libera/#maemo-leste/ Friday, 2022-01-21

Wizzupparazyd: I made backups of the droid4 images over 2021 on my own machine because amprolla ran out of space again, where should I copy them to00:09
Wizzupfreemangordon: so tomorrow should we try his patch first, or just go for buffer age?01:09
freemangordonI'd say - lets try the patch07:07
freemangordonwe may use it as a workaround for a while if it works07:07
tmlinduvos: if you happen to have adb logcat traces, i could use some gnss time init examples to add support for that08:52
tmlinduvos: grep -B3 MPDTIME= in the logcat output files if you have some, the time format is not utc or gps start time in utc seconds08:53
tmlindthe only example i have is GpsInterface_inject_time( 1549911987370, 65652, 76 ) translates to 0034AT+MPDTIME=287,1291573419,7608:56
tmlindi'll try to dump out some more examples too08:56
tmlindi think it might require enabling debugging for ts27010 under debufgs to see the at commands09:09
Wizzuprafael2k: looks like we might soon have some fix for the last rendering problems10:34
Wizzup(pp)10:34
Wizzupfreemangordon: shall I build mesa with enunes patch to see if it helps?10:34
freemangordonWizzup: in the repo?11:05
freemangordoncould you try locally first11:05
Wizzupfreemangordon: takes forever to compile, I'm building for -experimental in CI11:05
freemangordonok then11:06
Wizzuphuh telepathy-ring goes into offline if it is not connected to wifi11:10
Wizzuphmm11:10
Wizzuptmlind: droid4 image is here https://maedevu.maemo.org/images/droid4/20220120/11:40
freemangordonuvos: hmm, seems display state pipe is called twice per lock11:45
tmlindWizzup: ok i ended up using the older image last night, upgrade to devel is a pain currently :)12:08
uvosfreemangordon: well it should if its on->dim->off (it might allways do that to make sure the state is consistant i dont recall)12:10
uvosit its runing it twice with off as the argument12:11
uvostheres a bug somewhere, but it dosent matter mutch because every callback has to check if the callback isent run twice with the same argument anyhow12:11
Wizzuptmlind: hmm how is it a pain, we should fix that12:11
uvosas that can happen in various circumstances12:11
tmlindWizzup: mce and hildon-desktop refuse to upgrade, xorg won't restart, apt dist-upgrade needs to be finished in the emergency shell etc12:12
uvostmlind: i dont have any logs sorry12:13
uvostmlind: i can try and captue some tomorrow12:13
tmlinduvos: ok no problem, can't get the stock android to install, can't find a sim right now12:13
uvoslos should be fine right?12:14
uvosbtw this trick should work on d4 https://maedevu.maemo.org/images/bionic/README.txt12:15
tmlinduvos: no luck with the corner taps, setting up a new micro-sd card for d412:17
tmlinduvos: yeah lineageos works fine for capture too, i think you need to set the ts27010 debug level though12:18
tmlinduvos: now the corner trap trick worked :) it may need to be done on right away on the first language selection screen, and the top taps may need to be about 2cm lower than the top and multiple rounds12:29
Wizzuptmlind: hmmmm ok, because it reboots somehow during upgrade12:34
Wizzupfreemangordon: looks like the mesa build finished, now we'll still have to set the behaviour in clutter, I'll get my pp online in a bit12:36
uvosWizzup: lets just disable lifegurad for now, please12:53
Wizzupfreemangordon: what do you think, disable lg until we have fremantle-style system upgrades where it forces a reboot?13:13
Wizzupas in, it prevents you from doing anything but doing the upgade13:13
tmlinduvos: seems like the stock distro has buggy ts27010 debug, or los has enabled more features, not seeing the at commands with logcat after echo 0x7fffffff > /sys/module/ts27010mux/parameters/debug_level13:34
