Daanct12 | Good morning Aikon fellas | 02:57 |
---|---|---|
Daanct12 | https://0x0.st/-9aR.png | 03:22 |
Daanct12 | New update!! | 03:22 |
parazyd | :) | 08:58 |
Wizzup | Danct12: eventually we'll also make it work through the application manager | 11:45 |
uvos | ui-fonts seams like a not so great name for that package | 11:47 |
uvos | hildon-ui-fonts or leste-ui-fonts or something would be better | 11:47 |
uvos | "ui-fonts" is way to generic an prone to a namespace clash | 11:47 |
Wizzup | agreed | 11:49 |
buZz | jesster1234: hi :) | 11:52 |
* buZz was promoting maemo leste a bit in ##freenode today | 11:52 | |
* jesster1234 waves | 11:54 | |
Wizzup | :) | 11:56 |
parazyd | Wizzup: on the VM, I updated _only_ ke-recv via apt and it went fine. | 15:49 |
parazyd | With apt update && apt install ke-recv | 15:49 |
uvos | i proubly tires to use some serivce that has not finished starting up yet and crashes du to insufficant error checking | 15:59 |
uvos | this can happen with mce and gconf btw | 15:59 |
uvos | mce will just segfault if gconf is not available because of insufficant error checking | 16:00 |
uvos | (this is fixed in my branch) | 16:00 |
uvos | so it would only happen if you upgrade some other deamon at the same time.... | 16:00 |
parazyd | Quite possible, yeah | 16:01 |
parazyd | The thing with the reboot is that dsme tries to spawn it 10 times and then quits | 16:02 |
uvos | right | 16:02 |
parazyd | Maybe there should be some interval between retries | 16:02 |
uvos | mabye | 16:02 |
parazyd | Right now I think it's just without any waiting | 16:02 |
uvos | like the 10 one being 30 sec later | 16:02 |
parazyd | Yeah like incremental | 16:02 |
uvos | dsme should just try and respawn and and then give up and notify the user really | 16:03 |
uvos | the reboot behavior is more detrimental than benefical | 16:03 |
uvos | as seen offten by us allready... | 16:03 |
parazyd | I agree, and it's kinda risky | 16:03 |
uvos | right it breaks apt | 16:03 |
parazyd | For now it just renders unconfigured apt | 16:03 |
parazyd | But it could be worse | 16:03 |
uvos | right | 16:03 |
parazyd | It is a "clean" reboot, so files should sync, but still... | 16:04 |
uvos | yeah | 16:04 |
parazyd | And no-lg-reboots doesn't help here btw. | 16:04 |
uvos | there are several reboot reasons yeah | 16:04 |
uvos | lg file just stops one | 16:05 |
uvos | not the hartbeat one either | 16:05 |
parazyd | mhm | 16:05 |
uvos | anyhow btw i have ke-recv half implemented in mce, since they both do mostly the same stuff (abstract some kernel interfaces, they even do the same ones partially) i can just finish that and then we can drop ke-recv | 16:07 |
parazyd | https://github.com/maemo-leste/dsme/blob/master/modules/lifeguard.c#L879 | 16:08 |
uvos | looks like it should do this allready | 16:09 |
uvos | btw regarding ke-recv it dose these things: 1. sense for keyboard slide state 2. usb state and 3. memory state | 16:11 |
uvos | 1. mce allready dose too 2. i need to finish implmenting 3. ke-recv code dosent work on mainline kernels and in sfos they implmented this interface as a mce module we can mostly just cp | 16:11 |
parazyd | There's a bit more, but yeah | 16:13 |
parazyd | AFAIK it also does OTG and camera | 16:14 |
parazyd | And there was some FAT implementation, but I think that isn't necessary anymore. | 16:14 |
uvos | right btw im talking about the kec-recv binary not the repo, that contains a bunch of other tools | 16:14 |
parazyd | ok | 16:15 |
jesster1234 | so maemo being permanately landscape | 16:28 |
jesster1234 | seems like the perfect distro for the pinephone keyboard when that comes out | 16:28 |
uvos | jesster1234: leste can do portrait | 16:29 |
uvos | or even portait only if you ajust /usr/share/hildon/transtiton.ini | 16:29 |
jesster1234 | ah | 16:29 |
uvos | its just the n900 thats perma landscape | 16:30 |
uvos | (because of bugs in its pvr driver) | 16:30 |
jesster1234 | also the keyboard loks like `onboard` is that just a coincince? | 16:31 |
parazyd | What's onboard? | 16:35 |
Danct12 | osk for x11 | 16:35 |
lel | parazyd labeled an issue: https://github.com/maemo-leste/bugtracker/issues/524 (PowerVR DDK 1.17 Glamor/Xorg) | 16:36 |
parazyd | No, we have our own | 16:36 |
lel | parazyd closed an issue: https://github.com/maemo-leste/bugtracker/issues/543 (Import bulgarian translation) | 16:37 |
lel | parazyd assigned an issue: https://github.com/maemo-leste/bugtracker/issues/538 (Research mapphone kexecboot updates through a package) | 16:38 |
Danct12 | https://github.com/maemo-leste/ui-fonts/tree/master/fonts/nokia | 16:39 |
Danct12 | neat, looks like i can safely remove my nokia fonts :) | 16:39 |
