| 00:17.16 | *** join/#oe hrww (~hrw@redhat/hrw) |
| 00:17.44 | kroon | mhm.. strange |
| 00:18.43 | *** join/#oe challinan_ (~chris@173-10-226-189-BusName-WestFlorida.hfc.comcastbusiness.net) |
| 00:19.26 | *** join/#oe anunnaki (~anunnaki@hellsgate.pl) |
| 00:20.06 | kergoth | kroon: might want to open a bug. I'm guessing you have multiple sets of sstate floating around and it incorrectly grabbed one that was close to what you're building instead of the exact match, but it's richard's baby, he'd know better than i |
| 00:23.04 | *** join/#oe Crofton (~balister@75-146-16-53-Richmond.hfc.comcastbusiness.net) |
| 00:55.23 | kroon | kergoth, could you test if the output looks correct in your setup ? |
| 00:55.41 | kroon | kergoth, I see this output for basically any recipe I try |
| 00:57.08 | kroon | kergoth, I do use wipe-sysroot/cleanup-workdir/sstate-cache-management often, and do inc. rebuilds, so maybe its one of them that is confusing my setup |
| 00:58.04 | kroon | im using master branches btw |
| 00:58.06 | *** join/#oe kgilmer (~kgilmer@184.152.73.182) |
| 01:01.13 | *** join/#oe tobiash (~quassel@mail.bmw-carit.de) |
| 01:27.06 | *** join/#oe kgilmer (~kgilmer@184.152.73.182) |
| 01:31.41 | *** join/#oe NightMonkey_ (~NightrMon@173-164-139-193-SFBA.hfc.comcastbusiness.net) |
| 01:38.42 | *** join/#oe hillct (~hillct@cpe-071-070-219-043.nc.res.rr.com) |
| 01:50.17 | *** join/#oe flo_lap (~fuchs@Maemo/community/contributor/florian) |
| 02:24.06 | *** join/#oe s3v3n62mm (~s3v3n62mm@p3EE2F443.dip0.t-ipconnect.de) |
| 03:20.34 | *** join/#oe rogue2 (~pompomJui@41.0.38.138) |
| 03:34.07 | *** join/#oe NvrBst (~NvrBst@gateway/tor-sasl/nvrbst) |
| 03:35.05 | *** join/#oe _JokerDoom (~JokerDoom@24-196-46-114.dhcp.fdul.wi.charter.com) |
| 04:03.56 | *** join/#oe neabax (~neabax@192.77.239.81) |
| 04:21.10 | *** join/#oe Crofton (~balister@pool-108-44-91-50.ronkva.east.verizon.net) |
| 04:57.38 | *** join/#oe Crofton (~balister@pool-71-171-33-112.ronkva.east.verizon.net) |
| 04:58.03 | *** join/#oe Crofton|work (~balister@pool-71-171-33-112.ronkva.east.verizon.net) |
| 06:37.13 | *** join/#oe ant_work (~ant__@host54-128-static.10-188-b.business.telecomitalia.it) |
| 06:44.41 | dlan | how to force no clean ${WORKDIR}? I put inherit +="rm_work" into local.conf |
| 06:45.33 | *** join/#oe __JokerDoom (~JokerDoom@24-196-46-114.dhcp.fdul.wi.charter.com) |
| 06:47.20 | ant_work | remove it |
| 06:48.37 | ant_work | koen: hx4700 today? |
| 06:51.08 | ant_work | koen: in case, the irqs.h patch went upstream, remove it |
| 07:27.28 | *** join/#oe ao2 (~ao2@2001:1418:117::1) |
| 07:37.03 | rogue2 | The documentation says that FULL_OPTIMIZATION's default value is "-fexpensive-optimizations -fomit-frame-pointer -frename-registers -O2" |
| 07:37.25 | rogue2 | inside bitbake.conf it is FULL_OPTIMIZATION = "-O2 -pipe ${DEBUG_FLAGS}" instead? |
| 07:37.47 | rogue2 | I want to compile something with O1 |
| 08:10.33 | pompomJuice | awesome! |
| 08:10.47 | pompomJuice | I see this run-postinst package |
| 08:10.51 | pompomJuice | it solves all my problems |
| 08:11.13 | pompomJuice | now I can justify why I always say version bump is the best course of action |
| 08:11.21 | pompomJuice | gotcha |
| 08:19.44 | *** join/#oe rob_w (~rob@unaffiliated/rob-w/x-1112029) |
| 08:22.46 | pompomJuice | why do recipies point to these kinds of repos? |
| 08:22.55 | pompomJuice | ---> SRC_URI = "git://anongit.freedesktop.org/systemd/systemd;branch=master;protocol=git |
| 08:23.17 | pompomJuice | what if I want to make it lower, I cannot browse that url for tags, no idea where that repo comes from |
| 08:23.52 | pompomJuice | I guess I can check it out and see that way but... so much work |
| 08:25.43 | bencoh | khem: http://pastebin.notk.org/pastebin.php?show=d4cbea45 built for arm with -O2 (tested for cortexA9) |
| 08:25.48 | ant_work | pompomJuice: you can use SRCREV |
| 08:26.09 | pompomJuice | against github? |
| 08:26.23 | ant_work | against git |
| 08:26.39 | pompomJuice | checking... |
| 08:27.02 | bencoh | khem: (complete gcc call : arm-oe-linux-gnueabi-gcc -march=armv7-a -mthumb-interwork -mfloat-abi=hard -mfpu=neon -mtune=cortex-a9 --sysroot=/home/bencoh/oe-core/build/tmp-eglibc/sysroots/cubox-i -Wall -O2 -g -I. -I../.. -lrt -o dvb_gen_si dvb_gen_si.c ) |
| 08:27.38 | pompomJuice | ant_work: the problem is for example I want to modify that systemd_211.bb recipy to be systemd_204.bb |
| 08:27.40 | pompomJuice | oops |
| 08:27.52 | pompomJuice | for some reason bitbake recipies are not built that way anymore? |
| 08:28.02 | pompomJuice | PV from filename outdated? |
| 08:28.33 | ant_work | usually the versioned recipes d/l a tarball |
| 08:28.36 | pompomJuice | so I go into the file and change the SRCURI manually, but I can't because the SRC_URI points to a git repo that is not online |
