Wizzup | wonder if we can port this https://github.com/organicmaps/organicmaps | 12:03 |
---|---|---|
Wizzup | https://github.com/organicmaps/organicmaps/discussions/375 looks like it might be possible to build | 12:05 |
freemangordon | Pali: Do I get it right that you will post DM_USB patches? | 13:03 |
Pali | freemangordon: I'm not sure which patches are needed | 13:03 |
freemangordon | Pali: ok, I'll send a cumulative one to you | 13:04 |
freemangordon | to apply on top of master | 13:04 |
Pali | ok | 13:04 |
uvos | parazyd: btw the libsdl in devel can go into stable | 13:18 |
uvos | if not allready | 13:18 |
uvos | since libsdl12-compat dosent work for various reasons | 13:19 |
uvos | Wizzup: very interesting | 13:23 |
uvos | even just for desktop linux | 13:23 |
uvos | never mind leste | 13:23 |
uvos | https://github.com/organicmaps/organicmaps/blob/99c5b878a048b467d1c66cf5a5ff0ce04975c73b/platform/platform_linux.cpp#L267 hehe | 13:29 |
uvos | yeah right | 13:29 |
Wizzup | hehe | 13:34 |
Wizzup | well, still, it looks quite neat | 13:34 |
uvos | dosent suppot any gps backend at all on linux | 13:35 |
Wizzup | yeah, that would need to be added I imagine | 13:36 |
uvos | also is huge | 13:36 |
uvos | 5gb repo | 13:36 |
Wizzup | as long as directions and stuff work though, it'd be nice | 13:36 |
uvos | im not gona try to build that on d4 :P | 13:36 |
Wizzup | uvos: sure, yeah, but it does look quite advanced so could be nice, don't know about resource usage | 13:38 |
freemangordon | MB focus issue it still not fixed, right? | 13:40 |
uvos | freemangordon: no | 13:40 |
uvos | but i think the app maybe ending up with a grab if you start it via ssh while the display is locked is less of an issue than no sdl1.2 app woring at all | 13:40 |
uvos | so to stable it should go imo | 13:41 |
freemangordon | uvos: starting apps via ssh is the easiest way to trigger the bug, but not the only | 14:08 |
uvos | sure you can launch it and then lock the display before the app starts | 14:09 |
uvos | but i mean this becoming a problem is a remote posibility | 14:10 |
freemangordon | ok, if you say so | 14:10 |
uvos | you wuld have to 1. launch the app and lock the deivce before it starts. 2. the app would have to have a bug that locks it up | 14:10 |
uvos | then you have a problem | 14:10 |
freemangordon | the bug is not in the app, but in MB | 14:10 |
uvos | i know | 14:11 |
uvos | im not saying dont fix it | 14:11 |
freemangordon | I mean that it is not app specific | 14:11 |
uvos | but if the app is not locked up | 14:11 |
uvos | it dosent cause any real problem | 14:11 |
freemangordon | ok, ok | 14:11 |
freemangordon | but please, consider spending some time on that bug these days | 14:11 |
uvos | sure | 14:12 |
freemangordon | thanks | 14:12 |
freemangordon | Pali: please have a look at the mail, I think those need just a cover letter. And some testing on your side :) | 14:13 |
Pali | got it! | 14:13 |
freemangordon | I tested as much as I can, for some reason qflasher doesn;t work on leste amd64 | 14:13 |
Pali | qflasher requires /lib/i386-linux-gnu/libc.so.6 | 14:15 |
Pali | or rather 32bit version of libc.so.6 and 32bit ld loader (/lib/ld-linux.so.2) | 14:16 |
uvos | Wizzup: if you want to build that app | 14:17 |
uvos | bring 32GiB of ram | 14:17 |
uvos | or restrict treads alot | 14:17 |
Pali | freemangordon: just install 32bit libraries on your amd64 setup and it should work | 14:18 |
Pali | I will retest it | 14:18 |
freemangordon | Pali: I don;t want to break my leste VM too much :) | 14:28 |
Pali | ok :-) | 14:28 |
Pali | first test passed - boots in qemu and can boot 2.6.28 kernel in qemu | 14:28 |
freemangordon | :) | 14:29 |
Pali | now all tests via ./test/nokia_rx51_test.sh passed :-) | 14:31 |
