2025-08-27.log

gordon1there is no reason why trackball sensor must be on the bottom of the trackball, isn't there?00:07
joschgordon1: the more to the side you put the sensor the worse it will be to detect rotations around the axis perpendicular to that one00:10
joschgordon1: imagine the extreme case of putting the sensor not horizontal but vertically -- then you can no longer detect rotation around the horizontal axis through the sensor00:10
joschyou can fix this by having two sensors00:10
gordon1oh yeah, you're right00:11
gordon1didn't think about that00:11
- AnimaInvicta (QUIT: Ping timeout: 260 seconds) (~AnimaInvi@88-120-179-216.subs.proxad.net)00:47
+ AnimaInvicta (~AnimaInvi@88-120-179-216.subs.proxad.net)00:49
- AnimaInvicta (QUIT: Ping timeout: 245 seconds) (~AnimaInvi@88-120-179-216.subs.proxad.net)01:25
joschminute: i'm doing a bit of archaeology and i found this: https://source.mnt.re/reform/reform-system-image/-/merge_requests/129#note_1620201:28
joschminute: would you like me to remove reform-help being called from /etc/skel/.profile now?01:28
joschit's 3 months later and i don't know if this is still something you want01:29
+ AnimaInvicta (~AnimaInvi@88-120-179-216.subs.proxad.net)01:32
- mjw (QUIT: Ping timeout: 256 seconds) (~mjw@gnu.wildebeest.org)01:38
minutejosch: why do you have a mediatek patch? :3 https://source.mnt.re/reform/reform-debian-packages/-/merge_requests/134/diffs01:42
joschminute: mediatek users have said that they also experience display flicker in some of my reports for a311d flicker and i was poking in the dark to find out which commit fixed the display flicker in the kernel01:43
minutejosch: this won't do it because that driver is not used on our platforms01:44
joschyeah, i figured that out the hard way :)01:44
minutejosch: https://source.mnt.re/reform/reform-debian-packages/-/merge_requests/93/diffs01:44
minutejosch: you have a draft reverting my patch that should have fixed that flickering01:45
joschminute: that patch does fix it01:45
joschbut that's not how the kernel did it01:45
joschthe fix was even applied in the stable 6.12 series01:45
minutejosch: you mean they fixed the flickering in mainline, differently?01:46
minutejosch: for a311d as well?01:46
joschbut drivers/gpu/drm/meson/meson_plane.c did not get changed in a similar way as the revert does between v6.12.9 and v6.12.3801:46
joschyes, i was trying to find out what fixed it but it's not super high on my priority list because now it... just works XD01:46
minutejosch: it works with my patch removed?01:46
joschi was trying to find out what fixed it to leave a comment about it on the relevant bugs01:46
joschminute: yes, the revert patch did not get merged01:47
joschit also works fine in 6.16 without flicker01:47
minutejosch: not sure if i understand correctly. did you build a 6.16 with this one removed and the flicker is gone...? https://source.mnt.re/reform/reform-debian-packages/-/blob/main/linux/patches6.16/meson-g12b-bananapi-cm4-mnt-reform2/0030-WORKAROUND-meson-plane-disable-afbc-32x8.patch01:48
joschtalking about this is different because we talk about the removal of a removal :)01:48
minutejosch: my patch there is not a revert01:49
minutejosch: it removes some lines of code from the kernel yeah01:49
minutebut afaik it's not reverting some kernel commit01:49
joschsorry, i called it a removal because things broke once the lines you removed were added01:50
minuteok. and you have a special kernel build with those lines still in place and the flickering is gone?01:50
joschyes, but the flicker looks different than it looked back then (more frequent)01:51
joschso i have more investigation to do -- maybe it's not even the same flicker01:51
joschi got help with this in #dri-devel two days ago01:51
joschand it might or might not also be connected with hdmi being enabled on a311d borked the startup of sway01:52
