IRC logs for #automotive for Wednesday, 2015-08-19

*** waltminer has quit IRC00:04
*** lj50036 has joined #automotive00:27
*** lj50036 has left #automotive00:28
*** waltminer has joined #automotive00:38
*** jlrmagnus has joined #automotive02:15
*** waltminer has quit IRC02:17
*** jlrmagnus has quit IRC02:52
*** felipealmeida has quit IRC03:05
*** Dthiriez has quit IRC03:06
*** felipealmeida has joined #automotive03:12
*** FaH_auto has joined #automotive05:20
*** FabienH has joined #automotive05:21
*** fredcadete has joined #automotive06:57
fredcadeteCTtpollard: at AWTCE we recently upgraded from meta-ivi 7 to 807:08
fredcadeteit was mostly a painless upgrade07:09
fredcadetecorrection: from meta-ivi 6 to 807:09
fredcadetewe only had one small issue: the setting of the root password to "root" was moved from a base-passwd bbappend to the ivi-image; so in our own image root became passwordless07:11
fredcadeteit was easy enough to fix07:11
*** FabienH has quit IRC07:16
CTtpollardfredcadete: good to know07:35
CTtpollardI'm sorry I missed your response last night07:36
fredcadeteno problem07:42
fredcadetedid you get it working in the end?07:42
*** jacobo has joined #automotive07:42
CTtpollardfredcadete: I have just come back to my build this morning and it has failed on common07:43
CTtpollardI'm not sure if it's the contents of the append, or the location etc07:43
fredcadeteouch07:44
CTtpollardthe log it generated in buildspec isn't much use07:46
CTtpollardmy append file is http://paste.baserock.org/upeyitagex07:50
fredcadeteI'll bite the bullet and setup a GDP build environment07:50
CTtpollardlocated /meta-genivi-demo/recipes-extended/common-api/common-api-c++_2.1.6.bbappend07:50
CTtpollarddo you think I've placed the append in the right dir fredcadete ?07:54
fredcadeteit depends on your bblayers.conf07:55
fredcadetebut the best way to check is to call bitbake-layers07:55
fredcadetefrom the same shell and directory you would call `bitbake`, you can call `bitbake-layers show-appends`07:55
fredcadeteit will show you all appends that are being taken07:55
fredcadeteyou can check if your new append is there07:55
CTtpollardcommon-api-c++_2.1.6.bb:07:56
CTtpollard  /home/ubuntu/GDP/poky/../meta-genivi-demo/recipes-extended/common-api/common-api-c++_2.1.6.bbappend07:56
CTtpollardso it's taking it at least07:57
fredcadeteyes07:59
fredcadetethen what error do you get?07:59
CTtpollardone moment, I'll pull up the last trace07:59
CTtpollardhttp://paste.baserock.org/uxuqoxevel08:00
*** mdunford has joined #automotive08:02
fredcadetethat's something already, it did take your SRC_URI and compile everything08:02
fredcadetecan you give me /home/ubuntu/GDP/gdp-koelsch/tmp/work/cortexa15hf-vfp-neon-poky-linux-gnueabi/common-api-c++/2.1.6-/temp/log.do_package_write_rpm.24587 ?08:02
fredcadeteI have the feeling my advice yesterday about PR="" may be the problem08:03
CTtpollardyes sure08:03
*** jacobo has quit IRC08:04
CTtpollardhttp://paste.baserock.org/hugegolawi08:05
CTtpollardsame again really08:05
fredcadeteah08:05
fredcadeteerror: line 4: Empty tag: Release:08:05
*** jacobo has joined #automotive08:06
CTtpollardah, so yes maybe it doesn't like empty tags08:06
CTtpollardwell, an empty variable08:07
fredcadetehttp://www.yoctoproject.org/docs/1.8/ref-manual/ref-manual.html#var-PR08:09
fredcadetecould you try setting it to "r0" ? or simply not overriding it in the bbappend08:09
CTtpollardalthough the tag variable is ${PV} , as it is in the patch08:10
CTtpollardcan give it a shot08:10
CTtpollardit has passed through the task list, looking promising08:11
fredcadeteif you want to look at the sausage factory you can go into the work directory. It will probably have an intermediate file describing the package. And my guess is that in this file, line 4 is something like "Release : "08:12
fredcadeteok, I hope that settles it08:13
jeremiahNOTE: Tasks Summary: Attempted 4484 tasks of which 2948 didn't need to be rerun and all succeeded. <-- weee!08:13
jeremiahBuilt GDP w/ Renesas BSP08:13
CTtpollard:)08:14
CTtpollardwhich board jeremiah?08:14
jeremiahPorter08:14
CTtpollard:)08:14
jeremiahNow I have to flash it08:15
jeremiahOr is it installed onto an SD card?08:15
