| josch | minute: do we still need reform-firedecor or is it dead? | 00:15 |
|---|---|---|
| - potash (QUIT: Quit: The Lounge - https://thelounge.chat) (~potash@user/foghorn) | 00:17 | |
| + pomel0 (~pomel0@user/pomel0) | 01:07 | |
| minute | josch: dead, as we don't ship wayfire anymore | 01:27 |
| - lidstah (QUIT: Remote host closed the connection) (~lidstah@gateway/tor-sasl/lidstah) | 01:30 | |
| + lidstah (~lidstah@gateway/tor-sasl/lidstah) | 01:31 | |
| + paperManu_ (~paperManu@107.159.15.124) | 01:39 | |
| - paperManu (QUIT: Ping timeout: 264 seconds) (~paperManu@107.159.15.124) | 02:30 | |
| - mjw (QUIT: Ping timeout: 252 seconds) (~mjw@gnu.wildebeest.org) | 02:59 | |
| - kop316 (QUIT: Remote host closed the connection) (m-6f6zq6@static.138.159.90.157.clients.your-server.de) | 03:07 | |
| + kop316 (m-6f6zq6@static.138.159.90.157.clients.your-server.de) | 03:11 | |
| - paperManu_ (QUIT: Ping timeout: 265 seconds) (~paperManu@107.159.15.124) | 03:33 | |
| + spew (~spew@user/spew) | 03:34 | |
| - spew (QUIT: Quit: WeeChat 4.6.3) (~spew@user/spew) | 03:58 | |
| - bkeys (QUIT: Remote host closed the connection) (~Thunderbi@98.19.131.193) | 04:09 | |
| + bkeys (~Thunderbi@98.19.131.193) | 04:14 | |
| - bkeys (QUIT: Ping timeout: 252 seconds) (~Thunderbi@98.19.131.193) | 04:22 | |
| + bkeys (~Thunderbi@98.19.131.193) | 04:24 | |
| - pomel0 (QUIT: Remote host closed the connection) (~pomel0@user/pomel0) | 05:47 | |
| + pomel0 (~pomel0@user/pomel0) | 05:48 | |
| josch | minute: thank you -- i made a forum post to ask if anybody else cares but otherwise it'll get removed | 07:27 |
| + gustav25 (~gustav@c-78-82-54-254.bbcust.telenor.se) | 08:04 | |
| + kxtells (~kxtells@user/kxtells) | 08:41 | |
| kxtells | Hello, it's me, the keyboard guy (connection issues and finally total disconnect of the keyboard from the OS. Still can drive the OLED) | 08:58 |
| kxtells | Yesterday I got time finally to crimp the new cable and test if it improved something. I left it overnight turned on and came back this morning.. unfortunately with the same disconnection. | 08:59 |
| kxtells | I am out of ideas here. Last firmware, changed cable, any other suggestions? | 09:00 |
| + gidzit (~gidzit@gidzit.org) | 09:25 | |
| + andreas-e (~Andreas@2a02-8434-b6a3-e901-facc-8e87-8e54-890d.rev.sfr.net) | 09:44 | |
| - andreas-e (QUIT: Remote host closed the connection) (~Andreas@2a02-8434-b6a3-e901-facc-8e87-8e54-890d.rev.sfr.net) | 09:44 | |
| - gidzit (QUIT: Remote host closed the connection) (~gidzit@gidzit.org) | 10:09 | |
| minute | kxtells: uboot was already updated, right? | 10:24 |
| josch | minute: could i also send you a friendly ping for enabling the gitlab CI runner for vimjas reform-debian-packages branch? Thank you! https://source.mnt.re/vimja/reform-debian-packages/ | 10:58 |
| kxtells | minute yes, I updated uboot last week | 11:45 |
| + paperManu (~paperManu@107.159.15.124) | 12:12 | |
| + paperManu_ (~paperManu@107.159.15.124) | 12:28 | |
| + mjw (~mjw@gnu.wildebeest.org) | 13:03 | |
| minute | josch: i enabled some runner for vimja yesterday but maybe the wrong project? or different project? | 13:27 |
| minute | (on the phone until i reach office) | 13:27 |
| minute | kxtells: ok. then it's probably the crystal or the rp2040. we can warranty exchange the kbd. or if you're still up for an experiment, i can port the new usb report diffing code over to the pocket kbd and you can tell me if that makes a difference | 13:28 |
| kxtells | I am up for both! I enjoy experimenting :-D was thinking on sniffing the bus too to check for inconsistencies. | 13:32 |
