| - kensanata (QUIT: Ping timeout: 244 seconds) (~alex@user/kensanata) | 00:05 | |
| + kensanata (~alex@user/kensanata) | 00:06 | |
| + paperManu (~paperManu@198.58.139.163) | 00:17 | |
| - mjw (QUIT: Ping timeout: 264 seconds) (~mjw@gnu.wildebeest.org) | 00:30 | |
| josch | elb: wow this is really cool! I never saw a DIY pocket reform kit :D | 00:30 |
|---|---|---|
| josch | minute: might be of interest to you ^ | 00:31 |
| josch | on fedi: https://social.sdf.org/@elb/114889457424988546 | 00:38 |
| josch | elb: i found your comment about the charger board against the anodized aluminum frame interesting -- i haven't thought about that | 00:41 |
| elb | yeah I meant to bring that up at some point, but had forgotten until I processed that video | 00:41 |
| elb | I also had some comments about the assembly guide, I need to read it again | 00:41 |
| elb | it's very good, but ther eare one or two places where orientation wasn't entirely clear to me | 00:42 |
| josch | we need the pdf of the assembly guide in git somewhere :) | 00:42 |
| + paperManu_ (~paperManu@198.58.139.163) | 00:42 | |
| elb | it took me alomst exactly two months to process the footage from that video, maybe I'll do a usage impressions video by ... 2026 ;-) | 00:44 |
| josch | elb: and about your later comment about binary blobs -- there actually is a blob-free SoM for the reform: the LS1028A | 00:44 |
| - kensanata (QUIT: Quit: Ping timeout (120 seconds)) (~alex@user/kensanata) | 00:45 | |
| + kensanata (~alex@user/kensanata) | 00:46 | |
| elb | yeah, I was actually aware of that, but it's not really in the same class as the rockchip, unfortunately | 00:46 |
| josch | indeed | 00:47 |
| elb | and someone is going to take me to task for the fact that Raspberry actually DOES design their own chips | 00:47 |
| elb | just not the main SoC | 00:47 |
| josch | about the slop of the cables: this actually already came in handy for the classic reform where a later addition of a display adapter was only possible because the display cable was "too long" for the original setup | 00:47 |
| minute | elb: i skipped into your video earlier and coincidentally it was a point where you said something very interesting about solder mask being the only isolation under the charger board vs the aluminum. good point | 00:47 |
| elb | yeah it definitely avoids having to re-cable for upgrades/changes, and also allows for some floating usage during debugging or similar | 00:48 |
| elb | I thnk the only cable that really bugged me was the OLED cable | 00:49 |
| elb | because the offset it has in such a short cable seems like it's got to be torquing on the cable/connectors in an unfortunate way | 00:49 |
| elb | but both the keyboard and the OLED display are tagged down pretty tight, so the lack of meaningful movement should help | 00:49 |
| elb | I love, love, LOVE how tight the keyboard is | 00:50 |
| minute | elb: yeah the oled cable annoys me sometimes :D | 00:50 |
| - violet (QUIT: Remote host closed the connection) (~vi@user/meow/violet) | 00:51 | |
| - kensanata (QUIT: Quit: Ping timeout (120 seconds)) (~alex@user/kensanata) | 01:06 | |
| + kensanata (~alex@user/kensanata) | 01:06 | |
| - kensanata (QUIT: Ping timeout: 260 seconds) (~alex@user/kensanata) | 01:11 | |
| + kensanata (~alex@user/kensanata) | 01:12 | |
| - kensanata (QUIT: Ping timeout: 260 seconds) (~alex@user/kensanata) | 01:19 | |
| vkoskiv | I wrote about some linux debugging: https://vkoskiv.com/mesa-regression/ | 01:21 |
| vkoskiv | Not reform-related, but if there are factual mistakes in there, I think people here could spot those :D | 01:21 |
| vkoskiv | I do have several reform-themed posts ideas on my list, but I only started collecting proper notes on things I worked on around ~2023 | 01:23 |
| + kensanata (~alex@user/kensanata) | 01:24 | |
| - kensanata (QUIT: Ping timeout: 256 seconds) (~alex@user/kensanata) | 01:30 | |
| - paperManu_ (QUIT: Ping timeout: 264 seconds) (~paperManu@198.58.139.163) | 01:31 | |
| + kensanata (~alex@user/kensanata) | 01:32 | |
| - kensanata (QUIT: Ping timeout: 256 seconds) (~alex@user/kensanata) | 01:42 | |
| + kensanata (~alex@user/kensanata) | 01:42 | |
| + paperManu_ (~paperManu@198.58.139.163) | 01:51 | |
| - kensanata (QUIT: Ping timeout: 246 seconds) (~alex@user/kensanata) | 01:56 | |
| + kensanata (~alex@user/kensanata) | 01:56 | |
| - paperManu_ (QUIT: Ping timeout: 248 seconds) (~paperManu@198.58.139.163) | 01:58 | |
| - kensanata (QUIT: Ping timeout: 240 seconds) (~alex@user/kensanata) | 02:13 | |
| + kensanata (~alex@user/kensanata) | 02:14 | |
| - kensanata (QUIT: Ping timeout: 264 seconds) (~alex@user/kensanata) | 02:19 | |
| + kensanata (~alex@user/kensanata) | 02:21 | |
| - kensanata (QUIT: Client Quit) (~alex@user/kensanata) | 02:26 | |
| + kensanata (~alex@user/kensanata) | 02:26 | |
| - kensanata (QUIT: Ping timeout: 244 seconds) (~alex@user/kensanata) | 02:45 | |
| + kensanata (~alex@user/kensanata) | 02:47 | |
| - kensanata (QUIT: Ping timeout: 246 seconds) (~alex@user/kensanata) | 02:52 | |
| + kensanata (~alex@user/kensanata) | 02:55 | |
