47 |
| radiofree | nothing in that demo stratum build depends on it | 09:48 |
| radiofree | also, if we're going to attempt to composite this in wayland gtk2 isn't going to be helpful.... | 09:48 |
| radiofree | navit has a gtk ui, but we're not using that? | 09:48 |
| paulsherwood | i assume not | 09:48 |
| radiofree | jjardon: how long does gtk2 take to build? | 09:49 |
| paulsherwood | you mean on jetson, or x86? | 09:50 |
| radiofree | in general, let's say x86 | 09:50 |
| radiofree | though if you've built it on a jetson that would be nice to know | 09:50 |
| paulsherwood | i'm building it right now on x86 will let you know | 09:51 |
| paulsherwood | what's our position on reviews if, say someone gives -1 while two others give +1 ? | 09:51 |
| radiofree | i thought -1 was a veto? | 09:52 |
| paulsherwood | i'm thinking of my elapsed time patch, which would have been really useful in master over the last week, but is held up because richard_1aw doesn't like my style | 09:52 |
| paulsherwood | ok, assuming it's a veto, i disagree with him. how does that resolve itself? | 09:53 |
| * radiofree doesn't know enough about the structure of the baserock team to comment on that | 09:53 | |
| * paulsherwood neither :) | 09:54 | |
| radiofree | i suppose you could have someone else reply directly to the -1 and have them +1 it or something | 09:54 |
| paulsherwood | 2014-10-05 08:44:28 [Build 325/349] [gtk+] Elapsed time 00:05:53 | 09:54 |
| paulsherwood | (x86) | 09:54 |
| paulsherwood | we also seem to be building enlightenment here... do you think we need that? | 09:55 |
| radiofree | 8 minutes on jetson! | 09:55 |
| paulsherwood | you may be measuring a different thing, but great :) | 09:56 |
| radiofree | paulsherwood: i have no idea, i would have thought *not* but i haven't even attempted to build this demo stuff so may as well put that stuff in | 09:56 |
| paulsherwood | my elapsed time is from start of build (so includes creating staging area etc) | 09:56 |
| radiofree | ah | 09:57 |
| paulsherwood | radiofree: ack | 09:57 |
| paulsherwood | hmmm... i've tried reflashing this jetson repeatedly. keep getting bad CRC ... http://fpaste.org/139317/ | 10:00 |
| paulsherwood | is it the board, or me, or maybe the image is too big? | 10:00 |
| radiofree | it's failing to load the extlinux.conf file from the emmc | 10:03 |
| radiofree | how big was the system image you flashed? | 10:03 |
| paulsherwood | 4G | 10:03 |
| radiofree | that should be fine | 10:03 |
| paulsherwood | this is trying to use 14.40 image instead of radiofree magic image from before | 10:04 |
| radiofree | interrupt u-boot | 10:05 |
| radiofree | can you do btrls mmc 0 | 10:05 |
| paulsherwood | here's what i did http://fpaste.org/139318/ | 10:05 |
| radiofree | i have flashed 5G(+) images before so it's not the size | 10:06 |
| paulsherwood | Tegra124 (Jetson TK1) # btrls mmc 0 | 10:07 |
| paulsherwood | <DIR> systems | 10:07 |
| paulsherwood | <DIR> state | 10:07 |
| paulsherwood | Tegra124 (Jetson TK1) # btrls mmc 0 systems | 10:11 |
| paulsherwood | <SYM> default | 10:11 |
| paulsherwood | <DIR> factory | 10:11 |
| paulsherwood | Tegra124 (Jetson TK1) # btrls mmc 0 state | 10:11 |
| paulsherwood | <DIR> root | 10:11 |
| paulsherwood | <DIR> var | 10:11 |
| paulsherwood | <DIR> home | 10:11 |
| paulsherwood | <DIR> opt | 10:11 |
| paulsherwood | <DIR> srv | 10:11 |
| radiofree | paulsherwood: there's the problem then | 10:14 |
| radiofree | do btrls mmc 0 / again | 10:14 |
| radiofree | there's no extlinux.conf in / | 10:14 |
