| arno11 | Wizzup: sicelo: i just fixed a last minute UCM bug with earphone/jack plug. Seems to be fine now and the 2 main profiles are able to switch between speakers, earphone and jack | 21:00 |
|---|---|---|
| Wizzup | great! | 21:03 |
| arno11 | :) | 21:04 |
| arno11 | Wizzup: is there any known issues with realtime/timers/clock with Leste ? | 21:10 |
| Wizzup | do you mean kernel timers? | 21:14 |
| Wizzup | or something else | 21:14 |
| arno11 | yes from the kernel side | 21:14 |
| Wizzup | It's a tricky question | 21:21 |
| Wizzup | I am not aware of particular problems | 21:21 |
| arno11 | ok, maybe i'm wrong but the common point between PA call latency, audio video sync and ir-tx issues on n900 is timers | 21:21 |
| arno11 | that's why i'm asking | 21:22 |
| Wizzup | I think timers in general are just tricky :) | 21:32 |
| arno11 | yes indeed :D | 21:33 |
| arno11 | Wizzup: btw we can't load nokia modem on boot finally. too buggy without overclock | 21:44 |
| arno11 | the only 100% working combination is nokia-modem then sphone then cmt_pulse once hildon is loaded | 21:45 |
| arno11 | *already loaded | 21:46 |
| arno11 | (and omap_ssi and cmt_speech modules on boot) | 21:47 |
| Wizzup | ok | 21:53 |
| Wizzup | what does 'once hildon is loaded' mean exactly | 21:53 |
| arno11 | i mean when hildon desktop appears | 21:53 |
| Wizzup | ok, because there is a specific 'hildon is ready' dbus signal that is sent, which is why I asked | 21:55 |
| Wizzup | if sphone picks up on new modems then we don't need to (re)start sphone after loading the module | 21:55 |
| arno11 | oh nice | 21:55 |
| Wizzup | do you have the ucm changes ready for me? | 22:01 |
| arno11 | yes | 22:01 |
| arno11 | sending in few minutes | 22:04 |
| Wizzup | ty | 22:04 |
| arno11 | email sent | 22:10 |
| sicelo | thanks, got it | 22:16 |
| sicelo | btw, i think, at least later on, we should move as much of the `cset` commands that are in the EnableSequences to the RX-51 conf, under BootSequence | 22:18 |
| uvos | sphone dosent pickup new modems atm | 22:27 |
| uvos | but it dose pickup ofono (re)starting later | 22:27 |
| uvos | so as a workaround you could ensure ofono starts after the modem is ready | 22:28 |
| arno11 | ah ok | 22:28 |
| uvos | sphone will then pickup the modem at that time, no matter when ofono was started | 22:28 |
| uvos | *when it was started | 22:28 |
| uvos | ie start sphone first (or whenever) start nokia-modem and cmt_pulse and then ofono | 22:29 |
| Wizzup | arno11: I will push the UCM to devel | 22:31 |
| arno11 | ok :) thx | 22:31 |
| arno11 | uvos: ok thx but iirc cmt_pulse can't start before ofono because it uses dbus ofono events to work (otherwise cmt_pulse crashes) | 22:41 |
| uvos | ok, just the modem should also be enougth | 22:42 |
| uvos | point is that the modem must exist at ofono startup for sphone to pick it up | 22:42 |
| arno11 | ah ok | 22:42 |
| uvos | beyond that when sphone starts wont matter | 22:43 |
| uvos | so i gues its nokia-modem -> ofono -> cmt_pulse and sphone starts whenever | 22:43 |
| arno11 | ok sounds good | 22:45 |
| arno11 | and need all of this happening as late as possible to avoid nokia-modem crash and then ofono and cmt_pulse crash and finally sphone magical 'ofono is not ready message' lol | 22:51 |
| sicelo | this crash is related to voice, i guess? | 23:01 |
| sicelo | since at least there's no such crash in postmarketOS for me, and no need for special loading order. but i've not tested/used voice there | 23:01 |
| arno11 | ok interesting to test with voice | 23:09 |
| Wizzup | do you use PA in pmos? | 23:11 |
| Wizzup | or 3g data? | 23:12 |
| sicelo | no to both. but i always test that modem at least can be brought online, e.g. after a kernel update | 23:18 |
| sicelo | but will probably start using PA soon | 23:19 |
| sicelo | and seems 3g data is broken for some reason - i noticed it earlier today ... then again, ofono seems to behave strangely in pmOS for a while. e.g. can't search for network operators, while same version of ofono in leste has no issues | 23:20 |
| Wizzup | doesn't it work in leste (data)? | 23:20 |
| * sicelo suspects musl-related issues, but can't confirm it | 23:20 | |
| sicelo | last i tested, it worked fine under Leste | 23:21 |
| Wizzup | ok, yeah | 23:22 |
| arno11 | (indeed data is working very well under leste) | 23:22 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!