| 08:28.44 | pompomJuice | aah |
| 08:28.53 | pompomJuice | why tarball when it is better? |
| 08:28.56 | pompomJuice | oop |
| 08:28.57 | pompomJuice | or not |
| 08:28.58 | pompomJuice | nm |
| 08:29.02 | pompomJuice | git is not better |
| 08:29.14 | pompomJuice | ma bad |
| 08:29.50 | ant_work | see: http://tinyurl.com/pr8s3my |
| 08:30.43 | pompomJuice | makes sense because a rebuild destroys your companies internet for a while |
| 08:30.53 | pompomJuice | why cant they fix git to prune mode |
| 08:31.02 | pompomJuice | or gief head bytes only |
| 08:32.32 | pompomJuice | for example, systemd_211 contains this SRCREV=3a450ec5 |
| 08:32.43 | pompomJuice | I cannot find that rev on github/systemd |
| 08:33.22 | pompomJuice | cant find it on anongit either |
| 08:33.35 | pompomJuice | maybe this anongit spies on me? |
| 08:34.42 | pompomJuice | maybe this anongit version is recipy spesific? |
| 08:35.38 | ant_work | you can find the tags in github |
| 08:35.40 | ant_work | i.e. |
| 08:35.41 | ant_work | https://github.com/systemd/systemd/tree/v211 |
| 08:36.38 | ant_work | so https://github.com/systemd/systemd/archive/v211.zip |
| 08:36.54 | *** join/#oe bluelightning (~paul@83.217.123.106) |
| 08:36.54 | *** join/#oe bluelightning (~paul@pdpc/supporter/professional/bluelightning) |
| 08:37.03 | ant_work | but it is much easier to specify a SRCREV in your .bbappend |
| 08:37.19 | bencoh | I still don't know why we disallowed git tags in SRCREV btw ;) |
| 08:38.02 | ant_work | personally I don't like the github interface |
| 08:38.35 | ant_work | but yes, you can see the tags, hidden besides branch combo |
| 08:40.47 | ant_work | kernel.org cgit is much more clear imho |
| 08:41.42 | ant_work | wonders who asked for the picture of the committer on each github patch |
| 08:42.50 | bencoh | cgit <3 |
| 08:44.55 | bluelightning | morning all |
| 08:47.19 | pompomJuice | where is it morning now? |
| 08:47.26 | pompomJuice | You must be in the ocean |
| 08:47.33 | pompomJuice | or west coast |
| 08:47.46 | ant_work | http://www.total-knowledge.com/~ilya/mips/ugt.html |
| 08:47.55 | ant_work | gm bluelightning |
| 08:48.01 | pompomJuice | I am with bencoh on this one |
| 08:48.11 | pompomJuice | now I have to work with hashtags instea of human tags |
| 08:48.26 | pompomJuice | thats like going backwards? |
| 08:48.35 | *** join/#oe belen (~Adium@192.198.151.43) |
| 08:48.47 | pompomJuice | must be a security issue im sure |
| 08:48.50 | bluelightning | pompomJuice: I'm in the UK ;) |
| 08:48.58 | pompomJuice | you bastard |
| 08:49.02 | pompomJuice | late sleeper |
| 08:49.38 | pompomJuice | ;) |
| 08:50.00 | bluelightning | http://www.yoctoproject.org/docs/current/ref-manual/ref-manual.html#migration-1.6-matching-branch-requirement-for-git-fetching |
| 08:50.22 | *** join/#oe diego_r (~diego@host65-246-static.10-188-b.business.telecomitalia.it) |
| 08:51.07 | pompomJuice | ;nobranch=1 |
| 08:51.09 | pompomJuice | woop |
| 08:51.27 | pompomJuice | I just happen to get those validation errors now |
| 08:51.35 | pompomJuice | which is strange I did not expect them |
| 08:51.51 | pompomJuice | Fetcher failure: Unable to find revision 606c24e3bd41207c395f24a56bcfcad791e265a5 in branch v204 even from upstream |
| 08:51.58 | pompomJuice | git://github.com/systemd/systemd.git;branch=v204;protocol=https |
| 08:52.11 | pompomJuice | SRCREV = "606c24e3bd41207c395f24a56bcfcad791e265a5" |
| 08:52.20 | pompomJuice | that is legal, but got rejected |
| 08:53.05 | bluelightning | is that the hash of the tag or the hash of the revision it points to? |
| 08:53.09 | bluelightning | because the former will not work |
| 08:55.04 | pompomJuice | checking... |
| 08:55.16 | pompomJuice | what I normally do is |
| 08:55.20 | pompomJuice | go here --- > https://github.com/systemd/systemd/tree/v204 |
| 08:55.32 | pompomJuice | click on latest commit copy sha |
| 08:55.41 | pompomJuice | that is the sha that is not working |
| 08:55.58 | pompomJuice | so I donno about tags and things... don't think that sha is a tag at all... |
| 08:56.07 | pompomJuice | just HEAD in a branch |
| 08:56.31 | pompomJuice | but yea with the nobranch=1 |
| 08:56.33 | pompomJuice | it will work |
| 08:56.55 | *** join/#oe anarsoul (~anarsoul@178.124.194.242) |
| 08:57.25 | bluelightning | well, that should work... |
| 08:57.31 | pompomJuice | wait you said former wont work...? |
| 08:57.57 | pompomJuice | AFAIK you cant have SRCREV on something that is not a annotated tag? |
| 08:58.01 | bluelightning | I meant that tags have a hash all of their own, which is different from the actual code revision they point to |
| 08:58.03 | pompomJuice | in 1.6? |