| radiofree | what does your deploy morph look like | 10:15 |
| paulsherwood | ah... does your flash script use the cluster morph somehow? | 10:16 |
| radiofree | no | 10:16 |
| radiofree | put the generate rawdisk image doesn't have one | 10:16 |
| paulsherwood | ok this is a virgin jetson, flashing with 14.40 | 10:16 |
| radiofree | so it was at the deploy stage this error happened | 10:16 |
| paulsherwood | no deploy here | 10:16 |
| paulsherwood | just following http://wiki.baserock.org/guides/baserock-jetson/ | 10:16 |
| paulsherwood | which has been updated to include 14.40 | 10:17 |
| radiofree | http://git.baserock.org/cgi-bin/cgit.cgi/baserock/baserock/definitions.git/tree/clusters/release.morph | 10:17 |
| radiofree | the 14.40 release of jetson-devel won't work then | 10:17 |
| radiofree | it's missing | 10:17 |
| radiofree | DTB_PATH: "boot/tegra124-jetson-tk1.dtb" | 10:17 |
| radiofree | BOOTLOADER_CONFIG_FORMAT: "extlinux" | 10:17 |
| radiofree | BOOTLOADER_INSTALL: "none" | 10:17 |
| radiofree | see line 32 onwards | 10:18 |
| paulsherwood | right | 10:18 |
| paulsherwood | i guess we should patch that now, maybe frab a new tar? | 10:18 |
| radiofree | for the devel image it needs to be DTB_PATH: "tegra124-pm375.dtb" | 10:18 |
| radiofree | paulsherwood: yes +2 to modify that directly in master | 10:19 |
| radiofree | or should i do it? | 10:19 |
| paulsherwood | all my baserocks are busy | 10:19 |
| paulsherwood | please email a branch, though, so we can do a rushed review | 10:20 |
| radiofree | ok | 10:20 |
| paulsherwood | in other news... | 10:22 |
| paulsherwood | 11:15 < Kinnison> ALso, if you're doing the qt hmi, then you need to edit start_hmi.sh to launch with -platform wayland | 10:22 |
| paulsherwood | 11:15 < Kinnison> and then you run the *core* and it launches the hmi | 10:22 |
| paulsherwood | straycat: could we get a quick review on radiofree's patch please? | 10:27 |
| paulsherwood | (or pedroalvarez if you're around) | 10:27 |
| * paulsherwood wonders if he dares do a merge himself :) | 10:29 | |
| straycat | Sure | 10:31 |
| paulsherwood | straycat: if you like it, probably would be safer if you merge :) | 10:32 |
| straycat | Can't see anything I could object to, radiofree, has it been tested? | 10:34 |
| paulsherwood | well, we've tested that without it the image can't be flashed ;) | 10:35 |
| * paulsherwood doesn't know how to do a release test | 10:35 | |
| straycat | It looks fine, but since it's hard for me to know without testing, I'll deploy an image from that cluster and try it out. | 10:39 |
| paulsherwood | ooh, cool if you can do that | 10:39 |
| * paulsherwood still doesn't *quite* understand this deploy magic | 10:40 | |
| * paulsherwood feels like he's predicted the future | 10:41 | |
| paulsherwood | radiofree: remember http://fpaste.org/139309/ ? | 10:42 |
| paulsherwood | that's where you're going to crash :) | 10:42 |
| radiofree | paulsherwood: i guess it needs unistd.h? | 10:45 |
| straycat | Oh | 10:46 |
| paulsherwood | https://bugs.gentoo.org/show_bug.cgi?id=424707 | 10:46 |
| paulsherwood | radiofree: ^^ | 10:46 |
| paulsherwood | there's a patch, apparently | 10:46 |
| straycat | I can't deploy it because deploy assumes I've got an up to date cache. | 10:47 |
| paulsherwood | can you specify --artifact-cache-server? | 10:48 |
| paulsherwood | radiofree: yes, you're right, now i read through | 10:49 |
| paulsherwood | you're way ahead of me | 10:49 |
| * paulsherwood does wonder if we have the 'right' upstream for dbus-c++ | 10:52 | |
