| abr | the browser handles video very differently. it's the only place where colour format conversion is needed. | 05:48 |
|---|---|---|
| T42 | <b100dian> mal: prettey sure I remember browser needed https://github.com/sailfishos/droidmedia/pull/111 | 08:29 |
| T42 | <b100dian> https://piggz.co.uk/sailfishos-porters-archive/index.php?log=2024-12-23.txt#line30 | 08:31 |
| T42 | <nc1x72> Trying to setup my build environment after switching to a new cloud compute provider and now getting this error : | 09:13 |
| T42 | <nc1x72> ubuntu@b9209909-1295-4578-9ca2-d985810eb478-0:~$ sudo chroot $UBUNTU_CHROOT /bin/bash -c "chage -M 999999 $(id -nu 1000)" | 09:14 |
| T42 | <nc1x72> chage: user 'ubuntu' does not exist in /etc/passwd | 09:14 |
| T42 | <nc1x72> nevermind | 09:26 |
| mal | @b100dian ok, I need to recheck that | 10:12 |
| T42 | <b100dian> Here's my current status matrix to share my confusion | 10:20 |
| T42 | <b100dian> sake (A11): browser needs GECKO_CAMERA_DROID_FORCE_MEDIA_BUFFER + PR111, gst-droid works only with video/avc=0 | 10:20 |
| T42 | <b100dian> nagara(A13): browser needs GECKO_CAMERA_DROID_FORCE_MEDIA_BUFFER, gst-droid works with video/avc=no-media-buffers | 10:20 |
| T42 | <b100dian> fp5 (A14): browser needs GECKO_CAMERA_DROID_FORCE_MEDIA_BUFFER + PR111, gst-droid works with video/avc=no-media-buffers | 10:20 |
| mal | the no-media-buffers doesn't fix sake? | 10:24 |
| mal | I don't have GECKO_CAMERA_DROID_FORCE_MEDIA_BUFFER on fp5 for browser, not sure about PR111 | 10:24 |
| T42 | <b100dian> I inferred those are needed for fp5 from our previous discussion linked above | 10:49 |
| T42 | <b100dian> Yes. sake is weird | 10:51 |
| T42 | <Mister_Magister> sake's dead though | 10:52 |
| T42 | <Mister_Magister> unless you find exploit to unlock bootloader or asus goes through enlightement | 10:53 |
| T42 | <Mister_Magister> by the way <imagine some insanely offensive word> asus | 10:53 |
| T42 | <b100dian> It was my daily driver until a couple of weeks ago | 10:55 |
| T42 | <Mister_Magister> like i know it's sweet device but f asus | 10:57 |
| T42 | <Mister_Magister> miami is the goat now | 10:57 |
| T42 | <adampigg> dont ever say goat again š (re @Mister_Magister: miami is the goat no...) | 11:01 |
| T42 | <Mister_Magister> why not | 11:01 |
| T42 | <adampigg> because currently, my youtube adverts are for a pair of shorts that say that, and it annoys the fsck out of me š | 11:02 |
| T42 | <Mister_Magister> nah bro goat is goated word | 11:02 |
| T42 | <Mister_Magister> i'm not saying 6 7 or anything offensive like that | 11:02 |
| T42 | <abranson> is it the skibidi goat? | 11:03 |
| T42 | <Mister_Magister> no nothing like that | 11:03 |
| T42 | <abranson> phew | 11:04 |
| T42 | <Mister_Magister> goat just means greatest of all time | 11:04 |
| T42 | <Mister_Magister> so you could say mal is the goat | 11:04 |
| T42 | <abranson> Ī£ | 11:04 |
| T42 | <Mister_Magister> not the sigma ;-; | 11:04 |
| T42 | <adampigg> š | 11:04 |
| mal | @b100dian small style comment to your gst-droid PR, also needs rebasing | 18:41 |
| T42 | <b100dian> looking | 18:50 |
| T42 | <b100dian> rebase again? happy abr moves so fast:) | 18:51 |
| T42 | <b100dian> mal: about identation `if (gst_caps_features_contains` breaking on the next line is the problem? (I tried to keep it untouched but indented) | 18:51 |
| mal | @b100dian no, the next two lines, the content of the if and the closing } | 19:13 |
| mal | did you understand the issue? | 19:29 |
| T42 | <b100dian> ah, now I cannot unsee it. yes | 19:30 |
| T42 | <b100dian> updated. | 19:30 |
| T42 | <b100dian> So I think the gst-droid was too new for my sake that's why it blocked. Need an updated droidmedia for latest abr changes, no? | 19:31 |
