| josch | ch: ah this also explains why fex wants wine with another stack of patches on top | 00:28 |
|---|---|---|
| minute | ah neat @ WoA. i think this is kind of what hangover does? | 01:21 |
| minute | hangover also has precompiled dlls with fex + box64 in them | 01:22 |
| - mjw (QUIT: Ping timeout: 240 seconds) (~mjw@gnu.wildebeest.org) | 01:25 | |
| + bkeys (~Thunderbi@h69.128.19.98.dynamic.ip.windstream.net) | 01:25 | |
| - spew (QUIT: Quit: nyaa~) (~spew@user/spew) | 01:40 | |
| + chomwitt (~chomwitt@2a02:85f:9a42:1100:1ac0:4dff:fedb:a3f1) | 02:20 | |
| - chomwitt (QUIT: Ping timeout: 244 seconds) (~chomwitt@2a02:85f:9a42:1100:1ac0:4dff:fedb:a3f1) | 02:36 | |
| - paperManu (QUIT: Ping timeout: 240 seconds) (~paperManu@146.71.9.156) | 02:56 | |
| + paperManu (~paperManu@146.71.9.156) | 02:58 | |
| - pomel0 (QUIT: Ping timeout: 246 seconds) (~pomel0@user/pomel0) | 02:59 | |
| + pomel0 (~pomel0@user/pomel0) | 03:00 | |
| - paperManu (QUIT: Ping timeout: 240 seconds) (~paperManu@146.71.9.156) | 03:06 | |
| - nsc (QUIT: Ping timeout: 240 seconds) (~nicolas@i5C74DE29.versanet.de) | 03:11 | |
| + nsc (~nicolas@i5C74DD9D.versanet.de) | 03:13 | |
| + n0blenote (~n0blenote@user/n0blenote) | 03:56 | |
| ^alex | feeling like replacing all of the philips screws in our pocket reform with torx out of persnicketitude | 04:24 |
| kfx | switch to rivets. that's commitment. | 04:36 |
| + natalie (~natalie@user/natalie) | 04:37 | |
| + kensanata7 (~alex@user/kensanata) | 04:38 | |
| - Chwoka (QUIT: Ping timeout: 240 seconds) (~Chwoka@216.73.127.21) | 04:40 | |
| - kensanata (QUIT: Ping timeout: 240 seconds) (~alex@user/kensanata) | 04:40 | |
| - natalie- (QUIT: Ping timeout: 240 seconds) (~natalie@user/natalie) | 04:40 | |
| * kensanata7 -> kensanata | 04:40 | |
| - _justin_kelly713 (QUIT: Read error: Connection reset by peer) (~justinkel@user/justin-kelly/x-6011154) | 04:44 | |
| + _justin_kelly71 (~justinkel@user/justin-kelly/x-6011154) | 04:44 | |
| - AshCurry (QUIT: Ping timeout: 246 seconds) (~AshCurry@user/AshCurry) | 04:44 | |
| + natalie- (~natalie@user/natalie) | 04:46 | |
| + AshCurry (~AshCurry@user/AshCurry) | 04:47 | |
| + yqshao6 (~yqshao@user/yqshao) | 04:49 | |
| - natalie (QUIT: Ping timeout: 246 seconds) (~natalie@user/natalie) | 04:50 | |
| - aelius (QUIT: Ping timeout: 246 seconds) (~aelius@user/aelius) | 04:50 | |
| - sigrid (QUIT: Ping timeout: 246 seconds) (~sigrid@ftrv.se) | 04:50 | |
| - shdw_ (QUIT: Ping timeout: 246 seconds) (~shdw@static.218.156.216.95.clients.your-server.de) | 04:50 | |
| - yqshao (QUIT: Ping timeout: 246 seconds) (~yqshao@user/yqshao) | 04:50 | |
| - angelwood (QUIT: Ping timeout: 246 seconds) (~angelwood@user/angelwood) | 04:50 | |
| + angelwood_ (~angelwood@user/angelwood) | 04:50 | |
| * yqshao6 -> yqshao | 04:50 | |
| - n0blenote (QUIT: Remote host closed the connection) (~n0blenote@user/n0blenote) | 04:50 | |
| + n0blenote (~n0blenote@user/n0blenote) | 04:51 | |
| * angelwood_ -> angelwood | 04:51 | |
| + shdw (~shdw@static.218.156.216.95.clients.your-server.de) | 04:55 | |
| + sigrid (~sigrid@ftrv.se) | 04:55 | |
| - kensanata (QUIT: Read error: Connection reset by peer) (~alex@user/kensanata) | 04:56 | |