| paulsherwood | straycat: it has +2 | 10:53 |
| straycat | Okay | 10:54 |
| paulsherwood | radiofree: weird... SotK had already patched this? http://git.baserock.org/cgi-bin/cgit.cgi/delta/dbus-c++.git/commit/?h=baserock/genivi-demo-platform&id=8d351eec25430651acbac014e44ec011fac32d57 | 10:55 |
| radiofree | use that branch though | 10:55 |
| radiofree | i just cherry-picked his patches, he must have forgot to update the ref | 10:56 |
| paulsherwood | ok | 10:57 |
| straycat | Oh sorry I forgot that we don't upload the system artifacts to gbo, so no I wouldn't be able to deploy a jetson image from here. | 11:02 |
| paulsherwood | don't we? | 11:03 |
| paulsherwood | i thought we did | 11:04 |
| paulsherwood | pedroalvarez uploaded lots of stuff? | 11:04 |
| straycat | The only artifacts with -rootfs on gbo have date 'July 18' | 11:05 |
| straycat | I think we avoid uploading them for performance reasons, if morph finds the system artifact it will download it, where it may in fact be quicker to download the few artifacts you're missing and construct the system artifact yourself. | 11:06 |
| straycat | But in this case it means I need to find myself an arm box. | 11:07 |
| paulsherwood | straycat: that's probably more effort than it's worth? it's already +2, if you fancy merging it. pedroalvarez has said he'll try a release when he can | 11:08 |
| straycat | I don't know it wouldn't be too difficult, but if other people are confident it's correct then I guess I can merge it. | 11:09 |
| paulsherwood | straycat: it would be great to get proof :) | 11:12 |
| * straycat nods | 11:12 | |
| * straycat borrows tyrell | 11:12 | |
| paulsherwood | :) | 11:13 |
| radiofree | tyrell is old morph remember :\ | 11:16 |
| * straycat nods | 11:16 | |
| straycat | I'll change that briefly. | 11:17 |
| radiofree | boost is done! | 11:18 |
| paulsherwood | 2014-10-05 11:14:24 [Build 349/349] [genivi-plusplus-system-x86] Building system genivi-plusplus-system-x86 | 11:19 |
| paulsherwood | radiofree: you're ahead, but you'll need to fix that dbus-c++ | 11:19 |
| paulsherwood | s/ahead on jetson/ | 11:19 |
| paulsherwood | as will i. this is a bit like F1 | 11:20 |
| paulsherwood | where the fixes are pitstops | 11:20 |
| paulsherwood | 2014-10-05 11:16:51 Build ends successfully | 11:21 |
| paulsherwood | (x86) | 11:21 |
| * radiofree has just fixed dbus, building | 11:21 | |
| paulsherwood | 2014-10-05 11:22:08 [Build 333/351] [efl] Building chunk efl (jetson) | 11:22 |
| radiofree | do we have any small hdmi monitors in the office | 11:23 |
| radiofree | 9" or something | 11:23 |
| paulsherwood | radiofree: i'm going shopping | 11:23 |
| radiofree | do they exist? | 11:23 |
| paulsherwood | may buy a projector, and whatever other shiny i find :) | 11:23 |
| * paulsherwood would like to buy a touchscreen, but they don't seem to exist at retail | 11:24 | |
| radiofree | aria sell a 23" but that will be way too big | 11:24 |
| * paulsherwood fears this image will be too big to deploy as an upgrade | 11:25 | |
| radiofree | http://www.lilliputuk.com/monitors/hdmi/664-o-p/ | 11:25 |
| radiofree | hmm we have a small usb monitor thing in the office | 11:26 |
| radiofree | i'll give that a go | 11:26 |
| * paulsherwood really wonders what happened to the original jetson touchscreen | 11:26 | |