uvos"fremantle-style system upgrades where it forces a reboot" <- i honestly hate this13:35
uvosand there is no scenario where lg reboot helps anything anyhow13:36
freemangordonWizzup: this is fine for -devel, but not for -stable IMO13:36
freemangordonuvos: this was already discussed13:36
uvosnot really13:36
uvosyou just dissmiss it with no argument13:36
freemangordonuvos: I am not going to argue on simple common sense13:37
Wizzupwell there should be a sensible way for end users to do system upgrades13:37
Wizzupsimple dist-upgrade doesn't seem likely to cut it13:37
uvosfreemangordon: bullshit commen sense13:37
uvosWizzup: you can have lifeguard and apt13:37
uvosWizzup: just disable it before and reenable it after13:37
uvosno need to reboot or do anything else13:37
freemangordonuvos: so, Xorg crashes, how you;re going to restart if it happens to notice?13:37
uvosif dsme dident exist and hildon was a xdg shell you could just restart it13:38
uvoslike any deamon13:38
uvosanyhow13:38
Wizzupkeep in mind that what we do now doesn't have to be what we do ~3 months from now13:38
Wizzupif it's a real problem to upgrade for many people that's something of concern and more pressing than what we do later imho13:38
uvosso lifeguard's usefullness:13:38
uvosok lets pick a random deamon13:38
freemangordonWizzup: what about having no lg revoots for -devel only?13:39
uvosfreemangordon: it breaks all the time on stable too13:39
freemangordondo we know why?13:39
uvosyes random deamon fails (different one eatch time)13:39
uvosso please listen for a second13:39
freemangordonhow's that?13:39
uvosbugs in deamons13:39
freemangordonand we push those to -stable?13:40
uvosthey fail in untested configurations13:40
uvosso yes13:40
uvoswe cant test everything13:40
freemangordonmhm13:40
uvosso lifeguard's usefullness:13:40
freemangordonwell, lets disable WDs then, why not13:41
freemangordonwe can;t test kernel as well13:41
uvosits different13:41
uvosplease just listen for a while13:41
freemangordonsorry, I am out of this conversation13:41
uvossee13:41
freemangordonI don't find it useful13:41
uvosthis is how we "discuss" this13:41
Wizzupfreemangordon: I think enunes patch fixes the problem if we also request the bit on clutter13:43
freemangordonsaying "because we cannot code our daemons to be stable enough to not crash on restart, lets remove lg once and for all" is not something I am going to discuss13:43
uvosfreemangordon: thats not what im saying13:43
uvosat all13:43
Wizzupfreemangordon: wrt daemons it's a matter of prioritising13:43
uvosthe problem is that lifeguard dosent help13:43
uvosit hurts13:43
Wizzupwe can turn on lg later again, currently it's more of a problem than that it helps13:43
freemangordonsure, if we hit reboot loop13:43
uvosthere is no scenario where it helps at all13:43
uvosever13:43
freemangordonwhich means it is useless13:44
freemangordonwhich means it shall be removed13:44
uvosits worse thatn useless as it hurts13:44
uvosexcept where there are defficancys that make it impossible to restart deamons without restart13:44
uvosliek x13:44
uvosbut this is just because of architecture problems13:44
uvosand we should fix unrestartable deamons13:44
uvosto be restartable13:44
freemangordonthere are only several daemons that lead to system restart13:44
uvosquite a few do really13:44
freemangordonless than 1013:45
uvosthats quite a few13:45
uvosand again it dosent help anything for these 1013:45
uvosso lets go though an example13:45
freemangordonso, what do you do if something in /tmp gets messy? DBUS address, for example13:45
freemangordonno daemon restart will fix it13:45
freemangordonyou need to reboot the whole device13:45
uvosdeamon restart absoluly should clear the dbus addresses13:46
freemangordonnot really, becouse half of the system will use the old address13:46
