**** BEGIN LOGGING AT Tue Jun 18 02:59:57 2019
Jun 18 04:59:11 <JaMa> morning
Jun 18 05:06:31 <Herrie> Morning!
Jun 18 05:33:42 <JaMa> OSE with Yocto upgrade should be released this Friday
Jun 18 05:36:29 <Herrie> JaMa: Nice :)
Jun 18 05:36:44 <Herrie> Still curious why my googleapi includes aren't ending up at the right place :S
Jun 18 05:37:12 <JaMa> did you modify the recipes fro meta-webosose in any way?
Jun 18 05:49:37 <Herrie> JaMa: except for excessive inherits and regular PV/SRCREV not
Jun 18 06:09:53 <Herrie> JaMa: https://bpaste.net/show/0086bc049171
Jun 18 06:10:37 <Herrie> And removed the compatible_machines
Jun 18 06:11:41 <JaMa> you also removed whole do_install, why?
Jun 18 06:11:54 <JaMa> no surprise that headers weren't installed
Jun 18 06:26:55 <Herrie> Now builds
Jun 18 06:39:11 <Tofe> Morning
Jun 18 06:39:25 <Tofe> JaMa: good!
Jun 18 12:16:10 <saidinesh5> So looks like some firmware files were missing / didn't load
Jun 18 12:17:08 <Herrie> saidinesh5: Well that's in general a good status
Jun 18 12:17:21 <Herrie> And you know what's the problem, so that helps to debug ;)
Jun 18 12:17:23 <saidinesh5> Herrie: yup. although kind of surprising...
Jun 18 12:17:40 <saidinesh5> like i thought "include the meta-intel layer" and happily ever after is what would happen
Jun 18 12:17:49 <saidinesh5> but pretty sure this will get me wifi for sure
Jun 18 12:19:38 <Herrie> A lot of these things are small tinkering... We had same with our android targets at times as well
Jun 18 12:19:43 <Herrie> Building kernel modules etc
Jun 18 12:20:15 <Herrie> Yocto is great & powerfull, but learning curve is pretty steep as well at times;)
Jun 18 12:20:37 <saidinesh5> Yep. will have to search for yocto cheatsheet
Jun 18 12:22:16 <saidinesh5> but yeah firmware files for acpi were missing and i think the wifi modules didnt load properly either
Jun 18 12:22:39 <saidinesh5> bluetooth seems to have initialized but for some odd reason hcitool scan fails
Jun 18 12:22:42 <Herrie> Any logfiles that you can share are always welcome, some things might ring a bell ;)
Jun 18 12:22:45 <saidinesh5> and rfkill list also shows nothing
Jun 18 12:22:52 <saidinesh5> Ohh good idea
Jun 18 12:22:54 <saidinesh5> hold on
Jun 18 12:23:20 <Tofe> if the RFKILL config enabled in your kernel?
Jun 18 12:23:24 <Tofe> is*
Jun 18 12:23:39 <saidinesh5> the bestest thing about this port so far: Everything lives on the usb drive i can move between the tablet and laptop
Jun 18 12:23:52 <Herrie> :)
Jun 18 12:24:00 <saidinesh5> oh let me check , i didn't even edit kernel config tbh. i jsut used meta-intel
Jun 18 12:25:08 <saidinesh5> dmesg, journalctl and any other logs i can fetch?