| - jogu (QUIT: Remote host closed the connection) (~jogu@user/jogu) | 02:59 | |
| - kensanata (QUIT: Ping timeout: 264 seconds) (~alex@user/kensanata) | 03:01 | |
| + kensanata (~alex@user/kensanata) | 03:05 | |
| + jogu (~jogu@user/jogu) | 03:05 | |
| - kensanata (QUIT: Ping timeout: 240 seconds) (~alex@user/kensanata) | 03:09 | |
| + kensanata (~alex@user/kensanata) | 03:10 | |
| - kensanata (QUIT: Quit: Ping timeout (120 seconds)) (~alex@user/kensanata) | 03:15 | |
| + kensanata (~alex@user/kensanata) | 03:16 | |
| - kensanata (QUIT: Ping timeout: 256 seconds) (~alex@user/kensanata) | 03:21 | |
| + kensanata (~alex@user/kensanata) | 03:23 | |
| - paperManu (QUIT: Ping timeout: 240 seconds) (~paperManu@198.58.139.163) | 03:36 | |
| - kensanata (QUIT: Quit: Ping timeout (120 seconds)) (~alex@user/kensanata) | 03:45 | |
| + kensanata (~alex@user/kensanata) | 03:45 | |
| - kop316 (QUIT: Remote host closed the connection) (m-6f6zq6@static.138.159.90.157.clients.your-server.de) | 04:06 | |
| + kop316 (m-6f6zq6@static.138.159.90.157.clients.your-server.de) | 04:09 | |
| - kop316 (QUIT: Remote host closed the connection) (m-6f6zq6@static.138.159.90.157.clients.your-server.de) | 04:09 | |
| + kop316 (m-6f6zq6@static.138.159.90.157.clients.your-server.de) | 04:09 | |
| - kensanata (QUIT: Ping timeout: 256 seconds) (~alex@user/kensanata) | 04:32 | |
| + kensanata (~alex@user/kensanata) | 04:32 | |
| - kensanata (QUIT: Ping timeout: 260 seconds) (~alex@user/kensanata) | 04:46 | |
| elb | it looks to me like there are a few global variables in pocket-hid/src/remote.c that either aren't used at all (alert_low_battery, alert_blink) or are effectively actually local (term_x, term_y); is there something I'm missing, are there plans for those, or are those things I can clean up? | 04:46 |
| + kensanata (~alex@user/kensanata) | 04:46 | |
| elb | ohhhhh I see, term_x and term_y aren't actually local, they're ... kind of dangerous | 04:49 |
| - Ar|stote|is (QUIT: Quit: https://quassel-irc.org - Chat comfortably. Anywhere.) (~linx@149.210.6.130) | 05:05 | |
| - lidstah (QUIT: Remote host closed the connection) (~lidstah@gateway/tor-sasl/lidstah) | 05:13 | |
| + Ar|stote|is (~linx@149.210.6.130) | 05:14 | |
| + lidstah (~lidstah@gateway/tor-sasl/lidstah) | 05:14 | |
| - kensanata (QUIT: Ping timeout: 246 seconds) (~alex@user/kensanata) | 05:43 | |
| + kensanata (~alex@user/kensanata) | 05:44 | |
| - kensanata (QUIT: Ping timeout: 248 seconds) (~alex@user/kensanata) | 05:49 | |
| + kensanata (~alex@user/kensanata) | 05:50 | |
| - jacqueline (QUIT: Ping timeout: 248 seconds) (~jacquelin@user/jacqueline) | 05:56 | |
| + violet (~vi@user/meow/violet) | 05:56 | |
| + jacqueline (~jacquelin@user/jacqueline) | 06:11 | |
| - kensanata (QUIT: Ping timeout: 248 seconds) (~alex@user/kensanata) | 06:12 | |
| + kensanata (~alex@user/kensanata) | 06:13 | |
| + robin_ (~robin@user/terpri) | 06:36 | |
| - robin (QUIT: Remote host closed the connection) (~robin@user/terpri) | 06:36 | |
| - kensanata (QUIT: Ping timeout: 260 seconds) (~alex@user/kensanata) | 06:41 | |
| + kensanata (~alex@user/kensanata) | 06:41 | |
| - kensanata (QUIT: Client Quit) (~alex@user/kensanata) | 06:44 | |
| + kensanata (~alex@user/kensanata) | 06:44 | |
| - kensanata (QUIT: Ping timeout: 240 seconds) (~alex@user/kensanata) | 06:50 | |
| + kensanata (~alex@user/kensanata) | 06:51 | |
| - kensanata (QUIT: Ping timeout: 240 seconds) (~alex@user/kensanata) | 06:56 | |
| + kensanata (~alex@user/kensanata) | 06:57 | |
| - kensanata (QUIT: Ping timeout: 244 seconds) (~alex@user/kensanata) | 07:03 | |
| + kensanata (~alex@user/kensanata) | 07:03 | |
| - kensanata (QUIT: Ping timeout: 248 seconds) (~alex@user/kensanata) | 07:09 | |
| josch | vkoskiv: thank you for that article. It's interesting that you bisect with the arch packaging around mesa and not just plain mesa git. But I had similar issues with meson config option mess during the bisect. | 07:12 |
| + kensanata (~alex@user/kensanata) | 07:13 | |
| josch | vkoskiv: the reason i usually do not bisect the packaging is that the Debian packaging focusses around building *everything* and that makes each rebuild very slow and making the Debian build more minimal would be a lot of effort | 07:14 |
| josch | the second reason why i use plain mesa git is because of how easy it is to run anything against mesa from git instead of mesa from my system | 07:14 |
| josch | for example this was my last bisect recipe: | 07:15 |
| josch | meson setup build -Dgallium-drivers="['panfrost']" -Dvulkan-drivers="[]" -Dplatforms="['x11','wayland']" -Dglx-direct=true -Dgbm=enabled -Dprefix=/home/josch/git/mesa/prefix --buildtype=debugoptimized && time meson compile -C build && meson install -C build | 07:15 |
| josch | and to run an application against that build i then run something like this: | 07:15 |
| josch | LD_LIBRARY_PATH=/home/josch/git/mesa/prefix/lib/aarch64-linux-gnu/ gajim | 07:15 |
| josch | this is how i tracked down this bug in mesa in Debian stable: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1116427 | 07:18 |