| mal | droidmedia changes are not related to gst-droid changes | 19:32 |
| mal | @b100dian what issue do you have on sake with new gst-droid? | 19:33 |
| T42 | <b100dian> video not starting playing at all. I also had this issue with my mediabuffer branch on nagara, after rebasing it. | 19:34 |
| T42 | <b100dian> That's why I thought something in the latest gst-droid commits | 19:34 |
| T42 | <b100dian> let's forget sake, it's too complicated there since I have a half camera2 setup (just photos). But nagara with the latest gst-droid + no-media-buffers doesn't actually play videos, while it does with no-media-buffers applied over 0.20251025.0 | 19:36 |
| mal | I need to test on some fairphone that gst-droid to see that video playback works with those, need to check if the adaptive playback fixes need the other parts also | 19:39 |
| T42 | <b100dian> hmm playbin3 | 19:46 |
| mal | @b100dian the changes in couple of repos fix playback of adaptive videos in for example reddit app | 19:49 |
| T42 | <b100dian> I know, I'm a big fan of Andrew taking over that app | 19:50 |
| T42 | <b100dian> just trying to figure out if that's what I need (newer qtmultimedia) | 19:50 |
| T42 | <abranson> hang on i'll update the build of it on the community obs | 19:51 |
| T42 | <abranson> i had to add another package | 19:51 |
| T42 | <abranson> https://build.sailfishos.org/project/show/home:abranson:adaptivevideo | 19:54 |
| T42 | <abranson> that's all in devel, but should be fine to install on earlier | 19:54 |
| T42 | <b100dian> thanks! give me 10 mins, I have my device in testing for some reason and I wanted to fix that anyway | 19:55 |
| T42 | <abranson> it's not rebuilt yet anyway | 19:56 |
| T42 | <abranson> but if you get that on, and turn on adaptive in the quickddit settings, all the v.redd.it videos play pretty well, with sound! | 19:57 |
| T42 | <adampigg> mal: @Mister_Magister : did HBM ever get resolved? | 19:59 |
| T42 | <Mister_Magister> no | 19:59 |
| T42 | <Mister_Magister> but apparently they want HBM for jolla phone 2 so it maybe might get fixed | 20:00 |
| Mister_Magister | mal: the suspend/usb shenanigans you might remember we were debugging, it seems suspend is being blocked by usb because usb doesn't go into suspend | 20:26 |
| T42 | <b100dian> mal and abr: the latest gst-droid starts playing back after adding latest qtmultimedia š | 20:42 |
| T42 | <b100dian> obs is probably clogged with chum though ;P | 20:43 |
| T42 | <b100dian> OT: I have hints showing about camera usage each time I open it (and probably others). Is there some dconf / path not writable? Why don't they go away | 20:53 |
| T42 | <b100dian> `dconf write /sailfish/camera/camera_mode_hint_count 0` doesnt fix it (key was not present) | 21:05 |
| Mister_Magister | I get ptsd flashback whenever i see the nick abr | 21:05 |
| Mister_Magister | stupid google | 21:06 |
| Mister_Magister | (youtube implemented sabr which is server abr which is server adaptive bit rate which ruined my app) | 21:07 |
| Mister_Magister | which dummy is building zig for sfos | 21:10 |
| Mister_Magister | nephros u dummy | 21:10 |
| T42 | <b100dian> so abr broke microtube ;p | 21:13 |
| T42 | <b100dian> isnt' just the whole chum 5.1 building? | 21:13 |
| T42 | <abranson> The new one stops the zoomed frame when the stream switches. | 21:13 |
| T42 | <abranson> nemo-qtmultimedia-plugins | 21:14 |
| T42 | <abranson> Make sure you get that one too | 21:14 |
| Mister_Magister | give me back my workers | 21:14 |
| Mister_Magister | I want to hog the obs all to myself | 21:14 |
| T42 | <b100dian> abr - yes, thanks - I will, but was debugging why too-new gst-droid was not working | 21:14 |