Jun 18 12:27:26 <saidinesh5> https://bpaste.net/show/5d03e3e1010c
Jun 18 12:27:27 <saidinesh5> dmesg
Jun 18 12:27:53 <saidinesh5> journalctl: https://bpaste.net/show/f7e9a014f94e
Jun 18 12:28:26 <Herrie> You can also check /var/log/messages and /var/log/legacy-log
Jun 18 12:29:13 <saidinesh5> ohh wow didn;t  know those places are still in use
Jun 18 12:29:24 <saidinesh5> but dmesg itself is pretty useful imho. notice the firmware loading errors
Jun 18 12:29:38 <saidinesh5> Jun 18 17:54:03 intel64-tablet kernel: intel_sst_acpi 808622A8:00: Direct firmware load for intel/fw_sst_22a8.bin failed with error -2
Jun 18 12:29:49 <saidinesh5> Jun 18 17:54:03 intel64-tablet kernel: platform regulatory.0: Direct firmware load for regulatory.db failed with error -2
Jun 18 12:29:57 <saidinesh5> Jun 18 17:54:03 intel64-tablet kernel: brcmfmac mmc1:0001:1: Direct firmware load for brcm/brcmfmac43430a0-sdio.bin failed with error -2
Jun 18 12:31:30 <Tofe> regulatory.db --> https://github.com/webOS-ports/meta-pine64-luneos/commit/e894377ded58368974fef52466fca83f40560a7e
Jun 18 12:31:52 <Tofe> "-2" should mean "not found", usually
Jun 18 12:32:00 <saidinesh5> i know
Jun 18 12:32:17 <saidinesh5> was tryng to see what magical flags i have to enable in yocot to include them
Jun 18 12:32:22 <saidinesh5> ohhh
Jun 18 12:32:39 <saidinesh5> what about the firmware files themselves Tofe?
Jun 18 12:33:17 <Tofe> for them I'm not sure where they are
Jun 18 12:33:29 <Tofe> on pinephone I take from from a binary repo
Jun 18 12:33:34 <saidinesh5> they are supposed to be in linux-firmware i think
Jun 18 12:33:35 <saidinesh5> Ahh
Jun 18 12:33:40 <saidinesh5> and you manually added them?
Jun 18 12:33:44 <Tofe> yes
Jun 18 12:33:46 <saidinesh5> Ahh
Jun 18 12:34:03 <saidinesh5> is there anyway to force load firmware on the device?
Jun 18 12:34:30 <Tofe> unload and reload the module should do it
Jun 18 12:34:34 <saidinesh5> Ahh
Jun 18 12:34:41 <saidinesh5> what takes care of loading the firmware files? udev?
Jun 18 12:34:48 <Tofe> no, the kernel module
Jun 18 12:34:58 <saidinesh5> oh kernel module can access the filesystem paths?
Jun 18 12:35:39 <Tofe> it's often quite hardcoded to be in /lib/firmware/something
Jun 18 12:35:43 <saidinesh5> yep. but didn't know kernel modules can read files like that directly
Jun 18 12:35:57 <Tofe> ah, well, yes :)
Jun 18 12:36:19 <saidinesh5> will try first copying the firmware files to that path and booting from the usb stick once
Jun 18 12:38:54 <Tofe> yes, it should be enough to test
Jun 18 12:56:05 <saidinesh5> Tofe: you have any link for little opengles demos?
Jun 18 12:56:22 <saidinesh5> statically compiled preferred
Jun 18 12:56:33 <saidinesh5> AppImages to the rescue!
Jun 18 13:23:31 <Herrie> Tofe: Main issues seem to be the following 2 after adding email (regular) and Google Contacts & Calendar: 2019-06-18T13:03:31.798305Z [7] user.warning activitymanager [] <default-lib> LS_REQUIRES_SECURITY {"SERVICE":"com.palm.service.calendar.reminders","CATEGORY":"/","METHOD":"monitor"} Service security groups don't allow method call.
Jun 18 13:23:50 <Herrie> 2019-06-18T13:07:18.220029Z [233] user.warning activitymanager [] <default-lib> LS_REQUIRES_SECURITY {"SERVICE":"com.palm.service.contacts.linker","CATEGORY":"/","METHOD":"monitor"} Service security groups don't allow method call.