| - kensanata (QUIT: Ping timeout: 246 seconds) (~alex@user/kensanata) | 07:18 | |
| josch | turns out that was already known (and fixed) upstream: https://gitlab.freedesktop.org/mesa/mesa/-/issues/13191 | 07:19 |
| + kensanata (~alex@user/kensanata) | 07:19 | |
| josch | the fix is now backported and will be part of the next Debian stable update: https://salsa.debian.org/xorg-team/lib/mesa/-/merge_requests/62 | 07:21 |
| + FirefoxDeHuk (~FirefoxDe@109.108.69.106) | 07:30 | |
| + shtrophic (~m-hrdsqi@user/shtrophic) | 07:33 | |
| - FirefoxDeHuk (QUIT: Quit: Client closed) (~FirefoxDe@109.108.69.106) | 07:35 | |
| - kensanata (QUIT: Quit: Ping timeout (120 seconds)) (~alex@user/kensanata) | 07:39 | |
| + kensanata (~alex@user/kensanata) | 07:39 | |
| - kensanata (QUIT: Quit: Ping timeout (120 seconds)) (~alex@user/kensanata) | 07:47 | |
| + kensanata (~alex@user/kensanata) | 07:48 | |
| + FirefoxDeHuk (~FirefoxDe@109.108.69.106) | 08:05 | |
| - FirefoxDeHuk (QUIT: Quit: Client closed) (~FirefoxDe@109.108.69.106) | 08:09 | |
| - pomel0 (QUIT: Remote host closed the connection) (~pomel0@user/pomel0) | 08:56 | |
| - kensanata (QUIT: Ping timeout: 246 seconds) (~alex@user/kensanata) | 09:00 | |
| + kensanata (~alex@user/kensanata) | 09:01 | |
| - kensanata (QUIT: Ping timeout: 260 seconds) (~alex@user/kensanata) | 09:06 | |
| + kensanata (~alex@user/kensanata) | 09:07 | |
| - kensanata (QUIT: Client Quit) (~alex@user/kensanata) | 09:11 | |
| + kensanata (~alex@user/kensanata) | 09:12 | |
| - kensanata (QUIT: Client Quit) (~alex@user/kensanata) | 09:16 | |
| + kensanata (~alex@user/kensanata) | 09:16 | |
| - kensanata (QUIT: Quit: Ping timeout (120 seconds)) (~alex@user/kensanata) | 09:25 | |
| + kensanata (~alex@user/kensanata) | 09:25 | |
| - kensanata (QUIT: Ping timeout: 248 seconds) (~alex@user/kensanata) | 09:32 | |
| + kensanata (~alex@user/kensanata) | 09:34 | |
| - kensanata (QUIT: Ping timeout: 260 seconds) (~alex@user/kensanata) | 09:44 | |
| + kensanata (~alex@user/kensanata) | 09:44 | |
| - kensanata (QUIT: Ping timeout: 256 seconds) (~alex@user/kensanata) | 09:55 | |
| + kensanata (~alex@user/kensanata) | 09:56 | |
| - kensanata (QUIT: Ping timeout: 248 seconds) (~alex@user/kensanata) | 10:01 | |
| + kensanata (~alex@user/kensanata) | 10:02 | |
| + mjw (~mjw@gnu.wildebeest.org) | 10:11 | |
| - kensanata (QUIT: Ping timeout: 248 seconds) (~alex@user/kensanata) | 10:13 | |
| + kensanata (~alex@user/kensanata) | 10:14 | |
| - kensanata (QUIT: Quit: Ping timeout (120 seconds)) (~alex@user/kensanata) | 10:40 | |
| + kensanata (~alex@user/kensanata) | 10:40 | |
| - kensanata (QUIT: Client Quit) (~alex@user/kensanata) | 10:45 | |
| + kensanata (~alex@user/kensanata) | 10:45 | |
| - kensanata (QUIT: Ping timeout: 264 seconds) (~alex@user/kensanata) | 11:01 | |
| + kensanata (~alex@user/kensanata) | 11:02 | |
| - kensanata (QUIT: Ping timeout: 240 seconds) (~alex@user/kensanata) | 11:06 | |
| + kensanata (~alex@user/kensanata) | 11:10 | |
| - kensanata (QUIT: Quit: Ping timeout (120 seconds)) (~alex@user/kensanata) | 11:29 | |
| + kensanata (~alex@user/kensanata) | 11:29 | |
| + FirefoxDeHuk (~FirefoxDe@109.108.69.106) | 11:35 | |
| - kensanata (QUIT: Ping timeout: 260 seconds) (~alex@user/kensanata) | 11:42 | |
| + kensanata (~alex@user/kensanata) | 11:42 | |
| - kensanata (QUIT: Ping timeout: 240 seconds) (~alex@user/kensanata) | 11:46 | |
| + kensanata (~alex@user/kensanata) | 11:47 | |
| - kensanata (QUIT: Ping timeout: 250 seconds) (~alex@user/kensanata) | 11:52 | |
| dormito | for the mnt reform pocket: what is the logic voltage level of the internal uart header? Is it module specific? I'm looking at the datasheet for the i.mx8m+, and it looks like it's uarts can operate in either 1.8v or 3.3v. I also am not seeing any thing about that on the SOM-carrier schematics. | 11:52 |
| + kensanata (~alex@user/kensanata) | 11:52 | |
| - kensanata (QUIT: Ping timeout: 264 seconds) (~alex@user/kensanata) | 11:57 | |
| + kensanata (~alex@user/kensanata) | 11:58 | |
| - kensanata (QUIT: Ping timeout: 264 seconds) (~alex@user/kensanata) | 12:06 | |
| + kensanata (~alex@user/kensanata) | 12:11 | |
| - kensanata (QUIT: Ping timeout: 246 seconds) (~alex@user/kensanata) | 12:15 | |
| + kensanata (~alex@user/kensanata) | 12:17 | |
| + paperManu (~paperManu@198.58.139.163) | 12:20 | |
| - kensanata (QUIT: Ping timeout: 252 seconds) (~alex@user/kensanata) | 12:21 | |
| + kensanata (~alex@user/kensanata) | 12:23 | |
| - kensanata (QUIT: Ping timeout: 240 seconds) (~alex@user/kensanata) | 12:28 | |
| grimmware | vkoskiv: oh I read that recently and really appreciated it :) | 12:29 |
| - FirefoxDeHuk (QUIT: Quit: Client closed) (~FirefoxDe@109.108.69.106) | 12:34 | |
| - mjw (QUIT: Killed (iridium.libera.chat (Nickname regained by services))) (~mjw@gnu.wildebeest.org) | 12:46 | |
| * Guest3333 -> mjw | 12:46 | |
| + Guest7901 (~mjw@gnu.wildebeest.org) | 12:47 | |
| + kensanata (~alex@user/kensanata) | 12:53 | |
| + FirefoxDeHuk (~FirefoxDe@109.108.69.106) | 12:53 | |
| - kensanata (QUIT: Ping timeout: 264 seconds) (~alex@user/kensanata) | 12:58 | |
