[08:38] <alan_g> greyback: do you agree this is the right approach? https://bugs.launchpad.net/miral/+bug/1628630/comments/1
[08:39] <greyback> alan_g: I do agree
[08:39] <alan_g> thanks
[09:04] <zzarr> hello! is it possible to run the 16.10 mir version on 16.04 with an overlay (or other ppa)?
[09:05] <anpok> stable-phone-overlay gets the same mir version that is released to yakety
[09:05] <anpok> https://launchpad.net/~ci-train-ppa-service/+archive/ubuntu/stable-phone-overlay
[10:51] <zzarr> anpok, that works on my desktop?
[10:54] <zzarr> E: The repository 'http://ppa.launchpad.net/ci-train-ppa-service/landing-031/ubuntu xenial Release' does not have a Release file.
[10:55] <zzarr> ohh... that was another repo
[10:55] <zzarr> yes the one you linked is fine
[10:56] <zzarr> I should fix... but I feel that the release of Yakkey is to close to care
[11:21] <om26er> Hello! How reliable is fbset under Mir, shall I use it in production ready code to query screen resolutions ?
[11:31] <greyback> om26er: I don't think fbset would have anything to do with mir. fbset reads the geometry of the primary framebuffer, so the resolution of your "first" screen
[11:31] <greyback> om26er: what are you wanting to do?
[11:32] <om26er> greyback, I am trying to replace some code in autopilot that have hard-coded resolutions for some devices like mako, grouper etc. I want to make that dynamic by talking to the system
[11:33] <om26er> since I want the code to work both under Mir and X11, I probably have limited options, but fbset seems to work under both
[11:34] <greyback> om26er: if you want it display server independent, then yeah, fbset is best that I know of
[11:35] <greyback> fbset no good with multiple displays tho
[11:35] <greyback> MIR_SOCKET=/run/mir_socket mirout  only works if Mir running
[11:36] <greyback> om26er: coud you ask the toolkit? Toolkit should know the display geometry
[11:37] <greyback> danger with fbset might be on something like M10, whose display is rotated
[11:37] <om26er> greyback, hmm, for reliability I could have a hybrid xrandr/mirout solution. though if I can talk to the toolkit that would be cleaner though it would only be tied to Ubuntu then.
[11:38] <greyback> mode "1200x1920-0"   <- output on my M10
[11:38] <om26er> bummer
[11:39] <greyback> mirout reports the same
[11:39] <greyback> we have to fix how rotation is reported
[11:45] <greyback> unfortunately at the moment, I cannot provide you a reliable way of getting the actual screen geometry (incorporating rotation) - I fear for now you'll need a per-device rotation hack
[12:05] <zzarr> would a Intel Joule 570x thrive with Mir?
[12:16] <zzarr> my question is, would it be possible and a good idea to use Intel Joule as a platform for a phone?
[12:19] <kdub> zzarr, if it has a mesa driver (probably) then its alright, you might need to backport the mir patches to the egl triee
[12:19] <kdub> sorry, the mir patches to the egl bits in the mesa tree
[12:21] <zzarr> okey
[12:22] <zzarr> I'm thinking about designing a qwerty slider with a mouse-pointer feature
[12:22] <zzarr> (propably a gaming stick for mouse)
[12:40] <om26er> greyback, how can I check if the current display server is Mir ? Can I rely on /run/mir_socket mirout existence ?
[12:40] <om26er> */run/mir_socket existence
[12:41] <greyback> om26er: eh. That is a socket which can be located anywhere. On desktop, I think it at /tmp/lightdm-mir-0
[12:41] <greyback> om26er: check if MIR_SOCKET env var is set, probably safest
[16:01] <om26er> greyback__, it seems MIR_SOCKET is not in the environment at all
[16:01] <bregma> om26er, what is you environment?
[16:01] <om26er> I want to be very safe while detecting if the display server is based on Mir. For example it could be the demo server or unity8 server, is there lower level interface to check that
[16:02] <om26er> bregma, I tested on Arale and kvm
[16:05] <greyback__> om26er: it won't be set in the global env, but will be set in the env of any app that needs a display server
[16:13] <bschaefer> hmm the MIR_SOCKET should be unset since the default socket location will be under XDG_RUNTIME_DIR
[16:13] <bschaefer> you need MIR_SOCKET when you're running in kms/tty1 while X is running
[16:13]  * bschaefer is also a little unsure about U8 :)
[16:20] <om26er> I need to detect which display server is running in python/shell. There are quite a few ways to check if X the display server. For Mir, probably currently I could check if Unity8 is running.
[16:20] <bschaefer> om26er, can you just attempt to connection to the mir server?
[16:20] <bschaefer> if it fails then its not around?
[16:20] <om26er> bschaefer, How to ?
[16:21] <bschaefer> mir_toolkit/mir_connection.h:63:MirConnection *mir_connect_sync(char const *server, char const *app_name);
[16:21] <bschaefer> though not sure about python
[16:22] <bschaefer> the other thing is to check XDG_RUNTIME_DIR for a mir_socket or MIR_SOCKET
[16:22] <bschaefer> XDG_RUNTIME_DIR/mir_socket should exist or MIR_SOCKET should be around