CTtpollardheh, I call it flash08:16
jeremiahw00t08:39
CTtpollard'Attempted 4484 tasks of which 3136 didn't need to be rerun and all succeeded.'08:48
CTtpollardthe append looks to be working!08:48
*** txenoo has joined #automotive08:49
CTtpollardlet's see if it boots, I have being having problems with the koelsh and booting08:49
*** jonathanmaw has joined #automotive08:55
jeremiahw00t!08:58
jeremiah09:03
jeremiahla la la09:03
*** wschaller has joined #automotive09:05
fredcadeteCTtpollard: by chance are your booting problems a division by zero from the fb_flashcursor code in the kernel?09:09
fredcadeteor something like that in the serial trace09:09
CTtpollardI'll dump the trace one minute09:10
fredcadetebecause I have that on Lager once in a while09:10
CTtpollardfredcadete : http://paste.baserock.org/etiqoxayej09:12
fredcadetenope, that's rootfs problems09:12
CTtpollardI initially had the same problem with porter, but a second build seemed to avoid it09:12
fredcadeteI'll post my dump just in case someone is interested09:12
CTtpollardcool09:12
radiofreeare you sure you're setting the correct partition to mount as /?09:13
radiofree"Kernel panic - not syncing: No init found.  Try passing init= option to kernel. See Linux Documentation/init.txt for guidance."09:13
fredcadetehttp://pastebin.com/84jjQKnx09:14
CTtpollardradiofree: I think so, as it it's finding the kernel correctly under /boot09:14
fredcadeteCTtpollard: it's uboot that finds the kernel09:14
radiofreeit'll be your bootloader loading that09:14
fredcadetethen the kernel boots and it needs to find /sbin/init09:14
radiofreeinterrupt u-boot and and paste the bootargs09:14
radiofree"printenv"09:15
CTtpollardhttp://paste.baserock.org/ovobiyedad09:15
radiofreeso you're running bootcmd_sd and then bootcmd?09:16
radiofreeis your rootfs on the sd card?09:17
radiofreei'd guess that the sd card isn't being exposed as /dev/mmcblk0p1 to the kernel once booted, try changing root to /dev/sda109:17
*** txenoo has quit IRC09:19
CTtpollardradiofree: one thing of difference between my sd for porter and koelsch is that sbin is owned by root on the koelsch, and not on the porter09:19
CTtpollardcould that be an issue?09:19
CTtpollardas in, I can't navigate to that directory without root09:23
radiofreei wouldn't have thought that was the issue here09:23
radiofreewhere do you flash u-boot to? if it's not on the sdcard you could try your porter rootfs, just copy boot/uImage-r8a7791-koelsch.dtb to it, and boot/uImage (call it uImage-koelsch and update bootcmd_sd)09:25
CTtpollardI put them under /boot on the sd09:27
CTtpollardthis is just following the steps on http://wiki.projects.genivi.org/index.php/Hardware_Setup_and_Software_Installation/koelsch%26porter09:28
fredcadetethe ownership difference is weird09:30
radiofreewell i have no idea about koelsch or porter boards, but your rootfs should essentially be the same (graphics probably won't work)09:30
fredcadeteif /sbin/init is not executable on root it may cause that message09:31
fredcadete*by root*09:31
radiofreeso you can check to see if it's an issue with your koelsch sdcard by copying the koelsch kernel and devicetree to your porter sdcard09:31
fredcadete+1 radiofree09:31
*** txenoo has joined #automotive09:35
CTtpollardyup it boots that09:43
CTtpollardso there's something wrong when copying to the sd09:43
radiofreeok, well your koelsch sdcard is, to use the technical term, buggered09:43
radiofreelooking at those instructions is looks like they both have the same source base, so you can probably get away with dding that porter card to another one and just extracting the koelsch graphics drivers onto it09:46
CTtpollardthe gfx seem fine09:47
radiofreeoh right, i just assumed they'd have a different gpu considering there's two branches of instructions09:50
radiofreehowever the filenames seem to match, i guess there's hardware on the koelsch that needs additional drivers?09:51
radiofreethe scarily named copy_proprietary_softwares-v1*.sh09:51
CTtpollardheh09:51
fredcadetethose will probably be the same between koelsch and porter since they are both RCar-M209:53
fredcadetebut different for Lager09:53
*** apinheiro has joined #automotive10:00
jeremiahI hate fussing with the SD card, always an issue for me.10:08