| radiofree | paulsherwood: there's things like https://www.google.co.uk/shopping/product/11570598595658361690?q=displaylink+monitor&safe=off&client=firefox-a&hs=qnl&rls=org.mozilla:en-GB:official&channel=sb&biw=1366&bih=582&sqi=2&bav=on.2,or.r_cp.r_qf.&bvm=bv.76802529,d.d2s&tch=1&ech=1&psi=GisxVITxOovSaJeIgMAG.1412508443768.3&ei=IisxVIm-FYrcaKzLgMgG&ved=0CLsBEKYrMAU | 11:27 |
| radiofree | *but* i have no idea if these are going to work with a jetson | 11:27 |
| paulsherwood | right | 11:27 |
| radiofree | it has built! | 11:29 |
| * radiofree goes for lunch then will head into office | 11:29 | |
| paulsherwood | w00t!!! | 11:29 |
| jjardon | radiofree: gtk2 will take around 5 min in my laptop, don't know about arm | 12:01 |
| jjardon | FYI; if you use my baserock/jjardon/libdrm branch you will save 49 chunks in case you have to rebuild (you will have to remove the llvm dependency from mesa-common as seems it doesn't build on arm) | 12:03 |
| jjardon | radiofree: yeah, llvm is only needed by mesa but its currently in x-common | 12:04 |
| radiofree | jjardon: around 5 minutes on a Jetson! | 12:11 |
| * jjardon thinking about order a Jetson for Christmas | 12:15 | |
| jjardon | paulsherwood: I'd say he is the maintainer; so or you adapt the patch or you convince him your approach is simpler/better | 12:16 |
| paulsherwood | well his theoretical code would theoretically be better. however i've tried his suggestions and can't get them to work. i'm not a python wizard, though | 12:38 |
| paulsherwood | i assume it's not right that deploy hangs for 20 minutes at this point? http://fpaste.org/139337/ | 12:39 |
| * paulsherwood wonders if his deploy hang could have anything to do with "xfer-hole: Fix bug in copy_slice_from_file." | 13:15 | |
| straycat | That affects the "transfer image" stage, rather than the snapshot stage though | 13:28 |
| paulsherwood | i thought snapshots are not being used in upgrades at the moment? | 13:29 |
| *** wikicat [~wikicat@ec2-54-77-166-165.eu-west-1.compute.amazonaws.com] has joined #baserock | 13:31 | |
| radiofree | it's not all bad news, the bluetooth works out of the box | 13:53 |
| paulsherwood | cool! :) | 13:54 |
| radiofree | need to try smartdevicelink though | 13:54 |
| wikicat | Wiki change: Add link to general KVM deployment guide http://wiki.baserock.org/recentchanges/#change-ad6e73bbd07811fb4094ea83736e0d14c84343c5 | 13:57 |
| paulsherwood | straycat: i need to get to a working baserock-14.40-devel-system-armv7lhf-jetson.tar.gz on gbo somehow. first step is merging radiofree's patch. not sure if you're still intending to test that or can we just merge? | 14:45 |
| radiofree | paulsherwood: needs merge + revert version of u-boot | 14:48 |
| radiofree | for some reason in the plusplus image, the old kernel modules get loaded | 14:48 |
| radiofree | so no drm | 14:48 |
| paulsherwood | urggh :) | 14:48 |
| radiofree | will work on that, but i'm going to try some other things first | 14:48 |
| straycat | paulsherwood, baserock/straycat/fix-jetson-release has what we should need for a working system, I'm just building it now. | 14:50 |
| radiofree | paulsherwood: you can, however, build them manually on the board for now, if you want | 14:51 |
| * paulsherwood sees the same rsync hang on jetson | 14:55 | |
| paulsherwood | why revert the u-boot? | 14:57 |
| jjardon | what is the layer_management.git in genivi-x-generic and genivi-wayland-generic? Seems it didnt get any update since last year. Also genivi-wayland-generic is not used as a dependency in any system. And genivi-x-generic is used in non-genivi system si seems not necessary there | 15:42 |
