2025-11-03.log

- iank (QUIT: Ping timeout: 264 seconds) (~iank@fsf/staff/iank)00:15
* mark_ -> mjw00:22
+ iank (~iank@fsf/staff/iank)00:24
- mjw (QUIT: Ping timeout: 260 seconds) (~mjw@gnu.wildebeest.org)00:46
- paperManu (QUIT: Ping timeout: 256 seconds) (~paperManu@107.159.15.124)02:57
- TechnoWizard (QUIT: Remote host closed the connection) (~TechnoWiz@user/TechnoWizard)02:58
+ paperManu (~paperManu@107.159.15.124)03:00
+ aloo_shu_ (~aloo_shu@85.51.18.212)03:02
- aloo_shu (QUIT: Ping timeout: 244 seconds) (~aloo_shu@85.51.17.185)03:03
* aloo_shu_ -> aloo_shu03:03
- paperManu (QUIT: Ping timeout: 264 seconds) (~paperManu@107.159.15.124)03:13
+ paperManu (~paperManu@107.159.15.124)03:25
- paperManu (QUIT: Ping timeout: 252 seconds) (~paperManu@107.159.15.124)04:49
- cobra (QUIT: Ping timeout: 252 seconds) (~cobra@user/Cobra)05:20
+ plomlompom (~plom@139.59.138.73)05:23
+ cobra (~cobra@user/Cobra)05:24
- cobra (QUIT: Quit: ZNC 1.8.2 - https://znc.in) (~cobra@user/Cobra)05:35
+ cobra (~cobra@user/Cobra)05:38
+ TechnoWizard (~TechnoWiz@user/TechnoWizard)06:12
TechnoWizardI ordered my pocket reform in Nov of last year and got it around March, which means it is the version with WiFi but no bluetooth. I've tried picking up a few M.2 wireless cards that offer WiFi + Bluetooth, but none of them fit the connector. Is there a known upgrade option or do I need to get a new motherboard?06:21
TechnoWizardI've tried the ALFA AWM27921EU and the Intel AX210.06:21
+ chomwitt (~chomwitt@2a02:85f:9a5f:900:1ac0:4dff:fedb:a3f1)06:22
- colinsane (QUIT: Quit: bye) (~colinunin@97-113-142-187.tukw.qwest.net)07:01
- plomlompom (QUIT: Remote host closed the connection) (~plom@139.59.138.73)07:01
+ plomlompom (~plom@139.59.138.73)07:01
+ colinsane (~colinunin@97-113-142-187.tukw.qwest.net)07:04
- vkoskiv (QUIT: Ping timeout: 260 seconds) (~vkoskiv@109-204-186-7.cust.valoonet.fi)07:05
- plomlompom (QUIT: Remote host closed the connection) (~plom@139.59.138.73)07:05
+ plomlompom (~plom@139.59.138.73)07:05
+ rwa_ (0a82deb4eb@2a03:6000:1812:100::41b)07:32
- plomlompom (QUIT: Remote host closed the connection) (~plom@139.59.138.73)07:56
+ plomlompom (~plom@139.59.138.73)07:56
rwa_didn't know that sourcehut also has a hosted irc bouncer/client...08:05
chTechnoWizard: which part exactly didnt fit?09:02
joschthe alfa card is for the m.2 slot09:03
joschand the descrption sounds like it's only using usb signaling?09:03
joschthe second card is m.2 using both pcie (for wifi) and usb (for bluetooth)09:04
joschTechnoWizard: i don't think there is usb signaling on the slots09:04
- colinsane (QUIT: Quit: bye) (~colinunin@97-113-142-187.tukw.qwest.net)09:05
joschTechnoWizard: for rk3588 there is RCORE 2.0 carrier board which allows you to replace the board which had the headphone jack with a newer version which also has a M.2 slot where you can add the Intel AX210 for wifi+bt09:07
joschthis is as far as i understand it -- i have the upgrade on order but didn't receive it yet for my pocket09:08
TechnoWizardHere you can see the various chips side by side https://photos.app.goo.gl/sdvpJnoFWdryruu8909:26
TechnoWizardsounds like I need to get the new carrier board ordered as well, is that from the MNT website?09:27
chthe wwan slot is B-key09:30
chat least the sch says this09:31
chTechnoWizard: https://shop.mntre.com/products/mnt-pocket-reform-upgrade-bundle09:34
+ colinsane (~colinunin@97-113-142-187.tukw.qwest.net)09:36
TechnoWizardI'm checking my RCORE release version now. If it is 2+, can I get https://shop.mntre.com/products/mnt-pocket-reform-headset-switch-board-2-0 instead? it is much cheaper.09:37
ch> Please note that this Wi-Fi card slot only works with the RCORE RK3588 Release Version 209:40
+ mjw (~mjw@gnu.wildebeest.org)09:42
TechnoWizardokay, I have version 1, so you were right.09:43
chis the headphone v2 board in git?09:46
TechnoWizardat least I've already got the AX210, so I can just get an additional antenna and then I'll be good. 09:46
TechnoWizardhttps://www.mouser.com/ProductDetail/TE-Connectivity-Linx-Technologies/ANT-W63RPC1-UFL-200?qs=81r%252BiQLm7BTBe2gJuT0hJA%3D%3D09:55