minutehmm ok now i'm even more confused :001:52
minuteso with my patch there is a new kind of flicker?01:52
joschyes, i need to do more work on that topic01:52
joschmaybe, please let me investigate more before "wasting" your time with this :)01:52
minuteok, no problem :D01:52
joschbut i didn't take the time yet because yesterday when i started bisecting it broke with a nullptr error when booting...01:53
joschso i'll do that another time01:53
joschright now i'm preparing the reform-tools release01:53
joschminute: and i'd like your input on https://source.mnt.re/reform/reform-system-image/-/merge_requests/129#note_1620201:53
joschbecause reform-check now checks for this but the line never got removed01:53
minutejosch: i've forgotten about this, so whatever the current shipping behavior is, it's fine01:55
joschokay, then i revert the checks in reform-check01:58
+ emdevt (~emdevt@2001:7e8:fa17:7901:889:cfa1:9c37:6514)02:29
joschMNT Pocket Reform being sold in Germany (but already reserved): https://www.kleinanzeigen.de/s-anzeige/316869683802:31
- AnimaInvicta (PART: !!unknown attribute: msg!!) (~AnimaInvi@88-120-179-216.subs.proxad.net)02:37
joschbremner: i want to release the next reform-tools version soon -- maybe you can have a look at https://source.mnt.re/reform/reform-tools/-/issues/8 because as far as i understand you have kernel 6.17rcX running already?02:39
- paperManu (QUIT: Ping timeout: 258 seconds) (~paperManu@72.10.129.226)03:26
bremnerjosch: sure. Will try to do it in the next few days03:28
bremnerI wonder if I missed an email from gitlab...03:28
bremnernot a technical email problem, just a me not reading problem03:28
- chomwitt (QUIT: Ping timeout: 245 seconds) (~chomwitt@2a02:85f:9a74:ce00:1ac0:4dff:fedb:a3f1)03:50
- dok (QUIT: Server closed connection) (~dok@2a01:e0a:12:d860:3696:e5bc:8411:9bc5)03:56
+ dok (~dok@2a01:e0a:12:d860:3696:e5bc:8411:9bc5)03:57
- skipwich (QUIT: Remote host closed the connection) (~skipwich@user/skipwich)05:02
+ skipwich (~skipwich@user/skipwich)05:02
- skipwich (QUIT: Ping timeout: 256 seconds) (~skipwich@user/skipwich)06:06
+ skipwich (~skipwich@user/skipwich)06:21
- emdevt (QUIT: Quit: Leaving) (~emdevt@2001:7e8:fa17:7901:889:cfa1:9c37:6514)07:43
chhello from a train07:50
chor rather, tren07:50
- cobra (QUIT: Server closed connection) (~cobra@user/Cobra)07:51
+ cobra (~cobra@user/Cobra)07:52
- kuno (QUIT: Server closed connection) (sid299442@id-299442.helmsley.irccloud.com)08:01
+ kuno (sid299442@id-299442.helmsley.irccloud.com)08:02
joschhello! :)08:26
chorchello everyone! I'm running my pocket without the charging board, and getting this spamming dmesg: power_supply BAT0: driver failed to report `voltage_now' property: -1608:38
chorcthis is normal, right? and is there a quick way to silence it08:39
+ reform7228 (~rene@p200300d0ff0561007fc2100dbdd09829.dip0.t-ipconnect.de)08:56
* reform7228 -> rwa_08:56
- ZetaR (QUIT: Ping timeout: 248 seconds) (~user@c-98-208-141-90.hsd1.fl.comcast.net)09:00
- abortretryfail (QUIT: Ping timeout: 260 seconds) (~arf@146.ip-149-56-132.net)09:10
+ abortretryfail (~arf@146.ip-149-56-132.net)09:22
joschbremner: i just released the next reform-tools version, so testing the new codepath will be a bit easier for you :)10:17
asjacksonDoes the MT7612U wifi adapter also have bluetooth? I don't seem to have a bluetooth device 🤔 maybe the rf shield is still touching something (originally i didnt even have wifi because the shield was bent inwards slightly)10:34
chhttps://www.mediatek.com/products/broadband-wifi/mt7612u claims bluetooth should exist10:55
chi *guess* on USB though10:55
asjacksonYeah I have a feeling that bluetooth isn't supported, someone here will know for sure i guess :D11:02
chyou have a pocket with rk?11:02
asjacksonyep11:03
+ mjw (~mjw@gnu.wildebeest.org)11:11
+ andreas-e (~Andreas@2a02-8434-b6a3-e901-facc-8e87-8e54-890d.rev.sfr.net)11:12
