04:44 <vknecht[m]> bryanodonoghue: Hello ! pushed my WIP https://github.com/vknecht/msm8916-mainline-linux/commits/msm8939-camss ... still getting the same 1k junk blocks problem with TPG (8916 still fine)

07:50 <White> Hello Everyone, My question is how to do software R&D better?

07:57 <White> For example the ipc_logging feature in msm-5.10, is there any method to understand the whole feature faster?(from the designer view)?

08:01 <White> OK Fine RTFSC

10:13 <HappY[m]> Hello! In mainline I don't have charger support.. and my downstream driver if different.. so if I want to add charger support in mainline.. then how can I solve?

10:20 <z3ntu> HappY: if you're on sm8350 I'd expect this to use pmic-glink for battery & charger?

10:22 <HappY[m]> z3ntu: Pmic-glink never worked for me.. every time all device failed to connect..

10:23 <HappY[m]> And I don't know what is pmic-spmi but always probe failed

10:24 <z3ntu> HappY: pmic-glink has different things it does, one is battery&charger communication, another is DP altmode (connecting external display), one other is UCSI which does the USB role switching. The latter (UCSI) is quite unreliable for me also but the rest works fine

10:29 <HappY[m]> <z3ntu> "konradybcio: you mean some..." <- I have same driver in downstream.. so I was add .. fsa4480 & usb with pmic-glink like this https://github.com/Saikatsaha1996/linux-sm8350/blob/e4d90d13d227ba9cc88e7e56b6039224290117da/arch/arm64/boot/dts/qcom/sm8350-realme-porsche.dts#L258 but always failed to connect any device.. so just disabled..

10:30 <z3ntu> HappY: Maybe it's a good idea to just disable UCSI in the driver for now for your chipset, so you at least get the rest of the pmic-glink goodies

10:30 <HappY[m]> And battery always reporting to 100%

10:34 <HappY[m]> Failed like this

10:36 <HappY[m]> Same with failed to create device link with fsa4480

10:39 <z3ntu> this shouldn't be fatal

10:39 <z3ntu> HappY: What entries do you have in `/sys/class/power_supply/`?

10:40 <HappY[m]> z3ntu: I have to check.. I will update

10:40 <konradybcio> yeah this is not fatal

10:40 <konradybcio> it may be reporting 100% because your battery is fully charged

10:42 <HappY[m]> konradybcio: No not full charge .. my battery originally reported 62% in Android.. and then I flashed pmos to my device.. but in pmos battery reporting to 100%

10:44 <konradybcio> does it have 1 or 2 batteries?

10:45 <HappY[m]> konradybcio: 2 battery with parallel connection

10:46 <HappY[m]> * In reply to @konradybcio:matrix.org

10:46 <HappY[m]> does it have 1 or 2 batteries?

10:46 <HappY[m]> 2 batteries with parallel connection

10:46 <konradybcio> here you go :) perhaps you need a separate fuel gauge to get the real output

10:47 <HappY[m]> But parallel connection.. still I have to use dual fuel-guge ?

10:51 <HappY[m]> <konradybcio> "here you go :) perhaps you..." <- Without pmic-glink battery status reporting perfectly..

10:51 <konradybcio> depends how it's terminated

10:51 <konradybcio> i.e. whether it feeds 1 or 2 fuel guages

10:52 <HappY[m]> konradybcio: In stock dmesg reports single fuel-guge

10:54 <HappY[m]> Without pmic-glink

10:59 <konradybcio> maybe your downstream ignores pmic-glink?

11:10 <HappY[m]> <konradybcio> "maybe your downstream ignores..." <- This is my downstream pmic glink... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/uInfTScfcgEFtRpcTAFZndPj>)

11:49 <mal> minecrell: do you have any opinion about the watchdog slowness (watchdog triggers only after quite a while instead of the set timeout), should I somehow try to debug that or just send the dts patches first and look in the issue later?

11:49 <mal> not sure how common that issue is

11:49 <konradybcio> mal your client sent both messages twice

11:50 <mal> very odd, this is just normal irssi and has never misbehaved before

11:53 <konradybcio> well, it just did it again

11:57 <z3ntu_test> I wouldn't rule out the matrix bridge being weird either

11:57 <z3ntu> hm worked properly for this message 🤷

12:00 <mal> I can restart irssi just in case if this keeps happening

12:00 <konradybcio> please do, it keeps doubling

12:03 <lumag> konradybcio, mal: I don't see duplicates

12:05 <minecrell> konradybcio: well it seems like a Matrix issue to me (the messages are not duplicated on IRC) so maybe you need to fix things :D

12:05 <minecrell[m]> fun it duplicates all IRC messages on the Matrix side

12:09 <mal> does anyone still see duplicate messages from me?

12:10 <minecrell> no

12:10 <minecrell> now only mine are duplicated :p

12:11 <minecrell> test

12:11 <minecrell> strange somehow disconnecting fixes it on the Matrix side

12:20 <z3ntu> the bridge is a mysterious creature

12:25 <mal> minecrell: any opinion to my question about watchdog?

12:26 <minecrell> mal: iirc I tried to debug the watchdog slowness for at least a full day but didn't find any good hints why it happens. Somehow it works normally in LK but gets slow once Linux boots. I suspected something like cpuidle or so but that didn't help iirc

12:26 <minecrell> mal: I don't think you should add it upstream in the broken state

12:26 <minecrell> maybe consider adding the pmic watchdog with pm8916-wdt instead

12:26 <minecrell> I use that on 8909, it works reliably

12:31 <mal> ok

12:57 <z3ntu> Btw with pmic-glink for me upower considers two wrong devices to be batteries too so if the battery is 100% charged, the DisplayDevice is only 33%. The other devices seem to be battery_qcom_battmgr_usb and battery_qcom_battmgr_wls where upower thinks those are batteries too. The battery with correct data is battery_qcom_battmgr_bat

12:58 <z3ntu> Is there something wrong in dts/driver config maybe? Probably if you ask upower people they say to fix the kernel to provide correct data :)

15:51 <HappY[m]> <konradybcio> "maybe your downstream ignores..." <- With pmic-glink

15:52 <HappY[m]> Extra reporting real %

15:54 <konradybcio> maybe your downstream driver ignores battmgr reports?

15:59 <HappY[m]> <konradybcio> "maybe your downstream driver..." <- https://pastebin.com/Zm5Jtss2

16:00 <HappY[m]> Stock battery & charger related

16:19 <HappY[m]> Charger possible 🥹 for my device?

16:24 <HappY[m]> In my stock have batt manager firmware.. `battmgr.jsn` I don't know mey be it required?