2025-08-22.log

- Ar|stote|is (QUIT: Read error: Connection reset by peer) (~linx@149.210.0.217)00:01
+ Ar|stote|is (~linx@149.210.0.217)00:08
+ chomwitt (~chomwitt@2a02:85f:9a74:ce00:1ac0:4dff:fedb:a3f1)00:30
+ colinsane (~colinunin@97-113-135-188.tukw.qwest.net)00:33
- colinsane (QUIT: Ping timeout: 248 seconds) (~colinunin@97-113-135-188.tukw.qwest.net)00:38
+ erle (~erle@user/erle)00:48
+ colinsane (~colinunin@97-113-143-5.tukw.qwest.net)00:52
erlemy reform is making a slight sound when i tilt it. i hope it's not again a loose screw.00:53
erle(i am very happy about the transparent bottom because it allows me to visually diagnose if it is)00:54
- erle (QUIT: Quit: K-lined) (~erle@user/erle)00:55
- colinsane (QUIT: Ping timeout: 258 seconds) (~colinunin@97-113-143-5.tukw.qwest.net)01:13
+ emdevt (~emdevt@2001:7e8:fa17:7901:889:cfa1:9c37:6514)02:42
- paperManu (QUIT: Ping timeout: 244 seconds) (~paperManu@72.10.129.226)02:56
+ wickedshell (~wickedshe@c-174-56-48-20.hsd1.nm.comcast.net)03:11
- skipwich (QUIT: Remote host closed the connection) (~skipwich@user/skipwich)05:01
+ skipwich (~skipwich@user/skipwich)05:02
- Ar|stote|is (QUIT: Quit: https://quassel-irc.org - Chat comfortably. Anywhere.) (~linx@149.210.0.217)05:12
+ Ar|stote|is (~linx@149.210.0.217)05:13
- voltaire28 (QUIT: Server closed connection) (sid689360@id-689360.ilkley.irccloud.com)05:25
+ voltaire28 (sid689360@id-689360.ilkley.irccloud.com)05:25
- chomwitt (QUIT: Ping timeout: 248 seconds) (~chomwitt@2a02:85f:9a74:ce00:1ac0:4dff:fedb:a3f1)05:51
- Ar|stote|is (QUIT: Read error: Connection reset by peer) (~linx@149.210.0.217)06:27
+ Ar|stote|is (~linx@149.210.0.217)06:34
+ reform20474 (~rene@p200300d0ff4e92007fc2100dbdd09829.dip0.t-ipconnect.de)07:41
* reform20474 -> rwa_07:42
- emdevt (QUIT: Ping timeout: 248 seconds) (~emdevt@2001:7e8:fa17:7901:889:cfa1:9c37:6514)07:42
- Ar|stote|is (QUIT: Read error: Connection reset by peer) (~linx@149.210.0.217)08:39
+ Ar|stote|is (~linx@149.210.0.217)08:45
rwa_just curious: is RCORE-DSI the same as RCORE v2?09:55
minuterwa_: dsi means you use the dsi output for internal display instead of a hdmi converter (on classic reform)10:52
minutethe main difference between rcore v1+v2 is the signals on the wifi/bt port. on v1, it's sdio+usb. on v2, it's pcie+usb.10:53
joschminute: but i thought there was a hardware difference which means that those who upgraded their classic reform early with rcore v1 have to de-solder a resistor to convert it from non-dsi to dsi?10:55
+ hlep (~hlep@2a00:1398:9:fb03:3422:2a5:ffe3:7d77)11:18
minutejosch: that is another difference, right. v1 doesn't have the reset signal for the dsi to edp converter connected, v2 does11:20
joschah right11:26
joschrwa_: then i think "rcore-dsi" is the name for the device tree which an be used for both adapter boards (v1 and v2) but v1 needs a small hardware mod for the rcore-dsi dtb to work11:27
rwa_ hm11:28
+ chomwitt (~chomwitt@2a02:85f:9a74:ce00:1ac0:4dff:fedb:a3f1)11:52
- rwa_ (QUIT: Quit: Leaving) (~rene@p200300d0ff4e92007fc2100dbdd09829.dip0.t-ipconnect.de)12:05
- hlep (QUIT: Quit: Client closed) (~hlep@2a00:1398:9:fb03:3422:2a5:ffe3:7d77)12:11
amkdoes the pocket support hdmi/DP out on the USB-C ports? or is it just the microhdmi?12:43
joschamk: just microhdmi12:44
amkah cool, explains my docks not displaying :D12:44
+ paperManu (~paperManu@72.10.129.226)12:48
+ reform25737 (~rene@p200300d0ff4e92007fc2100dbdd09829.dip0.t-ipconnect.de)13:09
* reform25737 -> rwa_13:09
asjacksongot my mnt pocket kit and its all working now. had an odd issue with the wifi, wasnt showing up despite many reseats.  i noticed the shield was a bit squashed and bent it up, and now its working. think it might have been shorting slightly13:17
asjacksonthat took me ages to type. gonna take some time to get used to this tiny keyboard13:17
minuteasjackson: uh oh @ shield, glad it still works! and great the diy build works13:21
asjacksonit was great fun, really impressed with how everything was packaged. well done everyone at mnt and thanks for the cool tiny gay laptop13:23
grimmwareyou get used to the keyboard :)13:24
asjacksoni am slowly getting faster (and pressing backspace less) :D13:26
asjackson this thing is adorable 13:31
rwa_have fun :)13:33