| kxtells | at the same time, it is my daily driver and the keyboard crashes are a bit annoying | 13:32 |
| chorc | Hi everyone, I have a stupid Q: very happy about improved wifi performance with new Pocket adapter board, is it available for Reform Classic? I'm kinda eyeing updated Classic, and I need the slot for LTE modem there | 13:35 |
| minute | kxtells: yes, you'll get a replacement in any case. do you have a tool for usb sniffing? | 13:52 |
| minute | chorc: you have a PCIe LTE modem? | 13:53 |
| minute | mPCIe that is | 13:53 |
| kxtells | I don't have a specific USB sniffing tool, but I have a logic analyzer that I could plug into the cable. It is a 400Mhz analyzer so I am not sure if it would be enough | 13:55 |
| josch | minute: something is still not right with vimja's runner. I just pressed the retry button again and the jobs are still "stuck": https://source.mnt.re/vimja/reform-debian-packages/-/pipelines/5131/builds | 14:10 |
| josch | (i assume it's the runner but if you enabled those maybe it's something else?) | 14:11 |
| josch | chorc: the main problem with using a lte modem with the classic is, that the mini-pcie slot of the classic does not have a USB signal and most LTE/WWAN cards want USB | 14:12 |
| josch | chorc: i want the same thing but because of this limitation I think I rather need to figure out how to add another usb hub into my classic | 14:13 |
| josch | chorc: but if you find a LTE modem which does *not* require USB but works with plain PCI-E I'd be *very* interested in it :) | 14:13 |
| chorc | minute, josch: I have some kind of Quectel, the one that goes into Pocket slot on motherboard, would it work on Classic? | 14:17 |
| chorc | sorry, I didn't research it enough | 14:17 |
| minute | josch: ok, i probably didn't do the right thing then. doing the right thing now | 14:17 |
| minute | josch: should work now i think? | 14:18 |
| Zaba | even if a model supports PCIe you might need to use the USB interface to switch it into PCIe mode. they’re not really made to work conveniently | 14:18 |
| Zaba | modem* | 14:18 |
| minute | kxtells: 400mhz should be enough as we're running usb with 12mhz IIRC | 14:18 |
| chorc | so how people add modems to Classic Reforms? via USB adapter board? | 14:19 |
| minute | chorc: no, there's no USB on the classic on those slots unfortunately | 14:19 |
| minute | one could design a mPCIe pcb with a M.2 slot _and_ a PCIe to USB2 converter to fix this | 14:19 |
| minute | usb2 root hub that is | 14:19 |
| chorc | something like thist? https://www.amazon.com/Adapter-mPCIe-Ports-USB2-0-Converter/dp/B0BKQ5K54H | 14:21 |
| - mjw (QUIT: Ping timeout: 264 seconds) (~mjw@gnu.wildebeest.org) | 14:23 | |
| josch | minute: thank you, it works now! | 14:26 |
| josch | chorc: yes, i have a usb hub inside my classic reform to which i attached the lte modem: https://community.mnt.re/t/adding-lte-modem-and-other-usb-devices/1215/20 | 14:29 |
| chorc | josch: oh, okay, but you have that hub in nvme, and the storage in mpcie? (nice lineup of 18650s!) | 14:31 |
| josch | chorc: no, the hub is at the very far right, the green pcb inside the pink plastic with five white jst connectors | 14:31 |
| josch | chorc: i made "more" usb by adding a hub to the usb connector where the keyboard would go and then plug the keyboard into the hub | 14:32 |
| chorc | I see, but that's for some older CPU module, right? the new carrier board has one more mpcie, are you using it for wifi? | 14:34 |
| chorc | so much space in the classic reform! | 14:35 |