| jjardon | paulsherwood: did you have time to try the release_33 branch of llvm? Seems to build fine here in x86 | 16:42 |
| straycat | paulsherwood, the current u-boot doesn't work with the kernel the devel bsp has | 17:18 |
| radiofree | I found a test app for SDL | 17:40 |
| radiofree | but it's android, no idea how to compile and test this | 17:40 |
| radiofree | i have a lot of lorry requests | 17:55 |
| radiofree | hmm maybe not | 17:56 |
| paulsherwood | jjardon: i did not, will do no | 17:56 |
| paulsherwood | w | 17:56 |
| radiofree | https://github.com/otcshare/automotive-message-broker.git | 17:57 |
| radiofree | http://git.projects.genivi.org/lbs/positioning.git | 17:58 |
| radiofree | git://git.projects.genivi.org/lbs/navigation-application.git | 17:58 |
| radiofree | position is there sorry | 17:59 |
| paulsherwood | radiofree: put them in as full urls will be fine for now | 17:59 |
| paulsherwood | we can lorry them later | 17:59 |
| radiofree | http://git.projects.genivi.org/lbs/navigation-application.git | 17:59 |
| radiofree | paulsherwood: i will need to patch them | 18:00 |
| paulsherwood | ok | 18:00 |
| paulsherwood | i'll do the lorries | 18:00 |
| radiofree | paulsherwood: can you ask around for where an android demo app for SDL is as well? | 18:01 |
| radiofree | loads of boost errors with automotive-message-broker | 18:02 |
| radiofree | which is apparently fixed upstream in boost... | 18:03 |
| paulsherwood | http://git.projects.genivi.org/?p=smartdevicelink_tester_android.git;a=summary | 18:03 |
| paulsherwood | radiofree: ^^ | 18:03 |
| radiofree | yes i found that but have *no* idea how to compile it | 18:04 |
| paulsherwood | ok, regarding navigation-application - that's what we have as navigation.git | 18:06 |
| paulsherwood | their nvigation.git we have as poi-service.git | 18:07 |
| paulsherwood | i can't remember why | 18:07 |
| radiofree | poi-service is included in navigation by the looks of it | 18:08 |
| paulsherwood | http://git.baserock.org/cgi-bin/cgit.cgi/baserock/local-config/lorries.git/tree/open-source-lorries/genivi.lorry | 18:09 |
| paulsherwood | so it seems the only missing one is message broker | 18:10 |
| radiofree | and navigation-application | 18:10 |
| paulsherwood | no, that's already lorried - http://git.baserock.org/cgi-bin/cgit.cgi/delta/genivi/navigation.git/ | 18:14 |
| paulsherwood | same as http://git.projects.genivi.org/?p=lbs/navigation-application.git;a=summary | 18:15 |
| paulsherwood | can i get a quick review please? http://fpaste.org/139383/ | 18:15 |
| paulsherwood | anyways... http://git.baserock.org/cgi-bin/cgit.cgi/delta/automotive-message-broker.git/ | 18:20 |
| straycat | paulsherwood, looks syntactically fine to me | 18:20 |
| paulsherwood | tvm | 18:21 |
| * paulsherwood should learn to power jetsons off gently | 18:21 | |
| * paulsherwood has bad crc again | 18:21 | |
| straycat | I've not been able to deploy a jetson with the various fixes we have | 18:22 |
| straycat | So I don't think there's any reason to merge anything at the moment. | 18:22 |
| radiofree | the reversion of u-boot should have fixed it | 18:24 |
| jjardon | paulsherwood: Would not be better to put that inside the genivi/ namespace? | 18:30 |
| paulsherwood | it's not a genivi project | 18:31 |
| paulsherwood | straycat: i'm confused - how far did you get with deployment? | 18:53 |
| paulsherwood | (i agree, though... if it's not working) | 18:53 |
| straycat | built, deployed, flashed, kernel panics | 18:55 |
| paulsherwood | with the 3.10 kernel? | 18:56 |