freemangordonor, you want all the daemos to crahs and restart instead of doing a clean reboot?13:46
WizzupI think we're disregarding the more practical problem(s)13:46
uvosyes ofc13:47
freemangordonuvos: this might mae sense on server13:47
freemangordon*make13:47
freemangordonnot on mobile13:47
uvosfreemangordon: please let me explain for a second13:47
freemangordonok13:47
uvosso lets say we have a deamon with a bug13:47
uvosmaybe mis (dosent matter what deamon)13:47
humpelstilzchen[I wonder what your current target audience is - enduser or developer13:47
uvosit derefrences a nullpointer when you use it13:47
uvosso it chrashes13:48
uvosand dsme restarts it13:48
freemangordonhumpelstilzchen[: thats the issue,actually ;)13:48
uvosand it chrashes again when used etc13:48
Wizzupfreemangordon: -current- target audience :)13:49
Wizzupfreemangordon: btw, did you see my msg? pinephone problem seems fixed13:49
freemangordonWizzup: yes, but lets finish with lg first13:49
uvosthis is "fine" as lg will never reboot as the crashes are slow but we should give the user some indication on the problem as rn it will look like everything wokrs and this is imo worse than informing the user that something is wrong by the deamons features being missing13:49
uvosbut ok13:49
uvosno lg reboot needed13:49
uvosnow we have another bug, here it crashes right at startup if config option xyz is set13:49
uvoswe dont set xyz so we dident notice13:50
uvosour user now has a deamon that allways crashes on startup13:50
uvoswhat happens? device reboots deamon crashes again device reboot deamon crahses again13:50
uvosthe user cant use the device (even thoug the deamon might not be _absolutly_ critcal (maybe for emergency call at least))13:50
uvosand even worse13:50
uvosthe user cant give us the logs13:50
freemangordonuvos: it is as simple as:"we shall not push system critical daemons in -stable that are prone to such breakages"13:51
uvoswithout pluging in the sdcard (which is only possible if leste is on an sdcard)13:51
freemangordonalso, lg is an option13:51
uvosturning off lg in a boot loop is impossible13:51
freemangordonnot every daemon under dsme is set todo it13:51
uvosto point is that if a deamon is cosntantly crashing and restarting13:51
uvosa reboot is very unlikley to fix it13:51
freemangordonsee, it is mce, Xorg, h-d and few others that lead to l-g reboot13:51
Wizzupke-recv as well13:51
freemangordonyeah13:51
Wizzupand a few more iirc13:51
uvosand a boot loop makes it impossibe to debug or give us info13:52
Wizzupstatus applets, hildon-home13:52
Wizzupicd2 too13:52
uvosmce is not vital13:52
uvosyou can use the phone without mce13:52
freemangordonok, if we think those daemons shall not cause lg reboot, lets assess them one by one13:52
uvosand call emergancy13:52
Wizzupuvos: I disagree with you there but I do think we should not reset now if mce crashes many times13:52
uvosmce causing a reboot is detrimental13:52
Wizzupbut in practice it probably means you cannot unlock your phone :)13:52
freemangordonmhm13:52
Wizzupfreemangordon: the debian upgrade process also messes with these states and daemons13:53
uvosthe problem is that the likely hood of a reboot solveing anything is mutch mutch lower than a bootloop13:53
uvoswhat lg could do13:53
uvosis reboot _once_13:53
uvosand then never again untill the deamon in question starts correctly13:53
uvosthat would be fine13:53
uvosand hildon needs to become an xdg session13:53
freemangordonagain, if we think some daemons shall not cause lg reboot, lets just remove that freom their startup13:53
uvosso x can bre restarted13:53
uvosthats just a travisty atm13:53
freemangordonuvos: we don;t really need such thing on mobile13:54
uvosyes you do13:54
uvosfor recovery13:54