| josch | chorc: yes, this is a311d and with that you do not have the extra pcie+usb that you have with rcore R4 but you also don't need it because in contrast to rk3588, a311d has wifi+bt built-in | 14:36 |
| josch | when i upgrade my a311d to rk3588 (not done yet) i will *loose* wifi+bt and need a replacement for it. That's what the wifi/bt panel for the pocket can be used for. | 14:37 |
| josch | but on the motherboard itself i still just have pcie and not usb on the mini-pcie port. | 14:37 |
| chorc | got it | 14:37 |
| minute | josch: i merged reform-power-daemon to reform-tools but will test it now on pocket together with that new keyboard feature (brightness ctrl) | 14:46 |
| + spew (~spew@user/spew) | 14:51 | |
| - kxtells (QUIT: Quit: Client closed) (~kxtells@user/kxtells) | 14:53 | |
| josch | minute: sweet -- once you give the go i will create a new release | 14:55 |
| * Guest3342 -> mjw | 15:01 | |
| + kxtells5 (~kxtells@user/kxtells) | 15:58 | |
| + plomlompom0 (~plom@139.59.138.73) | 16:02 | |
| + kxtells_ (~AdminUser@79.117.156.55) | 16:03 | |
| - kxtells5 (QUIT: Client Quit) (~kxtells@user/kxtells) | 16:03 | |
| - kxtells_ (QUIT: Changing host) (~AdminUser@79.117.156.55) | 16:06 | |
| + kxtells_ (~AdminUser@user/kxtells) | 16:06 | |
| - kxtells_ (PART: !!unknown attribute: msg!!) (~AdminUser@user/kxtells) | 16:06 | |
| + kxtells_ (~AdminUser@user/kxtells) | 16:07 | |
| * kxtells_ -> kxtells | 16:07 | |
| - spew (QUIT: *.net *.split) (~spew@user/spew) | 16:08 | |
| - paperManu_ (QUIT: *.net *.split) (~paperManu@107.159.15.124) | 16:08 | |
| - Gooberpatrol66 (QUIT: *.net *.split) (~Gooberpat@user/gooberpatrol66) | 16:08 | |
| - plomlompom (QUIT: *.net *.split) (~plom@139.59.138.73) | 16:08 | |
| - manis (QUIT: *.net *.split) (01a66df340@185.72.67.185) | 16:08 | |
| - cobra (QUIT: *.net *.split) (~cobra@user/Cobra) | 16:08 | |
| - qbit (QUIT: *.net *.split) (~qbit@user/qbit) | 16:08 | |
| - rwa_ (QUIT: *.net *.split) (0a82deb4eb@2a03:6000:1812:100::41b) | 16:08 | |
| + paperManu_ (~paperManu@107.159.15.124) | 16:11 | |
| + spew (~spew@user/spew) | 16:11 | |
| + Gooberpatrol66 (~Gooberpat@user/gooberpatrol66) | 16:11 | |
| + cobra (~cobra@user/Cobra) | 16:11 | |
| + qbit (~qbit@user/qbit) | 16:11 | |
| + rwa_ (0a82deb4eb@2a03:6000:1812:100::41b) | 16:11 | |
| - Gooberpatrol66 (QUIT: Max SendQ exceeded) (~Gooberpat@user/gooberpatrol66) | 16:11 | |
| + Gooberpatrol66 (~Gooberpat@user/gooberpatrol66) | 16:11 | |
| minute | ok, i need to programmatically find the correct hidraw, otherwise mess | 16:11 |
| minute | ah, /dev/input/by-id | 16:12 |
| josch | we should do the same in reform-standby | 16:14 |
| josch | currently we just do for d in /dev/hidraw*; do | 16:14 |
| minute | reform-standby... haven't heard that name in a long time! | 16:14 |
| josch | heh :) | 16:14 |
| josch | dear usb-c enthusiasts, i have a question: does it matter which way round i plug in a usb-c cable? I have here a webcam which, if i plug it one way round into my pocket then it is listed as 5000MiB/s in "lsub -t" but if i plug it in the other way round it's listed as 480 MiB/s | 16:19 |
| josch | is that plausible? | 16:19 |
| minute | josch: totally | 16:19 |
| josch | oh no | 16:19 |
| minute | josch: it means there's a problem with usb superspeed muxing somewhere | 16:20 |
| [tj] | it shouldn't but it does | 16:20 |
| minute | josch: see here https://upload.wikimedia.org/wikipedia/commons/0/07/USB_Type-C_Receptacle_Pinout.svg | 16:20 |
| [tj] | the usb-c port has to detect orientation and configre rx/tx the right way | 16:20 |
| [tj] | this is left to "software" in the usb-c standard | 16:20 |
| josch | hah okay | 16:20 |
| minute | in reality it's mostly done by a hardware mux. | 16:20 |