Jun 18 13:24:06 <Herrie> And 2019-06-18T13:03:35.739804Z [11] user.warning mojodb-luna [] DB8 MOJ_SERVICE_WARNING {"sender":"com.palm.app.calendar","method":"batch","payload":{"$activity":{"activityId":31},"operations":[{"method":"find","params":{"query":{"desc":true,"from":"com.palm.calendarevent:1","limit":1,"orderBy":"eventDisplayRevset"}}},{"method":"find","params":{"query":{"desc":true,"from":"com.palm.cal
Jun 18 13:24:06 <Herrie> endarevent:1","limit":1,"orderBy":"eventDisplayRevset","where":[{"op":"=","prop":"_del","val":true}]}}},{"method":"find","params":{"query":{"from":"com.palm.calendarevent:1","orderBy":"eventDisplayRevset"}}}]},"error":"invalid parameters: caller='com.palm.app.calendar' error='property not allowed - '$activity''","reqErr":22}
Jun 18 13:25:38 <Tofe> saidinesh5: not that I can think of, unfortunately
Jun 18 13:25:50 <Tofe> JaMa: ^
Jun 18 13:26:26 <Tofe> JaMa: do you have something in your luneos minimal ui image maybe?
Jun 18 13:28:00 <Tofe> Herrie: activimanager may need to be in the public or private group to access this API
Jun 18 13:28:24 <Herrie> Tofe: Well I played around a bit, but with little luck so far
Jun 18 13:29:21 <Tofe> mmh
Jun 18 13:29:35 <Herrie> saidinesh5: We do have some opengl tests in /usr/palm/applications
Jun 18 13:29:43 <saidinesh5> Herrie: like?
Jun 18 13:29:46 <Herrie> sdl2_opengles1_test and sdl2_opengles2_test
Jun 18 13:29:59 <Tofe> Herrie: could we start a new tissot build, with the latest fixes, so that I can help with these ones
Jun 18 13:30:14 <Herrie> Tofe: Well if JaMa is happy with PDM we can merge that
Jun 18 13:30:20 <Tofe> saidinesh5: just know that these are wayland clients, iirc
Jun 18 13:30:22 <Herrie> I have a few others pending as well that I'll PR now
Jun 18 13:30:38 <Tofe> saidinesh5: but maybe they can be start directly on fb
Jun 18 13:30:50 <Herrie> It's weird your luna-next.service is failish
Jun 18 13:30:53 <Herrie> failing
Jun 18 13:31:06 <Tofe> Herrie: he replaced luna-next with bash :)
Jun 18 13:31:06 <Herrie> Jun 18 17:54:08 intel64-tablet systemd[674]: luna-next.service: Failed at step EXEC spawning /usr/sbin/bash: No such file or directory
Jun 18 13:31:13 <Herrie> Ah ok ;)
Jun 18 13:31:19 <Herrie> That explains ;)
Jun 18 13:32:02 <Tofe> Herrie: but even without the pdm merge I could reproduce these two errors, I guess?
Jun 18 13:33:13 <JaMa> there is luneui-example-image, but it doesn't work well yet
Jun 18 13:33:37 <JaMa> pdm changes shouldn't harm, but I'm surprised if it's any useful on other devices
Jun 18 13:33:50 <JaMa> pdm-plugin was supposed to hold rpi specific configuration for pdm
Jun 18 13:34:05 <JaMa> looking at the sources id doesn't really hold much, but still..
Jun 18 13:34:18 <Herrie> JaMa: Plugin doesn't do much itself ;)
Jun 18 13:34:24 <Herrie> The pdm service does a bit more
Jun 18 13:34:43 <Herrie> pdm-plugin is < 100 lines of code in each file :P
Jun 18 13:34:47 <Tofe> JaMa: could be useful to manage usb-host
Jun 18 13:34:50 <JaMa> yes, I know
Jun 18 13:34:59 <saidinesh5> Tofe: oh will look just copied linux-firmware from: https://git.kernel.org/pub/scm/linux/kernel/git/firmware/linux-firmware.git/about/
Jun 18 13:35:04 <Herrie> Tofe: yes for sure
Jun 18 13:35:05 <saidinesh5> okay fingers crossed now
Jun 18 13:36:04 <JaMa> Tofe: if it works, then fine, I'm just surprised that it does knowing the team who worked on that
Jun 18 13:36:13 <Tofe> :)
Jun 18 13:36:26 <Tofe> if it doesn't, we'll improve it! ;)
Jun 18 13:36:57 <JaMa> in the initial implementation pdm depended on MACHINE_ARCH pdm-plugin, not the other way around as one might expect from "plugin"
Jun 18 13:37:14 <Tofe> ah, yes, that's not a good start :)
Jun 18 13:38:16 <JaMa> and then they claimed that it can never work in qemux86, because pdm-plugin is _only_ for rpi3 at which point I've given up on that
Jun 18 13:39:36 <Tofe> "when in doubt, act as if you know what you're doing"
Jun 18 13:41:17 <saidinesh5> Herrie: the sdl2 opegnles2 test doesnt start : says no available video device
Jun 18 13:41:29 <saidinesh5> any crazy environment variables i should be aware of?