uvosx crahing need to cause a reboot and causing a reboot makes it hard to debug13:54
uvosso dont do it13:54
uvosand xdg session has many other advantages too13:54
freemangordonok, how hard is to go through the daemons and remove 'lg reboot' option from dsme startup line?13:54
uvos x crahing need not13:54
uvosfreemangordon: not very13:55
uvosbut thats hardly the point13:55
uvosits enabled by default atm13:55
freemangordonwrong13:55
uvosand if you mean during a boot loop13:55
uvosthen its very hard13:55
freemangordonit is explicitly enabled by dsme command line for a particular daemon13:55
uvosthat is default13:55
uvosfor the user13:55
freemangordonalso, I am using fremantle for the last 11 years - never hit a reboot look because mce is buggy or Xorg crashed or whatever13:56
freemangordon*loop13:56
uvosbecause nokia threw the thing over the wall13:56
uvosand it only runs on one device13:56
uvosit has mutch less variables13:56
freemangordonhow that matters?13:56
freemangordonah13:56
uvosand ofc we cant possibley do as mutch qc as nokia13:57
uvosfor eeatch device13:57
uvosas they did for n90013:57
freemangordonanyway, I am agains removing lg reboot13:57
uvosits just not realistic13:57
freemangordon*agains13:57
freemangordont13:57
freemangordonso, we can temporarily put no_lg_reboots until we think we have stable enough system13:57
freemangordonWizzup: ^^^13:58
WizzupI agree with that13:58
freemangordonto answer your question13:58
uvosi would not renable lg untill it only reboots once13:58
uvosperiod13:58
Wizzupso do we install that file from an update, or change the default in dsme? fine with either13:58
freemangordonhmm, image-builder?13:59
uvosi would move the very obscure file you have to touch13:59
uvosinto a proper config file13:59
uvosand make it not default13:59
freemangordonthis is developer, not end-user option anyways13:59
uvosa config .ini file13:59
uvosis not realy user facing13:59
uvosin a meaningfull way13:59
freemangordonwhy?13:59
uvostouch /etc/no_lg_reboots14:00
uvosis just bad14:00
uvoseven for developers14:00
freemangordonoh, come on14:00
freemangordonif it is documented14:00
uvosit should not have to be documented14:00
freemangordonBTW, right now TS on d4 does not react14:01
uvosan /etc/dsme.ini with lifeGuardReboot=false ad a # explantaion what this is is self documenting14:01
freemangordonthis is after I restarted mce a couple of times14:01
freemangordonwhile playing with quircs module14:01
freemangordon*quirks14:01
uvosif mce crashes while the display is off14:01
uvosthat is the result14:01
freemangordonwhy is that?14:01
uvosbecuase it disables the xinput devices14:02
uvosand it cant enable them on startup14:02
uvosbecause there are other disabled xinput devices it cant enable14:02
freemangordonbut it reenables them on startup, no?14:02
uvosso it dosent know what to enable14:02
Wizzupwe can move the file to /etc/dsme ?14:02
freemangordonsure, but how is that different?14:02
Wizzupmore sensible to ship in dsme package14:03
freemangordonI mean - lets focus on important things, not where a particular, developer-only use file is located14:03
WizzupI am just thinking about how to now make this no reset the default for now14:03
freemangordonI think image-builder shall create the file14:03
Wizzupso current users won't get it?14:03
freemangordonor, you want current users to have that too?14:03
WizzupI think so14:04
freemangordonwell, hildon-meta then14:04
Wizzupor leste-config14:04
freemangordonmhm14:04
Wizzupok, I'll try to make it work14:04
uvosnot meta14:04
freemangordonok, -config14:04
uvosbut i thing shiping  it in dsme makes more sense14:04
WizzupI agree with uvos but I also don't care enough to discuss it that much, either dsme package or leste-config14:04
Wizzupeither is fine14:04
Wizzupso - can we talk pinephone :)14:04