| minute | chip | 16:21 |
| minute | orientation is detected by comparing the CC1 and CC2 voltages | 16:21 |
| josch | at least i found this out by accident... it was very frustrating to have different resolutions on this webcam every time i plugged things together | 16:21 |
| [tj] | nah, it is a voltage divider on cc1 or cc2, only cc1 is carried by a cable | 16:21 |
| josch | thank you for confirming! | 16:21 |
| minute | and then a mux switches one tx/rw pair (actually a pair of pairs) to one side or the otehr | 16:21 |
| minute | [tj]: yes, but in practice it's done by comparing the voltages :D | 16:22 |
| [tj] | :D | 16:22 |
| [tj] | I have a debug mode adapter activator I made on my desk so I'm far too into this right now | 16:22 |
| [tj] | I'd like to be writing device drivers for something more sensible | 16:22 |
| minute | on the pocket we can read these voltages from fusb302 | 16:22 |
| minute | and on the camera it's done with a physical comparator chip | 16:23 |
| minute | (for example) | 16:23 |
| minute | on next we actually have to do the host-side lane switching in software, because fusb302 doesn't have a DIR output | 16:23 |
| minute | well, in firmware | 16:23 |
| minute | in the rp2350 | 16:23 |
| minute | btw in powertop the number 2 process is always usb_giveback_urb_hh... i need to check if that's better on the "report diffing" keyboard fw | 16:25 |
| minute | aha, on my next this process is basically gone from powertop | 16:27 |
| + wielaard (~mjw@gnu.wildebeest.org) | 16:42 | |
| rwa_ | reg usb-c: https://images.fotowolke.net/2025/11/18/KJov.jpg :D | 16:44 |
| minute | hehe | 16:45 |
| minute | ch: hmm when i add a tag to pocket-reform-keyboard-fw and build and flash locally, like 20251118, this ends up as 1118 in the keyboard, is that on purpose? | 16:46 |
| minute | hmm, maybe i should remove the hard fw version check from reform-power-daemon, this is just too painful to test | 16:47 |
| josch | rwa_: high quality meme :) | 16:57 |
| minute | wow, managed to soft-brick my pocket keyboard | 16:57 |
| minute | with picotool flash, it was half done and aborted due to "usb connection problem" or sth. then i did the wrong thing (standby switch toggle) | 16:58 |
| minute | fell into my own trap! | 16:58 |
| + colinsane (~colinunin@97-113-95-252.tukw.qwest.net) | 17:24 | |
| + Guest8 (~Guest8@61.74.148.82) | 17:30 | |
| - Guest8 (QUIT: Client Quit) (~Guest8@61.74.148.82) | 17:30 | |
| minute | josch: reform-tools with reform-power-daemon now releasable, as soon as you have the time! | 17:40 |
| ch | minute: "1118" is the fallback for local/untagged builds (11=month 18=day) | 17:49 |
| ch | minute: CI sets an envvar with the tag to make an "official" build | 17:49 |
| minute | ch: ah i see | 17:51 |
| minute | ch: thx | 17:51 |
| minute | is the gitlab UI quite slow or just here? | 17:54 |
| - wielaard (QUIT: Ping timeout: 244 seconds) (~mjw@gnu.wildebeest.org) | 17:56 | |
| minute | ch: lvfs doesn't seem to pick up my new tag, any idea what i could have done wrong? https://source.mnt.re/reform/pocket-reform/-/tree/main/pocket-reform-3d-printed-parts?ref_type=heads | 18:01 |
| ch | wrong link? | 18:25 |
| ch | i think the last time it needed a build trigger after the tag was pushed | 18:25 |
| ch | i really need to debug the ci config for that | 18:25 |
| ch | the gitlab ui is super slow | 18:29 |
| ch | maybe getting AI-dosed? | 18:29 |
| josch | minute: done and uploaded to unstable. This release includes a new binary package named plymouth-theme-reform which contains the monobar theme and the reform-y2k themee | 18:47 |
