| _Gary | Wizzup: I'm a knucklehead... now I realized when you said not to use bionic for the d3 I know now what you mean | 01:14 |
|---|---|---|
| _Gary | I loaded the latest from maedevu and looks like I have a working image! | 01:15 |
| _Gary | I able to bring up the maemo emergency shell now at least | 01:15 |
| _Gary | with that I'm going to make some wiki updates and corrections | 01:16 |
| Wizzup | _Gary: sweet, ty | 11:07 |
| uvos__ | im suprised you got anywhere with the bionic image on d3 | 11:16 |
| uvos__ | the d3 has half the ram | 11:16 |
| uvos__ | and thats not detectable at runtime | 11:16 |
| Wizzup | I think kexecboot specifically | 11:23 |
| uvos__ | well we boot the mainline kernel for kexecboot on d3/bionic | 11:26 |
| uvos__ | so same | 11:26 |
| Wizzup | freemangordon: uvos__: others: when should we start working on bookworm? | 11:43 |
| Wizzup | I imagine we want to finish a few other things and then go? | 11:43 |
| uvos__ | Wizzup: i gues it also depends on how intrusive bookworm will be | 12:17 |
| uvos__ | Wizzup: for sure first we should promote and then wait a bit | 12:18 |
| Wizzup | devel->stable you mean? | 12:18 |
| uvos__ | yeah | 12:18 |
| Wizzup | iirc bookworm still has gtk2 which is the biggest thing | 12:18 |
| uvos__ | gconf stll there? | 12:18 |
| Wizzup | https://packages.debian.org/bookworm/armhf/gconf2 | 12:18 |
| uvos__ | ok | 12:18 |
| uvos__ | maybe wont be too bad | 12:19 |
| uvos__ | did you try latest sphone? | 12:19 |
| Wizzup | I upgraded yesterday, didn't use it much yet, but didn't notice a problem yet | 12:19 |
| uvos__ | it needs a bit of vcm testing before it can be promoted | 12:19 |
| Wizzup | I do have to fix/upgrade my db entries from the past versions | 12:19 |
| Wizzup | anything in particular I should test? | 12:19 |
| uvos__ | mainly placeing and reciveing calls | 12:20 |
| uvos__ | and the logging | 12:20 |
| uvos__ | i also presume that sphone now starts mutch faster with vcm enabled | 12:20 |
| uvos__ | it should at least | 12:20 |
| uvos__ | esp on n900 | 12:21 |
| Wizzup | ok, I haven't checked yet, but sounds good | 12:23 |
| Wizzup | (had a busy few weeks since the new post) | 12:23 |
| uvos__ | Wizzup: debian testing dose now finally really really removed gconf btw | 12:26 |
| uvos__ | *has | 12:26 |
| uvos__ | https://packages.debian.org/search?keywords=gconf2&searchon=names&suite=testing§ion=all | 12:26 |
| uvos__ | so bookworm is end of the line here | 12:26 |
| Wizzup | not unexpected | 12:29 |
| Wizzup | lts support is until 2028 of bookworm though | 12:36 |
| arno11 | uvos__: you probably didn't see my previous msgs: sphone is now much much faster to start with vcm on n900 and backend switch is now ok | 13:16 |
| arno11 | Wizzup: at least on n900, all devel stuff should be moved to stable imo. (to make it usable ootb) | 14:18 |
| arno11 | and to avoid very slow first upgrade to devel (a real pita) | 14:19 |
| Wizzup | ok | 14:33 |
| Wizzup | let me do that now then | 14:37 |
| arno11 | cool | 14:37 |
| arno11 | oh, and ucm is still broken | 14:38 |
| Wizzup | can you explain how and what I need to do to fix it | 14:38 |
| arno11 | yep, a wrong typo in last ucm commit broke ucm | 14:40 |
| arno11 | we need /leste-config/pull/52 | 14:41 |
| arno11 | (my bad) | 14:41 |
| Wizzup | no worries, let me merge now | 14:43 |
| arno11 | ok ty | 14:43 |
| Wizzup | building for -devel now | 14:44 |
| Wizzup | ty | 14:44 |
| uvos__ | arno11: great | 14:47 |
| uvos__ | yeah unfortionatly the QApplication() constructor is really really really slow | 14:48 |
| Wizzup | bookworm also has qt5, so we're mostly ok there I think | 14:48 |
| uvos__ | sphone now avoids doing this by doing a (somewhat complex) chain of forks and sutch | 14:48 |
| arno11 | ok, it works really well now. ty | 14:52 |
| uvos__ | we might want to figure out why qt is stalling while inializing itself for so long | 14:53 |
| uvos__ | still | 14:53 |
| Wizzup | stalling, as in doing nothing, or it is busy? | 14:54 |
| uvos__ | Wizzup: app = QApplication(argv) in python with time suggests its stalling | 14:54 |
| Wizzup | you could try not using the maemo plugins for qt, see if it makes a difference | 14:55 |
| uvos__ | ofc sphone avoiding loading qt at all when signaling itself as it dose now is the best way to do it regardless but for other apps | 14:55 |
| Wizzup | (qpa and theming) | 14:55 |
| uvos__ | Wizzup: i think it makes more sense to look at strace first | 14:55 |