| + kensanata (~alex@user/kensanata) | 04:59 | |
| + aelius (~aelius@user/aelius) | 04:59 | |
| - AshCurry (QUIT: Ping timeout: 240 seconds) (~AshCurry@user/AshCurry) | 05:03 | |
| - sigrid (QUIT: Ping timeout: 240 seconds) (~sigrid@ftrv.se) | 05:03 | |
| + AshCurry (~AshCurry@user/AshCurry) | 05:04 | |
| + sigrid (~sigrid@ftrv.se) | 05:07 | |
| - Ar|stote|is (QUIT: Quit: https://quassel-irc.org - Chat comfortably. Anywhere.) (~linx@149.210.3.83) | 05:09 | |
| + Ar|stote|is (~linx@149.210.3.83) | 05:09 | |
| n0blenote | hiya all - I was wondering if folks knew about the crowdsupply order? is the plan for the preorder to be rolling (so new orders have a set buffer to allow for manufacturing), or is end of march a set date for all backers? sorry, i was struggling to find details! | 07:53 |
| * n0blenote -> noblenote | 08:06 | |
| - noblenote (QUIT: Quit: noblenote) (~n0blenote@user/n0blenote) | 08:12 | |
| + n0blenote (~n0blenote@user/n0blenote) | 08:13 | |
| - n0blenote (QUIT: Ping timeout: 260 seconds) (~n0blenote@user/n0blenote) | 08:39 | |
| + n0blenote (~n0blenote@user/n0blenote) | 08:42 | |
| + Chwoka (~Chwoka@216.73.127.21) | 08:52 | |
| josch | kfx: don't forget to fill the inside with epoxy resin for the true apple experience :) | 09:15 |
| + mjw (~mjw@gnu.wildebeest.org) | 09:42 | |
| - Chwoka (QUIT: Ping timeout: 256 seconds) (~Chwoka@216.73.127.21) | 09:44 | |
| lidstah | Hi all, I'm wondering if anyone tried the raspberry pi's NVMe SSDs with the reform classic/pocket - they're waaaaay cheaper right now than other brands (probably due to large stocks made before this winter) | 09:53 |
| - Ar|stote|is (QUIT: Quit: No Ping reply in 180 seconds.) (~linx@149.210.3.83) | 10:42 | |
| + Ar|stote|is (~linx@149.210.3.83) | 10:43 | |
| minute | lidstah: no, but i was planning to try them | 10:45 |
| minute | n0blenote: rolling more likely | 10:46 |
| lidstah | minute: I think I gonna get one 256GB (it's ~40€ rn, so cheaper than a brand new 128GB (!) NVMe which is ~50€) for when I get my Reform. If it doesn't work, it will go in my steamdeck (64GB eMMC LCD, the cheapest one back in the days. SD card works surprisingly well even for stuff like Skyrim or No Man's Sky) | 10:51 |
| - gianmarcogg03 (QUIT: Ping timeout: 240 seconds) (~quassel@user/gianmarcogg03) | 11:01 | |
| * voltaire28 -> voltaire28_ | 11:45 | |
| - lidstah (QUIT: Ping timeout: 252 seconds) (~lidstah@gateway/tor-sasl/lidstah) | 11:49 | |
| + lidstah (~lidstah@gateway/tor-sasl/lidstah) | 11:51 | |
| - n0blenote (QUIT: Ping timeout: 240 seconds) (~n0blenote@user/n0blenote) | 11:57 | |
| - mjw (QUIT: Ping timeout: 240 seconds) (~mjw@gnu.wildebeest.org) | 12:07 | |
| - Guest7711 (QUIT: Quit: Leaving) (~mjw@2001:1c06:2486:a800:a09a:fc1c:5a8:e74d) | 12:21 | |
| ch | minute: probably, checking out hangover in detail is on my endless todo list ;) | 12:25 |
| - pomel0 (QUIT: Ping timeout: 240 seconds) (~pomel0@user/pomel0) | 12:46 | |
| + paperManu (~paperManu@146.71.9.156) | 12:48 | |
| - paperManu (QUIT: Ping timeout: 246 seconds) (~paperManu@146.71.9.156) | 13:09 | |
| - elb (QUIT: Remote host closed the connection) (~elb@2600:4041:6682:ea00:5e78:effb:9afa:62f1) | 13:17 | |
