IRC logs for #automotive for Friday, 2016-04-08

*** jlrmagnus has quit IRC00:05
*** IOBN has quit IRC00:32
*** IOBN has joined #automotive00:33
*** rstreif_ has quit IRC02:09
*** rokrau_ has joined #automotive04:10
*** Sisco has quit IRC04:43
*** Sisco has joined #automotive04:43
*** AlisonChaiken has joined #automotive05:26
*** sanjeev has quit IRC05:52
*** gmacario1 has quit IRC06:54
*** Saint_Isidore has quit IRC06:57
*** Saint_Isidore has joined #automotive06:57
*** mps_ has joined #automotive07:03
*** Saint_Isidore has quit IRC07:03
*** jobol has joined #automotive07:03
*** gmacario has joined #automotive07:04
*** Saint_Isidore has joined #automotive07:08
*** Saint_Isidore has quit IRC07:12
CTtpollardit is listed as pedroalvarez in the kernel recipe due to the url source ref which was used in the meta-raspberry that tizen forked disappearing from anholts tree. Fortunately at the time baserock were lorrying this repo into their repos07:18
*** ctbruce has joined #automotive07:33
*** Tarnyko has joined #automotive07:48
*** ctbruce has quit IRC07:56
*** ctbruce has joined #automotive07:59
*** IvanS has joined #automotive08:16
*** jonathanmaw has joined #automotive08:48
*** gmacario has quit IRC09:02
*** gunnarx has joined #automotive09:56
*** Sisco has quit IRC10:15
*** gunnarx has quit IRC10:46
*** sanjeev has joined #automotive10:49
*** gunnarx has joined #automotive10:50
*** gunnarx has joined #automotive10:50
*** Sisco has joined #automotive11:01
*** sanjeev has quit IRC11:03
*** Tarnyko has quit IRC11:28
* CTtpollard updates meta-genivi-demo to ivi-extension-1.9.111:37
CTtpollardno more races! (hopefully)11:37
*** gunnarx has quit IRC12:10
*** gunnarx has joined #automotive12:14
gunnarxCTtpollard, no more races.  Sounds promising but I saw some error in navit on one of the builds.  I have the feeling there are similar race conditions hiding in quite a few yocto/bitbake recipes :(12:36
gunnarxgood if you fixed this one though12:37
CTtpollardthe navit one is one that I've never been able to produce locally, and sadly fixed on the qemu-ci pipeline one by increasing the parallel make and bb thread count12:39
*** gunnarx has quit IRC12:45
*** gunnarx has joined #automotive12:49
*** gunnarx has joined #automotive12:49
gunnarxWe're really pushing the limits of the build capacity today.   Not at all unexpected though.13:04
gunnarxHave to make another push to get build machines set up in the companies that have said they would13:04
CTtpollardgunnarx: I hope to have available time to contribute more, but the past few weeks of firefighting have not allowed it13:05
gunnarxSounds great.  I know there has been firefighting, but hopefully soon we should be able to lift the CIAT stuff another level.13:08
CTtpollardok, the experimental porter branch should have the last commit for the day13:42
CTtpollardunless slawr sends more stuff my way13:42
CTtpollardI believe chbae is off today, and wouldn't want to push commits to his branch without informing him13:52
*** persia has quit IRC14:34
*** persia has joined #automotive14:37
*** obdii_ has quit IRC14:39
*** sanjeev has joined #automotive14:48
*** jlrmagnus has joined #automotive14:57
*** jlrmagnus has joined #automotive14:58
*** rokrau_ has joined #automotive15:11
rokrau_jonathanmaw: Hey - quick question for you. If I simply edit the source code of hid-multitouch.c by hand and delete the offending lines - what is the correct way to get the image rebuilt? Will bitbake genivi-demo-platform pick up the changes automagically?15:13
jonathanmawrokrau_: not automagically.15:14
jonathanmawI believe the recommended way of doing it is to force-build the component you've manually hacked around with15:14
jonathanmawi.e. `bitbake -c compile -f linux-renesas`15:14
CTtpollardhowever the recipes now support this, so should pick up on it automagically :)15:15
rokrau_okay - i did bitbake virtual/kernel -c clean compile15:16
rokrau_followed by bitbake genivi-demo-platform15:16
rokrau_which took very long15:16
CTtpollardclean would reset manual changes, I think15:16
rokrau_i better check15:16
CTtpollardI think it sets the build source back to the state after it was fetched15:17
rokrau_why can't we apply Jonathan's patch ontop of the Changyeok's branch15:17
CTtpollardrokrau_: have you seen my latest email to the list?15:18
CTtpollardwe've pushed everything upstream in the experimental porter branch15:19
CTtpollardI believe chbae is away so is unable to merge latest updates to his raspi branch, they can be done manually though15:19
rokrau_yes but i need both raspberry pi and ported versions.15:19
rokrau_CTtpollard: yeah - i did it manually - it is just deleting a line in a file15:20
CTtpollardmeta-genivi-demo now supports this, it carries 2 different patches which for the porter or raspi215:22
rokrau_It is already weekend in Korea so I don't expect Changhyeok to do anything for the next two days15:22