Jun 18 13:42:28 <Tofe> does it accept command line arguments?
Jun 18 13:43:09 * saidinesh5 checks
Jun 18 13:43:25 <saidinesh5> not sure
Jun 18 13:43:39 <saidinesh5> even --help goes with Could not create window: no available video device
Jun 18 13:43:45 <Tofe> damn
Jun 18 13:46:04 <JaMa> saidinesh5: it might be easier to start with libdrm-tests
Jun 18 13:46:29 <saidinesh5> JaMa: oh where are they located?
Jun 18 13:46:41 <JaMa> built by libdrm recipe
Jun 18 13:46:42 <saidinesh5> and starting luna-next with eglfs isn't working ..
Jun 18 13:46:44 <saidinesh5> ohh
Jun 18 13:46:52 <saidinesh5> will have to rebuild that then
Jun 18 13:46:54 <JaMa> probably not installed in your image by default
Jun 18 13:47:00 <saidinesh5> yea
Jun 18 13:47:01 <JaMa> just install extra .ipk
Jun 18 13:47:17 <JaMa> meta-luneos/recipes-core/images/luneos-dev-image.bb:    libdrm-tests \
Jun 18 13:47:23 <JaMa> hmm should be in dev image
Jun 18 13:47:24 <saidinesh5> Ohh
Jun 18 13:47:37 <saidinesh5> rebooting
Jun 18 13:47:45 <saidinesh5> needs the regulatory db file for wifi anyway
Jun 18 13:47:52 <saidinesh5> it isn't in the linux-firmware i copied ther
Jun 18 13:48:14 <JaMa> I haven't read the backlog, so I don't know what device you're using, but if it has any special libdrm kernel module, then you might need to add it like I did with vboxvideo in ./meta-luneui/recipes-graphics/drm/libdrm_%.bbappend
Jun 18 13:49:01 <JaMa> for wifi see http://lists.openembedded.org/pipermail/openembedded-core/2019-June/283728.html
Jun 18 13:49:22 <JaMa> but you might be fine with the version from meta-networking
Jun 18 13:49:23 <saidinesh5> Ahh
Jun 18 13:50:33 <Tofe> ' #define PLUGIN_PATH "/usr/lib/libpdm-plugin.so.1" '  (first line in main cpp) I think there's a lot of room for improvements :p
Jun 18 13:52:21 <Tofe> anyway, as long as it doesn't crash or do stupid conflicting things, I'm fine with including it
Jun 18 14:15:06 <JaMa> FYI I've merged small oe-core upgrade which fixes some sstate signature issues, but causes quite big rebuild again
Jun 18 14:15:14 <JaMa> jenkins jobs already running
Jun 18 14:18:26 <Herrie> JaMa: We have a few more PR's pending :P
Jun 18 14:21:45 <Herrie> https://github.com/webOS-ports/webos-initscripts/pull/1, https://github.com/webOS-ports/db8/pull/7, https://github.com/webOS-ports/bootd/pull/2, https://github.com/webOS-ports/app-services/pull/10, https://github.com/webOS-ports/org.webosports.service.contacts.carddav/pull/83, https://github.com/webOS-ports/mojomail/pull/5, https://github.com/webOS-ports/mojoservice-frameworks/pull/2
Jun 18 14:21:57 <Herrie> And then we'd need to bump all SRCREVs as well
Jun 18 14:26:53 <JaMa> ok
Jun 18 15:32:55 <saidinesh5> okay will look at adding the regdb and other firmware files
Jun 18 17:01:14 <Herrie> Tofe: Updated testing protocol a bit
Jun 18 17:50:57 <saidinesh5> Herrie: how do i remove an item from a list?