| + kensanata (~alex@user/kensanata) | 12:59 | |
| - kensanata (QUIT: Client Quit) (~alex@user/kensanata) | 13:01 | |
| + kensanata (~alex@user/kensanata) | 13:01 | |
| + gustav25 (~gustav@c-78-82-53-228.bbcust.telenor.se) | 13:02 | |
| - kensanata (QUIT: Client Quit) (~alex@user/kensanata) | 13:06 | |
| + kensanata (~alex@user/kensanata) | 13:06 | |
| - kensanata (QUIT: Client Quit) (~alex@user/kensanata) | 13:11 | |
| + kensanata (~alex@user/kensanata) | 13:11 | |
| - kensanata (QUIT: Ping timeout: 248 seconds) (~alex@user/kensanata) | 13:15 | |
| + kensanata (~alex@user/kensanata) | 13:20 | |
| + erle (~erle@user/erle) | 13:27 | |
| - kensanata (QUIT: Ping timeout: 252 seconds) (~alex@user/kensanata) | 13:29 | |
| + kensanata (~alex@user/kensanata) | 13:30 | |
| - kensanata (QUIT: Ping timeout: 246 seconds) (~alex@user/kensanata) | 13:35 | |
| + kensanata (~alex@user/kensanata) | 13:36 | |
| - kensanata (QUIT: Ping timeout: 256 seconds) (~alex@user/kensanata) | 13:41 | |
| + kensanata (~alex@user/kensanata) | 13:41 | |
| - kensanata (QUIT: Ping timeout: 244 seconds) (~alex@user/kensanata) | 13:46 | |
| + kensanata (~alex@user/kensanata) | 13:47 | |
| - FirefoxDeHuk (QUIT: Quit: Client closed) (~FirefoxDe@109.108.69.106) | 13:47 | |
| - kensanata (QUIT: Ping timeout: 260 seconds) (~alex@user/kensanata) | 13:52 | |
| bremner | josch: if in doubt, blame gitlab | 13:54 |
| bremner | (re completion problems) | 13:56 |
| + kensanata (~alex@user/kensanata) | 13:57 | |
| - kensanata (QUIT: Ping timeout: 240 seconds) (~alex@user/kensanata) | 14:01 | |
| + kensanata (~alex@user/kensanata) | 14:03 | |
| - paperManu (QUIT: Ping timeout: 246 seconds) (~paperManu@198.58.139.163) | 14:05 | |
| - kensanata (QUIT: Ping timeout: 244 seconds) (~alex@user/kensanata) | 14:08 | |
| + kensanata (~alex@user/kensanata) | 14:09 | |
| - kensanata (QUIT: Quit: Ping timeout (120 seconds)) (~alex@user/kensanata) | 14:17 | |
| + kensanata (~alex@user/kensanata) | 14:17 | |
| - buckket (QUIT: Quit: buckket) (~buckket@vps.buckket.org) | 14:23 | |
| + buckket (~buckket@vps.buckket.org) | 14:24 | |
| - kensanata (QUIT: Quit: Ping timeout (120 seconds)) (~alex@user/kensanata) | 14:28 | |
| + kensanata (~alex@user/kensanata) | 14:28 | |
| + paperManu (~paperManu@modemcable141.205-200-24.mc.videotron.ca) | 14:40 | |
| grimmware | minute: I've got a workaround for the display sleep issue | 15:04 |
| grimmware | minute: https://source.mnt.re/grimmware/reform-debian-packages/-/commit/7b4a2f7c58b23dc9d076bb5b8a7c99c31e274fd1#eaf6e0dc1724268562cc82f38fff9d439f00aa41_668_666 | 15:04 |
| grimmware | ignore the stuff about MIPI_DSI_MODE_LPM, that's just me reverting a previous test | 15:05 |
| grimmware | but basically if init works fine but enable doesn't then just do a full init :P | 15:05 |
| grimmware | hence a workaround rather than a "fix" but it's kind of good enough | 15:05 |
| grimmware | the patch is weird because I'm directly editing the diff so I don't want to add or remove lines heh | 15:06 |
| grimmware | but essentially I'm booted into it now and it's working just fine | 15:06 |
| grimmware | this is based off fo your patch that fixes the flicker | 15:07 |
| grimmware | and I've just found a workaround for the qutebrowser bug I was facing too :3 | 15:23 |
| grimmware | a good day. | 15:23 |
| + pomel0 (~pomel0@user/pomel0) | 15:41 | |
| - plomtest (QUIT: Remote host closed the connection) (~plom@user/plomtest) | 15:56 | |
| + plomtest (~plom@user/plomtest) | 15:57 | |
| - qbit (QUIT: Remote host closed the connection) (~qbit@user/qbit) | 16:16 | |
| + qbit (~qbit@user/qbit) | 16:19 | |
| minute | grimmware: great | 16:33 |
| minute | grimmware: good idea @ full init, i can patch that into my patch (basically just the ...init() line instead of ...power_on(), right? | 16:34 |
| - chomwitt (QUIT: Ping timeout: 245 seconds) (~chomwitt@2a02:85f:9a5f:900:1ac0:4dff:fedb:a3f1) | 16:39 | |
| grimmware | yeah, just init instead of display_on | 17:06 |
| - Guest7901 (QUIT: Ping timeout: 260 seconds) (~mjw@gnu.wildebeest.org) | 17:43 | |
| minute | grimmware: hm, this looks like i should try init-in-enable in the dts for rk3588... i'll do that first | 17:52 |
| minute | (that workaround was for a311d initially) | 17:52 |
| grimmware | yeah that sounds like a much simpler solution | 17:53 |
| minute | not sure if it will just work though... | 17:53 |
| minute | well, i'll see in a few mins | 17:53 |
| grimmware | really glad I finally took the time to get my CI setup and reprepo going | 17:54 |
| minute | grimmware: awesome yeah | 17:57 |
| grimmware | josch: just mailed you the tracking number for the imx8mp | 18:00 |
| minute | aw, no more display for me | 18:06 |
| grimmware | minute: is that with the init-in-enable? | 18:10 |
| minute | grimmware: yeah, so it's not that simple. i'll try your change instead first | 18:11 |
| - colinsane (QUIT: Quit: bye) (~colinunin@97-113-71-58.tukw.qwest.net) | 18:14 | |
| josch | grimmware: hype! \o/ | 18:14 |
| grimmware | minute: yeah looks like you'll be missing the MIPI_DSI_MODE_LPM and picking up some imx8mp-specific clock setting | 18:16 |