+ chomwitt (~chomwitt@2a02:85f:9a74:ce00:1ac0:4dff:fedb:a3f1)11:30
- rwa_ (QUIT: Ping timeout: 255 seconds) (~rene@p200300d0ff0561007fc2100dbdd09829.dip0.t-ipconnect.de)11:40
minutech: asjackson: afaik it doesn't have bluetooth, it's a bit mysterious 12:01
chhmm12:01
chk12:01
asjacksonah right, no worries 12:03
minutethis is the card https://asiarf.com/wp-content/uploads/2023/08/M27612-BU3_V1-1P.pdf12:04
minuteso maybe if wifi is on usb, bt can't be on usb at the same time, and they didn't put an extra hub and serial-usb converter for bt on the card12:05
asjacksonthat makes sense.  thanks for confirming :)12:07
minutethere is a stamp module from ezurio that uses an older cypress/infineon/broadcom chip that has an integrated hub and has a mode where both wifi and bt can be on usb12:09
minuteunfortunately the mainline drivers (or fw, not sure) are very broken12:09
minutethat's the point where i gave up on usb for wifi and changed the design of rcore to yield another pcie+usb combo connector so we could use a proper card (ax210 and probably a bunch of others)12:10
minutebuuut it reminds me there is no bt solution for classic reform yet12:11
gordon1is there tusb thing but with 6 or 8 ports instead of 4?12:13
gordon1it would be soo nice to have just spare usbs hanging on headers somewhere on the mb12:13
+ paperManu (~paperManu@72.10.129.226)12:14
gordon1one extra can go to mini-pcie, and the rest available for mods12:14
- bkeys (QUIT: Ping timeout: 256 seconds) (~Thunderbi@98.19.131.193)12:35
bremnerjosch: ack re reform-tools12:40
+ reform15031 (~rene@p200300d0ff056100d042e7e3d691bc19.dip0.t-ipconnect.de)12:41
* reform15031 -> rwa_12:41
rwa_may classic reform shut down itself due to empty battery cells - this time the OLED does not show anything12:42
rwa_as minute suggested a few days ago i've plugged in a usb-c charger and left it alone12:42
rwa_its now charging almost an hour at a rate of 40-45 watts but the systemcontroller is still unresponsive - circle key provokes no reaction12:43
rwa_how long should i keep it on the charger before doing further steps?12:43
- rwa_ (QUIT: Quit: Leaving) (~rene@p200300d0ff056100d042e7e3d691bc19.dip0.t-ipconnect.de)12:52
- minute (QUIT: Server closed connection) (~mntirc@softboy.mntmn.com)13:36
+ minute (~mntirc@softboy.mntmn.com)13:36
- paperManu (QUIT: Ping timeout: 256 seconds) (~paperManu@72.10.129.226)14:13
+ rwa_ (~rwa_@pd9ffb856.dip0.t-ipconnect.de)14:33
rwa_it stopped charging, but system controller still unresponsive14:33
minuterwa_: sounds like keyboard is unresponsive14:34
minutenot sysctl14:34
rwa_jep, might have mixed things up 14:35
minuteyou could either: open the bottom and disconnect and reconnect both cables going to kbd, or, open the top and press reset button on kbd14:36
rwa_reset button on kbd worked14:39
- rwa_ (QUIT: Ping timeout: 258 seconds) (~rwa_@pd9ffb856.dip0.t-ipconnect.de)14:57
* mjw -> Guest734015:06
* Guest5996 -> mjw15:06
+ bkeys (~Thunderbi@h193.131.19.98.dynamic.ip.windstream.net)15:17
- bleb (QUIT: Server closed connection) (~cm@user/bleb)15:18
+ bleb (~cm@user/bleb)15:19
cwebberhewwo15:36
- bkeys (QUIT: Ping timeout: 260 seconds) (~Thunderbi@h193.131.19.98.dynamic.ip.windstream.net)16:08
+ emdevt (~emdevt@2001:7e8:fa17:7901:889:cfa1:9c37:6514)16:09
+ bkeys (~Thunderbi@66.110.201.50)16:11
- sad_plan (QUIT: Quit: WeeChat 4.7.0) (sadplan@tilde.club)16:27
+ ZetaR (~user@c-98-208-141-90.hsd1.fl.comcast.net)16:34
bremnerjosch: what version of reform-tools am I looking for?16:34
bremner(to test the 6.17 compat)16:34
minuterwa_: great16:36
minutecwebber: hi!16:36
bremnerjosch: assuming 1.77, should I expect to install that via apt?16:38
joschbremner: yes, but dinstall only ran now16:39
joschbremner: according to:16:39