freemangordonno, because changing that would require dsme upgrade14:05
Wizzupsure, ok14:05
freemangordonand this is one of the daemons you can't upgrade without rebooting the device14:05
freemangordonIIRC14:05
Wizzupfair enough14:05
Wizzupso pinephone ... it looks like the rendering problems are gone14:05
Wizzupbut there definitely seems to be a (serious) performance penatly14:05
uvoswe dont restart dsme on upgrade14:05
uvosthats it14:05
uvosotherwise you can upgrade it fine14:06
Wizzupas in, portrait mode is less smooth for sure than it was before14:07
Wizzupbut I don't see rendering problems in firefox or the qt web browser or the osso-xterm thing I used to reproduce the bug14:07
freemangordonsec, phone call14:08
Wizzupok, brb14:08
Wizzupyeah so the mesa patch + clutter change to request preserved buffers works14:15
freemangordonWizzup: maybe we shall take that patch temporarily until we implement buffer_age14:29
freemangordonI think it makes sense to test that mesa+clutter on d4/n90014:30
Wizzupok14:30
WizzupI can build the clutter patch for -experimental14:31
Wizzupwe would not expect perf change there right14:31
freemangordonmhm14:31
Wizzuphow do we think buffer age will help perf btw14:32
freemangordonhonestly I wonder how/why it works on SGX14:32
freemangordonWizzup: it will prevent uploads14:32
Wizzupmhm14:32
* freemangordon is back to mce d4 quirks module14:32
freemangordonuvos: nothing shall be done on   MCE_DISPLAY_DIM in that module, right?14:35
Wizzupfreemangordon: oh wait I still had use_fallback = TRUE14:37
freemangordon:)14:37
Wizzupthat explains poor perf14:38
freemangordonyeah14:38
freemangordonyeah, that's how I like it babe :). Double=press on d4 now losk immediately :)14:40
freemangordon*locks14:40
Wizzupneat14:41
Wizzupclutter-0.8 is building14:50
freemangordonuvos: do I get it right that we shall return from https://github.com/maemo-leste/mce/blob/maemo/beowulf-devel/src/modules/quirks-mapphone.c#L42 on  MCE_DISPLAY_DIM?14:52
uvosfreemangordon: no that it dose it on dim is a micro bug in the module yeah14:53
uvosthats obv part of the perf problem14:53
uvosif it shal disable modem rssi on dim or on off is a bit of a matter of taste14:53
uvosbut it should not do both as it dose rn14:53
freemangordonIt should on off, not on dim imo, as off follows dim in a second14:54
freemangordonalso, I think if old state was dim we shall not enable on 'on'14:55
freemangordonbecause it was not turned off14:55
Wizzupfreemangordon: I don't think I see a perf diff on the d415:11
freemangordongreat15:11
WizzupI think we can start pushing stuff from -devel to stable soon imho15:11
Wizzuprafael2k: hey15:11
freemangordonis it better with use_fallback = FALSE on PP?15:11
Wizzupfreemangordon: it's a bit better yeah, but still noticeable perf diff15:12
Wizzupbut it's not -that- bad15:12
freemangordonok15:12
Wizzuprafael2k: want to test a fix for pinephone rendering bugs?15:12
Wizzupfreemangordon: for me the main stable issue is the dmabuf leaks (or so it seems)15:13
Wizzupbut imho we're better than current stable15:13
Wizzupcurrent stable still has segfaults when vkb is visible and rotation changes15:13
Wizzup(d4 btw)15:13
Wizzuprafael2k: if you enable beowulf-experimental and reboot, you should have the fixes15:14
freemangordonWizzup: I'll have a look at those during the weekend for sure15:14
freemangordonI'll just fix this mce behaviour first as it drives me mad :)15:15
Wizzup:)15:16
WizzupI will look at adding the no lifeguard reboots to leste-config for now15:16
freemangordonok15:17
Wizzupbuilding15:22
Wizzupfreemangordon: shall I build this mesa and clutter for devel?15:41
freemangordoncould you make one more change in mesa while on it?15:42
freemangordonWizzup: https://github.com/maemo-leste-upstream-forks/mesa/blob/maemo/beowulf-devel/src/mesa/drivers/dri/pvr/pvrext.c#L7815:44