| + elb (~elb@2600:4041:6682:ea00:e4a:8e55:9f19:ddfa) | 13:17 | |
| - gordon1 (QUIT: Quit: WeeChat 4.7.2) (~gordon1@citadel.niflheim.info) | 13:19 | |
| + gordon1 (~gordon1@citadel.niflheim.info) | 13:19 | |
| + mjw (~mjw@2001:1c06:2486:a800:a09a:fc1c:5a8:e74d) | 13:31 | |
| jah | is there an alternative way to booting up the pocket than using hyper+enter on the keyboard? the reason I ask is because I'm going to hack on the keyboard firmware and I fear that I could run into a state where I do mess-up the power-on code (I'm trying really hard not too... but mistakes can always happen...) | 13:48 |
| ch | a complicated thing that should work (maybe there are others): enable the ACM_ENABLED build flag in the sysctl firmware, attach the sysctl usb lines to another machine. then you can use the emulated-uart to send a '1' | 14:01 |
| ch | maybe also an option: enable FACTORY_MODE, then you can trigger the power on by a full off/on cycle with the side switch. i think that has no other effects, but not sure | 14:02 |
| jah | okay, I'll dig into this later for now I'll just be super-careful and always flash back the original firmware after doing my tests ... I guess if I can still open the hyper+enter menu while the computer is on and my custom firmware is running, I should be pretty confident it should keep on working when powered-off | 14:23 |
| jah | btw I think I found an inconsistency, or maybe I don't understand how the time is computing in there | 14:23 |
| jah | but here, it says "2 seconds" but the magic number represents 1000ms: https://source.mnt.re/reform/pocket-reform/-/blob/main/pocket-reform-keyboard-fw/pocket-hid/src/main.c?ref_type=heads#L414-415 | 14:23 |
| jah | when using it, it feels like it is indeed 1 second unlike what is documented | 14:24 |
| jah | s/computing/computed/ | 14:24 |
| + wielaard (~mjw@gnu.wildebeest.org) | 14:29 | |
| ch | i agree with your analysis | 14:34 |
| + paperManu (~paperManu@modemcable141.205-200-24.mc.videotron.ca) | 14:52 | |
| - paperManu (QUIT: Client Quit) (~paperManu@modemcable141.205-200-24.mc.videotron.ca) | 14:56 | |
| + paperManu (~paperManu@modemcable141.205-200-24.mc.videotron.ca) | 15:02 | |
| + Chwoka (~Chwoka@216.73.127.21) | 15:24 | |
| + gianmarcogg03 (~quassel@user/gianmarcogg03) | 15:25 | |
| - jogu (QUIT: Remote host closed the connection) (~jogu@user/jogu) | 15:27 | |
| + jogu (~jogu@user/jogu) | 15:28 | |
| + siviq (~siviq@user/siviq) | 16:41 | |
| - siviq (QUIT: Client Quit) (~siviq@user/siviq) | 16:43 | |
| - Ar|stote|is (QUIT: Quit: No Ping reply in 180 seconds.) (~linx@149.210.3.83) | 16:46 | |
| + Ar|stote|is (~linx@149.210.3.83) | 16:51 | |
| jah | I would be happy to submit a patch for this as I'm hoping to upstream my work eventually, the question though is: | 17:07 |
| jah | should fix the documentation/comment (to 1s) or the magic number (to 2s)? | 17:07 |
| jah | should we* | 17:07 |
| jah | also I found a small bug (I guess it only really happen when debugging): the variable `menu_state` is not reset when doing `reset_keyboard_state` which can lead to an incoherent state | 17:24 |
| - Ar|stote|is (QUIT: Ping timeout: 250 seconds) (~linx@149.210.3.83) | 18:41 | |