CTtpollard(the kernels are different so we have to carry separate ones unfortunately)15:22
rokrau_CTtpollard: so what do i do there - switch meta-genivi-demo to a branch manually before building?15:23
rokrau_then resets it again15:23
CTtpollardyou can comment the submodule update out of the init script if you wish15:24
CTtpollardor do the checkouts after running it15:24
CTtpollardor run source poky/oe-init-build-env gdp-src-build which is the actual bitbake part of the script15:26
*** CTtpollard has quit IRC15:27
rokrau_so which branch is it then ? master or jonathans first or second faytech fix?15:30
jonathanmawrokrau_: meta-genivi-demo master has the latest changes15:35
jonathanmawi.e. the faytech fix, plus all the latest fixes for other things.15:35
rokrau_jonathanmaw: thanks!15:36
rokrau_jonathanmaw: and you tried building it with cbae/raspberrypi?15:36
jonathanmawrokrau_: not me. I believe CTtpollard has, but he's just left the office.15:37
gunnarxoh, is this where the action is? :)15:37
jonathanmawI'll check what version his RPi2 is currently running.15:37
gunnarxleaving the office, sigh... I would but it's so far to walk to the car.15:38
rokrau_gunnarx: all weekend long :-)15:38
*** mvick has joined #automotive15:38
jonathanmawrokrau_: looking at what's on the rootfs, I can confirm that it's got the faytech monitor fix and the RVI SOTA changes in15:45
jonathanmawI don't think wayland-ivi-extension provides enough version information for me to check whether that's changed as well.15:46
*** aeiche has quit IRC15:48
jonathanmawrokrau_: I've just been given access to CTtpollard's build machine. unless I'm looking in the wrong place, it looks like it hasn't been tried with the wayland-ivi-extension update.15:59
*** AlisonChaiken has quit IRC16:05
jonathanmawrokrau_: started a build of the RPi2, myself. I've got no official branch for what I'm trying (chbae/raspberrypi2, with meta-genivi-demo checked out to master and local.conf fiddled with to prefer wayland-ivi-extension 1.9.1 and USE_FAYTECH_MONITOR = "1")16:05
jonathanmawthe build's started successfully (so the basic recipe parsing has succeeded)16:05
jonathanmawI'll be leaving this build running over the weekend, and will be able to test it Monday morning, as long as the build doesn't consume all my disk space when left unattended16:08
*** jobol has quit IRC16:10
*** jbocklage has quit IRC16:18
*** sanjeev has quit IRC16:39
rokrau_jonathanmaw: Okay - my build for the RPi finished - will be testing it in one hour when i get to the office16:49
rokrau_jonathanmaw: SDK also built and installed - fingers crossed it's the "gold" version for me16:49
rokrau_jonathanmaw: renesas build still ongoing16:49
rokrau_jonathanmaw: that build box is a whopper and i will be dragging it with me to Paris16:50
jonathanmawrokrau_: ah, that'll be interesting to see16:54
jonathanmawalso, it's time for me to leave the office. have a good weekend :)16:54
rokrau_good weekend16:54
*** jonathanmaw has quit IRC16:55
*** rokrau_ has quit IRC16:59
*** ctbruce has quit IRC17:01
*** gunnarx has quit IRC17:01
*** Saint_Isidore has joined #automotive17:23
*** fredcadete_ has quit IRC17:37
radiofreerokrau: regarding your mac address issue17:39
radiofreei've used boards in the past that do the same thing, hence my suggestion to just hack the driver to accept a mac via the command line17:40
radiofreeit's a dreadful "solution", but at least preserves some level of sanity17:40
radiofreeyou could probably do it with some ifconfig script as well17:41
*** rokrau_ has joined #automotive17:45
rokrau_radiofree: okay thanks for the hint. I will try the setup for uboot to see if that fixes the issue - not keen to hack another driver's code17:45
radiofreetry it without "saveenv" first17:48
radiofreesetenv ethaddr fo:ob:ar:b.....17:48
radiofreethen boot17:48
radiofreeethaddr: Ethernet MAC address for first/only ethernet interface (= eth0 in Linux).17:48
radiofreeThis variable can be set only once (usually during manufacturing of the board). U-Boot refuses to delete or overwrite this variable once it has been set.17:48
*** rokrau_ has quit IRC17:59
rokrauradiofree: okay - thanks for the hint!18:57
rokrauradiofree: I am just worried about what we will find at the genivi amm for the boards. the configuration to set a static ip is in a connman file that is in a folder with the mac-address part of its name18:58
rokrauradiofree: unless there is another way to accomplish this via uboot config. not sure i udnerstand the interaction between uboot and OS here.18:59
rokraui still don't have touch on the Faytech :-(19:00
rokraugit lo19:53
*** IvanS has quit IRC20:00
rokraustill puzzled why the Porter board would create a new mac address on every boot22:37
*** mps_ has quit IRC22:46
*** jlrmagnus has quit IRC22:52
*** sanjeev has joined #automotive22:55
*** mvick has quit IRC23:10

Generated by 2.14.0 by Marius Gedminas - find it at!