libera/#maemo-leste/ Friday, 2021-06-11

sunshaviparazyd: Yes. It works. I needed to register first00:45
parazyduvos: Patches landed in -devel for mapphone kernel00:58
parazydI'll test tomorrow00:58
parazyd(If the vaccine doesn't make me tired, getting 2nd shot at 2pm)00:58
Wizzupuvos: xyboard and d4 are out for delivery today :)10:10
parazyd:)10:17
parazydMy Pinephone customs bill came too, just about +30 eur10:18
parazydWill pay it today and I should get it next week.10:18
Wizzupsweet10:22
inkyyay10:46
inkycongrats.10:46
inkyi also ordered pp.10:46
inkyforgot my usb condom at friends' house, and for a couple of weeks i am unable to use my d4.10:47
* mighty17[m] posted a file: Makefile.patch (0KiB) < https://libera.ems.host/_matrix/media/r0/download/matrix.org/noCvYOekuFtjgRgOAmIWLszP >11:21
* mighty17[m] posted a file: mms114.patch (2KiB) < https://libera.ems.host/_matrix/media/r0/download/matrix.org/faXKNkmFlUGZWcfdHkSEsOgJ >11:21
* mighty17[m] posted a file: omap2plus_defconfig.patch (3KiB) < https://libera.ems.host/_matrix/media/r0/download/matrix.org/oNosGJtWITrTKMJKusewKmnS >11:21
* mighty17[m] posted a file: omap4-samsung-espresso3g.patch (14KiB) < https://libera.ems.host/_matrix/media/r0/download/matrix.org/vnheWufkyuYSwXQBtFGErEaI >11:21
mighty17[m]uvos: have a look at these11:21
uvosinky: what do you need the condom for?12:03
uvosmighty17[m]: please dont send files like that12:04
uvosmighty17[m]: +# CONFIG_B43_BUSES_SSB is not set12:05
uvosdont add that kind of crap to omap2plus_defconfig12:05
uvosomap4-samsung-espresso3g.patch still contains way to long lines etc please run checkpatch.pl on it12:07
uvosall of the patches are missing headers with $subject lines and long descriptions and a signed-off by you12:07
uvosagain checkpatch will tell you this12:08
uvosplease follow everything up to step 4 here https://www.kernel.org/doc/html/v4.17/process/submitting-patches.html12:09
uvosor rather here https://www.kernel.org/doc/html/v5.12/process/submitting-patches.html12:10
uvos(up to date version)12:10
uvosWizzup: nice :), but likey wont be able to recive it today.12:12
uvosparazyd: i got my 1st shot on wensday, its certly "fun" i feal like im 80 yo suddenly :P12:12
parazyd:D12:13
parazydI got AZ, so doses are like 12 weeks separated12:13
parazydIn any case, tonight might be sleepytime12:13
uvosbtw upgrade went fine on bionic12:17
uvosafter you changed the kexecboot pkg12:17
mighty17[m]<uvos "dont add that kind of crap to om"> okay 😅12:17
uvosmighty17[m]: just add the absolue minimum config flags you really need12:18
mighty17[m]<uvos "mighty17: please dont send files"> so pastebin?12:18
uvosmighty17[m]: yeah or an email is even better12:18
mighty17[m]<uvos "mighty17: just add the absolue m"> i will need brcm for wifi12:20
uvosmighty17[m]: sure thats fine12:20
mighty17[m]and other config is for display and acclerometer12:20
uvossure all of that is fine12:20
uvosjust add the modules with =m12:20
mighty17[m]`User is blocking messages from unregistered users, and you are not registered.` :(12:21
uvosmighty17[m]: yeah you have to resigter with libera12:21
parazyduvos: ok @ update; then I'll promote to main soon12:21
uvosif you can read my message its fine12:21
mighty17[m]<uvos "just add the modules with =m"> i have tried to use as much modules as possible12:21
uvosand stuf like CONFIG_BRCM_TRACING12:21
uvosonly add that if you really really need it for the device12:21
uvoslike CONFIG_BRCMDBG12:22
uvosdrop that pls12:22
mighty17[m]well yes, the dbg arent that necessasy12:22
mighty17[m]necessary*12:22
uvosright12:22
uvosi want the patch to just include what is nessecary for functionality12:22
uvosnothing else12:22
uvosthat means just the modules themselves most likely12:22
mighty17[m]even b43leagcy should be dropped but im kins skeptical about taht12:23
mighty17[m]that*12:23
uvosif it doesent probe on your device or is coverd by another better driver please drop12:23
mighty17[m]<uvos "if it doesent probe on your devi"> device uses brcmfmac driver12:24
uvosthen drop bt4312:24
uvosjust the essentials please, i mean it ;)12:24