rwa_what i really like on my Classic Reform is that i can knock out anoying people and it still will work after that :D13:33
+ elb (~elb@2600:4041:6671:1300:f96b:f2a2:ea8c:4a45)13:35
- gsora (QUIT: Remote host closed the connection) (~gsora@user/gsora)13:45
joschrwa_: it will also survive 3 year olds jumping on it, falling off tables, bike crashes and first year students -- ask me how i know ;)13:46
joschi'm currently in a train between barcelona and sevilla and the spanish train system does an x-ray of all luggage -- i am slightly annoyed that they did not find either of the three reforms I carry...13:47
rwa_lol13:49
ch:D13:51
chwill you stay in sevilla for some time?13:51
joschyes, for about 23 minutes13:52
joschfinal destination is south of portugal13:52
joschch: why, did i miss a VAC message on debian-devel? ;)13:52
chrn on a plane to valencia, and next wed i'll take a train to malaga13:53
chcould also check out seville then :)13:53
chjosch: i didnt write one ;)13:53
chah 23min is a bit too short :)13:53
joschyeah, neither will i -- i'll be online too often for that to be necessary i think :D13:54
chyes13:54
joschand i took half my hardware in the hope that i get a few reform todo items done during "vacation" :)13:54
+ gsora (~gsora@user/gsora)13:55
chi was pondering to take some but i have enough work to take care off anyway13:56
joschwith imx8mq and a311d there was the point where no more work needed to be done because it was... finished :)13:59
joschjust a few more items left for pocket, rk3588 and mb3.0/keyboardv414:00
chcan you actually continue by train to portugal?14:13
chit seemed like the train network doesnt reach there14:13
- chomwitt (QUIT: Ping timeout: 244 seconds) (~chomwitt@2a02:85f:9a74:ce00:1ac0:4dff:fedb:a3f1)14:53
amkanother silly question, how might i start sway from a tty? when i do my trackball doesnt work (but usb mouse does), it works okay when started via gdm though14:56
chjust "sway" afaik. maybe the input devices in /dev/input dont have the correct permissions?15:03
* mjw -> Guest563315:05
- Guest5633 (QUIT: Killed (calcium.libera.chat (Nickname regained by services))) (~mjw@gnu.wildebeest.org)15:05
* mark_ -> mjw15:05
+ Guest5633 (~mjw@gnu.wildebeest.org)15:06
amki think it is actually working, but only if i move it in very small increments, so maybe its some sort of sensitivity issue15:06
+ chomwitt (~chomwitt@2a02:85f:9a74:ce00:1ac0:4dff:fedb:a3f1)15:16
- chomwitt (QUIT: Ping timeout: 248 seconds) (~chomwitt@2a02:85f:9a74:ce00:1ac0:4dff:fedb:a3f1)15:24
- rwa_ (QUIT: Remote host closed the connection) (~rene@p200300d0ff4e92007fc2100dbdd09829.dip0.t-ipconnect.de)15:27
joschch: that is correct. Me and two other poor souls are currently in a bus from Sevilla to Faro. They stopped all train connections between spain and portugal with covid...15:41
joschamk: if the trackball does not work there is usually some dirt on the sensor15:42
joschamk: try holding it up and then blow on the trackball -- doing that usually gets rid of the debris15:42
joschif that does not help, remove the round keys, unscrew the trackball lid, remove the trackball and visually inspect whether any dirt is covering the sensor15:43
amkhuh yep that worked, it just happened to coincide with me disabling gdm lol working again now, thanks!15:43
gordon1i did a bit of a disgusting discovery, if you take the ball (even the stock POM one) and wash and scrub it thoroughly with IPA, it stops tracking well for a while, but then it recovers after a while16:06
gordon1so i did handle SiN ball for 30 minutes just rolling it in my hands and it tracks way better now, though still not satisfactory16:07
gordon1okay, something interesting happened in dmesg https://0x0.st/Kipl.txt16:19
gordon1video stuttered for a bit in mpv, but after a restart it works fine again16:19
- [tj] (QUIT: Server closed connection) (sid609767@id-609767.lymington.irccloud.com)16:41
+ [tj] (sid609767@id-609767.lymington.irccloud.com)16:41
- bremner (QUIT: Server closed connection) (~bremner@notmuch/developer/bremner)16:44
+ bremner (~bremner@notmuch/developer/bremner)16:45
+ emdevt (~emdevt@2001:7e8:fa17:7901:889:cfa1:9c37:6514)17:25
bremnerI guess IPA is not beer in this context18:00