| uvos__ | but not right now | 14:55 |
| Wizzup | https://wiki.qt.io/Performance_Tip_Startup_Time hehe | 14:56 |
| Wizzup | they suggest boosters | 14:56 |
| Wizzup | (i.e. maemo launcher) | 14:56 |
| Wizzup | https://stackoverflow.com/questions/35564370/how-can-i-improve-the-startup-time-of-qt5-programs the bottom answer here is also interesting | 14:57 |
| Wizzup | not sure if relevant though | 14:57 |
| uvos__ | i mean thats maybe not a huge suprise given that this was written while qt was a nokia project | 15:00 |
| uvos__ | but yeah, things to explore | 15:00 |
| Wizzup | uvos__: did you ever release mce 1.10.10 ? | 15:01 |
| uvos__ | Wizzup: no idea | 15:02 |
| uvos__ | did i? | 15:02 |
| Wizzup | I don't think so | 15:02 |
| Wizzup | but you did bump the changelog | 15:02 |
| uvos__ | ok | 15:02 |
| uvos__ | hmm | 15:02 |
| uvos__ | maybe the build failed for some reason | 15:02 |
| Wizzup | well, the ci is operational again, so you can retry :) | 15:23 |
| Wizzup | freemangordon: telepathy-haze maemo/chimaera and maemo/chimaera-devel don't have the same history, can you check and fix? I think we can just replace maemo/chimaera with -devel, but there's a patch you may or may not want to include | 15:25 |
| Wizzup | we're thinking of applying to https://nlnet.nl/mobifree/ - any things that would make sense to propose? | 16:29 |
| Wizzup | we're looking at making a working qt mobile version of organicmaps (they have a desktop qt version) | 16:30 |
| Wizzup | probably also put the debian migration in there | 16:30 |
| Wizzup | sicelo: which devices did you recommend we try out because it was working decently with pmos? | 16:39 |
| sicelo | as of last few weeks, Pixel 3a seems to be the new star of the mainline world, it seems. also has camera working, very well for that matter. otherwise, OnePlus 6/6T (or some other SDM845) are also not far behind. going back some years, most of the stuff with MSM8916 too, e.g. Samsung A3/A5 (2015) | 17:33 |
| Wizzup | sicelo: do they have calls working? | 18:09 |
| inky | arno11: sorry i had to figure hhat out myself. | 18:46 |
| inky | _Gary: someone said d3 doesnt reboot for hours in emergency shell. so i fuess one of the services make it reboot. | 18:50 |
| Wizzup | wrong guess I'd say | 18:50 |
| inky | i'll try one day to just load wmaker and see dow long will it last. i need to edit sd card on pc and disable hildon. | 18:51 |
| Wizzup | you could try to disable the gpu module | 18:52 |
| inky | oh. how is it called? | 18:52 |
| sicelo | Wizzup: yes @ calls. some even have the eSIM working | 18:55 |
| sicelo | of course largely tested with modemmanager, but i think new ofono should support them perfectly fine as well | 18:56 |
| inky | interesting, that f and g are close to each other both on qwerty and dvorak. but on dvorak those are in the higher row. | 18:56 |
| Wizzup | sicelo: ok, cool | 18:57 |
| Wizzup | inky: something with powervr or sgx I am sure | 18:57 |
| inky | great, i'll try to find d3 and test. | 18:58 |
| inky | today. | 18:58 |
| _Gary | inky: yeah I tried to do a shutdown and after 30 minutes gave up and pulled the battery | 19:03 |
| _Gary | if you can get wmaker or matchbox to load on sd that would be great... | 19:03 |
| Wizzup | guys, the instability problem is not a service of some kind, it's related to voltage/frequency very likely | 19:04 |
| Wizzup | so whether you use window maker or something else doesn't matter | 19:04 |
| _Gary | okay got it...inky if there is anything d3 related to test I'm in... | 19:19 |
| freemangordon | Wizzup: ok, will have a look | 20:07 |
| Wizzup | freemangordon: also see msg regarding next debian | 21:33 |
| Wizzup | 11:43 < Wizzup> freemangordon: uvos__: others: when should we start working on bookworm? | 21:33 |
| Wizzup | freemangordon: also is connui-common ready for stable? | 21:51 |
| Wizzup | uvos__: nvm looks like 1.10.10 did make it, I must be blind | 21:52 |
| Wizzup | mce 1.10.10+m7 (> 1.10.8+m7.1 ) | 21:52 |
| arno11 | Wizzup: btw i spent a bit of time @vcm sip calls. i got great results for incoming calls, tweaking latency and buffer in farstream. but still got nothing for outgoing (no connection at all) | 22:27 |
| arno11 | sounds like it still fails to find a common codec | 22:30 |
| arno11 | weird as it works @incoming. PA seems not involved btw | 22:31 |
| Wizzup | arno11: that is weird yeah | 22:54 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!