freemangordonchange that to 815:44
freemangordonfrom 1415:44
freemangordonuvos: do you want me to do a PR for mce quirks fix?15:45
Wizzupfreemangordon: ok15:47
Wizzupfreemangordon: what does it fix, jw?15:47
freemangordonwell, it doesn't fix anything, but reporting 14 is wrong15:48
freemangordonblob report 8, so shall we do15:48
freemangordonwe'll have to implement FormatsQuery anyways15:49
freemangordonfor WL to be happy15:49
freemangordon(and tmlind to be happy :) )15:49
Wizzupok, mesa is building, prepped clutter for build15:53
freemangordonundim is also much faster now15:57
Wizzupnice, what did you change?15:57
lelfreemangordon opened a pull request: https://github.com/maemo-leste/mce/pull/52 (Mapphone quirks performance)15:57
* Wizzup checks15:57
lelfreemangordon review_requested a pull request: https://github.com/maemo-leste/mce/pull/52 (Mapphone quirks performance)15:57
freemangordonuvos: there is at least one more regression - if you slide-open the keyboard, device unlocks, it should re-lock if you close the kbd without touching anything else15:59
Wizzup+1 that'd be nice to have16:01
freemangordonugh, I was about to look at that h-s-m PR16:04
freemangordonhmm, lets have -things moved to stable first16:04
Wizzupfreemangordon: what is the ugh about?16:09
uvoshmm16:09
uvosthat worked untill very recently16:09
uvosill look into it16:09
uvosah i allready know what caused this problem16:10
uvoshmm16:10
uvosits not solvable16:10
uvoseasly16:10
freemangordongreat16:11
uvosso the problem is:16:11
freemangordonWizzup: it was about that I just remember about it16:11
uvoson mce ignores events from devices it conisders "switch" devices16:11
uvosfor this feature16:11
uvosand a couple of other things16:12
uvosin this case this means the slider switch itself dosent count as an aciton16:12
freemangordonuvos: there is a bug in PR, will push a fix in minute16:12
uvosproblem ist16:12
uvosthat the slider switch isent nesscarly on a different input device as everything else16:12
uvosthis is true on n90016:12
uvosand sorta on d4 but not really16:12
uvosparticualy the volume buttons are on the same device16:13
uvosso since i added the handlers for the volume buttons this broke as i had to make mce accept this input device16:13
uvoswe need to replace the per input device behavor changes16:14
uvoswith something that filters on keycodes16:14
freemangordonhmm, I think we already have that16:14
uvossadly no16:14
uvoswe dont16:14
uvosit filters event devices as a whole16:15
uvosbased on the keys they advertise16:15
uvosthis only works if the devices fit into its catigorys16:15
freemangordonyes, but you can have multiple listeners per device, no?16:15
uvosand there are even several devices to handle differently16:15
uvosno16:15
uvosand the listerners dont filter gennerally16:15
freemangordonhmm, I remember I implemented that correctly back than, but lemme check16:15
uvosthey often just do something whenever any key gets pressed on some device category16:15
uvossutch as cancle the autorelock16:16
uvosbut theres lots of sutch handlers16:16
freemangordonuvos: you can call mce_match_event_file_by_caps() as much as you want16:16
uvosright16:16
uvosbut that dosent help16:16
uvossince it expects there to be different event files16:17
uvosfor different caps16:17
uvosthis is a bad solution16:17
freemangordonnot really16:17
uvosyes really16:17
uvosyou cant make it work16:17
freemangordonyou may have more than one file with same caps16:17
freemangordonsame/matching16:17
uvosthe amount of files is detiermined by the kernel16:17
uvosand matching a device to multiple caps16:18
uvosis no use16:18
uvosas the handlers expect a cap list to contain only buttons they expect and no others16:18