gordon1correct, that is isopropyl alcohol or rubbing alcohol18:01
+ chomwitt (~chomwitt@2a02:85f:9a74:ce00:1ac0:4dff:fedb:a3f1)18:10
- emdevt (QUIT: Ping timeout: 248 seconds) (~emdevt@2001:7e8:fa17:7901:889:cfa1:9c37:6514)18:15
kop316bremner: I knew a guy who did spill that type of IPA on his laptop. We actually were able to clean it up with the right IPA and get it fixed18:20
fricklerregarding trackball: does anyone else have the issue on the pocket that it is getting almost unusable when the keyboard is tilted upright? like while lying on your back? even more than 45° become difficult18:23
frickleror did I maybe assemble it wrong?18:24
gordon1once i learned why you not supposed to use IPA in ultrasonic cleaner, surprisingly nothing burned down, even the cleaner still worked after that18:24
gordon1only new pair of trousers was required18:24
minutefrickler: it is definitely an achilles heel of the trackball that it can "fall" upwards18:25
fricklerminute: yes, that leads to big jumps when touching it and when removing the finger again. any solution to that would be great18:28
minutejosch: are you getting cell connection to pocket in that bus or how are you online?18:28
minutefrickler: yeah. if you find a solution, do tell18:29
minuteunipolar magnet sth sth18:29
gordon1need to solve quantum physics first, then we'll get nice things!18:32
gordon1is pocket ball rides on small bearing balls like classic reform?18:34
gordon1maybe adding more balls can help18:34
gordon1*more small balls18:34
minutegordon1: right, bearings on top could work.18:37
gordon1something like six balls in two concentric circles, five fixed and one that fits through a hole and is locked from the back, so you can drop the ball, put locking ball in and it makes everything captive18:40
gordon1not sure if i explained it well18:40
asjacksonmy mnt can raytrace in kicad faster than my intel laptop 🪛18:44
asjacksonamazed how fast it is18:44
minuteasjackson: nice! for me also the rk3588's GPU is faster in kicad than my dedicated intel gpu in my work pc... not sure why :D18:48
asjacksonpretty wild18:50
asjacksonmight say more about intel gpu :D18:51
minuteyeah. not sure what's going on there18:53
minuteironically i bought this GPU for an Ampere Altra build. but that didn't happen because my radeon died and i needed the gpu for work18:54
minuteanyway. MNT Reform Next motherboard revision 4 pcb samples finally ordered... whew18:54
- Gooberpatrol66 (QUIT: Quit: Konversation terminated!) (~Gooberpat@user/gooberpatrol66)19:32
+ Gooberpatrol66 (~Gooberpat@user/gooberpatrol66)20:03
- gachikuku_ (QUIT: Server closed connection) (~gachikuku@kamil.gr)20:14
+ gachikuku_ (~gachikuku@kamil.gr)20:14
joschminute: yes, the quectel modem was getting a connection in france, spain and now portugal without problems20:25
minutejosch: nice!20:26
- arminweigl (QUIT: Server closed connection) (~arminweig@sourcehut/user/arminweigl)20:39
+ arminweigl (~arminweig@sourcehut/user/arminweigl)20:39
jfredThe table on this newsletter page with shipment dates seems to go backwards for the Pocket Reform upgrade bundles (2025-06-12 to 2025-04-04?): https://mntre.com/media/reform_md/2025-07-31-july-update.html20:43
jfredIs the end date supposed to be 08-04? (If so, drat, guess I just missed that batch ordering on 8/8?)20:43
minutejfred: ah meh, sorry for that, i think it's 2024-06-1220:52
- L29Ah (PART: Disconnected: closed) (~L29Ah@wikipedia/L29Ah)20:52
minutejfred: will feedback this but it'll be updated only on monday i think20:52
jfredno worries!20:53
minuteafaik all of the ones ordered til 04-04 are shipped now20:53
minutethe current batch reaches to july iirc. i'll get this table updated on monday20:53
jfredmine was the new rcore board without the rk3588, if that makes a difference20:53
minutejfred: sorry, looking at this again, i guess the 2025-06-12 is correct, but the end date is wrong and i don't know the correct one by heart, but it'll be fixed20:55
jfredthat's okay, thank you :)20:55
bremnerIn case anyone is curious, I have rebased the reform patches on top of 6.17~rc2 https://salsa.debian.org/bremner/collabora-rockchip-358820:58
bremnerSeems to work OK, I haven't really stress tested it, but I am typing this on a pocket booted into that kernel20:58