TechnoWizardi think this should work09:55
TechnoWizardwoops I think it needs to be an MH4 antenna10:00
TechnoWizardwoops I think it needs to be an MH4 connector*10:00
+ vkoskiv (~vkoskiv@109-204-186-7.cust.valoonet.fi)10:24
- TechnoWizard (QUIT: Ping timeout: 252 seconds) (~TechnoWiz@user/TechnoWizard)10:26
+ enwu (~enwu@user/enwu)10:27
- erle (QUIT: Quit: K-lined) (~erle@user/erle)10:29
- mjw (QUIT: Ping timeout: 244 seconds) (~mjw@gnu.wildebeest.org)11:27
joschU-Boot 2022.04 MNT Reform 2 with i.MX8MP Module 2025-01-12-3-g54ab337-gb0e908b1-dirty (Nov 03 2025 - 10:29:45 +0000)11:52
joschgrimmware: i managed to flash a custom u-boot to your imx8m+ without bricking the board \o/11:52
joschminute: imx8m+ u-boot suffers from the same problem as rk3588 u-boot: since the console= parameter for serial is missing, there is no login prompt on serial11:54
joschi fixed that in https://source.mnt.re/reform/reform-imx8mp-uboot/-/merge_requests/6 with a bunch of other stuff but i wonder:11:54
joschminute: do you remember that you had to manually add the console=ttymxc1,115200 parameter to extlinux.conf to get a serial login on imx8m+?11:55
joschi also wonder why nobody else complained so far -- did imx8m+ users not use serial or did i miss messages here and in the forum?11:56
minutejosch: probably did11:56
joschi'm surprised this issue has existed for so many years and am questioning whether my conclusion is correct11:56
joschbut i just flashed the flash.bin created my the MNT gitlab CI to grimmware's imx8m+ and now it works11:56
joschnext i'll try a vanilla reform-system-any image which also didn't work before11:57
joschminute: you also remember the issue with the gsetting schemas from a few days ago which made the reform-system-image pipeline fail? I have a patch and it needs a new reform-tools release. Anything else you'd like me to add to that one?11:58
joschSystem Image v5: 2025-11-0312:09
joschperfect, reform-system-any now works with imx8m+ as well :)12:09
+ paperManu (~paperManu@107.159.15.124)12:15
minutejosch: very nice!12:24
+ mjw (~mjw@2001:1c06:2486:a800:158:adb1:a1e7:3c3f)12:42
joschLinux reform 6.17.6-mnt-reform-arm64 #1 SMP PREEMPT Debian 6.17.6-1+reform20251101T184857Z (2025-11-0 aarch64 GNU/Linux12:59
joschminute: i just tested the repository created as gitlab CI artifacts from the pipeline of this MR: https://source.mnt.re/reform/reform-debian-packages/-/merge_requests/144 And it works with RCORE-DSI RK3588 classic Reform at least.13:01
joschminute: if you find some time, it would be great if you could use your expert-eye and see if even more patches can be removed13:01
joschit would be great if for the 6.18 release we no longer just copypaste the full collabora patch stack but only cherry-pick those we actually need as you also have already hinted here a few days ago13:02
+ gustav25 (~gustav@c-78-82-52-240.bbcust.telenor.se)13:02
minutejosch: btw i've pondered the ssd install page of setup-wizard a bit more in the last days. thoughts: https://source.mnt.re/reform/mnt-reform-setup-wizard/-/issues/2213:07
- pomel0 (QUIT: Ping timeout: 264 seconds) (~pomel0@user/pomel0)13:19
joschminute: what you write makes sense to me. You want a user-friendly GUI tool to allow an easy setup without the terminal. The reform-setup-encrypted-disk script has to be improved with a good progress bar indeed. Using an imaging approach will not work as you should not copy a filesystem which is currently mounted.13:19
+ pomel0 (~pomel0@user/pomel0)13:19
- cobra (QUIT: Quit: ZNC 1.8.2 - https://znc.in) (~cobra@user/Cobra)13:20
+ cobra (~cobra@user/Cobra)13:28
- plomlompom (QUIT: Read error: Connection reset by peer) (~plom@139.59.138.73)13:32
+ plomlompom (~plom@139.59.138.73)13:32
- cobra (QUIT: Quit: ZNC 1.8.2 - https://znc.in) (~cobra@user/Cobra)13:36
minutejosch: ah good point @ imaging approach. can rsync be customized with options to make the progress linear or could we use one of the bazillion other modern backup-related tools?13:36
+ cobra (~cobra@user/Cobra)13:39