mighty17[m]also i needed to ask, the linux-firmware-brcm doesnt seem to have proper driver for bcm4330, so is it fine if i use downstream nvram and driver12:24
uvosby driver you mean firmware i assume12:25
uvosyeah thats fine12:25
mighty17[m]<uvos "just the essentials please, i me"> well there's no other way than i will have to test it myself, removing stuff gradually12:25
uvosmighty17[m]: sure just check what is loaded12:25
Wizzupmaybe look at what's loaded12:25
mighty17[m]<uvos "by driver you mean firmware i as"> yes https://github.com/TheMuppets/proprietary_vendor_samsung/tree/cm-14.1/espresso-common/proprietary/system/etc/wifi12:25
Wizzupmighty17[m]: did you manage to boot leste with 3d on the tablet?12:25
uvosmighty17[m]: thats fine12:25
mighty17[m]<Wizzup "mighty17: did you manage to boot"> nope, not yet12:26
Wizzupok12:26
uvoshaving a leste port with no pvr would be fine at first too12:26
uvosbtw12:26
mighty17[m]only pmos with 3d yet12:26
mighty17[m]uvos: gimme some time, i will try getting bare minimum modules and test it with pmOS12:27
uvosok12:27
mighty17[m]thanks :D12:27
Wizzupuvos: right12:27
mighty17[m]<uvos "omap4-samsung-espresso3g.patch s"> omg so many whitespace error :(12:31
uvosmighty17[m]: yeah12:39
uvosmighty17[m]: good news is that wen your done fixing that i think we can get your dts upstreamed12:39
mighty17[m]yay!!12:39
mighty17[m]but first let it run maemo xD12:40
uvostmlind: btw something esle is also breaking ilde randomly12:44
uvostmlind: somtimes my device just wont enter ret with omapconf showing GPMC is holding it awake12:45
uvosthis presits untill reboot12:45
uvosill try to find the time to diff the register state of this device]12:45
freemangordonyay, "no saved connections..." bug is fixed :). As I suspected, it turned out to be a REing bug.12:53
parazyd\o/12:54
uvosbtw mighty17[m] you need to add the panel-timing to the pannel driver instead of having it in dts12:59
Wizzupfreemangordon: great13:00
uvoslong term anyhow13:01
uvosits fine for now13:01
siceloty! @bug14:14
mighty17[m]<uvos "btw mighty17 you need to add the"> its the simple-lvds panel15:09
mighty17[m]oh right that also needs a patch15:09
freemangordonuvos: Wizzup: please upgrade icd2 and test if there are still issues15:17
freemangordon(the others as well if feeling brave)15:17
freemangordonmake sure you're on latest connui-common and connui-internet15:17
Wizzupfreemangordon: I apt upgraded and only got new icd2, but I suppose I already has new connui15:21
Wizzuphad*15:21
freemangordonyeah15:21
freemangordonthis note was more for "the others" :)15:21
mighty17[m]<uvos "btw mighty17 you need to add the"> https://github.com/MightyM17/linux/commit/a203576cc87cd9628a738e13da6995b8ede90ae8#diff-b0efb865b9590da1c62bb09d5b597a61bcc769888ce286b5c8a9f72d35fc3788L3298 something like this? its unnecessary afaik15:21
freemangordonWizzup: the only remaining issue I am aware of is this endless blinking icon when you try to connect to non-existing hidden AP15:22
freemangordonbut that shall be fixed in icd plugin IIUC15:22
Wizzupok15:24
Wizzupyes15:24
tmlinduvos: gpmc on droid4 keeping device busy?15:29
tmlinduvos: also just found my mz609, i guess i'll try installing bootloader to it now15:30
freemangordonWizzup: hmm, now it seems to always behave correctly15:30
freemangordonoh, wait15:30
freemangordonyeah, my bad, I was entering correct hidden AP SSID so it was rejecting the assoc15:32
freemangordonif you enter wrong one, it never stops blinking neither it gives error message15:32
tmlinduvos: yup i did fastboot flash cache boot.img, booted android and as root wrote utags-mz609-32-mmcblk1p6-boots-mmcblk1p18-kexecboot.bin to /dev/block/mmcblk1p6 with dd and mz609 booted to kexecboot just fine :)15:44
tmlindhmm i guess i need to build and upload proper droid4-kexecboot images at some point now15:46
tmlinduvos: i'll wait with the images until you've confirmed xt875 boots ok and maybe let's remove boot.cfg for xt875 on first boot if you have such a patch15:47
Wizzupfreemangordon: I will try to fix that soon15:53