| - aloo_shu (QUIT: Ping timeout: 264 seconds) (~aloo_shu@85.51.16.241) | 18:43 | |
| + Ar|stote|is (~linx@149.210.3.83) | 18:47 | |
| ^alex | kfx, we don't have a rivet gun :( | 18:54 |
| elb | jah: welcome to keyboard hacking! What are you doing? | 19:10 |
| elb | uh the menu state thing is my fault | 19:10 |
| elb | also I agree that that's a bug | 19:11 |
| + aloo_shu (~aloo_shu@85.51.17.216) | 19:14 | |
| ^alex | urgh our keyboard firmware is incredibly out of date because we have a custom matrix.h >.> | 19:19 |
| josch | same XD | 19:19 |
| josch | ^alex: there are ideas to allow for persistent settings in the (pocket) keyboard firmware -- in the long run, maybe that's how things should get customized | 19:20 |
| ^alex | we would be somewhat interested in a Kaleidoscope port because reasons | 19:31 |
| ^alex | (it's the only keyboard firmware with "TopsyTurvy", lets us invert the sense of the shift keys for the number row) | 19:32 |
| jah | elb: thanks! okay cool, I'll propose a patch for that as well | 19:36 |
| jah | I just finished implementing a first working version of "mod_tap", so you can take a key and make it act as a modifier when hold | 19:36 |
| jah | I'm not 100% happy, yet I need to polish a few things, for example right now it is working with a delay (TAPPING_TERM is qmk) but I also want to make it work without delay just by checking if user is making a chord | 19:36 |
| * mjw -> Guest446 | 19:38 | |
| - Guest446 (QUIT: Killed (calcium.libera.chat (Nickname regained by services))) (~mjw@2001:1c06:2486:a800:a09a:fc1c:5a8:e74d) | 19:38 | |
| * wielaard -> mjw | 19:38 | |
| + Guest446 (~mjw@2001:1c06:2486:a800:a09a:fc1c:5a8:e74d) | 19:38 | |
| elb | jah: oh nice | 20:00 |
| elb | I was going to do some serious keyboard fw hacking over winter break, but winter break is burning up fast | 20:00 |
| elb | I started a branch to unify the v4 keyboard firmware with the pocket firmware, in part to backport the update suppression logic there | 20:01 |
| elb | (which I want to change significantly) | 20:01 |
| elb | but then, you know, *waves hands* | 20:01 |
| ch | popping off the yak stack: bisecting binutils | 20:04 |
| josch | nice, i'm bisecting linux today :) | 20:06 |
| jah | elb: yeah no worries :) I did plan to do that two months ago, so I'm well aware how such things can shift indefinitely in time! | 20:06 |
| jah | that is good to know, and at some point I should try to see if I get my change on the v4 keyboard firmware but as I don't have the hardware to test that might be risky... that update supression logic is more up-to-date in pocket or v4? | 20:06 |
| ch | minute: slightly cleaned up test branch is here: https://source.mnt.re/zeha/pocket-reform/-/tree/zeha/charging-c - if you want to give it a try. maybe dont leave it out of your sight tho? | 20:08 |
| elb | I don't have a v4 either, which is part of what slows me down | 20:10 |
| elb | and the update suppression is on the v4 branch | 20:10 |
| josch | elb: i have one hear without switches -- if you want to hack the firmware, i can send it over? | 20:10 |
| josch | *here | 20:11 |
| ch | josch: do you happen to know how debians u-boot package is built from git? | 20:12 |