| T42 | <b100dian> so: qtmultimedia update needed :check: | 21:15 |
| Mister_Magister | since when do we have 5.1? | 21:15 |
| T42 | <b100dian> I need to catch up with forums, but repos are there (?) | 21:15 |
| Mister_Magister | ughhhhhh i want my workers | 21:16 |
| Mister_Magister | there's like 300 scheduled packages before mine gets in it will take ages ;-; | 21:18 |
| Mister_Magister | instead of fifo we need to implement magisterifo :D | 21:21 |
| mal | btw, community port releases should not be done before actual release in my opinion | 21:25 |
| Mister_Magister | ye ye we know :) | 21:26 |
| Mister_Magister | so something is very broken with the usb, it won't go to sleep, it won't detect state, nothing, nada, zilch | 21:33 |
| Mister_Magister | mal: i fixed usb suspend by disabling usb-moded | 21:36 |
| Mister_Magister | so usb-moded writes something to the usb controller that it does not like | 21:37 |
| Mister_Magister | not at all | 21:37 |
| Mister_Magister | but it still doesn't change state | 21:38 |
| Mister_Magister | it's sleeping like a baby now | 21:40 |
| Mister_Magister | so I'll have to (i really really really really really hope mal helps) debug what is that usb-moded writes to usb that usb doesn't like, but first i need to figure out why usb blocks listing of the wakeup_sources and why doesn't the state change | 21:43 |
| mal | what do you mean state doesn't change? | 21:47 |
| Mister_Magister | it doesn't detect that usb is connected | 21:47 |
| mal | and blocking wakeup_sources? | 21:47 |
| Mister_Magister | reminder, debugfs is not enabled | 21:47 |
| Mister_Magister | reminder, you couldn't cat it anyway | 21:47 |
| Mister_Magister | and it's not wakeup_sources fault | 21:47 |
| mal | which detection do you use? android_usb or extcon? which one shows correct state? | 21:47 |
| Mister_Magister | neither | 21:48 |
| mal | ? | 21:48 |
| Mister_Magister | neither of them changes state | 21:48 |
| T42 | <Mister_Magister> not android_usb nor extcon | 21:50 |
| T42 | <Mister_Magister> and usb-moded writes something to usb that stops it from suspending, wakeup_sources has nothing to do with this | 21:50 |
| T42 | <Mister_Magister> the usb device fails to suspend if usb-moded is enabled | 21:50 |
| T42 | <Mister_Magister> i also had to echo disabled > /sys/devices/platform/soc/4e00000.ssusb/power/wakeup in order to unblock autosleep thread | 21:52 |
| Mister_Magister | as a reminder, autosleep worker function is blocked, cat'ing wakeup_count hangs, and if debugfs was enabled cat'ing wakeup_sources would also hand all due to shared piece of code that lists wakeup sources and never finishes. It loops forever | 21:54 |
| Mister_Magister | but the above echo somehow unblocks that loop | 21:54 |
| Mister_Magister | so 2 things need to be established, how to make usb-moded NOT break the usb and how to fix usb detection | 21:55 |
| mal | can you read it if you keep usb connected all the time after boot, by making sure it never tries to suspend | 21:56 |
| Mister_Magister | no | 21:56 |
| Mister_Magister | also if usb is connected at boot the android_usb will be locked to connected state | 21:57 |
| mal | and plugging usb and disconnecting doesn't help with suspending etc? | 21:57 |
| mal | I remember some device had issue that usb needed to be connected once for suspend to work | 21:57 |
| Mister_Magister | no | 21:57 |
| Mister_Magister | it's my and your devices | 21:58 |
| Mister_Magister | pipa and some of your devices | 21:58 |
| Mister_Magister | usb detection is completely broken | 21:58 |
| mal | so suspend works if usb-moded is disabled? | 21:59 |
| Mister_Magister | and if i do the echo disable yes | 21:59 |