jeremiahI think the GPU, versioned as M2, is the same in Porter and Koelsch10:10
CTtpollardjeremiah: do you follow the tar option for copying onto the sd as the guide says? I think this is my issue10:11
radiofreewiki should be updated then, for the Koelsch it states "If you do not already have it as part of a working relationship with Renesas, the R-Car M2 graphics driver can be obtained by contacting Renesas. " whereas it appears you can just use the instructions for the porter10:12
radiofreealso, that probably means you can simply the instructions to just one set, the only difference will be the device tree you use at the end?10:12
radiofrees/simply/simplify/10:12
CTtpollardI agree they can probably be shortened / simplified into one in certain places10:16
*** wschaller has quit IRC10:30
*** wschaller has joined #automotive11:11
CTtpollardkoeslch build booting \o/ although I still think I've got a permissions difference, a new sd card looks to have remedied it11:12
fredcadeteCTtpollard: if you don't untar with "sudo" the file permissions will not be properly set11:12
fredcadeteeven if your user has write permissions in the SD card11:13
fredcadeteI have made that mistake once or twice11:13
CTtpollardfredcadete: yep I'm usign sudo, in the exact method as listed11:13
fredcadetemust be another thing then11:13
fredcadeteI like to avoid sudo when I can but in this case I was not able to11:14
CTtpollardI can see at a quick glance in nautilus that the file permissions differ between the two rootfs11:14
*** wschaller has quit IRC12:08
*** wschaller has joined #automotive12:30
*** josemic has joined #automotive12:56
jeremiahCTtpollard: I usually use dd, but that can be, umm, problematic.13:14
*** wschaller has quit IRC13:24
*** waltminer has joined #automotive13:27
*** wschaller has joined #automotive13:35
*** txenoo has quit IRC14:02
*** jlrmagnus has joined #automotive14:46
waltminerHi fredcadete nice to speak to you on the SC call. Thanks for participating14:47
jlrmagnusMorning14:47
waltminergood morning Magnus14:49
fredcadetewaltminer: same here :)14:49
jlrmagnusMorning Walt.14:50
CTtpollardI've emailed genivi-project with my patch proposal to get gdp building again for the time being14:59
jeremiahw00t15:00
jeremiahjlrmagnus: morn15:00
* paulsherwood apologises for missing the sc call15:00
* fredcadete thinks paulsherwood missed a hell of a party15:01
jeremiaho_O15:02
CTtpollardsounds intriguing15:05
* waltminer is still enjoying the coffee and biscuits that were part of the SC meeting15:29
*** txenoo has joined #automotive15:38
*** rosch has joined #automotive15:40
*** txenoo has left #automotive15:41
CTtpollardjeremiah: are you still wanting genivi to have repo with the functionality of gunnar's sub-module approach?15:54
*** Dthiriez has joined #automotive16:12
myselfit must be morning on the west coast!16:26
jlrmagnusIt is.16:27
jlrmagnusMornign David.16:27
fredcadetethat means I should already have gone home16:27
fredcadetegood evening all16:27
*** fredcadete has quit IRC16:27
jlrmagnusDthiriez: I've updated my local rvi_core master so that it works a bit better.16:27
jlrmagnusI'm currently validating the certs sent out by Rudi's backend server code.16:27
*** Egy has joined #automotive16:31
*** Egy has quit IRC16:33
*** Egy has joined #automotive16:33
*** Egy has quit IRC16:46
*** Egy has joined #automotive16:49
*** josemic has left #automotive17:06
*** Dthiriez has quit IRC17:09
*** Dthiriez has joined #automotive17:09
*** wschaller has quit IRC17:14
*** apinheiro has quit IRC17:59
*** waltminer has quit IRC18:21
*** Egy has quit IRC18:23
*** jacobo has quit IRC18:32
*** Egy has joined #automotive18:40
*** Egy has quit IRC18:56
*** Egy has joined #automotive19:12
*** waltminer has joined #automotive19:23
jlrmagnusDthiriez?19:23
*** RzR is now known as rZr19:23
*** waltminer has quit IRC19:54
*** Egy has quit IRC20:05
*** waltminer has joined #automotive21:11
*** waltminer has quit IRC21:55
*** waltminer has joined #automotive22:10
*** jlrmagnus has left #automotive23:37
*** waltminer has quit IRC23:46
*** waltminer has joined #automotive23:52

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!