**** BEGIN LOGGING AT Wed Jan 19 02:59:57 2022 Jan 19 08:58:21 Morning! Jan 19 09:00:36 Tofe: Seems that for suspend to work properly with Megi's kernel on PP with crust we need 2 (additional) patches in TFA still as per conversation on the Pine64 Dev Telegram channel. This is what they suggest at their end in terms of packaging: https://termbin.com/1wbn We should add the 2 patches for TFA I guess to our TFA 2.6 recipe. Since they're patching sunxi files, we could apply the Jan 19 09:00:36 patch for all targets so we can have a single build of TFA. Jan 19 09:09:32 Herrie: right, I just saw that too Jan 19 09:09:52 I didn't test suspend that much on our side, so I don't even now what is our status on that topic :) Jan 19 09:10:13 Also seems they're using the "no-stack-protector" patch, not sure what that one does Jan 19 09:11:12 Seems some option for GCC Jan 19 09:11:27 Tofe_: Well our state machine needs a bit of tweaking Jan 19 11:23:23 Tofe_/JaMa: You know if there's a way to see what directly and indirectly depends on trusted-firmware-a? Jan 19 11:54:00 Seems we might need to build with ENABLE_STACK_PROTECTOR=none for PP and not provide this for PPP. Jan 19 12:24:15 Herrie: you could to a bitbake -g luneos-dev-image and tinker with the dot file Jan 19 12:24:24 s/to/do/ Jan 19 13:31:14 or just PNBLACKLIST it and see what fails Jan 19 15:55:15 JaMa: well u-Boot fails which is expected, if I build with bb -k it should show everything right? Or ... ? Jan 19 15:56:35 yes, it should show all recipes which depend or rdepend on it Jan 19 15:56:45 and are included in the image Jan 19 15:57:11 you can use "bitbake -kn world" to show also the recipes not included in the image Jan 19 16:14:39 anyone out there? Jan 19 18:01:15 Herrie: https://github.com/webOS-ports/meta-pine64-luneos/pull/29 this builds fine, not tested yet on my PP Jan 19 18:08:47 JaMa: so, I removed commit on the initramfs suffix, and now luneos-dev-image:do_rootfs fails again (makes sense :) ) Jan 19 18:16:30 And I'm not sure what I should look for, actually Jan 19 18:17:44 anyone out there? Jan 19 18:18:05 codepoet: Hi Jan 19 18:18:39 hey! is this the canonical home of webos-ports since the freenode fall-out? Jan 19 18:19:11 codepoet: yes it is :) Jan 19 18:19:27 ok, cool Jan 19 18:20:00 i was wondering if it would be permissible to bridge this chat to our webOS Discord chat server? Jan 19 18:20:17 There's about a dozen or so of us using either webOS or LuneOS on that channel Jan 19 18:22:33 it also has a bridge to a webOS chat app called SimpleChat, but I could keep that in another channel Jan 19 18:23:50 ah, well, why not; however here it's more about LuneOS, we don't really deal with webOS issues itself (but the two topics are often a bit entangled of course) Jan 19 18:25:01 ok, cool. i don't really know how to do it yet, but there's some OSS in GitHub, so I think I can figure it out. Jan 19 18:25:26 Understood about being LuneOS focused. I've been developing apps for webOS for the past 2-3 years, and this year I'm aiming to start writing for Lune Jan 19 18:25:41 Herrie: an opinion about this proposal? Jan 19 18:26:34 Oh no Discord 😢 Don't care about me though please, I don't matter here 😉 Jan 19 18:27:34 codepoet: what kind of discussion traffic do you have in the discord chat? here it's not that much, but for some chans on telegram I realize that the discussions are so heavy that I can't even keep up Jan 19 18:27:57 you wouldn't have to use Discord -- but users on the Discord server would be able to chat with you (and vice versa) Jan 19 18:49:17 Tofe_: do you have ".rootfs" in "bitbake-getvar -r initramfs-uboot-image IMAGE_NAME_SUFFIX" ? Jan 19 18:59:59 JaMa: yep Jan 19 19:00:35 but maybe initramfs-uboot-image doesn't use it ? Jan 19 19:04:03 JaMa: I have initramfs-uboot-image-pinephone-0-0.rootfs.cpio.gz.u-boot, but the symlink is just called initramfs-uboot-image-pinephone.cpio.gz.u-boot Jan 19 19:09:26 Tofe_: you're right now I can see that with my oe-core changes removed, not sure why I haven't reproduced it before Jan 19 19:10:53 JaMa: ah good, I'm not crazy :) Jan 19 19:11:35 but maybe there's a way to have a PR compatible both with before your changes and after them ? Jan 19 19:35:22 sorry, I missed the question about discussion. real job interrupted. Jan 19 19:35:53 up to you guys how we integrate. if you want lots of chatter (80-100 messages a day) I could integrate the main SimpleChat channel Jan 19 19:36:19 discussion ranges from what apps still work, apps I'm building/people are using, to general discussion about smart phones and the industry Jan 19 19:36:59 if you just want to hear from LuneOS users, I could integrate a separate channel, and people could self-select where they direct their conversation Jan 19 19:37:36 we'll all be newbies, but probably only a couple messages a week that way Jan 19 19:46:25 codepoet: I'd say, let's begin with the #luneos-irc discord channel and see if it works Jan 19 19:49:04 ok, thanks! hopefully this helps get people more engaged with LuneOS! I'll work on it. Jan 19 19:52:45 great :) Jan 19 21:50:52 Tofe_: I think in your PR we should also include the stack protector flag Jan 19 21:51:25 It seems this is however only needed for PP, with A64 platform, not with rk3399 Jan 19 21:52:14 Seems only u-Boot really depends on TFA so it shouldn't be a big problem to build this device specific Jan 19 21:53:40 HerrieTP: shouldn't we use a configure flag instead ? Jan 19 21:54:26 Tofe_: Yeah see the discussion in other channel should be the configure flag just like we do platform etc now Jan 19 21:54:43 Just seems it's A64 so PP specific Jan 19 21:54:57 So we should do a device specific append Jan 19 21:55:14 This replaces the patch that was in some other repos for this Jan 19 21:56:00 HerrieTP: but if that has no impact on rk3399, why have it device specific ? note that the recipe isn't currently MACHINE_ARCH, so we'd have to add that Jan 19 21:56:51 Tofe_: Well we'd want to have stack protector on rk3399 since it doesn't cause havoc Jan 19 21:57:11 Only issue seems to be A64 which doesn't behave without it appearantly Jan 19 21:57:48 Proper fix should be done upstream, but seems there's no agreement (yet) on how... Jan 19 21:58:11 See the cross-distro channel for some thoughts and links Jan 19 22:01:20 https://gitlab.manjaro.org/manjaro-arm/packages/core/uboot-pinephone/-/blob/master/PKGBUILD#L109 yes, they did the change Jan 19 22:04:05 About 1 hour ago :P Jan 19 22:04:34 We were already on TFA 2.6 so not sure why we didn't get the error but well Jan 19 22:14:44 HerrieTP: is it supposed to not build ? or is there an issue at runtime ? Jan 19 22:14:51 because I didn't test runtime yet Jan 19 22:22:40 Tofe_: Based on some of the links I would say build Jan 19 22:23:23 Herrie: that's also what I deduce from https://github.com/ARM-software/arm-trusted-firmware/commit/7af195e29a4213eefac0661d84e1c9c20476e166 Jan 19 22:23:48 I think we don't need to do anything if we don't want SSP Jan 19 22:24:44 however, activating it requires at least two variables: STACK_PROTECTOR_ENABLED=1 and ENABLE_STACK_PROTECTOR=default **** ENDING LOGGING AT Thu Jan 20 02:59:56 2022