| Mister_Magister | and cat /sys/power/wakeup_count finishes currently when device goes to suspend | 21:59 |
| Mister_Magister | it doesn't ever finish otherwise | 21:59 |
| Mister_Magister | if i enable the usb-moded after all the wakeup-sources are exhausted i get error that usb failed to suspend because it's not in LPM | 22:00 |
| Mister_Magister | and since usb can't suspend the whole suspend is cancelled | 22:01 |
| Mister_Magister | but that does not happen if usb-moded is masked | 22:01 |
| Mister_Magister | I maybe will drop all the weird usb stuff i copied from piggz and see if that fixes it | 22:04 |
| Mister_Magister | as to usb detection i got some android guy maybe he can help | 22:04 |
| mal | what did you copy? | 22:07 |
| mal | where is the droid-config repo? | 22:08 |
| Mister_Magister | well it's not in droid-config repo so that won't help you but here you go https://github.com/VerdandiTeam/droid-config-miami | 22:08 |
| mal | but what did you copy related to usb and from what device+ | 22:10 |
| mal | what is this supposed to be https://github.com/VerdandiTeam/droid-config-miami/blob/master/sparse/etc/gstdroidcodec.conf | 22:10 |
| mal | that is not where that should be | 22:10 |
| Mister_Magister | it's supposed to be a typo :) | 22:10 |
| Mister_Magister | https://github.com/HelloVolla/droid-config-halium-mimir/blob/master/sparse/usr/bin/droid/setup-configfs.sh https://github.com/HelloVolla/droid-config-halium-mimir/tree/master/sparse/etc/usb-moded | 22:11 |
| mal | why would you need some configfs setup script? | 22:11 |
| mal | I assume you don't have halium based build? | 22:12 |
| Mister_Magister | I don't know i'm doing as I'm told but the idea is to use ncm | 22:12 |
| Mister_Magister | i don't have halium based build yes | 22:13 |
| mal | it doesn't have rndis? | 22:13 |
| Mister_Magister | rndis is deprecated so the idea is to use something that isn't | 22:13 |
| mal | even in that case such script should not be needed | 22:15 |
| Mister_Magister | well i'm dumbass and piggz linked said script so blame him | 22:15 |
| mal | but same behavior with that script? | 22:16 |
| mal | I mean without it | 22:16 |
| Mister_Magister | I currently don't have it running yes | 22:16 |
| Mister_Magister | and yes same behaviour | 22:16 |
| Mister_Magister | In fact i don't think I have any of those changes now that i'm looking at it so this entire exercie was pretty much pointless :D | 22:17 |
| Mister_Magister | sorry | 22:17 |
| mal | check some things, you have only one udc? | 22:18 |
| Mister_Magister | udc and dummy iirc from last time | 22:18 |
| mal | just in case force udc in usb-moded config | 22:18 |
| Mister_Magister | uhhh, how? | 22:19 |
| mal | https://github.com/mlehtima/droid-config-fp5/blob/master/sparse/etc/usb-moded/usb-moded-fp5.ini#L5 | 22:19 |
| Mister_Magister | is that enough? https://paste.opensuse.org/pastes/5474c8ebc59b | 22:21 |
| Mister_Magister | and yes that's my device | 22:21 |
| mal | assuming you use suitable gadget_conf_directory, b.1 is the default | 22:22 |
| Mister_Magister | i have c.1 so i'll add that line | 22:23 |
| Mister_Magister | https://paste.opensuse.org/pastes/f186996a303f | 22:23 |
| Mister_Magister | all good? | 22:24 |
| mal | see what happens | 22:24 |
| Mister_Magister | mmm it panicked the kernel :) | 22:24 |
| Mister_Magister | so now it's rebooting | 22:24 |
| Mister_Magister | mal: nope, it breaks the usb | 22:26 |
| mal | hmm | 22:26 |
| mal | something then wrong in the config | 22:26 |
| Mister_Magister | well everything beside what i linked above is standard | 22:27 |
| mal | does it use usb or rndis interface name? | 22:29 |
| mal | on device side | 22:29 |