joschcurl --silent http://deb.debian.org/debian/dists/unstable/main/binary-arm64/Packages.xz | unxz | grep --after-context=1 --line-regexp --fixed-strings 'Package: reform-tools'16:39
joschat least my local mirror still has 1.76...16:40
joschand yes, via apt but as you know, going from "git push" to "apt install" can take 6+ hours...16:40
bremnerAh OK. I'll try again later today16:40
bremnernot sure if I just bumped the serial console cable or there is some bad software juju, but apt went into block drawing character mode: https://0x0.st/Kopn.png16:51
bremnerI had to quit the tio session and run reset in the containing (alacritty) terminal16:51
joschbremner: i had a very similar experience when upgrading from bookworm to trixie -- everything became unreadable16:53
joschbremner: copying the console content to gedit turned the garbage into readable characters16:53
- Guest7340 (QUIT: Ping timeout: 256 seconds) (~mjw@gnu.wildebeest.org)16:53
bremnerhuh. 8th bit problem maybe. Was yours also in a serial console?16:53
joschno, with xfce-terminal16:54
bremnerhuh^2.16:54
+ bleb_ (~cm@user/bleb)16:57
- bkeys (QUIT: *.net *.split) (~Thunderbi@66.110.201.50)17:01
- bleb (QUIT: *.net *.split) (~cm@user/bleb)17:02
- minute (QUIT: *.net *.split) (~mntirc@softboy.mntmn.com)17:02
- switchy (QUIT: *.net *.split) (~switchy@mechboards/switchy)17:02
- brennen (QUIT: *.net *.split) (~brennen@user/brennen)17:02
* bleb_ -> bleb17:02
+ mark_ (~mjw@gnu.wildebeest.org)17:02
+ bkeys (~Thunderbi@66.110.201.50)17:03
+ minute (~mntirc@softboy.mntmn.com)17:07
+ switchy (~switchy@mechboards/switchy)17:08
+ brennen (~brennen@user/brennen)17:10
- emdevt (QUIT: Ping timeout: 256 seconds) (~emdevt@2001:7e8:fa17:7901:889:cfa1:9c37:6514)17:26
joschbremner: never mind, the package is still (for more than 6 hours now) in "Needs-Build" state: https://buildd.debian.org/status/package.php?p=reform-tools17:50
joschso it will hit the archive in 4.5 hours at the earliest17:50
- colinsane (QUIT: Ping timeout: 256 seconds) (~colinunin@97.113.128.255)17:56
- bkeys (QUIT: Ping timeout: 258 seconds) (~Thunderbi@66.110.201.50)18:23
+ bkeys (~Thunderbi@h193.131.19.98.dynamic.ip.windstream.net)18:24
+ vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:50)18:39
joschminute: i'm looking into making a simple plymouth theme for the next reform-tools release. Would you like the plain MNT logo or rather display the existing wallpapers like y2k?18:49
joschwe can also prepare multiple themes, for example a non-free one with the logo and a free one with y2k or something (we just need names)18:54
+ AnimaInvicta (~AnimaInvi@88-120-179-216.subs.proxad.net)19:03
minutejosch: oh cool! your second idea sounds good, i'd like to test when you have something ready19:23
joschokay, for example this is the "spinner" theme but with y2k as background: https://mister-muffin.de/p/6VLd.jpg19:26
joschi'll create a system image with it19:27
joschnote though that this should only work on rk358819:27
- chomwitt (QUIT: Ping timeout: 245 seconds) (~chomwitt@2a02:85f:9a74:ce00:1ac0:4dff:fedb:a3f1)19:32
minuteimx8mplus kernel 6.15.3 was pretty broken eh19:35
minutejosch: cool19:36
minutemaybe we could get rid of ubsan warnings for qcacld2 like this https://github.com/openzfs/zfs/blob/fd51786f86e4e39ce3aa0a4ab8c44935496ff7c2/module/os/linux/zfs/zio_crypt.c#L23319:53
minutevia https://github.com/openzfs/zfs/issues/16278#issuecomment-219775993619:53
minutejosch: btw why does plymouth work only on rk3588? (just curious)20:06
- bkeys (QUIT: Ping timeout: 248 seconds) (~Thunderbi@h193.131.19.98.dynamic.ip.windstream.net)20:16
gordon1what is emmc offset of uboot env on rk3588?20:20
+ bkeys (~Thunderbi@66.110.201.50)20:22
gordon1CONFIG_ENV_IS_NOWHERE=y no :(20:23
- jfred (QUIT: Server closed connection) (c9b58a1025@libera/sponsor/jfred)20:34
+ jfred (c9b58a1025@libera/sponsor/jfred)20:34
+ sad_plan (sadplan@tilde.club)20:37