Wizzupfreemangordon: I also want to make sure the dialog isn't grey when it shouldn't be15:54
WizzupI felt it was sorted before, but with the different scan method it changed maybe?15:54
freemangordonno idea, I have dummy nets here so never gray16:05
freemangordon*grey16:05
uvostmlind: ok ill try and build kexecboot today16:07
uvosi dont have time to make a patch atm16:08
uvostmlind: yes gpmc keeps it awake16:08
parazyduvos: Let's think about how we could ship kexecboot as .deb16:09
parazydFor updates16:09
uvosparazyd: easy just package the image and dd it over on postins16:09
uvosits perfectly safe worst case the user has to fastboot flash it again if we crash during update16:10
parazydCare figuring out the dd params?16:10
uvossure see https://uvos.xyz/maserati/Partitions.txt16:11
uvosits dd if=image.img of=/dev/mmcblk2p13 bs=2M16:11
parazydAwesome16:12
parazydThanks16:12
uvoswe might want to guard it by mounting it first and checking for kexecboot16:12
parazydFor sure16:12
uvosjust in case the kernel changes the mmcblkX again16:12
parazydYep16:12
uvosand we end up flashing to sd or something16:12
uvosi think checking for the partiton lable should be enough16:12
uvosor for kexecboot binary16:13
uvosor both16:13
parazydokay16:14
uvostmlind: https://uvos.xyz/maserati/os_idle_uc_audit_details.txt16:16
uvostmlind: omapconf report16:16
uvostmlind: the diff to a report where its ideling correctly is | GPMC                    | FAIL16:17
uvosin MODULES SYSCONFIG AUDIT16:17
uvosalso nice to see that the amount over all amount errors in omapconf has reduced signifcantly16:18
uvoswith just 4 now on a working system16:18
uvos(was 8 in 5.8)16:18
uvosand the remaining errors seam moslty silly16:20
tmlinduvos: ok interesting that gpmc is not idling16:20
uvoslike it complaining about clocks that are lower than it expects16:20
tmlindheh ok16:20
uvosill keep you updated if it happens again16:22
uvosi have seen the device fail to idle often recently16:22
uvosbut this is the only time i checked omapconf16:22
tmlindweird, i have uptime since -rc2 and no issues idling16:23
uvosmight just be 5.11 i dont know if i saw it with 5.13 yet16:23
tmlindoh only 13 days, anyways a while16:23
uvosi dont think so16:23
mighty17[m]<uvos "mighty17: yeah or an email is ev"> so can i upload there? coz ml doesnt accept attachments16:23
uvosmighty17[m]: your supposed to just append the patch to the email16:24
mighty17[m]sorry for the silly question xD16:24
mighty17[m]but how can i append it (using gmail ig)16:25
tmlinduvos: do you need mz609 utags and boot test images btw?16:25
uvostmlind: sure soon16:26
uvosi can also make my own ofc16:26
uvosmighty17[m]: like this https://marc.info/?l=linux-omap&m=162341806303648&w=216:27
uvosidk if using gmail is a good idea16:27
uvos(its not)16:27
tmlinduvos: ok well probably best you make your own then, that way you can patch away that xt875 default boot.cfg file :p16:27
uvosbut you can just paste the patch after your email message16:27
mighty17[m]alrighty thanks16:27
uvosmighty17[m]: you can use git format-patch to write the email for you btw16:29
mighty17[m]<uvos "mighty17: you can use git format"> oh thats better i'll look into that (ps already sent email)16:31
uvosmighty17[m]: still needs a proper [subject] line and maybe a bit more of a long form description than just Adding support for Samsung Galaxy Tab 2 (7 inch)16:33
uvosshould be something like for instance [PATCH] ARM: dts: <whatever dts file>: add this stuff16:34
uvosalso cc tm16:35
uvostmlind:16:35
uvosalso mighty17[m] you cant use gmail16:36
uvosits breaking the patch by formating it16:37
tmlindmighty17[m]: do you have a tc358765 lcd bridge also on espresso?16:40
tmlindhmm maybe it's there only on tablets16:49
tmlindanyways, need to go pick up my wife, ttyl16:49
uvosWizzup: i did manage to get your pacakge16:52
uvosWizzup: its in bad shape16:52
uvosWizzup: did you forget to add a hw4x battery or did it fall out? (package has a big hole)16:53
uvosotherwiese the xyboard and the d4 seam to have made it in one piece16:53