| 08:58.40 | bluelightning | btw, v204 is a tag, not a branch, so specifying that as the branch is where things are going wrong... |
| 08:58.49 | pompomJuice | aah |
| 08:58.50 | pompomJuice | take it slow |
| 08:59.33 | pompomJuice | completely messed that up |
| 08:59.36 | pompomJuice | lol |
| 09:06.07 | pompomJuice | I'm struggling to change systemd_211 to build 204 |
| 09:06.11 | pompomJuice | now im getting |
| 09:06.12 | pompomJuice | <PROTECTED> |
| 09:13.19 | bluelightning | that's a bug |
| 09:13.39 | bluelightning | could you help me to reproduce it? |
| 09:14.43 | pompomJuice | trying to understand the code and docs, what is a SCM? |
| 09:14.52 | pompomJuice | basically |
| 09:15.35 | pompomJuice | to reproduce -> copy systemd from oe-core to mata-copmany/recipies |
| 09:16.04 | pompomJuice | rename systemd_211.bb systemd_204.bb |
| 09:17.58 | bluelightning | SCM = source code manager (i.e. version control system) |
| 09:17.59 | pompomJuice | let me get a patch |
| 09:20.05 | pompomJuice | http://hastebin.com/ubucewasig.diff |
| 09:20.29 | pompomJuice | patch systemd_204.bb with that |
| 09:20.35 | pompomJuice | then try to build the recipy |
| 09:22.30 | pompomJuice | that patch applies on 211, needs to be 204 |
| 09:22.33 | pompomJuice | but yea |
| 09:22.44 | ant_work | pompomJuice: if you create your .bb why don't you d/l the tarball? |
| 09:22.46 | pompomJuice | that just says 211 cause that is how I made the patch |
| 09:22.56 | pompomJuice | uuhm |
| 09:23.03 | pompomJuice | shit |
| 09:23.04 | pompomJuice | ok |
| 09:23.05 | pompomJuice | lol |
| 09:23.11 | ant_work | pompomJuice: if you add a .bbappend you only have to care about SRCREV |
| 09:23.28 | pompomJuice | serious |
| 09:23.31 | pompomJuice | lol |
| 09:23.35 | pompomJuice | ok let me try that |
| 09:23.38 | ant_work | well, and PV |
| 09:23.47 | ant_work | depends on the reicpe |
| 09:24.03 | pompomJuice | yea |
| 09:24.14 | pompomJuice | could be that I tried that and it failed |
| 09:24.17 | pompomJuice | let me try it again |
| 09:25.35 | bluelightning | pompomJuice: aha, got it... it's the trailing semicolon you've added in the SRC_URI line |
| 09:25.48 | bluelightning | it shouldn't choke on that, but that's why it is at the moment |
| 09:25.56 | bluelightning | shall fix |
| 09:25.57 | pompomJuice | lol |
| 09:26.05 | pompomJuice | pesky |
| 09:26.28 | pompomJuice | ant_work |
| 09:26.40 | pompomJuice | systemd_211.bb does not lend itself to be appended |
| 09:26.46 | pompomJuice | take the SRC_URI for example |
| 09:26.50 | pompomJuice | id have to copy it all |
| 09:26.53 | pompomJuice | and that might change |
| 09:26.57 | pompomJuice | not cool |
| 09:27.04 | pompomJuice | but yea I hear ya |
| 09:27.17 | bluelightning | why does the SRC_URI need changing? |
| 09:29.00 | pompomJuice | well, if I append the recipy |
| 09:29.26 | pompomJuice | id need to change/override the SRC_URI... that URI contains more than just the git url, but some patches too |
| 09:29.32 | pompomJuice | say tomorrow someone adds another patch |
| 09:29.35 | pompomJuice | I wont get it |
| 09:30.02 | ant_work | I suspect it will be easier for you to create a 204.bb using the tarball and set your PREFERRED_VERSION accordingly |
| 09:30.03 | pompomJuice | since I overwrite it with yesterdays SRC_URI that was adapted to take in the tarball instead of the git repo url |
| 09:30.14 | pompomJuice | I agree |
| 09:30.21 | bluelightning | pompomJuice: I understand that you're trying to change the first line of SRC_URI, I'm just not sure why... |
| 09:30.27 | pompomJuice | oh |
| 09:30.29 | pompomJuice | hehe |
| 09:30.48 | pompomJuice | I inherited this project that got baked into kernel 2.6.37 and that kernel only supports up to 204 |
| 09:30.50 | pompomJuice | disaster |
| 09:31.10 | bluelightning | right, but can't 204 be fetched from the original repository? |
| 09:31.14 | pompomJuice | so we are running on borrowed time here |
| 09:31.44 | pompomJuice | No I had trouble because of that semicolon;) |
| 09:32.19 | pompomJuice | wait |
| 09:32.22 | pompomJuice | there is a way |
| 09:32.26 | bluelightning | I'm just thinking you probably shouldn't need to change SRC_URI at all |
| 09:32.33 | bluelightning | just PV and SRCREV |
| 09:32.40 | pompomJuice | oh yea |
| 09:32.49 | pompomJuice | but that brins me to my earlier issue |
| 09:33.12 | bluelightning | (I mean, unless the current SRC_URI includes stuff that is incompatible with 204, such as patches specific to 211 - not sure if that is the case) |
| 09:33.34 | pompomJuice | yea |
| 09:33.39 | pompomJuice | it does |
| 09:33.42 | pompomJuice | now I remember |
| 09:33.49 | pompomJuice | I actually took the last working 204 recipy |