Pali | anyway, when modifying configs/nokia_rx51_defconfig file, run make savedefconfig | 14:31 |
Pali | it generates a new "canonical" file: ./defconfig | 14:32 |
freemangordon | ah | 14:32 |
Pali | and move this canonical file defconfig to configs/nokia_rx51_defconfig | 14:32 |
freemangordon | Ah, I see | 14:32 |
freemangordon | well, just amend that last commit and edit it | 14:32 |
Pali | I will fix your second patch to use canonical | 14:32 |
freemangordon | mhm | 14:32 |
Pali | yes! | 14:32 |
Wizzup | uvos: that will stress our CI for sure ;) | 14:40 |
mighty17[m] | Just for Fun, can I install gdm3 and run it xD | 14:47 |
Wizzup | not sure what the point would be | 14:47 |
parazyd | uvos: ack @ libsdl | 14:49 |
Pali | freemangordon: Azure CI testing for all N900 patches started: https://github.com/u-boot/u-boot/pull/83 | 15:12 |
Pali | once it finish and all tests pass I will send them to ML | 15:12 |
Pali | link to n900 tests: https://dev.azure.com/u-boot/u-boot/_build/results?buildId=2399&view=logs&j=9a06d2a9-1498-5de0-2a01-be581d48ba67 | 15:13 |
mighty17[m] | not sure what the point would be > Wizzup is it necessary to hv a point, somethings are just done for fun :P | 15:21 |
Wizzup | true, but gdm would require some work, and I think you will just a see login screen? :P | 15:23 |
mighty17[m] | gdm wont work out of box? hm weird | 15:26 |
uvos | dsme and the use of xsession breaks this essentaly | 15:27 |
Wizzup | right, we execute our own xsession automatically | 15:27 |
uvos | we need to rm dsme or make it a session manager and turn the xsession scripts in a xdg-session | 15:27 |
uvos | then gdm etc would work | 15:28 |
uvos | and you could choose between hildon and sway or whatever at login | 15:28 |
Wizzup | it does make emergency calls and such real tricky | 15:28 |
uvos | not really | 15:28 |
uvos | the device as we ship it boots to hildon by default | 15:28 |
uvos | if the user wants to install something else and disable autologin thats on him | 15:29 |
mighty17[m] | it does make emergency calls and such real tricky > step 1 make it work, step 2 fix other stuff :D | 15:29 |
mighty17[m] | we need to rm dsme or make it a session manager and turn the xsession scripts in a xdg-session > oh so we have a custom script for this : | 15:30 |
mighty17[m] | :0* | 15:30 |
uvos | mighty17[m]: yes the way hildon starts is a huge and terribe hack | 15:33 |
mighty17[m] | so what if we remove the custom xsession | 15:36 |
freemangordon | Pali: CONFIG_WDT is already in next, why it is in your PR too? | 16:22 |
Pali | becase it is based on master and I just wanted to run test pipeline | 16:23 |
freemangordon | ok | 16:23 |
Pali | after it finish, I will close PR and send remaining DM USB patches to u-boot ML | 16:23 |
Pali | opening pull request on github is the easiest way how to run test pipeline | 16:24 |
freemangordon | ok | 16:24 |
freemangordon | yeah, got it | 16:24 |
freemangordon | Pali: please cc tom when sending USB patches | 16:50 |
Pali | yes! I have it prepared | 16:50 |
freemangordon | ok | 16:50 |
freemangordon | because I was the other patches are sent to so TI guy | 16:50 |
freemangordon | hmm, did I wrote that?!? :D | 16:58 |
Pali | I put Tom into recipients list | 17:13 |
Pali | Lokesh is also in recipients list as til now all n900 patches are propagated throw Lokesh branch | 17:14 |
freemangordon | ok | 17:15 |
cockroach | /var/run/docker.sock | 17:17 |
cockroach | sorryk, wrong window | 17:17 |
uvos | who was the guy who made a dialer that directly talked at commands? | 22:53 |
uvos | was than on qt? id like to make him aware of libqofono | 22:54 |
sicelo | iirc he is aware of it | 23:01 |
sicelo | and if i'm not too mistaken, should be rafael2k | 23:02 |
uvos | ok | 23:04 |
uvos | great then | 23:04 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!