parazyd | uvos: Do you know if mce handles screen timeout or is there some other component? | 16:41 |
parazyd | uvos: Timeout being: turning off the screen after N seconds idle | 16:42 |
uvos | mce dose this | 16:42 |
uvos | and i compleatly rewrote the code that dose this too | 16:42 |
uvos | why? | 16:42 |
uvos | is it not working right? | 16:42 |
Wizzup | you up for a swim, or did you already go in the -clear | 16:43 |
Wizzup | oops :) | 16:43 |
Wizzup | (that was meant for a pm) | 16:43 |
parazyd | uvos: No I just want to solve this: https://github.com/maemo-leste/bugtracker/issues/506 | 16:43 |
Wizzup | parazyd: I see ke-recv | 16:43 |
uvos | Wizzup: sure im up for a swim :P | 16:43 |
parazyd | lol | 16:43 |
Wizzup | uvos: :p | 16:43 |
parazyd | 32C here | 16:44 |
uvos | https://github.com/maemo-leste/mce/blob/master/modules/inactivity.c | 16:44 |
uvos | parazyd: ^^^ | 16:44 |
parazyd | uvos: So the schema in cpdisplay can be patched like this: http://ix.io/3qrS | 16:45 |
parazyd | And then there we could support it | 16:45 |
uvos | sure | 16:45 |
parazyd | Could you make a patch for this here? https://github.com/maemo-leste/mce/blob/master/modules/inactivity.c#L226 | 16:45 |
parazyd | I can also if you don't have the time | 16:46 |
uvos | https://github.com/maemo-leste/mce/blob/eae9aa8897e9d90fd50eef050820749bc8795c38/modules/inactivity.c#L233 | 16:46 |
uvos | just remove that and dont | 16:46 |
uvos | https://github.com/maemo-leste/mce/blob/eae9aa8897e9d90fd50eef050820749bc8795c38/modules/inactivity.c#L237 | 16:46 |
uvos | if its 0 | 16:46 |
parazyd | uvos: So, http://ix.io/3qrU ? | 16:47 |
parazyd | Also the 30 there should probably be the DEFAULT_TIMEOUT defined on top of inactivity.c | 16:50 |
uvos | yeah sure that should work | 16:50 |
parazyd | Awesome | 16:50 |
parazyd | I think it's a valid usecase | 16:50 |
parazyd | For example it could be utilized to keep the screen on when you're navigating | 16:50 |
uvos | im not sure sure we allready have https://github.com/maemo-leste/mce/blob/master/modules/inactivity-inhibit.c | 16:51 |
parazyd | ah ok, sure | 16:51 |
parazyd | Still, useful when I'm working on the device :) | 16:51 |
parazyd | Will push the patches | 16:51 |
uvos | ok | 16:51 |
lel | parazyd assigned an issue: https://github.com/maemo-leste/bugtracker/issues/506 (Feature: Infinite backlight timeout) | 16:59 |
parazyd | uvos: Seems to work | 17:53 |
lel | parazyd closed an issue: https://github.com/maemo-leste/bugtracker/issues/506 (Feature: Infinite backlight timeout) | 17:53 |
freemangordon | Pali: now what? | 21:46 |
sicelo7 | ping N900 | 21:47 |
Pali | freemangordon: you can answer that we can do full board conversion after u-boot reduce size of usb dm code and also after reduce size of all dts components as it currently is too big and cannot be enabled for n900 | 21:49 |
Pali | freemangordon: also you can write that because full board conversion is not simple task and reviewing/merging simple patches for n900 took more than year, it means that full board conversion would not be merged into u-boot sooner than in the end of year 2022 | 21:50 |
Pali | it is simple reality | 21:51 |
Pali | that discussion does not make sense | 21:51 |
freemangordon | mhm | 21:51 |
Pali | either they want to remove n900 board code and do everything for it | 21:52 |
freemangordon | I'll ask | 21:52 |
Pali | or want to move forward, and the easiest way is by simple steps | 21:52 |
Pali | like that your patch | 21:52 |
Pali | small patch = less review = sonner merge | 21:53 |
Pali | thats is why I really like your patch! | 21:53 |
Pali | it is elegant solution for this issue, how to migrate code to USB DM without need to rewrite everything | 21:53 |
Pali | based on fact how much time took reviewing and merging last n900 patches, I'm very sceptical that full board migration patches would be merged sooner that in 12 months | 21:55 |
Pali | plus, this work is non-trivial and I guess developing and testing would take lot of time too | 21:56 |
Pali | so 1.5 year horizont is really real | 21:56 |
Pali | ... but the main question is, why to do full board conversion at all? | 21:57 |
Pali | this is not required, and there is no plan to remove boards without dts | 21:58 |
Pali | so why it should be requirement for n900? and why now? | 21:58 |
Pali | also... I do not think that your patch is hack, it is designed in the same way like other uboot drivers with those #ifdefs | 22:02 |
freemangordon | yeah | 22:05 |
freemangordon | Pali: I can't do much more, if my last mail don't convince them... well,I guess I'll vent a bit, but not sure it will help | 22:47 |
Pali | freemangordon: thanks for this! | 22:47 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!