| 09:34.16 | pompomJuice | that broke in 1.6 because the postinst stuff got moved to a package instead ( this is what I am after, this is why I am upgrading from 1.5 to 1.6) |
| 09:35.47 | pompomJuice | I am seeing some strange things in systemd_211.bb |
| 09:36.01 | pompomJuice | file://0001-uClibc-doesn-t-implement-pwritev-preadv.patch |
| 09:36.24 | pompomJuice | Index: systemd-209/src/libsystemd/sd-bus/test-bus-memfd.c |
| 09:36.34 | pompomJuice | that patch refers to 209? |
| 09:36.36 | pompomJuice | anyway |
| 09:36.44 | pompomJuice | maybe it goes from src upwards |
| 09:37.08 | bluelightning | since the patch is effectively applied with -p1 that part of the path is ignored |
| 09:37.13 | pompomJuice | aah |
| 09:37.18 | pompomJuice | so it goes from src |
| 09:37.23 | pompomJuice | makes sense |
| 09:37.29 | bluelightning | it is possibly an indicator that the patch is designed for v209 and upwards though, so it may fail to apply to v204 |
| 09:37.38 | pompomJuice | absolutely |
| 09:37.54 | pompomJuice | angstrom yocto 1.6 might fail with systemd 204 |
| 09:38.00 | pompomJuice | we are walking a tightrope here |
| 09:38.06 | pompomJuice | I dont like it |
| 09:38.12 | pompomJuice | but that is the kinds of things we deal with |
| 09:38.20 | pompomJuice | I am suprized I got this far |
| 09:38.29 | pompomJuice | systemd possibly has 5000 bugfixes a day |
| 09:38.34 | pompomJuice | yet I can roll it back |
| 09:38.39 | pompomJuice | bizarre |
| 09:38.55 | ant_work | using sysvinit is not an option? |
| 09:39.02 | pompomJuice | I use angstrom |
| 09:39.23 | pompomJuice | not sure is that the def way to go there? systemd is nice and does more for me that sysvinit does at the moment |
| 09:39.36 | pompomJuice | systemd is actually the core of everything we do on the device |
| 09:39.52 | ant_work | before was udev, now systemd is the nightmare ;) |
| 09:39.54 | pompomJuice | the irony being it is the one damn program that I cannot upgrade |
| 09:40.11 | bluelightning | systemd 204 should be buildable - but it does seem like overriding the entire SRC_URI value is the right way to go actually |
| 09:40.13 | pompomJuice | have you printed one of those "graphs" yet |
| 09:40.15 | pompomJuice | haha |
| 09:40.19 | pompomJuice | I printed it |
| 09:40.22 | pompomJuice | then I saw it |
| 09:40.23 | bluelightning | if patches are added to the core recipe I think you actually won't want those anyway |
| 09:40.25 | pompomJuice | then I gave up |
| 09:40.36 | *** join/#oe hillct (~hillct@cpe-071-070-219-043.nc.res.rr.com) |
| 09:41.16 | pompomJuice | systemd dependicy graphs |
| 09:41.32 | pompomJuice | absolute madness |
| 09:44.12 | *** join/#oe jbrianceau_away (uid10952@gateway/web/irccloud.com/x-nokcbxuknqfqwkol) |
| 09:49.59 | pompomJuice | ok it is building again... |
| 09:51.17 | pompomJuice | best thing I can do now is migrate anything that does not patch systemd211 to my systemd204 since those might be oe/angstrom spesific |
| 09:52.41 | *** join/#oe eballetbo (~eballetbo@43.Red-2-139-180.staticIP.rima-tde.net) |
| 10:15.29 | *** join/#oe kristoffer (~kristoffe@ua-83-227-162-207.cust.bredbandsbolaget.se) |
| 10:34.06 | *** join/#oe Crofton|work (~balister@pool-71-171-33-112.ronkva.east.verizon.net) |
| 11:07.33 | kroon | bluelightning, hey. are you familiar with the output of bitbake -S printdiff <recipe> ? I wonder if there is a problem in master since the output looks kind of strange, at least on my setup here |
| 11:08.18 | bluelightning | kroon: not completely (I had occasion to try it yesterday), but what are you seeing? |
| 11:10.13 | kroon | bluelightning, the output says bitbake cannot reuse certain cached artefacts from sstate, still, when I bitbake the recipe it seems to run completely from sstate cache |
| 11:10.34 | bluelightning | hmm, that's odd |
| 11:10.48 | bluelightning | are you using SSTATE_MIRRORS at all? |
| 11:12.05 | kroon | no. but I am big fan of wipe-sysroot/cleanup-workdirs/sstate-cache-management, and I almost always do incremental builds |
| 11:12.50 | kroon | so i thought maybe i'm hitting some rare corner case here |
| 11:13.17 | pompomJuice | success angstrom yocto 1.5 to 1.6 only took 2 days |
| 11:13.23 | pompomJuice | good job men! |
| 11:13.55 | bluelightning | kroon: hmm... well it would appear to be broken somehow by the sounds of it |
| 11:14.22 | bluelightning | does it report what the actual difference is? |
| 11:15.12 | kroon | bluelightning, yeah i get reports about things that changed for a couple of depending packages |
| 11:16.29 | kroon | it looks ok on a fresh daisy build |
| 11:17.12 | kroon | oh well. i'll see if I can do something about it, otherwise perhaps file a bug |
| 11:21.18 | pompomJuice | can someone explain how this run-postinsts package works? |
| 11:25.53 | *** join/#oe crxz0193 (~crxz0193@203.247.149.152) |