| grimmware | otherwise the code is roughly functionally the same as my change | 18:16 |
| + colinsane (~colinunin@97-113-71-58.tukw.qwest.net) | 18:17 | |
| grimmware | if you want to reuse init-in-enable I think there's going to have to be something else in the device tree to indicate it's imx8mp | 18:17 |
| grimmware | I know you were talking about refactoring this code, I think structuring it around the matrix of potential hardware combinations is probably a good idea | 18:19 |
| grimmware | it means that as more modules come out or more screens have to be supported it becomes an additive thing | 18:20 |
| erle | josch did you find time to try out my proposed patch for the fn neo2 thing? | 18:20 |
| grimmware | my gut feeling is that this is where DRY can become painful | 18:20 |
| erle | grimmware DRY is not only about elegance and understanding, IMO it is mostly about maintenance. so if you never need to update something, then you are allowed to sin a little IMO ;) | 18:22 |
| erle | seriously offtopic, but my fav example for violating DRY is a magic card that has 3 lines of text and each of them says “target creature gets +1/+1.” … i have so far not found any way to make *that* DRY. | 18:25 |
| erle | (since you can choose the same target multiple times, you could give one creature +3/+3, or two creatures +2/+2 and +1/+1 respectively, or three creatures +1/+1 each. much more text if you try to DRY.) | 18:26 |
| grimmware | I mean clarity trumps all other concerns in this space I feel :) | 18:28 |
| erle | i also thought about refactoring the handling of the fn keys for the keyboard stuff to prevent bugs like the one i encountered with the fn lock switching keys around, but then i thought: nope, it is already “good enough”, there is nothing to be gained if the hw revision is not even produced anymore. | 18:29 |
| - pomel0 (QUIT: Ping timeout: 250 seconds) (~pomel0@user/pomel0) | 18:31 | |
| + pomel0 (~pomel0@user/pomel0) | 18:32 | |
| * robin_ -> robin | 18:32 | |
| ch | downloading artifacts for jobs/14554 now, then i'll see what the new kernel does | 18:40 |
| grimmware | ch: you're also v1 display right? | 18:41 |
| ch | yeah | 18:41 |
| grimmware | good things coming :) | 18:42 |
| minute | grimmware: you're right. my original plan was to make the driver as customizable as possible through dts so the quirks for each soc could be configured in the respective dts. | 18:43 |
| minute | but sometimes worse is better... or at least ok for now | 18:43 |
| minute | not saying your solution is "worse", i mean my current rube goldberg code in there | 18:43 |
| grimmware | I get what you mean :) | 18:43 |
| - erle (QUIT: Quit: K-lined) (~erle@user/erle) | 18:46 | |
| + andreas-e (~Andreas@2a02-8434-b6a3-e901-facc-8e87-8e54-890e.rev.sfr.net) | 18:48 | |
| minute | now installing grimmware linux... | 18:48 |
| grimmware | grimmux | 18:48 |
| minute | display is back on | 18:49 |
| grimmware | ... | 18:51 |
| + rwa_ (~rene@2001:9e8:3388:8300:1d49:28f3:f844:247c) | 18:51 | |
| minute | grimmware: works! | 19:02 |
| grimmware | EYYYYY | 19:02 |
| minute | grimmware: good job! now how to make this releasable. | 19:02 |
| ch | minute: so the kernel from !140 cannot wake up my display | 19:12 |
| minute | grimmware: i have a new theory. i think my patch just wasn't fully baked, and the 2nd init being required could explain it: mipi_dsi_dcs_exit_sleep_mode() is missing in the "wake up" path | 19:12 |
| minute | ch: ok, that was kind of expected, good. my panel is just a bit different then than the one you and grimmware have. | 19:12 |
| ch | they are from different batches, right? | 19:13 |
| minute | probably | 19:13 |
| - rwa_ (QUIT: Remote host closed the connection) (~rene@2001:9e8:3388:8300:1d49:28f3:f844:247c) | 19:20 | |
| grimmware | minute: what's incredibly frustrating is that that's what I was looking for (missing steps from enable) but because it's the *first* one I just wasn't seeing it | 19:34 |
| + Guest7901 (~mjw@gnu.wildebeest.org) | 19:34 | |
| grimmware | like, I'm kind of astounded I missed that | 19:34 |
| grimmware | minute: are you working on patching in the mipi_dsi_dcs_exit_sleep_mode()? | 19:37 |
| grimmware | goodness me the screen on this thing is very nice when it's not flickering | 19:38 |
| minute | grimmware: yeah i'm trying that in the branch, just to see if my hunch was correct or not | 19:39 |
| minute | grimmware: yes, non-flickering screen is much better :D | 19:39 |
| grimmware | nice | 19:39 |
| minute | ok i should also try some other experiment now... driving displayport directly from rk3588 | 19:40 |
| grimmware | I can test out your new branch tonight :) That is how I would like to spend my evening. | 19:40 |
| minute | neat. let me see if it works for me first though, to save you some frustration :D | 19:41 |
| grimmware | fair :) | 19:41 |
| grimmware | really optimistic about this charging board as well, it's been a good month for MNT | 19:41 |
| minute | now testing https://source.mnt.re/reform/reform-debian-packages/-/jobs/14594/artifacts/browse/changes-arm64/ | 20:06 |