| elb | josch: I was offered one from mnt at one point as well, if I wanted to tackle things, but I hate to do pre-commitment that way, because i'm notoriously unreliable ;-) | 20:13 |
| elb | what with job and kids | 20:13 |
| josch | ch: i am not involved in debian's u-boot packaging but vagrantc is usually online in a few hours | 20:13 |
| elb | josch: I do appreciate the offer | 20:13 |
| elb | at this point basically I feel like it's safe for me to create an integrated branch that definitely works on the pocket, but then at some point it'll need vetting on the v4, and we'll cross that bridge when we get there | 20:14 |
| josch | elb: no strings attached and i got mine from mnt for hacking purposes as well and would just send it on to be potentially useful for the next person :) | 20:14 |
| josch | just keep it in mind and ping me in case/if/when you would like to take me up on the offer :) | 20:15 |
| elb | sounds good! | 20:16 |
| josch | elb: job and kids also made the board lie around unused a long time before i got to do something with it XD | 20:16 |
| elb | I saw smoke signals of a pocket v2 keyboard at some point, I think; if that happens I'll get one of those and that will simplify things, too | 20:16 |
| elb | right now I live in fear at every firmware push ;-) | 20:16 |
| elb | (I mean obviously it's all recoverable ,but I tend to do things like self-hosted hacking sitting at a coffee shop) | 20:17 |
| ch | i guess the FACTORY_MODE sysctl fw might be nice in this case? dont have to open the case and all | 20:21 |
| ch | assuming you also bring an additional usb keyboard ;-) | 20:22 |
| elb | yeah I always have an additional keyboard | 20:28 |
| elb | and a screwdriver | 20:28 |
| elb | but still, there's painful and then there's painful ;-) | 20:28 |
| ch | josch: whats your linux bisect about? | 20:30 |
| ch | E: u-boot-qcom: statically-linked-binary [usr/lib/u-boot/qcm6490/u-boot.elf] | 20:42 |
| ch | very smart, lintian | 20:42 |
| + siviq (~siviq@user/siviq) | 20:59 | |
| - siviq (QUIT: Client Quit) (~siviq@user/siviq) | 21:02 | |
| josch | ch: ls1028a lost display support somewhere between trixie and unstable | 21:33 |
| ch | fun :( | 21:37 |
| josch | ch: nah this *is* fun compared to "it doesn't work at all" and then you have to figure out why :) | 21:38 |
| josch | hrm... FACTORY_MODE is a good idea... i don't think classic reform motherboard firmware has this | 21:38 |
| d_m | it's certainly nicer to debug issues with binary search than the alternative | 21:40 |
| josch | like the first-time platform bring-ups m.inute is doing all the time -- not my kind of fun at all XD | 21:43 |
| + voltaire28 (~jlafon@28.162.2.93.rev.sfr.net) | 21:43 | |
| josch | the last big issue was to fix suspend on imx8mq but at least it used to work and i just had to bisect compared to rk3588 right now, for example | 21:44 |
| d_m | right. | 21:44 |
| rick_ | ohh i see.. h264 decoding in mpv/ffmpeg need some patches. installed the packages from v4l2request branch. it works now <3 | 21:51 |
| rick_ | and it halfs the process cpu usage by 40-50% for my stream ^^ | 21:52 |