- plomlompom (QUIT: Remote host closed the connection) (~plom@139.59.138.73)13:39
+ plomlompom (~plom@139.59.138.73)13:39
joschswitching out rsync for something else should be no problem. We only need to take care that whatever the tool is, it is able to correctly copy things like named pipes, extended attributes, sockets and other special things13:41
+ gidzit (~gidzit@gidzit.org)13:48
- gidzit (QUIT: Remote host closed the connection) (~gidzit@gidzit.org)13:49
- plomlompom (QUIT: Remote host closed the connection) (~plom@139.59.138.73)13:49
+ gidzit (~gidzit@gidzit.org)13:49
+ plomlompom (~plom@139.59.138.73)13:49
- gidzit (QUIT: Remote host closed the connection) (~gidzit@gidzit.org)13:51
+ gidzit (~gidzit@gidzit.org)13:51
- plomlompom (QUIT: Remote host closed the connection) (~plom@139.59.138.73)13:51
joschmaybe even rust/qt comes with a sensible implementation of "copy things, preserving attributes" and then it could even be considered to re-implement reform-setup-encrypted-disk in rust, copypasting from the shell script what is needed and avoid having to set up some sort of progress-communication from the shell script to the GUI frontend in rust13:51
joschbut these are just idle thoughts as i'll not be the one writing it :)13:51
+ plomlompom (~plom@139.59.138.73)13:52
+ paperManu_ (~paperManu@107.159.15.124)13:55
- paperManu_ (QUIT: Ping timeout: 264 seconds) (~paperManu@107.159.15.124)14:00
- paperManu (QUIT: Ping timeout: 264 seconds) (~paperManu@107.159.15.124)14:04
- pomel0 (QUIT: Remote host closed the connection) (~pomel0@user/pomel0)14:09
minutejosch: hm right. there is https://lib.rs/crates/fs_extra14:21
minute> This function will also copy the permission bits of the original files to destination files (not for directories).14:22
minute"not for directories" not great 14:22
bkeysminute: No rush, the tracking said y'all got my Reform, I just want to confirm14:26
minutebkeys: i just got it!14:26
bkeysOkay glad to hear, sadly I did overwrite the Fedora install cause I was worried it was a software issue14:28
bkeysIt should still have uboot on the emmc though14:28
+ paperManu (~paperManu@modemcable141.205-200-24.mc.videotron.ca)14:52
- martins (QUIT: Changing host) (924a802ee9@2001:bc8:1210:2cd8::15c)15:00
+ martins (924a802ee9@user/martins)15:00
* martins -> grunskis15:06
* grunskis -> martins15:07
grimmwarejosch: you can see why I thought that board would be better in your hands than mine :)15:17
minutei have current blender running on rk3588 reform with amd gpu 15:20
grimmware:O15:20
grimmwareyou fucking what15:20
grimmwareis it effort?15:20
minuteyeah it's mindblowing15:20
minuteneeds patched kernel and oculink adapter15:21
grimmwareyou gonna put the patches into reform-debian-packages?15:22
minutehas high fps too15:22
minutegrimmware: already di15:22
minutedid15:22
grimmwareol15:22
grimmware*lol15:22
grimmwareamazing15:22
minutehttps://source.mnt.re/reform/reform-debian-packages/-/merge_requests/14515:22
minuteso the minisforum DEG-1 does _not_ work15:23
minutebut this one works https://www.amazon.de/dp/B0CRR7HMX5?ref=fed_asin_title15:23
minuteJMT "DOCK-OC4"15:24
[tj]so now we can have reform desktop?15:25
amkpocket with an oculink port would be quite the device :o15:25
amospallaHi, is openrgb running on pocket a thing, and what are the features of openrgb regarding color leds manipulation? I am planning o write something to easy keyboard leds manipulation, but maybe clashes with openrgb.15:25
grimmwareminute: did you have a chance to play with any of the bpftrace stuff by the way?15:27
minutegrimmware: not yet, but will15:27
grimmwareminute: if you identify something you'd like me to work on to make it more useful to you or alternatively if you just want me to write shit for you to use for debugging I'm more than happy to do that15:28
minutegrimmware: cool, many thanks15:28
grimmwarelike, obvs away from home for the next 2 weeks without my pocket but I've got a work project coming up early next year that could really do with me refreshing my memory anyway15:29