| 11:33.17 | pompomJuice | if I am on oe-angstrom build I can safely ignore any uClibc patches right? we use elibc |
| 11:42.38 | *** join/#oe tlwoerner (~trevor@linaro/tlwoerner) |
| 11:43.05 | *** join/#oe ldnunes (~ldnunes_@177.35.70.10) |
| 12:06.20 | *** join/#oe hillct (~hillct@cpe-071-070-219-043.nc.res.rr.com) |
| 12:17.36 | *** join/#oe phdeswer (~phdeswer@2001:16d8:cc63:0:1177:36fd:512:5871) |
| 12:22.38 | ant_work | pompomJuice: be prepared, soon it will be musl libc |
| 12:24.37 | pompomJuice | lol |
| 12:25.08 | ant_work | I guess glibc will be for LSB comaptibility |
| 12:25.20 | ant_work | only |
| 12:28.46 | pompomJuice | this is a longshot but |
| 12:29.58 | koen | ant_work: ask yourself for which archs LSB is defined |
| 12:30.02 | pompomJuice | I moved now fro 1.5.1.6 and I get basic kernel panic unable to mount root -> http://hastebin.com/galeforigu.sm |
| 12:30.16 | pompomJuice | how can that be? |
| 12:30.23 | pompomJuice | did not change the kernel at all |
| 12:30.28 | pompomJuice | or my image |
| 12:30.35 | *** join/#oe cbrake (~Thunderbi@cable-pool6-cs-200.doycomm.com) |
| 12:30.39 | pompomJuice | anyone else get this... maybe I get a freeby here |
| 12:30.42 | koen | pompomJuice: which filesystem is on p2 ? |
| 12:30.46 | *** join/#oe adelcast (~adelcast@130.164.62.249) |
| 12:30.59 | pompomJuice | double checking |
| 12:31.16 | pompomJuice | ext4 |
| 12:31.18 | pompomJuice | yea |
| 12:31.51 | pompomJuice | created like this: EXTRA_IMAGECMD_ext4 += " -i 4096 -O ^huge_file -O ^has_journal -E stride=2,stripe-width=256 -b 2048" |
| 12:32.02 | koen | and you set rootfstype=ext4 on the cmdline? |
| 12:32.08 | koen | and ext4 is builtin? |
| 12:32.16 | pompomJuice | I thought my tuning broke it so I took it back to this -> EXTRA_IMAGECMD_ext4 += " -i 4096 -O ^huge_file" |
| 12:32.18 | pompomJuice | still same error |
| 12:32.19 | koen | and if it's a module, is it present in th erootfs? |
| 12:32.37 | pompomJuice | it should be as this worked on 1.5 |
| 12:32.41 | pompomJuice | id be suprosed |
| 12:32.57 | pompomJuice | surprised if I changed aything that broke that, or I cant see how |
| 12:33.11 | koen | that error you posted says "failed to mount", which usually means "fs driver not installed" |
| 12:33.13 | pompomJuice | I only changed minor cosmetic issues with some recipies, other than that 1.6 works out of the box |
| 12:33.18 | pompomJuice | yea I know |
| 12:33.23 | pompomJuice | same kernel as I said |
| 12:33.33 | pompomJuice | let me check though |
| 12:33.43 | pompomJuice | maybe sunspot activity changed my kernel config |
| 12:34.58 | pompomJuice | checking... |
| 12:35.27 | pompomJuice | something is up |
| 12:35.31 | pompomJuice | I cant mount it in linux either |
| 12:37.45 | pompomJuice | seems to be garbage |
| 12:39.24 | pompomJuice | where can I see in the logs what it was doing when creating that FS? |
| 12:40.04 | pompomJuice | when I grep mmcblk0p2 in my image work folder I get nothing |
| 12:40.44 | pompomJuice | it's like the partition is made but no mk2fs.ext4 is being run.. |
| 12:41.41 | pompomJuice | <koen> and you set rootfstype=ext4 on the cmdline? |
| 12:41.44 | pompomJuice | testing that now |
| 12:41.51 | pompomJuice | but I doubt it |
| 12:43.59 | pompomJuice | aah wait |
| 12:44.15 | pompomJuice | I have one more part I forgot to backport, ... |
| 12:45.21 | pompomJuice | yea |
| 12:45.48 | pompomJuice | there is a meta-angstrom/conf/distro/angstrom-v2014.06.conf that I need to backport |
| 12:47.14 | pompomJuice | wait i dont have to anymore |
| 12:47.30 | pompomJuice | that redicilous line that restricted systemd to a particular version has been removed |
| 12:47.30 | pompomJuice | woop |
| 12:47.54 | koen | don't get excited |
| 12:48.05 | pompomJuice | hehe |
| 12:48.05 | koen | there will be a line for 213/214 soon |
| 12:48.17 | pompomJuice | ok then I will not remove the tech I built to override that |
| 12:48.27 | pompomJuice | I wanted to check by you if my trick is legit.. |
| 12:49.17 | pompomJuice | I create a meta-company/conf/distro/angstrom-v2014.06.conf that contains the override and the include to the original file inside meta-angstrom...? |
| 12:49.20 | pompomJuice | is that legal? |
| 12:49.22 | pompomJuice | it worked |
| 12:49.38 | pompomJuice | include ${ANGSTROM_BASE}/conf/distro/angstrom-v2014.06.conf |
| 12:49.38 | pompomJuice | PREFERRED_VERSION_systemd = "204" |
| 12:54.03 | pompomJuice | where in the logs can I find the part where it is creating that partition? |
| 12:54.15 | pompomJuice | I am unclear as in what part it happens |
| 12:54.23 | pompomJuice | I assumed it was in the do_rootfs on the image part |
| 12:54.28 | pompomJuice | but I dont see anything.... |