Jun 18 17:51:10 <saidinesh5> like IMAGE_FS_TYPES -= "wic" doesn't seem to be working
Jun 18 17:52:30 <saidinesh5> also how do i clear old images when doing a new build? (They hog a lot of space!)
Jun 18 18:01:08 <Herrie> saidinesh: They're in tmp-glibc/deploy/images
Jun 18 18:01:22 <Herrie> Sometimes it's good to do a rm -rf tmp-glibc in general
Jun 18 18:01:26 <Herrie> Will clean quite some bits
Jun 18 18:01:39 <Herrie> Since there's sstate cache, rebuilding it won't take that long
Jun 18 18:02:27 <saidinesh5> Herrie: did the rm -rf tmp-glibc 2 days ago when building this
Jun 18 18:02:40 <saidinesh5> surprised how long even the rm -rf takes tbh
Jun 18 18:04:15 <Herrie> Well there's a lot in there
Jun 18 18:04:23 <Herrie> Takes even a few minutes on my super fast SSD ;)
Jun 18 18:04:26 <saidinesh5> yep
Jun 18 18:05:04 <Herrie> Got myself a Threadripper 2990WX with 128GB of RAM and 1TB M.2 SSD in December ;)
Jun 18 18:05:10 <Herrie> Build time < 1 hr from scratch ;)
Jun 18 18:05:14 <saidinesh5> okay so now need to figure out how to include linux-firmware and that regulatory.db file
Jun 18 18:05:23 <saidinesh5> Oh wow
Jun 18 18:05:38 <Herrie> Replaced my previous builder from 2010 & desktop PC ;)
Jun 18 18:05:39 <saidinesh5> my build time was around 2-3 hours i think from scratch
Jun 18 18:05:49 <saidinesh5> Ryzen 7, 32GB ram and slow as hell HDD
Jun 18 18:06:03 <saidinesh5> need to buy a new HDD though. will probably go for the 4TB 7200rpm one
Jun 18 18:06:04 <Herrie> With current prices M.2 SSD of 1TB are doable
Jun 18 18:06:20 <saidinesh5> yeah but 4TB sounds like a lot more.. lol
Jun 18 18:06:26 <Herrie> Yeah ;)
Jun 18 18:06:34 <Herrie> And got 750/750 fibre in the new house as well ;)
Jun 18 18:06:39 <Herrie> So that helps as well ;)
Jun 18 18:06:41 <saidinesh5> oh wow
Jun 18 18:06:57 <saidinesh5> i am on 150Mbps but pretty sure my router is already a bottleneck
Jun 18 18:07:35 <saidinesh5> wanted to downgrade to 100Mbps for just that reason but they decided to give me 2 free months per year on  my current plan
Jun 18 18:11:13 <saidinesh5> okay so need to figure out how to add linux-firmware
Jun 18 18:23:43 <Herrie> Well Tofe has it in meta-pine64
Jun 18 18:25:02 <saidinesh5> Oh
Jun 18 18:31:22 <saidinesh5> Herrie: so how can i find warrior version of this file?http://git.yoctoproject.org/cgit/cgit.cgi/poky/tree/meta/recipes-kernel/linux-firmware/linux-firmware_git.bb
Jun 18 18:34:52 <saidinesh5> Also i did a make update
Jun 18 18:34:57 <saidinesh5> and i am stuck with this: ERROR: Required build target 'luneos-dev-image' has no buildable providers.