| minute | plomlompom0: take note, this reform-tools release, once rolled out, should alleviate most of the "spontaneous reset" issues for pocket reform rk3588 users | 18:53 |
| minute | josch: many thanks! and cool that the plymouth themes made it | 18:53 |
| minute | ch: sorry yes, wrong link | 18:54 |
| minute | ch: ah yes, so it's not only me @ gitlab. i'll check if i can see something in the log, or maybe we've been throttled by netcup | 18:54 |
| minute | > Traffic (current month) 1,754 GiB | 18:59 |
| minute | hm, network traffic is quite little | 19:00 |
| minute | funny stuff in dmesg on that machine | 19:01 |
| minute | > [2606137.881272] virtio_balloon virtio2: Out of puff! Can't get 1 pages | 19:01 |
| minute | (upgrading gitlab and doing health checks and will also power cycle the machine) | 19:06 |
| - pomel0 (QUIT: Read error: Connection reset by peer) (~pomel0@user/pomel0) | 19:23 | |
| + pomel0 (~pomel0@user/pomel0) | 19:27 | |
| minute | ah we're getting a lot of hits from googlebot | 19:41 |
| minute | also, for each request, in production.log > GraphQL-Ruby's complexity cost system is getting some "breaking fixes" in a future version. See the migration notes at https://graphql-ruby.org/api-doc/2.5.11/GraphQL/Schema.html#complexity_cost_calculation_mode_for-class_method | 19:42 |
| minute | lol | 19:42 |
| minute | mhm https://gitlab.com/gitlab-org/gitlab/-/issues/567326 | 19:44 |
| minute | uff, there are 136354 hits from Googlebot in production_json.log | 19:45 |
| minute | since yesterday | 19:45 |
| + wielaard (~mjw@gnu.wildebeest.org) | 19:58 | |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:20) | 20:14 | |
| - mjw (QUIT: Killed (tantalum.libera.chat (Nickname regained by services))) (~mjw@2001:1c06:2486:a800:7602:5eff:dc71:a72c) | 20:59 | |
| * wielaard -> mjw | 20:59 | |
| + Guest2191 (~mjw@2001:1c06:2486:a800:7602:5eff:dc71:a72c) | 20:59 | |
| cwebber | minute: wow | 21:18 |
| cwebber | I wonder if actually from Google or if from some non-Google AI company | 21:18 |
| minute | not sure... said Googlebot in the user agent | 21:18 |
| minute | and IP range of google | 21:18 |
| cwebber | maybe Google is trying to compete with the other AI scrapers by giving up on being polite | 21:19 |
| cwebber | Google, for all its faults, at least used to be a relatively polite crawler for all it did | 21:19 |
| cwebber | but the old norms of the internet are falling apart thanks to greed :( | 21:20 |
| - spew (QUIT: Quit: WeeChat 4.6.3) (~spew@user/spew) | 22:03 | |
| + TechnoWizard (~TechnoWiz@user/TechnoWizard) | 22:08 | |
| - TechnoWizard (QUIT: Remote host closed the connection) (~TechnoWiz@user/TechnoWizard) | 22:10 | |
| - gustav25 (QUIT: Quit: Quit) (~gustav@c-78-82-54-254.bbcust.telenor.se) | 22:15 | |
| - GNUmoon2 (QUIT: Remote host closed the connection) (~GNUmoon@gateway/tor-sasl/gnumoon) | 22:59 | |
| + GNUmoon2 (~GNUmoon@gateway/tor-sasl/gnumoon) | 22:59 | |
| - mjw (QUIT: Ping timeout: 264 seconds) (~mjw@gnu.wildebeest.org) | 23:18 | |
| amospalla | https://github.com/amospalla/reform/tree/main/mleds | 23:37 |
| amospalla | I wrote a program to play things on the keyboard leds | 23:37 |
| amospalla | If somebody wants to play with it. | 23:38 |
| - chomwitt (QUIT: Ping timeout: 250 seconds) (~chomwitt@2a02:85f:9a5f:900:1ac0:4dff:fedb:a3f1) | 23:40 | |
| josch | ch: the intex ax210 works \o/ | 23:47 |
| josch | now i need to find space for the antennas XD | 23:47 |
| amospalla | I was trying to do some quick script and ended writing this :/ | 23:51 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