| 12:58.27 | bluelightning | log.do_rootfs for the image I would suggest... |
| 12:58.37 | pompomJuice | ok so it is there.. |
| 12:58.38 | pompomJuice | checking... |
| 13:09.55 | pompomJuice | the rootfs image seems to work... |
| 13:09.59 | pompomJuice | when I mount the file... |
| 13:16.42 | pompomJuice | qemu-img convert -f raw "${SDIMG}" -O vmdk ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.sdcard.vmdk |
| 13:17.21 | *** join/#oe hillct (~hillct@cpe-071-070-219-043.nc.res.rr.com) |
| 13:25.50 | pompomJuice | looks like these dudes made a new image.bbclass to make an SDCARD image that is vmdk |
| 13:25.56 | pompomJuice | I suspect this has been done in the meantime |
| 13:25.59 | pompomJuice | how? |
| 13:31.21 | *** join/#oe eballetbo (~eballetbo@43.Red-2-139-180.staticIP.rima-tde.net) |
| 13:33.53 | *** join/#oe Anthony__ (~textual@5ec06809.skybroadband.com) |
| 13:37.03 | pompomJuice | I see the vmdk.bbclass uses ext3? is that better than ext4 for sdcards? |
| 13:37.15 | pompomJuice | or is ext3 ext4 w/o journalling? |
| 13:38.14 | koen | ext3 has journalling |
| 13:38.23 | koen | ext4 has extents and other things |
| 13:38.58 | pompomJuice | since our fstype is ext4 with journalling.. pretty vanilla like you would have on your desktop... surely that is not correct? |
| 13:39.22 | koen | sorry |
| 13:39.27 | pompomJuice | I have in the meantime removed journalling... but I have no idea what is the best setup for a embedded device that boots of an SDCARD... |
| 13:39.32 | pompomJuice | I added noatime |
| 13:39.32 | koen | ext4 has journalling and extents |
| 13:39.42 | pompomJuice | I see |
| 13:40.04 | pompomJuice | would you use ext4 or ext3 on SDCARD that has low io? |
| 13:40.11 | koen | ext4 |
| 13:40.22 | pompomJuice | higer io still ext4? |
| 13:40.26 | koen | es |
| 13:40.28 | koen | yes |
| 13:40.36 | koen | ext4 is pretty much always better |
| 13:40.48 | pompomJuice | then why does the vmdk.bbclass use ext3? |
| 13:40.53 | koen | dunno |
| 13:41.25 | pompomJuice | actually I have no idea what I am reading here.. |
| 13:41.32 | koen | I generate sd card images outside OE since I need root privs for some operations |
| 13:41.38 | pompomJuice | it has a dependency on ext3, its not using ext3 |
| 13:41.48 | pompomJuice | hectic |
| 13:42.01 | pompomJuice | what kind of image creation commands require root, and why would they? |
| 13:42.23 | koen | GPT partitioning |
| 13:42.34 | koen | gpt installs the backup block at the end of the disk |
| 13:42.42 | koen | so you need a real sdcard |
| 13:42.45 | koen | among other things |
| 13:43.16 | pompomJuice | I added this to my oinage |
| 13:43.20 | pompomJuice | IMAGE_FSTYPES = "ext4 vmdk" |
| 13:43.33 | pompomJuice | but I see no vmdk being produced... what am I doing wrong |
| 13:43.48 | koen | never used vmdk images |
| 13:43.54 | pompomJuice | shit |
| 13:43.56 | pompomJuice | ok |
| 13:44.06 | pompomJuice | takes long to flash onto the sdcard? |
| 13:44.06 | koen | and most of the sdcard images produced are useless |
| 13:44.20 | koen | a 40MB root partition |
| 13:44.25 | koen | so I need to resize it |
| 13:44.38 | koen | and resizing takes longer than creating it from scratch |
| 13:44.48 | pompomJuice | well it seems at the moment that my partition is being magled by this vmdk process that was in this custom sdcard-image.bbclass in our project... |
| 13:44.56 | pompomJuice | can I just replace it with stock image.bbclass...? |
| 13:45.07 | koen | no idea |
| 13:45.10 | *** join/#oe JokerDoomWork (~JokerDoom@12.179.209.115) |
| 13:45.12 | pompomJuice | ok.. |
| 13:45.44 | pompomJuice | if you do a bitbake myinage |
| 13:45.53 | pompomJuice | what file to you normally flash to your sdcard then? |
| 13:46.20 | ant_work | you extract the tar.gz on a partition - that's all |
| 13:46.46 | pompomJuice | we used to have winimage |
| 13:49.25 | *** join/#oe kgilmer (~kgilmer@2604:2000:84c7:2d01:d806:bf01:2d6b:5d5b) |
| 13:49.51 | ant_work | koen: if you are short in time I can build for hx4700 and upload on github |
| 13:50.02 | ant_work | tolchain is already built |
| 13:50.05 | ant_work | +o |
| 14:04.08 | *** join/#oe kgilmer (~kgilmer@184.152.73.182) |
| 14:06.37 | *** join/#oe jkridner (~jkridner@pdpc/supporter/active/jkridner) |
| 14:11.20 | *** join/#oe adelcast (~adelcast@130.164.62.249) |
| 14:16.28 | pompomJuice | I disabled journalling on our ext4 that runs on sdcard |
| 14:16.39 | pompomJuice | but then someone yanked the power and the fs broke |
| 14:16.47 | pompomJuice | an fsck.ext4 solved the problem... |
| 14:17.03 | pompomJuice | what would be the reccomendation if you want to extend your sdcard image life |
| 14:17.30 | pompomJuice | our sdcard is 8GB in size, 4 GB of it acts like a rolling buffer |