| minute | grimmware: yay! | 20:06 |
| minute | grimmware: my patch doesn't work... funnily, the display backlight works though and it shows 1 line of colored pixel noise | 20:11 |
| minute | kind of like video stream is not turned on | 20:12 |
| grimmware | hmm. | 20:13 |
| minute | so yeah, maybe it just needs the full init (or some parts of it, but i think not worth it to dissect that now) after sleep+poweroff | 20:14 |
| grimmware | yeah the re-init is so fast anyway | 20:14 |
| elb | I have questions about the keyboard firmware and style and development practices | 20:15 |
| minute | on disp v1 it's kind of slow, but on v2 it's fast yeah | 20:15 |
| elb | I don't want to step on anyone's toes or make changes just for the sake of making changes, but there are some things I would change in here for just best embedded development practices, like removing unused variables, declaring (virtually all) globals static, etc. | 20:16 |
| elb | does that sound ... acceptable? | 20:16 |
| elb | like main.c and remote.c both have a fair number of global variables that are non-static but also not used in any other file, those are vulnerable to global variable unification problems in the future (but also many are namespaced, so ...) | 20:17 |
| minute | elb: sounds good to me! | 20:24 |
| minute | elb: meaning, these changes are welcome | 20:24 |
| elb | minute: awesome | 20:26 |
| ch | lmk if there's a new kernel to test | 20:27 |
| elb | would you rather one MR with a bunch of unrelated but mostly cleanup and simplfiication changes, or a number of smaller MRs with smaller and mostly-similar changes in each | 20:27 |
| ch | personal opinion: for cleanups its nice to get MRs out more quickly because there's usually no need to group them or wait for anything | 20:30 |
| minute | ch: just pushed a new version, pipeline will take a while of course. i did it like grimmware now but just removed the whole other branch in enable(). will need to test with imx8mp and dispv2 | 20:30 |
| minute | elb: i agree with ch on that | 20:30 |
| ch | btw, 39C3 yay/nay? | 20:31 |
| rick_ | if i may burst in here. i'm at 39c3 and bring my pocket reform with me <3 | 20:34 |
| - pomel0 (QUIT: Ping timeout: 240 seconds) (~pomel0@user/pomel0) | 20:39 | |
| + pomel0 (~pomel0@user/pomel0) | 20:44 | |
| - pomel0 (QUIT: Ping timeout: 240 seconds) (~pomel0@user/pomel0) | 20:48 | |
| minute | ch: yay! | 21:01 |
| minute | rick_: nice | 21:01 |
| minute | i just need to book hotel :0 | 21:01 |
| minute | (and be lucky to get a ticket of course) | 21:01 |
| minute | hmm, oops in panel_edp_probe+0x394/0x568 | 21:02 |
| ch | cool. yeah i also need everything still | 21:02 |
| minute | ah drivers/gpu/drm/panel/panel-edp.c:814 | 21:02 |
| minute | (trying direct eDP from rk3588) | 21:02 |
| + chomwitt (~chomwitt@2a02:85f:9a5f:900:1ac0:4dff:fedb:a3f1) | 21:03 | |
| minute | ah lol | 21:03 |
| minute | > We're using non-optimized timings and want it really obvious that someone needs to add an entry to the table, so we'll do a WARN_ON splat. | 21:03 |
| minute | > panel-simple-dp-aux aux-fdec0000.edp: Unknown panel CMN 0x1239, using conservative timings | 21:03 |
| minute | 0x1239 is chimei innolux, so aux seems to work | 21:03 |
| minute | something just crashes though, hmmm | 21:04 |
| + pomel0 (~pomel0@user/pomel0) | 21:04 | |
| - pomel0 (QUIT: Ping timeout: 252 seconds) (~pomel0@user/pomel0) | 21:09 | |
| - chomwitt (QUIT: Ping timeout: 252 seconds) (~chomwitt@2a02:85f:9a5f:900:1ac0:4dff:fedb:a3f1) | 21:10 | |
| minute | oh wow, got it to work (eDP) | 21:16 |
| grimmware | :O | 21:19 |
| grimmware | Fuckin TUESDAY YEAH. | 21:19 |
| minute | haha | 21:20 |
| minute | weird, the two desktop icons/illustrations in setup wizard are super mini smol on this system image | 21:22 |
| ch | eDP on next, i assume? | 21:25 |
| minute | ch: currently on classic (was easiest to test) | 21:25 |
| josch | i can confirm | 21:25 |
| minute | ch: photo https://mastodon.social/@mntmn/115413774387210452 | 21:25 |
| minute | josch: mini small icons? | 21:25 |
| josch | yes | 21:26 |
| minute | josch: i wonder what happened :00 | 21:26 |
| josch | maybe it was introduced as a result of some of the recent library bumps | 21:26 |
| minute | could be | 21:27 |
| grimmware | oh could it be a dpi scaling thing? | 21:28 |
| josch | i also wanted to share my recent usb-c contraption connected to my pocket reform: https://mister-muffin.de/p/c0sa.jpg -- took several USB-A to USB-C and vice-versa adapters to connect everything today | 21:32 |
| josch | the last time my pocket rebooted was back in september, so before the recent firmware changes -- no problem since then | 21:33 |
| + chomwitt (~chomwitt@2a02:85f:9a5f:900:1ac0:4dff:fedb:a3f1) | 21:33 | |
| josch | i'm using it to take 1.5 hour lecture recordings and then encode the raw video on the pocket as well (no problems so far) | 21:33 |
| sigrid | even your adapter has an adapter | 21:33 |
| josch | the pocket can now record video with such high quality that it is too much for a311d to play back XD | 21:34 |