uvosthe solution is to just have a single event que with all event devices bound to16:19
uvosand then have something that filters eatch event centraly16:19
uvoslike xcb works16:19
uvosalso libinput16:19
uvoswe could use libinput acctually16:19
uvosthat would be best16:19
freemangordonbut you can attach different handler for every fd returned16:19
uvosbut that dosent help16:20
uvosso there are categorys16:20
uvoslike switches16:20
freemangordonand filter in the handler itself16:20
uvosthose get ignored for timeout16:20
uvosand for autoreloc etc16:20
freemangordonhmm16:20
uvosall of them16:20
uvosso if a device matches that cagiory it gets ginored16:20
freemangordonI see16:20
freemangordonanyway, we'll have to fix that16:20
uvosthat makes sense if its an input device with just the slider switch or something16:20
uvosbut if its an input device with the silder switch and a keyboard16:21
uvosit breaks16:21
uvosthere is nothing you can do about that16:21
uvosexcept replace how this works entriely16:21
tmlindfor the issues with reboots, maybe all those watchdog features could be enabled by a package that can also be uninstalled if desired16:21
freemangordonuvos: ok, will have a look eventually16:22
freemangordon"nothing you can do" is a bit of an exaggregation :)16:22
freemangordonlemme fix the current PR first16:22
uvoswell you have to rewirte how the filtering works16:22
freemangordonwill do, if it is needed16:23
lelfreemangordon synchronize a pull request: https://github.com/maemo-leste/mce/pull/52 (Mapphone quirks performance)16:33
freemangordonuvos: please have a look ^^^ when you have time16:33
lelIMbackK closed a pull request: https://github.com/maemo-leste/mce/pull/52 (Mapphone quirks performance)17:02
uvosfreemangordon: ill test and make a release later17:02
uvosbut looks good17:02
uvosthanks17:02
humpelstilzchen[Regarding Pinephone Keyboard: Currently faq https://xnux.eu/pinephone-keyboard/faq.html recommends userspace driver instead of kernel driver because kernel driver blocks access from userspace utilities.17:15
uvosa userspace driver for a keyboard sounds lika a travesty imo17:15
humpelstilzchen[lol, but for the time being easier to work with17:16
humpelstilzchen[it seems to be a good idea to read some additional values from the keyboard17:17
humpelstilzchen[so either I switch to userspace on mine or develop a kernel driver for reading these values17:17
uvoswhat values are these?17:17
humpelstilzchen[e.g. voltage and current - the keyboard contains a battery.17:18
uvosshould be pretty easy to add a power_supply driver for it17:18
humpelstilzchen[Also there seems to be the option to enable/disable charging from this battery17:18
uvosso it just needs a ps framework driver ;)17:18
Wizzuphumpelstilzchen[: yeah we'd rathre stick to kernel imho17:19
Wizzupuhoh so soon we have to show capacity for two batteries? :)17:20
humpelstilzchen[afaik there is no capacity information17:20
Wizzupwell if you can get voltage you can estimate it17:20
uvos+ current17:21
Wizzupright17:21
uvosoh btw17:21
humpelstilzchen[more or less. Also the datasheet for the battery has no discharge curve17:21
uvosi mesured the voltage drop on mapphones17:21
uvosthe battery current dependent one17:21
uvosits 153 mOhm17:22
uvos(very large)17:23
uvosso we can use that to improve estimation17:23
uvosbut we need something that can tell us aboutthis per device17:23
uvos(really kernel should be able to, but it cant)17:23
uvosit also cahnges with capcaity a bit17:24
uvosthis is at 40% charge17:24
Wizzuprafael2k: ok, going to look at ofono today17:45
Wizzuprafael2k: so we need these patches? https://gitlab.com/postmarketOS/pmaports/-/tree/master/temp/ofono17:46
Wizzuprafael2k: anything else?17:48
freemangordonuvos: ok, thanks (mce)17:54