| 14:18.55 | koen | 1) don't powercycle the card |
| 14:18.59 | koen | 2) don't write to the card |
| 14:19.14 | koen | it seems powercycling kills the cards faster than writing to them |
| 14:23.44 | bencoh | a rolling buffer is prolly a sure way to kill it ;) |
| 14:25.02 | pompomJuice | wait |
| 14:25.07 | pompomJuice | let me explain further |
| 14:25.17 | pompomJuice | because this issue is a massive bother for me |
| 14:25.26 | pompomJuice | our device logs electricity |
| 14:25.29 | pompomJuice | at a huge rate |
| 14:25.37 | pompomJuice | then spits the data into a local sqlite3 db |
| 14:25.44 | pompomJuice | that is the "rolling buffer" |
| 14:25.55 | pompomJuice | then the data is uploaded to the mother server |
| 14:26.09 | pompomJuice | I have noticed that our devices come back after 3 months or so with corrupt filesystems |
| 14:26.26 | pompomJuice | I presume it is because the sdcard was warn out... but I am a noob I have no feeling for these things... |
| 14:26.46 | pompomJuice | in this channel however I am sure someone knows the answer |
| 14:26.50 | bencoh | "huge" rate ? |
| 14:26.55 | pompomJuice | well |
| 14:27.00 | pompomJuice | it depends |
| 14:27.19 | pompomJuice | let me ask the data dude |
| 14:27.38 | pompomJuice | depends on how it is logging |
| 14:28.05 | *** join/#oe Crofton (~balister@pool-108-44-88-252.ronkva.east.verizon.net) |
| 14:28.20 | *** join/#oe Crofton|work (~balister@pool-108-44-88-252.ronkva.east.verizon.net) |
| 14:28.31 | pompomJuice | I would say |
| 14:28.33 | pompomJuice | worse case |
| 14:28.36 | *** join/#oe Crofton|work (~balister@pool-108-44-88-252.ronkva.east.verizon.net) |
| 14:28.44 | pompomJuice | it will fill up that 4GB in 6 months or so |
| 14:28.55 | pompomJuice | but |
| 14:29.01 | pompomJuice | it could also fill that up in 2 days |
| 14:29.12 | pompomJuice | so we wont deploy a device with such settings |
| 14:30.33 | pompomJuice | but the device must go into a factory and we dont wanna maintain it for the next 5 years |
| 14:35.35 | pompomJuice | back to the hole vmdk issue |
| 14:36.28 | pompomJuice | I make development builds on a Oravle VBOX VM on my windows PC... heresy I know... |
| 14:36.35 | pompomJuice | the USb interface is dog slow |
| 14:36.46 | pompomJuice | so normally I would copy the vmdk out into windows and write it |
| 14:36.51 | pompomJuice | but now vmdks are bad |
| 14:37.00 | pompomJuice | what would be the alternative? |
| 14:37.12 | *** join/#oe Crofton|work (~balister@pool-108-44-88-252.ronkva.east.verizon.net) |
| 14:37.27 | *** join/#oe Noor (~quassel@110.93.212.98) |
| 14:37.56 | *** join/#oe Crofton|work (~balister@pool-108-44-88-252.ronkva.east.verizon.net) |
| 14:42.09 | *** part/#oe _chase_ (~a0271661@nat/ti/x-vbvzelrswkakdcai) |
| 14:54.53 | *** join/#oe kgilmer (~kgilmer@65.209.23.146) |
| 15:03.04 | *** join/#oe eFfeM (~frans@c73189.upc-c.chello.nl) |
| 15:26.15 | *** join/#oe blilly (~blilly1@c-50-170-54-92.hsd1.wa.comcast.net) |
| 15:41.33 | *** join/#oe _chase_ (~a0271661@nat/ti/x-kquhpogdwpbucnxr) |
| 16:00.49 | *** join/#oe dvhart (~dvhart@134.134.137.75) |
| 16:02.50 | *** join/#oe belen (~Adium@192.198.151.44) |
| 16:17.58 | *** join/#oe nitink (~nitink@134.134.137.71) |
| 16:39.21 | *** join/#oe JokerDoomWork (~JokerDoom@12.179.209.115) |
| 16:47.55 | *** join/#oe belen (Adium@nat/intel/x-wqpyppruvvxspxwh) |
| 16:52.27 | *** join/#oe dvhart (dvhart@nat/intel/x-dzofniiczhkyvlap) |
| 17:18.17 | blilly | Any chance someone can tell me how to force the rootfs to complete despite conflicts? |
| 17:18.25 | *** join/#oe florian (~fuchs@Maemo/community/contributor/florian) |
| 17:19.18 | *** part/#oe JokerDoomWork (~JokerDoom@12.179.209.115) |
| 17:19.47 | bluelightning | blilly: what is the conflict? |
| 17:22.48 | blilly | bluelightning: turns out that qt4-x11 and qt4-embedded will happily coexist, unless you build qtserialport with dev, then it pukes on qt-assistant and friends |
| 17:23.41 | bencoh | (qt4-embedded ? undead from the past ?) |
| 17:23.57 | bluelightning | blilly: it sounds like you have a dependency on the main qt4-x11 or qt4-embedded package, which almost certainly is not what you want (because that pulls in absolutely every package that Qt builds, rather than specifically the packages you would normally want) |
| 17:24.09 | bluelightning | bencoh: still "current" in Qt4 |
| 17:24.13 | koen | qtserialport is cmake based, right? |
| 17:25.02 | blilly | bluelightning: turns out I'm ok with having it pull everything in, since the image is meant for demonstration/development |
| 17:25.08 | blilly | koen: yes, I believe so |
| 17:25.24 | koen | the stock findQT cmake macros suck |
| 17:25.47 | koen | (isn't qtserialport qt5?) |
| 17:26.47 | blilly | koen: my understanding is that exists in both places |