mighty17[m]<tmlind "mighty17: do you have a tc358765"> hm? the lvds bridge is doestek dtc34lm85am17:29
mighty17[m]<uvos "also mighty17 you cant use gmail"> ohk, git format-patch is the way17:30
mighty17[m]<uvos "mighty17: still needs a proper ["> we arent mainlining it yet are we?17:30
uvosmighty17[m]: i dont see why not maybe add just [RFC PATCH] to signify to the ml that your not intending to mainline rn17:32
mighty17[m]i cant edit it now 😅17:32
uvosmighty17[m]: otherwise if you fix the formating and the other nitpicks i dont see why we should not try and mainline immidatly17:32
uvosalso the dts goes and defconfig changes go to linux-omap17:34
uvosbut the code changes to the input driver go nowhere since they are in kernel allready17:34
uvosjust send that to me privately17:34
mighty17[m]uvos: reason is, i dont think its mainline ready, the pixel clock is half of whats in downstream (basically its wrong) and so is the pwm in backlight node, mmc is bugged (sometimes its mmc0 and 1 then 1 and 2 its random in dmesg), usb is broken, need to send a twl6032 patch as well17:36
mighty17[m]and he left the room wut?17:36
sicelomighty17[m]: we have logs ;-)18:01
Wizzupuvos: hrm18:09
Wizzupuvos: battery for what?18:10
Wizzupuvos: I did not provide a battery with the d4, the tablet should work as-is18:11
uvosWizzup: we had discussed a non extended battery, because i wanted to test the cpcap changes with that18:14
uvoswrt autodetection of the battery type18:14
uvosbut its ok its not so important18:14
uvosthe xyboard is missing the screws in the back, they seam to be M1.318:14
uvosbu i dont know the length, and as i can see the backlight through the holes i dont want to try it18:15
uvoscould you unscrew one from your device (or tmlind) and tell me how long it is so i can get a replacement?18:15
uvosnon extended battery for the bionic btw18:15
uvossomeone really had a go at that d4 you sent me :P18:16
uvosevery other screen is striped or in the wrong place or missing18:16
Wizzupuvos: ok yeah I forgot that @ battery18:16
uvosWizzup: ok :)18:16
uvosWizzup: at least it dident get lost in the post :)18:16
WizzupI definitely didn't open up the tablets, but can see about the screws18:16
uvosits just the screws on the metal back plate18:17
uvosthere all missing18:17
Wizzuphrm @ someone had a go at the d4 that I sent18:17
uvosidk whats holding the plate in18:17
uvosand the somone who really dident know what they are doing tried to solder the usb port18:17
uvosit has all the pins shorted :P18:17
Wizzupright, that's the non-working usb port18:17
uvosanyhow no matter the display is fine so i can use it to repair the other one :)18:17
Wizzupbtw, regaridng damage to the package, that's weird, I think I was quite careful18:17
uvosWizzup: you where18:18
uvosbut it was droped or something18:18
Wizzupaha18:18
uvoslooked like someone droped something heavy on one end really18:18
uvosbut everything survived thanks to the bubbel wrap18:18
Wizzupglad to hear18:24
uvos__old d4 works again :)18:27
uvos__i swaped the upper assembly with the one you sent me and it works18:27
uvos__tmlind: if you ever need me to test something on the unit where wifi only works on mainline with the pullup disabled this device exists again.18:28
mighty17[m]<sicelo "mighty17: we have logs ;-)"> ah right makes sense18:29
uvos__mighty17[m]: ok sounds sane18:30
Wizzupuvos: great18:30
uvos__mighty17[m]: leave the ml out then18:30
mighty17[m]okayy18:33
mighty17[m]so nowbuilding maemo first?18:39
uvos__mighty17[m]: yeah so maemo steps 1. fix the patches 2. send them to me 3. if i think they are ok ill give to parazyd to include in the kernel and we will start building the espresso dtb. then you can do the image builder + hildon meta and leste config stuff18:42
uvos__and also figure out what the bootloader situation is going to be18:42
uvos__then we can start building images for the device18:43
uvos__i would not worry about sgx to mutch if ddk1.17 works on pmos ddk1.9 will very very likely just work with the droid4 packages18:44
uvos__bootloader situation must involve you booting the mainline or vendor kernel to kexecboot pretty mutch18:44
