libera/#maemo-leste/ Monday, 2022-08-01

Wizzupuvos: I made a few calls and the speaker and handset switching works fine )on the d4 at least)18:03
uvosWizzup: great :)18:50
uvoscould you try the pp?18:51
uvosand if that works ill promote sphone18:51
uvosmaybe after your rtcomm patch18:51
uvosif thats comeing today18:51
Wizzupshould, but my day's been filled with work and meetings18:51
uvosok18:52
uvosbut you have a pp around?18:52
Wizzupyes18:52
uvosok18:52
WizzupI will also test the other thing you asked me to test out18:52
uvosthanks yeah18:52
WizzupI just want to do the rtcom el patch first18:52
uvosthat ones wierd, it worked on pp before and still works on bionic and d418:52
uvosbut multiple people have reported it18:52
Wizzupbtw, I've "received" the same sms about 9 times today on my d418:53
uvosyeah i have seen that18:53
Wizzupjust got it again :D18:53
uvosofono is a mess still18:53
WizzupI think it relates to me turning on the screen or something18:53
Wizzupah, got it 3 times18:53
Wizzupodd18:53
uvosmight be mce kicking the modem18:53
Wizzupright18:53
uvosand ofono failing to tell the modem its been recieved or somehting18:54
WizzupI am not sure if the problem is ofono or the kernel18:54
uvosdunno yeah18:54
Wizzupwe should probably try it with the power consuming ofono as well, just to see if that works fine18:54
Wizzup(the other ofono driver)18:54
uvosthe whole modem infra also behaves really wierd in a different way18:55
uvosso i have a contract with 2 sim cards18:55
uvosand often when the d4 is with the sim unlocked the other device wont be able to make or recive calls or send or recive sms18:55
uvosas if the operator thinks the d4 is sending something on gsm18:56
uvos(other than data)18:56
uvosthis even presits if the device is turned off18:56
uvosthe only way to make it go away is to reboot the d4 into android and connect to the operator there18:56
uvosor wating a couple of h18:56
Wizzupodd.18:56
sicelothe issue with droid-4 refusing to charge if booted from cable is a bit annoying (because how else do you boot from a depleted battery?). anyway ... most times, the kernel is able to start, but then you have to know when to remove cable, and reinsert it (i think buZz explained this a couple of days/weeks ago).18:58
uvosi dont see this18:58
uvosbut i can see why it would happen18:59
uvosdepending on when the charge enable interrupt comes in it will fail to work18:59
siceloi was wondering - would a forced reset in kernel be helpful *after* the module has loaded. clearly reinserting usb cable works, so i suppose a forced reset in s/w should help too18:59
uvosdepending on when cpcap-battery probes realatvie to cpcpa-mfd18:59
uvossicelo: not sure what you mean by reset18:59
uvossicelo: the way it works is cpcap fires an intterupt when the interrupt is enabled19:00
uvosdepending on how it goes timeing wise thats before the kernel is ready to start charging19:00
uvosim not sure if theres a register you can read to see if the usb cable is plugged in19:01
uvosiirc alot of the other negotaing stuff goes by the fw19:01
uvoswhich we dont understand at all19:01
uvosid have to check19:02
uvosa temporary fix might be to just build the charger module into the kernel19:02
uvosthat would help with booting with a low battery anyhow since charging would start earlyer19:02
sicelomaybe that's what should be done then19:02
siceloor delay loading the charging module until things have settled19:03
uvosthe problem is that the intterupt infa is enabled by the mdf code19:03
uvosnot the charging module19:03
uvos*mfd19:03
sicelommm. this is kernel code you're talking about? or firmware that we have no control over?19:04
sicelo(mfd code, i mean)19:05
uvoskernel code19:06
uvosanother way that might also work ist to just enable charging unconditionally on probe19:08
uvosand just let the vbus low irq disable it again19:08
uvosif vbus is low19:08
uvosie 0v19:08
sicelooh btw, this happens even when battery has power (just in that situation, you can simply boot without the cable)19:13
siceloanyway ... hope someday a solution will be found19:13
uvossicelo: yeah the driver just tries to read CPCAP_REG_INTSx at boot20:32
uvosof you cant expect that to show the current state, or even to contain valid data at all if you just read the irq state registers without wating for cpcap to fire an irq20:33
uvosno suprise its buggy and behaves inconsistantly20:33
uvoslooks like android itterogates the fw20:35
uvoswe cant do that20:35
uvosofc that dosent mean its the only way to gain the information20:35
uvoslooks like while i cant repo this on d4 i can on xt87520:45
uvosWizzup: please kick linus21:45
uvos*x21:45
uvosfor devel21:46
Wizzupdone21:54
uvoshttps://phoenix.maemo.org/job/droid4-linux-source/182/console22:07
uvosWizzup: this is pretty hard to read22:07
uvosupps forgot to push the tag22:07
uvosWizzup: ok pls kick again22:08
uvossicelo: when this drops please check if it fixes the issue for you22:08
siceloit's this? https://github.com/maemo-leste/droid4-linux/commits/maemo/beowulf-devel22:15
sicelo i don't see a commit which can fix anything there. you didn't push or something?22:16
siceloseems the build failed too22:17
sicelohttps://phoenix.maemo.org/job/droid4-linux-source/182/consoleText22:17
* sicelo makes a note to build 5.19 for N90022:20
uvosthe source is not in that branch22:20
uvosjust the packaging22:20
uvosyes it failed i forgot to push the tag as mentioned above22:20
Wizzupuvos: yes I think that hapened in the latest update22:35
Wizzupof jenkins22:35
uvossicelo: its in the repo now23:23
uvospls test23:24
uvosWizzup: btw i merged in your additional force enables23:24
uvosand that seams to work ok23:24
Wizzupalso for headset?23:26
uvosyes i squahsed exp3 and merged it in23:27
uvos(without the regmap hack)23:27

Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!