| josch | rick_: you installed packages from this MR? https://source.mnt.re/reform/reform-debian-packages/-/merge_requests/148 | 21:54 |
| rick_ | yep | 21:54 |
| josch | nice! thank you for the feedback! | 21:55 |
| rick_ | currently works like a charm <4 | 21:56 |
| rick_ | <3 | 21:56 |
| gordon1 | is it new set of patches with av1? | 21:58 |
| rick_ | and thanks for patching this. i use mpv to watch streams quite often ^^ | 21:58 |
| josch | gordon1: i integrated the last patch stack you linked me with the top commit | 21:59 |
| gordon1 | nice, good to know it works elsewhere | 21:59 |
| gordon1 | the only thing's missing now is vp9 iirc | 22:00 |
| gordon1 | not from ffmpeg but from kernel | 22:00 |
| gordon1 | yep https://gitlab.collabora.com/hardware-enablement/rockchip-3588/notes-for-rockchip-3588/-/blob/main/mainline-status.md | 22:02 |
| gordon1 | not even a patch ready yet | 22:02 |
| sigrid | you can try https://github.com/chewitt/linux/commits/rockchip-6.18.y/ but who knows if it's working | 22:10 |
| gordon1 | i'm not that desperate | 22:12 |
| rick_ | mhh or maybe it doesn't completly fine. at firest the process usage was lower (around 15%), without the patch it was around 40-50%.. after a few mins its now way higher.. currently 98% cpu process usage | 22:13 |
| rick_ | after killing mpv and starting again its back to 15%. but i do not see any errors in the console output | 22:13 |
| rick_ | jfi | 22:13 |
| rick_ | ill observe it a bit ^^ | 22:14 |
| gordon1 | rick_: if you run your video with mpv -v you can verify if it is using hwdev, you should see something like this: [vd] Using hardware decoding (v4l2request). | 22:16 |
| gordon1 | *hwdec | 22:18 |
| rick_ | yeah its using it. but i totally forgot about -v, maybe i see now more why it behaves like this | 22:18 |
| + notsoweird (~notsoweir@user/NotSoWeird) | 22:35 | |
| cararemixed | Now I want to 3d print magnetic demon horns 😈 https://usercontent.irccloud-cdn.com/file/Gh5jCRkk/IMG_7184.JPG | 22:37 |
| josch | haha would be glorious :D | 22:50 |
| - voltaire28 (QUIT: Ping timeout: 240 seconds) (~jlafon@28.162.2.93.rev.sfr.net) | 22:50 | |
| minute | cararemixed: ahaha | 22:51 |
| ch | hah | 22:52 |
| + n0blenote (~n0blenote@user/n0blenote) | 23:00 | |
| - n0blenote (QUIT: Client Quit) (~n0blenote@user/n0blenote) | 23:02 | |
| + n0blenote (~n0blenote@user/n0blenote) | 23:02 | |
| - n0blenote (QUIT: Client Quit) (~n0blenote@user/n0blenote) | 23:07 | |
| + n0blenote (~n0blenote@user/n0blenote) | 23:07 | |
| - paperManu (QUIT: Ping timeout: 250 seconds) (~paperManu@modemcable141.205-200-24.mc.videotron.ca) | 23:11 | |
| minute | chuckling at > Unsupported platform 'MNT Quasar 8550 2026-01-03' | 23:12 |
| josch | minute: do you mean "unsupported machine:" in the initramfs hook? | 23:16 |
| josch | minute: if yes, can you turn the machine name in the dtb to "Reform 2 with Quasar 8550 Module" before the wrong string ends up somewhere? :) | 23:17 |
| minute | josch: yeah no worries :D will do | 23:18 |
| rick_ | gordon1 yeah i think it was sth differnt.. runs fine now since over an hour ^^ | 23:31 |
| + paperManu (~paperManu@146.71.9.156) | 23:32 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