grimmwarecurrently in Seoul :O15:29
minutejosch: hmm @ rsync > Along with --info=progress2 you can also use --no-inc-recursive option (or its shorter --no-i-r alias) to disable incremental recursion. 15:31
grimmwareminute: do you think that Blender is likely to ever work natively on the rk3588 without significant effort? I'm not particularly in-the-know about graphics stuff, it's an OpenGL 3 thing right?15:31
joschnice, --no-inc-recursive sounds exactly what we want15:31
joschminute: i'll just add this option to the rsync call in reform-tools now. Should not have any other negative effect.15:32
minutegrimmware: it migh work on vulkan after this is done https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3746215:41
minutejosch: awesome, thanks! 15:41
minutegrimmware: some vulkan/GL features need to be emulated with compute shaders for blender to fully work15:42
minutegeometry shaders have been mostly removed from blender codebase but they're still used to emulate barycentrics which are also not supported by arm mali / panvk15:43
grimmwaregotcher15:47
- cobra (QUIT: Quit: ZNC 1.8.2 - https://znc.in) (~cobra@user/Cobra)15:47
grimmwareI love the idea of being able to do some simple mesh work on the sofa15:47
grimmwaretrackpads are very much not the one for that15:47
grimmwareI'm traveling with my GPD P2 Max and the keyboard and trackpad fill me full of bile and vitriol15:48
+ cobra (~cobra@user/Cobra)15:50
[tj]trackball is excellent for kicad15:52
[tj]until you hit the 3d viewer having different button binds to all the other editors15:53
grimmwareI love doing 3d work with a trackball15:53
[tj]openscad and kicad do not have pleasant interfaces 15:54
[tj]but I blame those tools15:54
grimmwareOpenSCAD has an okay interface if you get it to just rerender on save and use an external editor15:55
grimmwareFreeCAD can get in the sun15:56
[tj]I find moving in openscad on the pocket hellish15:56
grimmwareThat is one of the most hateful pieces of software I think I've ever used15:56
minutegrimmware: well. it's getting better. also, you have to set the input mode to blender mode.15:57
grimmwaretrick question, it waits for input ;)15:57
grimmwarewhoops15:58
grimmwarewrong window15:58
grimmwarebut yeah, FreeCAD is already Not The One for doing constraint solving in a violently single-threaded program15:58
grimmwareand then there's the actual object model...15:58
grimmwarehonestly if I have to know that much about software internals it better just give me an API15:59
grimmwarewith OpenSCAD I like the fact I can learn about 3 tricks and then use them to beat a problem into submission16:00
grimmwareand with that it's midnight here so I'm going to bed16:01
grimmwarelaters16:01
joschamospalla: https://gitlab.com/CalcProgrammer1/OpenRGB/-/merge_requests/3006/16:12
amospallajosch: thank you, the thing is, I'm writing a thing, but I'm wondering if it has any sense, if openrgb works.16:13
amospallaFor example, my plan is, this daemon would manage both background efects/whatever, while notifications like battery, high cpu load, etc would appear, and then back to the background.16:14
joschamospalla: openrgb should work (last time i tried was in september) but openrgb is also an extremely complex piece of software16:15
joschso the argument could also be made that some NIH is okay to avoid the openrgb complexity16:16
amospallaOk, thank you, that is the response I needed :D16:16
joschamospalla: that's why this exists: https://source.mnt.re/reform/reform-tools/-/blob/main/examples/keyboard_rainbow.py?ref_type=heads16:17
josch(even though you can also render a rainbow using openrgb)16:17
amospallaMy idea has always been around the OLED, but given the pocket firmware does not accept bitmaps, I abandoned this idea. When I saw this exact script, I recovered what I wanted to do with the OLED and translated somehow to the keyboard leds.16:19
amospallaBut both, as an interface, fill different purposes. With the OLED I planned it as text, but text does not work with keyboard leds (already tried, not easy to read).16:20