| straycat | If that's what we have in the devel bsp now then yes. | 18:57 |
| straycat | Someone else could always try the branch I made, I might have made an error somewhere, but I have double checked things. | 18:58 |
| paulsherwood | it appears to be | 18:58 |
| paulsherwood | i would love to, as soon as i resuscitate a jetson :) | 18:58 |
| paulsherwood | straycat: is the built artifact available somewhere? | 18:59 |
| straycat | I could upload it, if you build from that branch it should trigger a rebuild of u-boot and linux. | 19:01 |
| paulsherwood | ok no problem | 19:03 |
| paulsherwood | nvidia's cdn seems to be really throttling that file | 19:04 |
| straycat | paulsherwood, Are you going to rebuild or shall I upload? | 19:04 |
| paulsherwood | i'll try to rebuild, after i've re-flashed | 19:06 |
| straycat | Okay | 19:06 |
| paulsherwood | radiofree: could you push your latest stuff somewhere, please? | 19:19 |
| radiofree | paulsherwood: it's all local, trying to get navit to build correctly | 19:19 |
| radiofree | but i'll be able to base it off a baserock genivi platform | 19:19 |
| radiofree | there's things that haven't been pushed for navit | 19:20 |
| radiofree | glu/glut etc | 19:20 |
| paulsherwood | 'haven't been pushed'? | 19:24 |
| radiofree | as far as i can tell, to get the opengl backend you need glu and glut | 19:25 |
| paulsherwood | ok, they're lorried, at least | 19:26 |
| radiofree | can we have | 19:37 |
| radiofree | http://cgit.freedesktop.org/mesa/glut/ | 19:37 |
| radiofree | http://cgit.freedesktop.org/mesa/glu | 19:37 |
| paulsherwood | they're already there | 19:39 |
| paulsherwood | http://git.baserock.org/cgi-bin/cgit.cgi/?q=glu | 19:39 |
| radiofree | right you are, sorry | 19:39 |
| paulsherwood | np | 19:39 |
| paulsherwood | flashing is *way* faster on lenovo than in my mac vm | 19:40 |
| radiofree | perhaps some issue with the passthrough on mac | 19:41 |
| paulsherwood | yup | 19:42 |
| * paulsherwood has hit crc error 4 times on flashing a specific jetson... is starting to believe it's busted | 19:52 | |
| radiofree | stuck on boost now | 19:57 |
| radiofree | and lots of other things actually | 19:59 |
| paulsherwood | stuck? or building? | 20:02 |
| * paulsherwood is an idiot | 20:03 | |
| paulsherwood | how do i stop the jetson treating sdcard as something to boot from? | 20:04 |
| * paulsherwood would like to use SDcard as /src | 20:04 | |
| radiofree | it shouldn't choose that if its btrfs | 20:08 |
| paulsherwood | i may not have formatted it properly yet | 20:08 |
| paulsherwood | so, if it is mmcblk1p1, it's formatted btrfs. | 20:10 |
| paulsherwood | do i need to format mmcblk1 too? | 20:10 |
| radiofree | i don't think so | 20:15 |
| wikicat | Wiki change: Add link to general Virtual Box deployment guide http://wiki.baserock.org/recentchanges/#change-2c735fe10c1885b52429354ba4ca63bebbab4233 | 20:16 |
| paulsherwood | any chance wikicat could show a link to the diff? | 20:19 |
| * paulsherwood wants the moon, as always | 20:19 | |
| paulsherwood | radiofree: i guess i'll just settle for popping the sd card to let it boot | 20:23 |
| radiofree | need a new version of boost | 20:25 |
| paulsherwood | https://github.com/boostorg/boost ? | 20:27 |
| paulsherwood | looks like the upstream | 20:29 |
| paulsherwood | i'll create a lorry if no-one objects? | 20:29 |
| radiofree | please | 20:29 |
| paulsherwood | http://fpaste.org/139405/ | 20:32 |
| paulsherwood | http://git.baserock.org/cgi-bin/cgit.cgi/delta/boost.git/ | 20:34 |