| Mister_Magister | in ip a it's usb0 if that's what you're asking | 22:29 |
| mal | on device? | 22:30 |
| Mister_Magister | yes | 22:30 |
| Mister_Magister | we live in 2026 who needs usb anyway | 22:31 |
| Mister_Magister | wifi and qi charging all the way | 22:31 |
| mal | which sfos version do you build? | 22:32 |
| Mister_Magister | semi-latest | 22:32 |
| T42 | <Mister_Magister> 5.0.0.67 | 22:32 |
| mal | too old for what I want you to use | 22:33 |
| Mister_Magister | xdd | 22:33 |
| mal | update to .76 | 22:33 |
| mal | or .77 | 22:33 |
| Mister_Magister | well i could build newer if someone didn't hog all the obs to themselves | 22:33 |
| mal | why is obs the problem? | 22:33 |
| Mister_Magister | im just joking | 22:33 |
| mal | https://github.com/mer-hybris/s19mps-droid-configs/blob/master/sparse/etc/connman/main.conf.d/10-s19mps.conf excluding the last line, needs at least 5.0.0.73 | 22:34 |
| mal | it's possible to do the same with older version but it's a bit more complicated | 22:35 |
| Mister_Magister | don't you worry | 22:36 |
| mal | https://github.com/mlehtima/droid-config-fp5/blob/master/sparse/etc/connman/main-custom.conf and https://github.com/mlehtima/droid-config-fp5/blob/master/sparse/var/lib/environment/connman/override-main-config.conf | 22:36 |
| mal | those work with older releases | 22:36 |
| Mister_Magister | i'm updating i'm updating | 22:36 |
| Mister_Magister | I wonder if there's any sfos device with newer kernel than 5.4 | 22:39 |
| Mister_Magister | like sfos port | 22:39 |
| Mister_Magister | added the one file | 22:40 |
| mal | not yet released fp6 has 6.1, jolla phone has 6.12 | 22:41 |
| Mister_Magister | no issues with usb detection on those? | 22:41 |
| mal | nope | 22:41 |
| Mister_Magister | do you mind dropping their SOC'sā¦ well i guess you can't for jolla phone and i can google for fp6 so nvm | 22:42 |
| Mister_Magister | mal: okay what now | 22:43 |
| mal | jolla phone has mediatek dimensity 7100 | 22:44 |
| mal | so how does it behave now? | 22:44 |
| Mister_Magister | the same | 22:45 |
| mal | no usb at all working? | 22:45 |
| Mister_Magister | nope, usb detection is broken so usb-moded wouldn't do anything anyway no? and usb device fails to suspend yes | 22:46 |
| mal | yeah | 22:46 |
| mal | which platform is that? | 22:46 |
| mal | which kernel version? | 22:47 |
| Mister_Magister | miami, edge 30 neo, Qualcomm SM6375 Snapdragon 695 5G 5.4 | 22:47 |
| mal | hmm, wait | 22:47 |
| Mister_Magister | someone at qualcomm got REALLY HIGH when changing the naming scheme | 22:47 |
| Mister_Magister | some serious drugs were in works | 22:47 |
| mal | a14 base? | 22:47 |
| Mister_Magister | 15 | 22:48 |
| Mister_Magister | latest and greatest unless you finished a16 adaptation | 22:48 |
| mal | on fp5 this was needed https://github.com/mlehtima/android_kernel_fairphone_qcm6490/commit/873c6024bd45d637e8aa7768d33b9964066ea020 | 22:52 |
| Mister_Magister | also for context, usb device fails to suspend because it gets to here https://github.com/VerdandiTeam/android_kernel_motorola_sm6375/blob/lineage-22.2/drivers/usb/dwc3/dwc3-msm.c#L5989, I don't see anything setting in_lpm to true, only the dwc3_msm_suspend that's below the if. After i set the if two lines above to false so that it never gets there, it does suspend, but then usb-moded breaks it, but comment would suggest that | 22:52 |
| Mister_Magister | some runtime power management sets the in_lpm to true | 22:52 |
| mal | are the android_usb and extcon both saying disconnected all the time? | 22:53 |
| Mister_Magister | yes | 22:53 |
| mal | does your device use that qti_battery_charger? | 22:54 |
| Mister_Magister | pardon my wording but "hell if I know" | 22:54 |