minuteamospalla: yeah openrgb is not as nice as we thought it would be16:22
amospallaSo I'm trying to write something in a way the user easily can set keyboard leds (sending bitmaps with text you can write on notepad).16:23
joschminute: MNT gitlab CI currently fails for i386 and amd64 because the mntre-builder machine only has a few gigs of free space which is not sufficient for building linux16:23
amospallaok, thank you.16:23
minutejosch: lets disable i386 and amd64 for now if possible? until further notice16:24
joschokay16:24
minutealso, wondering where that space went16:24
josch/ramdisk has 37G in it16:25
josch(investigating...)16:25
joschit's eaten up by /ramdisk/docker/overlay2/16:26
joschi'm disabling it for now. I have too little clue about docker16:26
minuteok16:29
minutei'll reboot it16:30
minuteit's a ramdisk anyway16:30
minutejosch: the arm64 builder has enough space in any case16:30
- Guest1173 (QUIT: Quit: Leaving) (~mjw@2001:1c06:2486:a800:7602:5eff:dc71:a72c)16:35
joschminute: did you just reboot it? Now jobs show this error:16:45
joschERROR: Job failed: failed to pull image "registry.gitlab.com/gitlab-org/gitlab-runner/gitlab-runner-helper:x86_64-d1f69508" with specified policies [always]: open /ramdisk/docker/tmp/GetImageBlob1687262251: no such file or directory (manager.go:203:1s)16:45
joschminute: was there not some initial command that needed to be run on that box after a reboot?16:45
joschah no that was just about the zram16:45
- amk (QUIT: Remote host closed the connection) (~amk@user/amk)16:51
+ amk (~amk@user/amk)16:51
minutejosch: i did that a few minutes after rebooting17:21
minutejosch: like, the zram mount17:21
minutejosch: maybe the runner service needs a restart17:22
minutejosch: hmm, for this you made a fix right? what do i have to do to have this fix in my branch? https://source.mnt.re/reform/reform-system-image/-/jobs/15160#L1173517:22
minute> No such key “screen-brightness-down-static” in schema “org.gnome.settings-daemon.plugins.media-keys” as specified in override file “/usr/share/glib-2.0/schemas/20_reform.gschema.override” and --strict was specified; exiting.17:23
joschyes, i have it locally in my reform-tools git repo17:23
joschif you need to build sysimages now, you can either comment out the offendling line17:23
minutejosch: i need to 17:23
minutejosch: so the key was just changed or...?17:24
joschor you apply the commit from this MR: https://source.mnt.re/reform/reform-system-image/-/merge_requests/14317:24
minuteok, will try that...17:24
joschyes, the key was moved to a different schema17:24
minutebut the key still works?17:24
minuteIIRC the old one was already there before but for some reason i had to use -static17:25
minuteerm, s/old/non-static17:25
joschi didn't try that yet but since the schema compiles without problems it "should"17:25
joschminute: after you built a system image with the patch from above MR, i'd be great if you could try and see if brightness control still works and suggest a different configuration which works and which i can then publish with the next reform-tools version17:25
minutejosch: ok i'll try that17:26
minutewhy is this gitlab UI slow like molasses... ah maybe because i had to turn on ipv6 here earlier17:27
- gidzit (QUIT: Ping timeout: 256 seconds) (~gidzit@gidzit.org)17:58
joschminute: i temporarily threw out amd64 and i386 so that things build again: https://source.mnt.re/reform/reform-debian-packages/-/commit/e469c2ad131054c7afe4f9f059eb2eb575545b1a18:06
josch(because i'm preparing the reform-tools release right now i want to test CI artifacts)18:06
+ AnimaInvicta (~AnimaInvi@88-120-179-216.subs.proxad.net)18:23
- amospalla (QUIT: Quit: WeeChat 4.7.1) (~jordi@user/amospalla)18:38
joschminute: i just had an idea. What would stop me in practice from takin the new port board for the pocket reform with the m.2 slot and stick it into the *classic* reform with rcore version 2 or later? Suppose I 3D print myself a fitting holder, would this mean that my classic reform suddenly has an additional pcie slot?18:41