Jun 18 18:34:57 <saidinesh5> Missing or unbuildable dependency chain was: ['luneos-dev-image', 'packagegroup-webos-extended', 'com.webos.service.pdm', 'fuse-utils']
Jun 18 18:36:01 <saidinesh5> ERROR: Nothing RPROVIDES 'fuse-utils' (but /src/LuneOS/webos-ports-env/webos-ports/meta-webos-ports/meta-luneos/recipes-webos/com.webos.service.pdm/com.webos.service.pdm.bb RDEPENDS on or otherwise requires it)
Jun 18 18:37:22 <Herrie> https://github.com/openembedded/openembedded-core/tree/warrior/meta/recipes-kernel/linux-firmware
Jun 18 18:44:50 <Tofe> saidinesh5: looks like a layer is missing
Jun 18 18:45:10 <Tofe> meta-filesystem
Jun 18 18:45:27 <Herrie> For fuse-utils: You need to add my latest change merged: https://github.com/webOS-ports/webos-ports-setup/commit/89fa3d5bec8b349aa39507a63349a3b1ad2bd22e
Jun 18 18:46:14 <Tofe> (sorry if I'm not accurate, I just had a wine testing course :) )
Jun 18 18:46:38 <Herrie> Tofe: I thought the French are natural experts :P ? No need for courses LOL
Jun 18 18:46:40 <Tofe> tasting*
Jun 18 18:47:18 <Tofe> ah, if only! we're just natural drinkers :)
Jun 18 18:48:21 <Tofe> tonight was Moldavian wine, very interesting
Jun 18 18:49:47 <saidinesh5> ohh
Jun 18 18:49:51 * saidinesh5 looks
Jun 18 18:50:10 <Tofe> https://photos.app.goo.gl/CjMMVuPnzK1qDYxb7
Jun 18 18:51:29 <Tofe> good luck to buy some, though; most is sent to Russia
Jun 18 18:52:05 <saidinesh5> Herrie: how come that linux-firmware isn't present in the tree we have?
Jun 18 18:52:13 <saidinesh5> we also include meta-openembedded right?
Jun 18 18:52:31 <Herrie> saidinesh5: Well you need to include it somewhere so thta it builds
Jun 18 18:52:33 <Herrie> And deploys
Jun 18 18:52:35 <Herrie> Let me check
Jun 18 18:53:27 <Herrie> You need to add it to machine.conf
Jun 18 18:53:54 <Herrie> https://github.com/webOS-ports/meta-pine64-luneos/blob/master/conf/machine/pinephone.conf#L18
Jun 18 18:54:34 <Herrie> https://github.com/shr-distribution/meta-smartphone/blob/warrior/meta-hp/conf/machine/tenderloin.conf#L37
Jun 18 18:54:51 <Herrie> But then with the right tree name for yours
Jun 18 18:55:05 <Herrie> Tofe: Had an ex-gf from Moldova ;)
Jun 18 18:55:11 <Herrie> Nice wines indeed, hard to get here
Jun 18 18:55:46 <Herrie> Volume is nothing compared to French, Spanish, Italian, South African, Argentinian, Chilean and Australian wines, so hard to find.
Jun 18 18:55:52 <Herrie> Georgian ones are quite nice too
Jun 18 18:56:16 <saidinesh5> what's the difference between rdepends and depends?
Jun 18 18:56:51 <saidinesh5> MACHINE_EXTRA_RDEPENDS = " \
Jun 18 18:56:51 <saidinesh5>     linux-firmware
Jun 18 18:56:51 <saidinesh5>     wireless-regdb
Jun 18 18:56:51 <saidinesh5> "
Jun 18 18:56:53 <saidinesh5> added this
Jun 18 18:57:36 <Herrie> RDEPEND = runtime depends
Jun 18 18:57:40 <Herrie> DEPENDS = required for build
Jun 18 18:58:06 <Herrie> You need to have the specific firmware for your chip
Jun 18 18:58:11 <Herrie> Not sure which one you have?
Jun 18 18:58:36 <saidinesh5> Ahh
Jun 18 18:58:44 <saidinesh5> added that to RDEPENDS itself
Jun 18 18:59:01 <saidinesh5> and i think my wireless chip is a broadcom one though
Jun 18 18:59:24 <Herrie> What is your device again?
Jun 18 18:59:26 <saidinesh5> How do i remove the .wic images?