| mal | check kernel config and/or sysfs, power_supply stuff for example | 22:55 |
| Mister_Magister | CONFIG_QTI_BATTERY_CHARGER is not present in config.gz | 22:56 |
| Mister_Magister | So no, I don't think so | 22:57 |
| Mister_Magister | there's CONFIG_MMI_CHARGER | 22:57 |
| Mister_Magister | but charging state detection works | 22:58 |
| mal | ok | 22:58 |
| mal | and you are sure all needed modules are loaded? | 22:58 |
| Mister_Magister | yes I am sure | 22:59 |
| mal | how many extcon devices does it have? | 23:00 |
| Mister_Magister | 3 | 23:00 |
| Mister_Magister | 0, 2, 4 | 23:00 |
| Mister_Magister | huuh wait | 23:03 |
| Mister_Magister | it magically started suspending | 23:04 |
| Mister_Magister | with usb-moded running | 23:04 |
| Mister_Magister | wait i got theory | 23:04 |
| mal | I wonder if writing empty to udc helps? | 23:07 |
| mal | usb-moded does that when disconnecting usb | 23:08 |
| mal | I think | 23:08 |
| Mister_Magister | okay mal with these changes, after fresh boot, and echo disabled > /sys/devices/platform/soc/4e00000.ssusb/power/wakeup, the suspend doesn't work, but it works if I plug and unplug usb (despite usb connection never actually being detected) | 23:13 |
| mal | Mister_Magister: so all of the extcon device state files just show USB=0 or something | 23:13 |
| Mister_Magister | yes | 23:14 |
| mal | Mister_Magister: so the same plug once issue | 23:14 |
| Mister_Magister | so with these changes and echo, we're in same position as pipa yes | 23:14 |
| Mister_Magister | but to get to this i had to modify kernel and disable usb wakeup | 23:14 |
| mal | you still need that echo? | 23:14 |
| Mister_Magister | yes | 23:14 |
| mal | modify kernel how? | 23:14 |
| Mister_Magister | i described it above, look at the "also for context" message | 23:14 |
| Mister_Magister | i can repaste it if you wish | 23:15 |
| mal | and you are sure it's needed? and it wasn't some issue because usb-moded was broken? | 23:15 |
| mal | since something seems to have changed now with usb moded configured | 23:15 |
| Mister_Magister | welel I'll do some testing and answer that | 23:15 |
| mal | yeah, try again without kernel changes and echo, just to be sure | 23:16 |
| Mister_Magister | one by one, first i'll try without echo | 23:16 |
| Mister_Magister | if you change 2 variables at the time you won't know which one caused it :D | 23:16 |
| Mister_Magister | echo is very much needed yes because the autosleep function won't ever get unlocked | 23:19 |
| Mister_Magister | i'll now try without kernel change | 23:19 |
| Mister_Magister | but with echo | 23:19 |
| Mister_Magister | mal: and it doesn't suspend either, as I mentioned "msm-dwc3 4e00000.ssusb: Abort PM suspend!! (USB is outside LPM)" | 23:24 |
| mal | where is your kernel source | 23:24 |
| Mister_Magister | I linked it multiple times | 23:24 |
| Mister_Magister | even in the message with context | 23:24 |
| Mister_Magister | [00:52] <Mister_Magister> also for context, usb device fails to suspend because it gets to here https://github.com/VerdandiTeam/android_kernel_motorola_sm6375/blob/lineage-22.2/drivers/usb/dwc3/dwc3-msm.c#L5989, I don't see anything setting in_lpm to true, only the dwc3_msm_suspend that's below the if. After i set the if two lines above to false so that it never gets there, it does suspend, but then usb-moded breaks it, but | 23:25 |
| Mister_Magister | comment would suggest that | 23:25 |
| Mister_Magister | some runtime power management sets the in_lpm to true | 23:25 |
| Mister_Magister | you even have exact line of that message :) | 23:26 |
| mal | can you check usb mode? something like /sys/class/udc/*/device/../mode | 23:30 |