bremnerthat branch (reform-patches) also has some PCI reset patches on it 21:00
minutebremner: oh nice, futuristic kernel21:03
bremneryes ;)21:03
joschminute: do you want to hit the merge button on 6.16.1? :)21:05
minutejosch: yeah21:06
joschit also has the display freq fix but untested for older innolux displays21:06
joschon the other hand, those people who i know have upgraded their old classic reform with rk3588 like vagrantc or gsora are also not online right now21:07
joschoh sorry, gsora is idling21:07
joschjust vagrant is not here21:07
gordon1last time i tried this pci reset patch was waking up nvmes from hibernate21:23
joschgordon1: with 6.17?21:24
gordon1if we're talking about the same thing21:24
gordon1no, it was 6.15 something iirc21:24
gordon1it was a while ago21:24
joschwe might've merged "that patch" with 6.16.1 i think?21:24
bremnergordon1: it's not waking up something for me, it never comes back21:24
gordon1bremner: black screen after return from hibernate?21:25
joschdid you try without pcie? just booting from sd-card? pcie is "just" for nvme or wifi, no?21:25
gordon1i'm pretty sure there is something else with panfrost21:25
gordon1josch: i never tried it on reform, tried it on rock5b+ from initramfs and checked which devices woke up21:26
bremnergordon1: yes, some messages about unable to resume PHY on the serial console21:26
gordon1now that i have display on my reform finally - i can look at what happens with panfrost21:26
bremnerI can grab the messages, 1 sec21:27
gordon1just need another machine, was too hasty to decommission my thinkpad21:27
gordon1(plan is to have serial console on uconsole, excuse the pun, but whenever it comes - nobody knows)21:28
bremnerhttps://paste.debian.net/1392902/21:28
gordon1bremner: that's with 6.17, isn't it?21:29
bremneryes.21:29
bremner+ a few patches from mani@kernel.org21:29
gordon1yeah, i thought this thing was fixed by link reset patch, but apparetnly it didn't21:29
bremnerI have been meaning to mention on the appropriate mailing lists that this isn't working, but the messages are pretty much the same as the first time I asked (and was suggested an an additional series of 4 patches), so I am not too optimistic.21:32
bremnerjosch: interesting idea. I'm not sure if the kernel needs modding too, or just not use the devices. nvme is currently idle and wifi module is removed.21:36
- andreas-e (QUIT: Quit: Leaving) (~Andreas@2a02-8434-b6a3-e901-facc-8e87-8e54-890d.rev.sfr.net)21:47
- mjw (QUIT: Killed (tantalum.libera.chat (Nickname regained by services))) (~mjw@2001:1c06:2486:a800:7602:5eff:dc71:a72c)21:53
* Guest5633 -> mjw21:53
+ Guest9812 (~mjw@2001:1c06:2486:a800:7602:5eff:dc71:a72c)21:53
- op_4 (QUIT: Remote host closed the connection) (~tslil@user/op-4/x-9116473)22:38
+ op_4 (~tslil@user/op-4/x-9116473)22:38
- chrcav (QUIT: Quit: leaving) (~chrcav@user/chrcav)22:50
elbhmmm I installed an NVMe SSD in my Pocket Reform and the machine seems to have hard locked at 70% of reform-migrate23:06
elbnot responding to the keyboard, trackball, or pings23:06
elbok, simply mounting a filesystem on the device and doing a lot of I/O results in a lockup23:15
joschelb: ouch -- what nvme is that?23:15
elbsk hynix gold p3123:16
elbI'm going to try to ssh in and leave a dmesg going, see if it squeezes out any errors on the way23:16
elbunfortunately nothing that made it out via ssh, at least23:24
elbis there a serial console enabled by default on the .1" header on the motherboard?23:30
elbthe handbook doesn't seem to say23:31
elboh wait yes it does, I was in the wrong chapter23:31
elbhuh the pinout in the pocket reform handbook is wrong23:43
elbman I'm getting really terrible copy off this port, I'm not sure my adapter is good for 1.5 Mbps23:46
elbwhen I opened it up to plug in a serial cable, I reseated the ribbon cable on the SOM adapter board at both ends (I assume it's PCIe, I didn't look), the SOM adapter board, and the NVMe, and so far I've written out 7 10 GiB files without problems, maybe it was a connection issue?  The back also isn't on, and it does put a fair bit of pressure on the SOM, I wonder if problems will show up if I reinstall the back plate23:54
elbok never mind, just got a spontaneous reboot23:57

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