Jun 18 18:59:31 <Herrie> We might be able to find it
Jun 18 18:59:39 <saidinesh5> oh it is in the log above
Jun 18 18:59:40 <saidinesh5> hold
Jun 18 18:59:47 <Herrie> .wic is something Tofe or JaMa could answer I guess
Jun 18 18:59:50 <saidinesh5> Jun 18 17:54:03 intel64-tablet kernel: brcmfmac mmc1:0001:1: Direct firmware load for brcm/brcmfmac43430a0-sdio.bin failed with error -2
Jun 18 18:59:53 <saidinesh5> this is the wifi chip
Jun 18 19:00:16 <saidinesh5> i am just adding linux-firmware because it seems useful even otherwise
Jun 18 19:03:39 <Herrie> I would make it linux-firmware-bcm43430a0
Jun 18 19:04:19 <Herrie> As per https://github.com/openembedded/openembedded-core/blob/warrior/meta/recipes-kernel/linux-firmware/linux-firmware_git.bb#L264
Jun 18 19:04:25 <JaMa> saidinesh5 remove wic from IMAGE_FSTYPES variable
Jun 18 19:04:59 <saidinesh5> okay rebuilding the image with firmware and wireless regdb
Jun 18 19:05:09 <saidinesh5> i think it was whining about one more file but let's see that after this boots
Jun 18 19:05:42 <Herrie> You're making solid progress :D
Jun 18 19:05:48 <saidinesh5> haha thanks
Jun 18 19:06:52 * elvispre built a set of images last night... and is already building the qt stack again.
Jun 18 19:10:01 <Herrie> elvispre: JaMa bumped some stuff ;)
Jun 18 19:10:11 <Herrie> That caused that. I merged a bunch of things as well ;)
Jun 18 19:10:39 <Herrie> But my things shouldn't cause drastic rebuild ;)
Jun 18 19:10:44 <elvispre> Herrie: I saw some of your stuff. That's what I thought I was picking up.
Jun 18 19:10:58 <saidinesh5> they did rebuild a lot after i bumped this
Jun 18 19:11:09 <Herrie> It's JaMa's changes
Jun 18 19:11:13 <saidinesh5> Ahh
Jun 18 19:11:19 <Herrie> That caused Qt rebuild
Jun 18 19:11:26 <Herrie> Mine rebuild stuff as well, but not Qt ;)
Jun 18 19:11:54 <saidinesh5> btw. WARNING: keymanager-0.1.0+gitAUTOINC+7b50f4160d-r0 do_packagedata: QA Issue: Package version for package keymanager-src went backwards which would break package feeds from (0:0.1.0+git0+560ad64823-r0 to 0:0.1.0+git0+7b50f4160d-r0) [version-going-backwards]
Jun 18 19:12:01 <saidinesh5> messages like that can be safely ignored right?
Jun 18 19:12:10 <saidinesh5> i mean 0.1.0 to 0.1.0 is still the same
Jun 18 19:12:15 <saidinesh5> except maybe the git commit changed
Jun 18 19:14:33 <JaMa> as long as you don't expect package feed to work correctly you can ignore it
Jun 18 19:14:58 <JaMa> if you want opkg upgrade to work for you, then you need to set PRserv
Jun 18 19:15:53 <saidinesh5> PRserv? but is there any public repo for this architecture?
Jun 18 19:16:38 <JaMa> PRserv is just service which increments 0 in git0 to 1 when you change the git commit (and increments it every time it sees new git sha)
Jun 18 19:16:49 <saidinesh5> Ahh
Jun 18 19:17:11 <JaMa> elvispre: systemd was rebuilt because oe-core bump, so lets blame systemd devs :)
Jun 18 19:17:15 <saidinesh5> how do i enable it/where do i set it?
Jun 18 19:17:33 <saidinesh5> okay ouch qt5base is rebuilding now
Jun 18 19:17:52 <JaMa> it reuses 80% of sstate, but the remaining 20% are the big guys
Jun 18 19:17:54 <JaMa> NOTE:   do_populate_sysroot: 80.3% sstate reuse(456 setscene, 112 scratch)
Jun 18 19:17:56 <elvispre> JaMa: OK :) (They're used to taking heat, I think.)