| Mister_Magister | i think it was like in normal or smth | 23:30 |
| mal | echo peripheral to it | 23:31 |
| mal | and see if anything changes | 23:31 |
| Mister_Magister | hmm I can't find mode | 23:31 |
| Mister_Magister | /sys/devices/platform/soc/4e00000.ssusb/mode here it is | 23:32 |
| Mister_Magister | it is peripheral | 23:32 |
| mal | ok | 23:32 |
| Mister_Magister | at least now device is usable, I never ever use usb, other than for charging (which works, and there's also qi charging) so technically i could finally use this device :d | 23:33 |
| Mister_Magister | :D* | 23:33 |
| Mister_Magister | btw don't let me hold you if you want you can go ahead to sleep | 23:34 |
| mal | the usb connect disconnect is still needed? | 23:35 |
| Mister_Magister | well if we want usb network/mtp yes | 23:35 |
| Mister_Magister | or like usb OTG | 23:36 |
| mal | I meant for it to suspend | 23:36 |
| Mister_Magister | yeye | 23:36 |
| Mister_Magister | connect/disconnect or disabling usb-moded | 23:36 |
| mal | what happens if you echo host to the mode | 23:36 |
| Mister_Magister | entire device froze | 23:38 |
| Mister_Magister | oh and there it goes, kernel panic | 23:38 |
| mal | ok | 23:38 |
| mal | what happens if you write something else | 23:39 |
| mal | you do the writes via wlan? | 23:39 |
| Mister_Magister | ye ssh | 23:40 |
| Mister_Magister | https://paste.opensuse.org/pastes/f0dcbb00a4d5 I follow instructions :))) | 23:40 |
| Mister_Magister | oh OH! IT suspends without echo now! | 23:41 |
| Mister_Magister | shall it try without kernel change? | 23:42 |
| Mister_Magister | and it suspends way more cleanly now | 23:42 |
| Mister_Magister | cleanly as in after 2 fails not 200 | 23:43 |
| mal | did usb detection change? | 23:44 |
| Mister_Magister | did not check went into changing kernel | 23:44 |
| mal | can you check on device what kind of usb related *.rc files you have and check what things those write | 23:47 |
| Mister_Magister | mal MAL IT SUSPENDS WITHOUT CONNECT/DISCONNECT, ECHO OR KERNEL CHANGE | 23:47 |
| Mister_Magister | HYPE | 23:48 |
| Mister_Magister | i did check in past but didn't find anything remotely interesting | 23:48 |
| mal | so the peripheral mode does something fishy | 23:48 |
| mal | check if any of the .rc files write mode | 23:49 |
| Mister_Magister | https://paste.opensuse.org/pastes/27f7c5661db4 | 23:49 |
| Mister_Magister | these are in vendor | 23:49 |
| mal | no writes there to mode | 23:50 |
| Mister_Magister | https://paste.opensuse.org/pastes/6e1e37b306df /system | 23:51 |
| Mister_Magister | neither here | 23:51 |
| Mister_Magister | i literally did find /system -name \*usb\*rc | 23:51 |
| Mister_Magister | same for vendor | 23:51 |
| mal | getprop sys.usb.typec.mode | 23:53 |
| Mister_Magister | empty | 23:53 |
| Mister_Magister | ah you wanted usb detection | 23:54 |
| Mister_Magister | nothing changed | 23:55 |
| Mister_Magister | mal: wait, i connected the device and wrote peripheral and have usb connection and state changed to configured in android0 | 23:56 |
| Mister_Magister | https://paste.opensuse.org/pastes/d62dc3cfe216 | 23:56 |
| mal | does it suspend? | 23:57 |
| Mister_Magister | not at all :) | 23:57 |
| Mister_Magister | well it's to be expected i think mce blocks suspend while charging no? | 23:57 |
| mal | I of course mean disconnecting usb and checking if suspend works now in peripheral mode | 23:58 |
| Mister_Magister | aaaaand nope, it doesn't | 23:58 |
| Mister_Magister | it doesn't even try to suspend so autosleep thread is blocked | 23:59 |
| Mister_Magister | once i wrote none i didn't even manage to type dmesg before it suspended | 23:59 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!