- mjw (QUIT: Killed (tantalum.libera.chat (Nickname regained by services))) (~mjw@2001:1c06:2486:a800:7602:5eff:dc71:a72c)20:51
* mark_ -> mjw20:51
+ Guest6424 (~mjw@2001:1c06:2486:a800:7602:5eff:dc71:a72c)20:51
- wose (QUIT: Server closed connection) (wose@mail.zuendmasse.de)21:06
+ wose (wose@mail.zuendmasse.de)21:06
+ colinsane (~colinunin@97.113.128.255)21:11
+ chomwitt (~chomwitt@2a02:85f:9a74:ce00:1ac0:4dff:fedb:a3f1)21:22
- colinsane (QUIT: Ping timeout: 248 seconds) (~colinunin@97.113.128.255)21:24
joschminute: it didn't work out-of-the-box on a311d nor on imx8mq21:27
joschgordon1: i can give you the offset of u-boot but not of the environment21:28
- bkeys (QUIT: Ping timeout: 245 seconds) (~Thunderbi@66.110.201.50)21:29
+ colinsane (~colinunin@97.113.128.255)21:30
elbthe pocket reform handbook says that the USB PD port is usable for USB, but as far as I can tell, nothing I plug in actually works21:43
elbshould it?21:43
joschelb: it is not usable for usb-data -- where in the handbook does it say that?21:44
elbunder "USB 3.0 hub" it says two internal and two external USB ports, and then under right side ports it says "2x USB-C:" and then "Both ports support the connection of USB 3.0/2.0/1.0 devices."21:45
elbboth on the hardware page21:45
elbif it's not expected to work, that's workable for me, I just wasn't sure if I was doing something wrong based on the documentation21:46
joschhuh, does it actually work? one sec...21:49
joschnope, the usb-c charging port does not do anything when i plug devices into it21:52
joschminute: ^21:52
elbI haven't checked the schematics yet21:54
+ bkeys (~Thunderbi@66.110.201.50)21:55
gordon1josch: yeah, i know the u-boot offset, but env does not exist22:12
gordon1which is very sad because the only thing i wanted is to increase u-boot boot timeout without re-flashing the whole u-boot22:17
jfredTIL about Zink... could it be possible to use that with PanVK to get more complete OpenGL support on rk3588?22:27
gordon1i would love to test it with paraview but paraview is broken for python3.13 right now22:37
gordon1it needs geometry shaders22:39
gordon1>WARNING: Some incorrect rendering might occur because the selected Vulkan device (Mali-G610) doesn't support base Zink requirements: feats.features.fillModeNonSolid feats.features.shaderClipDistance 22:50
gordon1not a great start, but glxgears work22:50
+ paperManu (~paperManu@72.10.129.226)22:50
- bkeys (QUIT: Ping timeout: 258 seconds) (~Thunderbi@66.110.201.50)22:55
- andreas-e (QUIT: Quit: Leaving) (~Andreas@2a02-8434-b6a3-e901-facc-8e87-8e54-890d.rev.sfr.net)22:57
jfredah, drat22:59
+ bkeys (~Thunderbi@98.19.131.193)23:01
gordon1superslicer works, freecad works as well23:01
gordon1kicad works23:04
- jbmorley_ (QUIT: Server closed connection) (sid21953@id-21953.tinside.irccloud.com)23:09
- colinsane (QUIT: Quit: bye) (~colinunin@97.113.128.255)23:09
+ jbmorley_ (sid21953@id-21953.tinside.irccloud.com)23:10
gordon1$ vulkaninfo|grep geom23:11
gordon1geometryShader                          = false23:11
gordon1ACTION immediately lost interest23:12
+ colinsane (~colinunin@97.113.128.255)23:12
joschgordon1: the variables bootdelay is at offset 899977123:34
gordon1i wanted to use u-boot-tools to change it with fw_setenv, not sure if i'm brave enough to edit my only way of booting the maching with r223:39
joschgordon1: it is also entirely possible that the stuff around u-boot (it's not only u-boot that gets flashed but also other stages of the bootloader chain) contain a checksum of u-boot which make this sort of hack impossible but i couldn't say -- just something to look out for23:40
- colinsane (QUIT: Ping timeout: 258 seconds) (~colinunin@97.113.128.255)23:40
gordon1josch: i really doubt that but i don't want to find out :D23:57
gordon1gonna just live with it now, at least until when i'll get permanent second device23:57

Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!