minutejosch: yes, if you have MRCOREXXR02 at least18:42
joschwooooah!18:42
joschmaybe i've found what would make me switch for real from a311d to rk3588 :D18:42
joschi need to make another order :)18:42
minutejosch: handy table https://shop.mntre.com/products/mnt-pocket-reform-headset-switch-board-2-0?taxon_id=19#compatibility18:42
minutejosch: many thanks @ x8618:42
joschlol okay, i was not the first one to think about this possibility XD18:42
joschminute: motherboard 3.0 will also re-appear in the shop soon-ish, no? Then i'll place the order then together with it.18:45
minutejosch: yes19:01
joschsplendid :)19:02
minutetesting new system image...19:07
joschmine is still building -- please cancel my jobs if they interfer with yours19:10
- mjw (QUIT: Ping timeout: 244 seconds) (~mjw@2001:1c06:2486:a800:158:adb1:a1e7:3c3f)19:14
+ mjw (~mjw@gnu.wildebeest.org)19:15
minutejosch: that brightness shortcut doesn't seem to work btw19:17
minutehyper-f1/f2 work though (but because they're media keys i guess)19:18
minuteah and we still don't have firefox in the dock19:18
joschhuh that is odd because glib-compile-schemas will complain for invalid settings19:18
joschso the setting does exist but has not the desired effect -- what happens when removing "-static" from the end?19:18
minutealternate character key is also unset, so altgr+aeous for äëüß doesn't work with eurkey19:19
minutejosch: it looks like brightness shortcut was just removed from gnome19:21
minutejosch: so we have to do it like in sway19:21
minuteyeah, i'll add that in...19:22
joschminute: it was removed from gnome? But in this MR, screen-brightness-up and screen-brightness-down got added to gnome-shell with the default being XF86MonBrightnessUp and XF86MonBrightnessDown, respectively: https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/375119:25
minute> BrightnessManager19:26
minute> More a question for the design team, but we may want to expose the shortcuts in Settings19:27
joschafk now for a few hours -- ttyl!19:27
minute> They aren't listed there currently19:27
minutejosch: ok cya19:27
f_btw, how thin/thick is reform next compared to reform?19:28
f_I'm wondering if I should wait for Reform Next and hopefully my current laptop dies by then ;)19:28
minutejosch: when i manually set org.gnome.shell.keybindings screen-brightness-down, it works19:32
f_and sorry if I'm asking way too many questions lol (feel like I might be asking way too many)19:33
minutef_: np, i just don't have time to look it up atm. it might be written on the reform next page19:33
minuteon crowd supply19:34
minuteyou can also see the laptop in the video19:34
minutevs the classic reform video 19:34
f_cool!19:34
f_Looks a bit thinner than my current laptop19:38
f_good enough!19:38
- spew (QUIT: Remote host closed the connection) (e9e8135d87@user/spew)19:39
f_thanks ^^19:39
minutejosch: your patched screen-brightness-down/up schema overrides seem to be ignored, even if they are correct. this is so frustrating19:41
minuteidk how to fix this... incredible that small stuff like this is a showstopper19:41
f_Honestly really cool what y'all are doing, excited to get one of these for sure19:42
f_I wanted something as modular as my current laptop, but this might honestly actually be *more* modular than that19:42
bremnerminute: did you get a chance to summarize your s2idle experiments somewhere, or just the IRC logs?19:43
minutebremner: irc logs19:43
bremnerack19:43
minuteat least i found the reason for altgr not working19:44
+ mark_ (~mjw@gnu.wildebeest.org)19:47
- mjw (QUIT: Ping timeout: 244 seconds) (~mjw@gnu.wildebeest.org)19:49
frickleroh, I had something similar recently when trying an external german keyboard on the pocket, everything was fine except for AltGr having no effect. so I'm curious for the solution19:49
+ vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:50)19:55
- plomlompom (QUIT: Remote host closed the connection) (~plom@139.59.138.73)20:08
+ plomlompom (~plom@139.59.138.73)20:08
+ pomel0 (~pomel0@user/pomel0)20:17
* mark_ -> mjw20:23