| minute | josch: wow ok :D | 21:37 |
| ch | somewhat surprising this doesnt run into a usb hub limit :D | 21:38 |
| minute | (meanwhile installing new kernel build on the pocket re:dispv1 flicker stuff) | 21:38 |
| josch | ch: i'm surprised as well but i checked with lsusb because i had to make sure the camera got the 5 GiB/s for 1080p@60 MJPEG recording :) | 21:39 |
| minute | pocket display v1 on rk3588 works fine re:display off/on issues with this build https://source.mnt.re/reform/reform-debian-packages/-/jobs/14621/artifacts/browse/repo/pool/main/l/linux/ | 21:42 |
| minute | now i have to test with imx8mp... | 21:42 |
| rick_ | oh regarding usbc stuff... should usbc dp alt mode work poket reform? dmesg always tells me 'maybe you have a bad cable?' i tried multiple usbc cables, docks and so one.. and nothing did work. it not that important for me, just asking if it should work xD | 21:42 |
| ch | no | 21:43 |
| rick_ | okay i see, so i don't need to debug it further. used mini hdmi after my tests, which works perfectly fine ^^ | 21:44 |
| minute | rick_: there's no alt mode support in the current motherboard | 21:44 |
| minute | but yeah, i also use microhdmi @ home with it | 21:44 |
| rick_ | yeah thats also enought for me. thanks for the info! :D | 21:45 |
| ch | should write down somewhere what the situation is on the rk. i forgot this like 3 times already | 21:46 |
| minute | theoretically one could make a bizarro port board for pocket that splices into the charging stuff and muxes the eDP to another USB-C port | 21:49 |
| minute | but i think i'd rather add some kind of alt-mode functionality to the motherboard later | 21:50 |
| minute | also not sure how compatible the eDP is to DP | 21:50 |
| ch | the switchable pins are on the wrong port, was that it? | 21:50 |
| minute | ch: yeah | 21:51 |
| ch | mh | 21:51 |
| minute | i need to do a v2 pocket reform motherboard relatively soon though....... | 21:51 |
| minute | grimmware: ok so with that double init patch, the display v1 still works on imx8mp | 21:58 |
| grimmware | amazing. | 21:58 |
| minute | will try dpms toggle on that machine... | 21:59 |
| minute | uh oh | 22:00 |
| grimmware | D: | 22:00 |
| minute | grimmware: the flicker is back, on imx8mp :DD | 22:00 |
| grimmware | what the fuck | 22:00 |
| minute | but maybe... | 22:00 |
| grimmware | this is an emotional rollercoaster | 22:01 |
| minute | grimmware: welcome to my job! lol | 22:01 |
| elb | ch: minute: OK, well, then the first one is waiting :-) !67 | 22:03 |
| minute | ok so imx8mp dts has the "init-in-enable" | 22:03 |
| minute | and only imx8mp has that | 22:04 |
| minute | ah well. it's clear why | 22:07 |
| ch | the watchdog reset on pocket-hid somehow seems to interfere with flashing | 22:09 |
| minute | ch: oh no | 22:09 |
| * mjw -> Guest8392 | 22:09 | |
| - Guest8392 (QUIT: Killed (lead.libera.chat (Nickname regained by services))) (~mjw@2001:1c06:2486:a800:7602:5eff:dc71:a72c) | 22:09 | |
| * Guest7901 -> mjw | 22:09 | |
| minute | ch: isn't it turned off in the reset-for-flashing handler? | 22:10 |
| + Guest8392 (~mjw@2001:1c06:2486:a800:7602:5eff:dc71:a72c) | 22:10 | |
| ch | need to checl | 22:11 |
| - gustav25 (QUIT: Quit: Quit) (~gustav@c-78-82-53-228.bbcust.telenor.se) | 22:15 | |
| ch | i dont understand what HsvToRgb really does | 22:38 |
| minute | ch: you mean _how_ it does what it does? | 22:38 |
| ch | mostly i'm wondering if it aims to produce RGB in normal colorspace, or something else | 22:39 |
| minute | ch: it aims to convert h/s/v to rgb 0-255 :D | 22:39 |
| ch | hmm | 22:39 |
| minute | ch: is there a problem with it? | 22:40 |
| ch | so i wondered if i can stop hardcoding 0x004040 | 22:40 |
| ch | which i assumed was in RGB | 22:40 |
| minute | yeah that's rgb | 22:40 |
| ch | but converting that back using an online converter, i get some blue | 22:40 |
| ch | not pink | 22:40 |
| minute | converting that... back? to hsv? | 22:40 |
| ch | https://www.colorhexa.com/004040 | 22:41 |
| minute | ch: ah sorry it's not RGB | 22:41 |
| minute | ch: it's BGR or so | 22:41 |
| minute | no | 22:41 |
| minute | GRB? | 22:41 |
| minute | ok let me check :D | 22:42 |
| ch | 400040 might match | 22:42 |
| ch | so RBG? | 22:42 |
| minute | something is definitely odd | 22:43 |
| minute | void led_task(uint32_t color) { | 22:44 |
| minute | uint32_t pixel_grb = color; | 22:44 |
| minute | lol | 22:44 |
| minute | i wonder if i had fixed that issue in keyboard4 | 22:44 |
| minute | ch: i guess so far it didn't matter because there's no fixed value hsv input, just a wheel, so if it's out of phase one didn't notice | 22:45 |
| ch | https://www.colorhexa.com/400040 | 22:45 |
| ch | i think color maybe, brightness no | 22:46 |
| ch | so maybe more is going on | 22:46 |
| minute | ch: well LED colors will look differently | 22:46 |
| + pomel0 (~pomel0@user/pomel0) | 22:46 | |
| ch | right | 22:46 |
| ch | do you know what the color was supposed to be in RGB? | 22:47 |
| minute | ch: i cleaned up some of that stuff in keyboard4-fw, we should port that over | 22:47 |