| straycat | It would be useful to have a link to the diff. | 20:43 |
| paulsherwood | ?i don't understand? the fpaste was a link to the diff | 20:47 |
| paulsherwood | imagine i have a jetson, with an empty cache, and on the same network a jetson with a full one... is there a magic rsync command i could use to push the whole directory across? | 20:48 |
| paulsherwood | straycat: i'm interesed in your patches to the artifact resolver. should they make it faster, or slower, do you know? | 20:51 |
| straycat | paulsherwood, Sorry I was replying about wikicat :) | 20:55 |
| straycat | paulsherwood, It's probably easiest to scp the directory across | 20:56 |
| straycat | paulsherwood, Performance won't be changed at all, it really is just a tidy up. I think richard_1aw has plans to resolve sources rather than artifacts as well. | 20:58 |
| * straycat disappears | 21:00 | |
| paulsherwood | straycat: thanks, understood | 21:10 |
| wikicat | Wiki change: some build times http://wiki.baserock.org/recentchanges/#change-7ac9862eba4406a34059ba79bbcdfdd4804af083 | 21:12 |
| radiofree | i'm about to push a system that branches from our genivi-baseline | 21:14 |
| paulsherwood | ok | 21:14 |
| radiofree | this *will not* contain the navigation app, or SDL yet | 21:14 |
| radiofree | i'll work on that on monday night/tuesday | 21:15 |
| radiofree | there's quite a lot of changes in navigation master (i.e it actually builds....) | 21:15 |
| radiofree | which we haven't rebased against in a while... | 21:15 |
| paulsherwood | ok | 21:16 |
| radiofree | i'll leave this building overnight | 21:17 |
| radiofree | has new boost in glut etc | 21:17 |
| paulsherwood | if you push it, i'll build it too | 21:18 |
| radiofree | will also build it on an x86 vm, so i can have sometihng to work on on the plane | 21:18 |
| paulsherwood | me too :) | 21:18 |
| radiofree | baserock/james/genivi-demo | 21:31 |
| radiofree | -platform | 21:31 |
| * radiofree heads home | 21:31 | |
| radiofree | back in 10 | 21:31 |
| wikicat | Wiki change: add more jetson times http://wiki.baserock.org/recentchanges/#change-6fbccc493e38dff079f7085c0410b8d463a06772 | 21:38 |
| wikicat | Wiki change: add more mac times http://wiki.baserock.org/recentchanges/#change-912285da53045c619acf6f4453dccb27b335a70a | 21:42 |
| jjardon | paulsherwood: nice wiki page. I noticed some components are in several different lines though; do they mean different things? | 21:43 |
| paulsherwood | jjardon: i've done multiple builds. i suppose i could strip them out, just settle for slowest for each component, but it seems that the variations are useful to know | 21:46 |
| paulsherwood | i have some times from a cloud-based system too | 21:47 |
| jjardon | paulsherwood: ah, ok. makes sense. Maybe you its good to add that info to avoid confusion? | 21:50 |
| paulsherwood | will do | 21:50 |
| jjardon | paulsherwood: did you have any luck with llvm? | 21:53 |
| * jjardon wonder why coreutils takes so much time in the Jetson | 21:55 | |
| paulsherwood | jjardon: i'm still investigating llvm, sorry | 22:00 |
| * paulsherwood is a bit confused | 22:00 | |
| wikicat | Wiki change: add some explanation http://wiki.baserock.org/recentchanges/#change-71d9a64557e6ed5746487ce91db2df1879269c5f | 22:02 |
| paulsherwood | ok, it's building now... will let you know in about...25 minutes' time :) | 22:03 |
| jjardon | paulsherwood: thanks, FYI I built the release_33 branch in x86 fine (and no need for groff) | 22:05 |
| radiofree | paulsherwood: sorry, had missing files in the branch | 22:09 |