- pomel0 (QUIT: Ping timeout: 244 seconds) (~pomel0@user/pomel0)21:07
+ pomel0 (~pomel0@user/pomel0)21:07
+ thelounge4202 (~thelounge@gidzit.org)21:20
+ gidzit (~gidzit@gidzit.org)21:21
minutefrickler: but are you using gnome?21:21
cwebber<minute> i have current blender running on rk3588 reform with amd gpu  [09:20]21:21
cwebberWHAT21:21
cwebberYES21:21
minutecwebber: yeah21:21
cwebberYESSSSSSSSSSS21:21
minutecwebber: tiny bit more details https://mastodon.social/@mntmn/11548630737596968421:22
cwebberhow weird/hard is it to specify an egpu when running a program?!21:22
cwebber(what on earth happens if you unplug it @_@)21:22
minutecwebber: hmm good question, normally the compositors i know bind a display to a gpu21:22
minutecwebber: i haven't yet tried to use the external gpu for internal display21:23
minutecwebber: the gpu has like 5 display connectors on its own :D21:23
cwebberwow21:24
cwebberrelatedly I wonder if my pocket is just, fully borked :(21:24
minuteone could do some kind of semi automatic external/internal displayport switcheroo though. that would be quite cursed 21:24
cwebberunrelatedly?21:24
cwebberI dunno21:24
minutecwebber: what happened?21:24
cwebberlooking forward to the charger board getting here but21:24
cwebberI did the thing from the debug page where you borrow a jumper, plugged it in manually21:25
cwebberand it did boot!21:25
cwebberwithout the batteries21:25
cwebberthen I took out the batteries, used the external charger I bought21:25
cwebberand tried charging them and plugging them back in and turn it on to see if it would work if I had manually charged them21:25
minutecwebber: ah, and with charger board it's borked yes?21:26
cwebberno luck, so I put it back to with the jumper for manual-power-on21:26
cwebberbut now it won't boot at all21:26
cwebberI think y'all are shipping me a new charger board but 21:26
cwebberno idea why that is21:26
minutecwebber: hmmm you double-sure that the jumper went to the right pins? (2+3?) because it's pretty fiddly21:26
cwebberI think so21:27
cwebberbut also it may be even more fiddly than that21:27
cwebberbecause21:27
minutecwebber: if it went to 1+2 or 3+4 it would just do nothing21:27
cwebberthe jumper separated21:27
cwebberinto the plastic and the metal21:27
minuteuh21:27
cwebberso I pushed it back together21:27
minuteoh no21:27
cwebberand tried putting it in21:27
minutewell, ok21:27
cwebbermaybe I should try with the other jumper21:27
minutebtw i recently found there are jumpers like this with a longer handle21:27
cwebberanyway I don't actually know there's a problem for sure21:27
minutecwebber: i think it's unlikely that your board just broke21:28
cwebberI just got kinda bummed out that it didn't boot again. maybe I should plug in the other jumper and see21:28
cwebberminute: I hope that's the case21:28
cwebberI do love the pocket21:28
minutecwebber: like, that would be very rare21:28
minutecwebber: the charger board v1 is unfortunately quite breakable though. so it makes sense that it works with the jumper21:28
cwebberI'll try with the other jumper21:28
minutecool21:29
cwebberspeaking of blender21:29
cwebberit's unrelated21:29
cwebberbut I'm excited to use blender with the reform21:29
- gidzit (QUIT: Remote host closed the connection) (~gidzit@gidzit.org)21:29
cwebberblender has become my go-to for *all* artwork recently21:29
+ gidzit (~gidzit@gidzit.org)21:29
cwebber2d and 3d21:29
chminute: is the new headphone board pcb/sch in git? couldnt find it yesterday21:31
minutech: yes, hang on21:31
minutech: here it is https://source.mnt.re/reform/pocket-reform/-/tree/main/pocket-reform-wifi?ref_type=heads21:32
choh21:33
chminute: thanks!21:33
chits cool to have a kicad viewer in the browser, unfortunately not super mobile friendly21:36
minuteomg nice, setting the favorite apps worked on first try on new system image build21:46
minutesooo much better21:46
minutefirefox AI stuff is also nuked by default now21:47
minutegnah, there's another checkbox i didn't notice21:48
minute"Use AI to suggest tabs and a name for tab groups"... wtf21:48