| minute | ch: well 0x400040 is correct. i think it was 0x800080 before and i halved the values, and it's swizzled because it's not really rgb in that codebase, but grb | 22:47 |
| minute | in reform-keyboard4-fw i think i untagled most of the rgb/hsv mess | 22:48 |
| minute | and it also remembers the last chosen color | 22:48 |
| minute | ch: https://source.mnt.re/reform/reform/-/blob/master/reform2-keyboard4-fw/src/leds.c?ref_type=heads | 22:48 |
| ch | https://source.mnt.re/reform/pocket-reform/-/merge_requests/66 in the meantime this constrains power draw from the leds | 22:49 |
| - schalken (QUIT: Ping timeout: 256 seconds) (~schalken@117-118-178-69.gci.net) | 22:49 | |
| - paperManu (QUIT: Ping timeout: 256 seconds) (~paperManu@modemcable141.205-200-24.mc.videotron.ca) | 22:49 | |
| minute | ch: lgtm :D how did you come up with the threshold? | 22:49 |
| + schalken (~schalken@117-118-178-69.gci.net) | 22:50 | |
| ch | to my eyes at around that threshold there is a wide gap where the brightness does not -visibly- change until you go up to almost the meax | 22:50 |
| ch | the max | 22:50 |
| minute | ch: oh nice | 22:50 |
| ch | so that seemed good | 22:50 |
| minute | ch: thanks, set to auto-merge | 22:51 |
| ch | and i think lower than the threshold didnt change much power wise | 22:51 |
| minute | cool | 22:51 |
| ch | https://source.mnt.re/reform/pocket-reform/-/merge_requests/67 from elb - i guess this needs a runner assigned? | 22:52 |
| minute | ch: let me attempt a quick refactor of the leds stuff | 22:52 |
| ch | k | 22:52 |
| minute | oh yeah @ elb | 22:52 |
| - pomel0 (QUIT: Ping timeout: 244 seconds) (~pomel0@user/pomel0) | 22:52 | |
| ch | i was gonna check out elb's MR and then the led color again | 22:52 |
| minute | ah no the runner is already there, probably just needs a rerun | 22:53 |
| + pomel0 (~pomel0@user/pomel0) | 22:57 | |
| ch | sway output power off/on works with the newest kernel build | 23:00 |
| josch | hehe, i have some reform-system-image changes but i see you are busy :D I'll push later to not hog the gitlab runner time :) | 23:01 |
| minute | ch: nice. now i just need to fix that for imx8mp again ;/ | 23:12 |
| - schalken (QUIT: Ping timeout: 256 seconds) (~schalken@117-118-178-69.gci.net) | 23:12 | |
| + schalken (~schalken@117-118-178-69.gci.net) | 23:14 | |
| grimmware | ch: \o/ | 23:20 |
| minute | ch: yet untested https://source.mnt.re/reform/pocket-reform/-/merge_requests/69 | 23:24 |
| ch | seems sensible, gonna test | 23:26 |
| - chomwitt (QUIT: Quit: WeeChat 3.8) (~chomwitt@2a02:85f:9a5f:900:1ac0:4dff:fedb:a3f1) | 23:28 | |
| ch | WFM | 23:30 |
| ch | i think the min brightness on poweron could even be less | 23:32 |
| minute | noice | 23:34 |
| minute | it should be 0x200020 so... need to test that myself also :D | 23:35 |
| ch | https://source.mnt.re/reform/pocket-reform/-/merge_requests/69/diffs#2260c41e482b89a87ab5c37958f1cfa1440397ba_117_120 | 23:36 |
| ch | imo this can be < 0x6 | 23:36 |
| minute | ch: ok, feel free to edit and i can test later! think i need to go home and get some rest | 23:37 |
| minute | but nice session here today! | 23:37 |
| ch | i dont think i can commit into your MR, but left a note | 23:38 |
| ch | the code changes all make sense to me | 23:38 |
| minute | ah, so i can merge first | 23:38 |
| minute | ah no, i created a conflict m)) ok tomorrow | 23:39 |
| ch | (: | 23:39 |
| ch | get some rest | 23:39 |
| - andreas-e (QUIT: Quit: Leaving) (~Andreas@2a02-8434-b6a3-e901-facc-8e87-8e54-890e.rev.sfr.net) | 23:39 | |
| josch | was nice to read along :) | 23:40 |
| - schalken (QUIT: Ping timeout: 244 seconds) (~schalken@117-118-178-69.gci.net) | 23:48 | |
| + schalken (~schalken@117-118-178-69.gci.net) | 23:49 | |
| ch | https://source.mnt.re/reform/pocket-reform/-/merge_requests/65 also now tested and imo good | 23:52 |
| ch | the green led on the rcore carrier could also be less bright ;-) | 23:53 |
| ch | josch: you dont happen to have an esim module that can go into the sim card slot and can then be managed from linux? | 23:55 |
| josch | i would have to look up online what an "esim" is :D | 23:55 |
| josch | i have a normal sim card as far as i know | 23:56 |
| ch | right | 23:58 |
| ch | so normal sim cards will go away eventually | 23:58 |
| ch | and instead you can download a sim profile ('esim') into a security chip your device already has | 23:58 |
| ch | some vendors figured out you can put that chip into a form factor where it looks like a normal simcard | 23:59 |
| elb | yeah I'd like all of the internal LEDs to be less bright | 23:59 |
| elb | that's on my list of things to look at | 23:59 |
| elb | like if I can tell it's on _at all_ that's enough of my battery for an LED ;-) | 23:59 |
| josch | please do this on your pocket reforms -- i need my classic reform as a torchlight at night! | 23:59 |
| ch | blue+red leds on the right side should already be less bright than before | 23:59 |
| elb | I mean the classic reform has a clear bottom, that's fair ;-) | 23:59 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