| radiofree | repushed | 22:09 |
| radiofree | added x86_32 ++ system | 22:15 |
| paulsherwood | radiofree: do you have an ESTA? | 22:18 |
| radiofree | i got one when i went to san francisco | 22:23 |
| radiofree | paulsherwood: is there an artifact cache server i can use? | 22:24 |
| radiofree | morph master on my jetson is stuck on "2014-10-05 22:19:50 Resolving artifacts" | 22:24 |
| radiofree | x86 wants to rebuild the world | 22:26 |
| paulsherwood | radiofree: resolving takes a long time | 22:33 |
| paulsherwood | jjardon: llvm 33 builds fine :-) | 22:33 |
| radiofree | paulsherwood: is there a artifact-cache-server i can use for x86? | 22:35 |
| radiofree | paulsherwood: how long, 10 minutes? 20? | 22:35 |
| paulsherwood | to build llvm? | 22:37 |
| radiofree | no to "resolve artifacts" | 22:39 |
| paulsherwood | oh, resolving... actually, i can't be sure... i don't build verbose | 22:39 |
| radiofree | i'm building verbose and it's doing *nothing* | 22:39 |
| radiofree | nothing in the log either | 22:39 |
| radiofree | it never usually takes that long | 22:39 |
| radiofree | also [23:36:04] <radiofree> paulsherwood: is there a artifact-cache-server i can use for x86? | 22:39 |
| paulsherwood | i would drop verbose, it's not helpful. you can get that info from the log anyway. | 22:39 |
| paulsherwood | you mean for this work, not 14.40? | 22:40 |
| radiofree | yeah, the time it did work just then, it wanted to rebuild from core | 22:41 |
| paulsherwood | i have a machine that has the artifacts on it, maybe you can scp them? | 22:41 |
| wikicat | Wiki change: add cloud times http://wiki.baserock.org/recentchanges/#change-2b37591ec91320f022ed6973d14522187ea739be | 22:50 |
| jjardon | paulsherwood: \o/ i will prepare a new patch series. Thanks again for testing! | 22:50 |
| paulsherwood | yw | 22:53 |
| radiofree | shouldn't *something* be going on in morph.log when it's resolving artifacts? | 22:53 |
| radiofree | morph is using 100% of the cpu though... | 22:53 |
| paulsherwood | this definitely needs an overhaul | 22:54 |
| radiofree | does 2014-10-05 22:52:03 WARNING No _validate_cross_refs_for_chunk | 22:54 |
| radiofree | mean anything? | 22:54 |
| paulsherwood | not to me | 22:54 |
| radiofree | i think something is going wrong | 22:58 |
| radiofree | when it has worked it returns almost immediatly and says it's returning cache keys | 22:59 |
| paulsherwood | ok | 22:59 |
| paulsherwood | i'm trying to build it on both jetson and cloud now | 23:00 |
| paulsherwood | earlier today, on cloud... plusplus system... | 23:01 |
| paulsherwood | 2014-10-05 08:19:36 Stratum [lua-runtime] version is 093a32d | 23:01 |
| paulsherwood | 2014-10-05 08:27:29 [Build 27/349] [cmake] Building chunk cmake | 23:01 |
| paulsherwood | 12 minutes resolving i think | 23:01 |
| paulsherwood | sorry, 8 | 23:01 |
| * paulsherwood cant count any longer | 23:01 | |
| radiofree | ah | 23:02 |
| radiofree | well i'll just leave it | 23:02 |
| radiofree | actually | 23:02 |
| radiofree | yes, i'll just leave it | 23:02 |
| paulsherwood | ERROR: Are you trying to cross-build? Host architecture is x86_64 but target is x86_32 | 23:03 |
| paulsherwood | just get some sleep :) | 23:03 |
| paulsherwood | what time are you aiming to be at the airport? | 23:03 |
| paulsherwood | (pls remember to bring two jetsons) | 23:04 |
| paulsherwood | radiofree: ^^ | 23:04 |
| radiofree | paulsherwood: yep i will, around 8 i think | 23:08 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!