Wizzuptmlind: I've rebased the ofono code we have on ofono 1.34, and will also add the pinephone patches now (assuming it all works well)18:11
Wizzupuvos: ^18:11
uvosWizzup: great18:13
uvosfigureing out whats wrong with the motorola modem ofono driver would be pretty great also18:13
uvoswith sgx mostly out of the way this at least is what furstrates me the most when using the device18:14
WizzupI don't have time do to that atm, but yes18:19
WizzupI agree 100%18:19
Wizzupuvos: hmmm looks like it might even solve some problems19:41
Wizzupalthough it maybe uses some of pavels work19:42
Wizzup(modem is called droid_0)19:43
tmlindWizzup: sounds good, if anybody has cycles to work on ofono we should use ell api directly to manage the mdm6600 ports as it's packet based after19:53
sicelowho/what decides modem name in ofono? i get n900_2 on n900, which looks odd sometimes. doesn't matter too much, but i do wonder20:19
Wizzupsicelo: I think in this case other driver is being used tbh20:29
Wizzupwill investigate a bit later20:30
uvosif its using palis hacked at driver for the d4 we really dont want that20:32
uvosas it uses the at interface20:32
Wizzuppavel not pali20:32
Wizzupand yes I'll investigate in a bit20:32
uvosright20:33
humpelstilzchen[modrana seems to be broken20:45
Wizzupwhat do you see? any this is om the pinephone?20:52
humpelstilzchen[yes, on pine. It does not download any map.20:53
humpelstilzchen[http://paste.debian.net/1227958/20:53
humpelstilzchen[strange, it works after deleting all files modrana in /home/user21:02
humpelstilzchen[this should have been the first time I have started modrana..21:02
Wizzuphumpelstilzchen[: hm, weird. good to know for sure21:12
huckgI just tried installing modrana on my bionic and it downloaded a map without any tweaking.21:13
WizzupI will definitely say I only gave it light testing :)21:15
Wizzupuvos: so you think this should be reverted? https://git.kernel.org/pub/scm/network/ofono/ofono.git/log/?qt=grep&q=droid+421:16
huckgHowever, the Vim app in the Debian "folder" does not load for me.21:19
Wizzupit's possible it doesn't specify that it wants a terminal21:20
Wizzuphm, it works for me21:20
huckgIs there a terminal command to launch that app?21:21
Wizzupare you on -devel or on stable?21:21
huckgstable21:21
WizzupI think there may have been some fixes to the launching that haven't made it all the way to stable yet21:22
Wizzupusually parazyd did these things but he's been preoccupied21:22
huckgok, good to know that it has been fixed.21:23
Wizzupyeah there is a diff in maemo-launcher it seems21:23
uvosWizzup: yeah probubly21:24
uvosWizzup: but it would be good to talk with Pavel21:24
Wizzupsure21:24
uvoshuckg: well your in luck vim is impossible to use on bionc anyhow :P21:24
Wizzupprimary goal is to get things working on pp with same ofono21:24
WizzupI figured I would also just rebase (had to fix some api changes)21:24
uvosright21:25
uvosbut that driver really needs to go as the at commands are quite broken21:25
Wizzupyeah21:25
Wizzupfor the next two weeks I'm really focussed on tp and the looming deadline :)21:25
Wizzupjust wanted to get the pine ofono stuff fixed21:25
uvosright21:25
uvosso just revert for now21:26
Wizzuptrying so now21:26
uvosso that it uses tmlinds qmi based driver21:26
uvosagain21:26
uvosnot that that isent broken :P21:26
Wizzupit works decently but yeah there are some problems21:27
Wizzupuvos: btw I pushed some work here https://github.com/maemo-leste-upstream-forks/ofono-d4/tree/maemo-ofono21:38
Wizzupstill toying with it21:38
Wizzupseems to work at least for pin and online21:43
Wizzup3g too21:45
Wizzupbuilding for experimental21:49
Wizzuprafael2k: there is an ofono in beowulf-experimental if you want to try it on the pp22:06

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