minutealso bing and perplexity in the default search engines... mega sigh21:48
minutealt-gr umlauts are fixed... yes21:49
minuteok brightness also works21:50
- thelounge4202 (QUIT: Quit: The Lounge - https://thelounge.chat) (~thelounge@gidzit.org)21:51
minutejosch: sorry, gnome screen brightness actually works the way you intended--not sure why it works now when baked into reform-tools (vs the hotfix patch)21:57
- gidzit (QUIT: Ping timeout: 256 seconds) (~gidzit@gidzit.org)21:59
joscho022:01
joschi have no explanation22:02
joschminute: you tried the result of the pipeline i started a few hours ago or did you try something else?22:02
- gustav25 (QUIT: Quit: Quit) (~gustav@c-78-82-52-240.bbcust.telenor.se)22:15
minuteah, the checkbox is browser.tabs.groups.smart.userEnabled22:16
minutejosch: i merged that into mine but maybe i held it wrong somehow, or pipeline race22:16
minutejosch: anyway it's solved22:17
+ amospalla (~jordi@user/amospalla)22:19
joschindeed, no reason to spend more time on it if it works -- lets hope it still works when it installs the schema file from the package :)22:19
- AnimaInvicta (PART: !!unknown attribute: msg!!) (~AnimaInvi@88-120-179-216.subs.proxad.net)22:40
- gachikuku_ (QUIT: Read error: Connection reset by peer) (~gachikuku@kamil.gr)22:47
minutejosch: should :D22:47
joschokay, funny "bug": since bmaptool will avoid copying blocks without data, flashing the reform-system-any bootloader did *not* wipe rk3588 u-boot from my sd-card which is why things which i wanted to try out with my local rk3588 u-boot builds did not make any sense (how do you still boot even though i write all zeroes to eMMC???)22:52
joschs/bootloader/system image/22:52
minuteah, there's one more annoying default i want to change, gnome hard-grouping window+their modals together22:55
minuteorg.gnome.mutter attach-modal-dialogs false22:57
- paperManu (QUIT: Ping timeout: 264 seconds) (~paperManu@modemcable141.205-200-24.mc.videotron.ca)22:59
joschnice, --no-inc-recursive gives a nice percentage-based progress output which steadily climbs up to 100% :)23:06
minutejosch: oh nice nice23:07
minutei'm thinking how to marry ptyxis --new-window to x-terminal-emulator... might not be possible easily23:08
minutei can't just do x-terminal-emulator --new-window23:08
minutebecause some terminals won't understand that23:08
vagrantcsomewhere there is documentation about what arguments "x-terminal-emulator" has to support23:09
vagrantcfor exactly this sort of reason23:10
dormitoI've only seen the pocket: but the rp2040 micro is a really neat touch. would be kind of cool if it had of been wired to some boot strapping pins that could be adjusted in the menu. For things like deving your own OS/u-boot/atf builds.23:16
minutevagrantc: aha! will check if there is something there23:17
minuteinteresting https://github.com/kovidgoyal/kitty/discussions/855423:18
minutehaha i love how they don't wanna bolt on single dash options23:19
+ paperManu (~paperManu@107.159.15.124)23:20
minutehell yes, xdg-terminal-exec opens a new window!23:21
josch\o/23:22
minuteextremely un-frustrating23:22
joschminute: you pushed some content into reform-tools git. I'm about to make a new release. Anything else you are working on? Should I wait with releasing?23:34
minutejosch: i just pushed some more comment changes, i might be done with reform-tools but i'm looking at the state of sway right now... sigh23:37
minutein sway, xdg-terminal-exec starts uxterm of course23:37
minutehaha, and if i uninstall xterm (which is required by reform-desktop-full because of legacy reasons), it launches ptyxis, but not in a new window23:39
+ paperManu_ (~paperManu@107.159.15.124)23:40
minutethat's because there's no /usr/share/xdg-terminal-exec/sway-terminals.list23:40
minutejosch: can you remove xterm from reform-desktop-full deps? :323:42
joschsure thing23:47
minuteoh? https://github.com/Alexays/Waybar/blob/161367d9617673a4ef9caf8299411dc5153464d1/src/modules/wlr/taskbar.cpp#L59323:55
minuteah that's local to the taskbar m)) whyyy23:57

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