Jun 18 19:18:41 <saidinesh5> i hope this doesn't rebuild webengine too
Jun 18 19:18:56 <Herrie> It does ;)
Jun 18 19:18:58 <JaMa> saidinesh5: it's enabled by default in our webos-ports-setup (with PRSERV_HOST = "jenkins.nas-admin.org:8585" in local.conf), but I don't know how you set your build
Jun 18 19:19:13 * saidinesh5 checks
Jun 18 19:19:44 <saidinesh5> JaMa: Ah i had to comment it out a long time ago
Jun 18 19:19:45 <JaMa> but as long as you're flashing whole images, you can ignore this
Jun 18 19:19:50 <saidinesh5> I am
Jun 18 19:19:56 <saidinesh5> well at least for now
Jun 18 19:20:13 <saidinesh5> JaMa: how do i disable creation of wic images?
Jun 18 19:20:22 <saidinesh5> like i tried to do IMAGE_FSTYPES -= wic but that causes error
Jun 18 19:21:38 <JaMa> there isn't any -=
Jun 18 19:21:46 <JaMa> find out where it's added to IMAGE_FSTYPES in first place
Jun 18 19:21:55 <JaMa> you can use bitbake -e your-image to find out
Jun 18 19:22:04 <saidinesh5> Ohh will do
Jun 18 19:22:11 <JaMa> and if you cannot remove it where it's added, then IMAGE_FSTYPES_remove = "foo"
Jun 18 19:22:22 <saidinesh5> Ahh
Jun 18 19:22:30 <JaMa> where foo is the actual wic type (it probably isn't just "wic")
Jun 18 19:22:35 <saidinesh5> oh of course
Jun 18 19:22:58 <saidinesh5> where do i get to find out all these <VARIABLETYPE>_<operation> manual?
Jun 18 19:23:08 <JaMa> bitbake manual
Jun 18 19:23:17 <saidinesh5> Ahh
Jun 18 19:23:44 <JaMa> there are only 3 _operation = _append,_prepend,_remove
Jun 18 19:24:24 <JaMa> the rest are _<override> and that's like conditional and you can use any word which appears in OVERRIDES variable as an override, and you can combine them as well
Jun 18 19:24:47 <JaMa> e.g. _arm_luneos or _append_arm
Jun 18 19:25:23 <JaMa> and then there are +=,=+,.=,=.,=,?=,??=
Jun 18 19:25:27 <saidinesh5> Ahh and what about the commonly defined <VARIABLETYPE>
Jun 18 19:26:04 <JaMa> VARIABLETYPE? it's possible to define typed variables, but that's probably not what you mean, because that's used quire rarely
Jun 18 19:26:36 <JaMa> e.g. ERR_REPORT_UPLOAD_FAILURES[type] = "boolean"
Jun 18 19:26:42 <saidinesh5> like IMAGE_FSTYPES is defined somewhere right? where? and what are the important global varaibles like that i can poke around?
Jun 18 19:26:47 <saidinesh5> yeah
Jun 18 19:26:54 <JaMa> ah, so just common variable
Jun 18 19:27:11 <saidinesh5> yeah
Jun 18 19:27:25 <JaMa> they could be defined everywhere, but bitbake -e is _really_ useful
Jun 18 19:27:30 <saidinesh5> Ahh
Jun 18 19:27:33 <saidinesh5> will check bitbake -e
Jun 18 19:28:00 <JaMa> it will show you not only all the variables used for given recipe, but also where were they defined/redefined/appened-to etc
Jun 18 19:28:13 <JaMa> so you see whole history and final processed value
Jun 18 19:28:17 <saidinesh5> Ahh
Jun 18 19:28:19 <saidinesh5> nifty!
Jun 18 22:02:41 <JaMa> Herrie: new issue from pdm-plugin http://jenkins.nas-admin.org/job/LuneOS/view/testing/job/luneos-testing_workspace-compare-signatures/135/console
**** ENDING LOGGING AT Wed Jun 19 02:59:57 2019