uvos__since thats the only way your device can share a kernel build with our other devices18:45
uvos__since the android bootloader wont be able to let you set cmdline or dtb18:45
uvos__vendor kernel rebuilt with kexec built in booting to kexecboot18:47
uvos__is likely best18:47
mighty17[m]<uvos__ "and also figure out what the boo"> about that for now, can we have another branch with that cmdline? we are not in the stage when we can call it booting so 1st step would be to get stuff working imo18:47
uvos__as this avoid any buggyness in kexecboot that might get in the way of development18:47
mighty17[m]<uvos__ "i would not worry about sgx to m"> yeah ddk isnt a big issue, if it works on d4 it will work here18:47
uvos__mighty17[m]: no i dont think we will include your device will a different branch18:47
uvos__ofc you can go and work on leste like that18:48
uvos__by building your own kernel18:48
uvos__but untill you figure out the bl situation so that you dont have to force the cmdline we cant accept the device18:48
mighty17[m]<uvos__ "mighty17: no i dont think we wil"> noo not like that, for initial testing then we'll figure out a way18:48
mighty17[m]<uvos__ "by building your own kernel "> yeah till it boots we can do that?18:49
uvos__sure but this dosent require "me" to do anything18:49
uvos__just base apply your patches to our kernel then and try boothing the bionic rootfs18:49
uvos__should mostly work18:49
mighty17[m]i mean yes but i have no clue on how to build maemo18:50
uvos__you dont have to build anything really18:50
uvos__just use the bionic rootfs18:50
uvos__and then uninstall the leste-config and hildon meta once its booted18:50
uvos__and then you just have to make and build those 2 packages18:51
uvos__and once you have the bl stuff figured out well add the device to our packages18:51
mighty17[m]flash boot.img (with leste kernel) and rootfs nice18:52
uvos__yeah dont forget to add the needed modules to the rootfs after dding it to the sdcard18:54
uvos__otherwise that should work18:54
mighty17[m]maemo seems to be down18:59
* mighty17[m] uploaded an image: (43KiB) < https://libera.ems.host/_matrix/media/r0/download/matrix.org/asbIbyKJRBZqomnDxrhIGfvu/image.png >18:59
uvos__no its not19:00
uvos__https://maedevu.maemo.org/images/bionic/20210530/19:00
uvos__its your end19:00
uvos__parazyd: btw this weeks image faild to build19:02
uvos__parazyd: because you ran out of space19:02
uvos__parazyd: but it still created a directory https://maedevu.maemo.org/images/bionic/20210606/19:02
Wizzupmighty17[m]: let me know if that problem (accessing maedevu) persists19:09
mighty17[m]uvos__: can i use pmos envkernel to build boot.img or is there something else needed?19:20
uvos__mighty17[m]: should be fine as long as you specify the right cmdline19:23
uvos__pmos dose something wierd with the cmdline and its initramfs19:23
uvos__but i dont really know any pmos internals19:24
parazyduvos__: ack will clean up tomorrow21:05
parazydNeed some rest now. Seems vaccine is doing its thing, albeit much less than first dose21:06
adc_parazyd: get well soon :)21:26
parazydFor sure :) Thanks21:27
uvos__parazyd: yeah get well! At least soon you, and more generally we all, will have this covid thing in the bag :)21:29
parazydYes!21:29
parazydLong fight, but we're getting there21:29
parazydhttps://twitter.com/juanbuis/status/140290804331514265721:30
uvos__Wizzup: tmlind: ok im gona try the "wrong" mbm.bin21:30
uvos__lets hope im not gonna brick my device allmost immidatly :|21:30
* uvos__ is somewhat worried :P21:30
uvos__tmlind: btw any idea what the mbmbackup partiton is supposed to do?21:32
uvos__i gues its automatic? ie mbmloader checks mbm.bin and if its not happy it uses mbmbackup?21:33
uvos__it worked21:34
uvos__allow-mbmloader-flashing-mbm.bin from eu xoom2 works fine on xyboard21:34
uvos__dosent help us any21:39
uvos__that mbm dosent allow utag flashing either21:39
uvos__VRZ_9.8.2OT_127_MZ609_CFC_1FF.xml mbm isent more permissive either :(21:41
* enyc meows23:14

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