| 17:27.06 | blilly | or can |
| 17:30.23 | bluelightning | blilly: so the short answer is there's no way to force do_rootfs to ignore conflicts, you will have to resolve them |
| 17:30.53 | bluelightning | one way to do that is not to install all of qt; just install those packages you need and exclude the overlapping packages |
| 17:31.27 | blilly | ultimately I need the includes, and couldn't care less about the rest |
| 17:31.52 | blilly | but am not sure how to accomplish that |
| 17:33.50 | bluelightning | probably just explicitly install all of libqt*-dev |
| 17:34.47 | bluelightning | where you can find the actual names under packages-split in the workdir for the Qt recipe i.e. bitbake -e qt4-x11-free | grep ^WORKDIR= |
| 17:34.59 | bluelightning | bbl |
| 17:35.06 | blilly | bluelightning: thanks |
| 18:06.07 | *** join/#oe tgall_foo (~tgall@70.35.96.184) |
| 18:06.07 | *** join/#oe tgall_foo (~tgall@linaro/tgall-foo) |
| 18:06.23 | *** join/#oe woglinde (~henning@g225136067.adsl.alicedsl.de) |
| 18:11.41 | *** join/#oe dvhart (dvhart@nat/intel/x-qawqpdjbaglyszdw) |
| 18:13.05 | *** join/#oe neabax (~neabax@173.247.199.210) |
| 18:16.21 | *** join/#oe Crofton|work (~balister@pool-71-171-41-23.ronkva.east.verizon.net) |
| 18:17.01 | *** join/#oe Crofton (~balister@pool-71-171-41-23.ronkva.east.verizon.net) |
| 18:32.14 | *** join/#oe neabax (~neabax@173.247.199.210) |
| 18:35.38 | *** join/#oe neabax (~neabax@173.247.199.210) |
| 18:40.57 | *** join/#oe neabax (~neabax@173.247.199.210) |
| 18:47.55 | *** join/#oe nitink (nitink@nat/intel/x-pwmvdfiypvpnfalt) |
| 18:51.23 | *** join/#oe NightMonkey_ (~NightrMon@173-164-139-193-SFBA.hfc.comcastbusiness.net) |
| 19:04.16 | *** join/#oe dvhart (dvhart@nat/intel/x-eqvinooobpegzzmc) |
| 19:09.58 | *** join/#oe sr105|away (~sr105@65349hfc19.tampabay.res.rr.com) |
| 19:10.43 | *** join/#oe florian (~fuchs@Maemo/community/contributor/florian) |
| 19:24.01 | *** join/#oe sr105|away (~sr105@65349hfc19.tampabay.res.rr.com) |
| 19:50.22 | *** join/#oe belen (~Adium@5751cae3.skybroadband.com) |
| 19:56.37 | *** join/#oe ohama (ohama@cicolina.org) |
| 19:57.31 | *** join/#oe hillct (~hillct@cpe-071-070-219-043.nc.res.rr.com) |
| 20:00.45 | gokukameha | hi |
| 20:01.33 | gokukameha | i need help in making an os for my beaglebone board |
| 20:01.54 | gokukameha | i am new to oe could any one help me getting started |
| 20:03.22 | *** join/#oe phdeswer (~phdeswer@a88-112-199-213.elisa-laajakaista.fi) |
| 20:03.50 | *** join/#oe nitink1 (~nitink@134.134.139.76) |
| 20:03.55 | *** join/#oe jackmitc1 (~Thunderbi@cpc22-cmbg15-2-0-cust21.5-4.cable.virginm.net) |
| 20:09.58 | *** join/#oe s3v3n62mm2 (~s3v3n62mm@141.45.250.75) |
| 20:18.34 | *** join/#oe Leatherface- (~leatherfa@2001:470:dd7d:0:c1a0:10f0:6e14:cfc6) |
| 20:32.50 | *** join/#oe nitink (nitink@nat/intel/x-ebkplrspsfucssza) |
| 20:36.52 | *** join/#oe ant_home (~andrea@host160-222-dynamic.7-79-r.retail.telecomitalia.it) |
| 20:41.08 | *** join/#oe dvhart (dvhart@nat/intel/x-tnnotetwfpeioifm) |
| 21:15.40 | *** join/#oe s3v3n62mm (~s3v3n62mm@p3EE2F443.dip0.t-ipconnect.de) |
| 21:31.54 | blilly | bluelightning: Thanks again, that did the trick. |
| 21:37.46 | blilly | gokukameha: You should look here: https://www.yoctoproject.org/download/texas-instruments-arm-cortex-a8-development-board-beaglebone , here: beagleboard.org/project/yocto-project/ , and the #beagle IRC channel |
| 21:38.28 | blilly | gokukameha: http://www.crashcourse.ca/wiki/index.php/Building_for_BBB_using_OpenEmbedded and here also |
| 21:53.33 | *** join/#oe dvhart (~dvhart@134.134.137.75) |
| 21:55.17 | darknighte | signs off for a long birthday weekend |
| 22:01.35 | *** part/#oe _chase_ (~a0271661@nat/ti/x-kquhpogdwpbucnxr) |
| 22:13.51 | *** join/#oe hbeck (~hbeck@38.114.125.83) |
| 22:26.29 | ant_home | JaMa: we owe you all a beer for the 14 days build ;) |
| 22:31.06 | JaMa | ant_home: hehe, thanks |
| 22:31.38 | JaMa | <PROTECTED> |
| 22:32.46 | ant_home | at least for june ;) |
| 22:35.09 | *** join/#oe neabax (~neabax@173.247.199.210) |
| 22:36.34 | JaMa | ant_home: you haven't seen how we dring good czech beer :) |
| 22:36.49 | JaMa | ups, first typo after 6 beers :) |
| 22:37.03 | ant_home | heh |
| 22:45.40 | *** join/#oe neabax (~neabax@173.247.199.210) |
| 22:57.55 | *** join/#oe ftagius (~fagius@99-178-168-231.lightspeed.rlghnc.sbcglobal.net) |
| 23:03.16 | *** join/#oe NvrBst (~NvrBst@gateway/tor-sasl/nvrbst) |
| 23:08.33 | *** join/#oe neabax (~neabax@173.247.199.210) |
| 23:10.29 | *** join/#oe hillct (~hillct@cpe-071-070-219-043.nc.res.rr.com) |
| 23:16.29 | *** join/#oe dvhart_ (dvhart@nat/intel/x-lcbxsrisiadyeonq) |
| 23:45.04 | *** join/#oe neabax (~neabax@173.247.199.210) |