| - chomwitt (QUIT: Ping timeout: 250 seconds) (~chomwitt@2a02:85f:9a74:ce00:1ac0:4dff:fedb:a3f1) | 00:38 | |
| + bkeys (~Thunderbi@98.19.131.193) | 02:12 | |
| - paperManu (QUIT: Ping timeout: 255 seconds) (~paperManu@198.16.179.107) | 02:28 | |
| - antti (QUIT: Quit: auf Wiedersehen) (~antti@user/antti) | 02:50 | |
| + emdevt (~emdevt@2001:7e8:faf6:f101:70f9:10a4:805e:9fe1) | 03:07 | |
| - nsc (QUIT: Ping timeout: 248 seconds) (~nicolas@i5C74DEEB.versanet.de) | 03:36 | |
| + nsc (~nicolas@23-96-142-46.pool.kielnet.net) | 03:38 | |
| - op_4 (QUIT: Remote host closed the connection) (~tslil@user/op-4/x-9116473) | 04:05 | |
| + op_4 (~tslil@user/op-4/x-9116473) | 04:05 | |
| - vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:50) | 04:23 | |
| + antti (~antti@user/antti) | 04:51 | |
| - antti (QUIT: Client Quit) (~antti@user/antti) | 04:53 | |
| + antti (~antti@user/antti) | 05:00 | |
| - skipwich (QUIT: Remote host closed the connection) (~skipwich@user/skipwich) | 05:01 | |
| + skipwich (~skipwich@user/skipwich) | 05:01 | |
| - antti (QUIT: Quit: auf Wiedersehen) (~antti@user/antti) | 05:19 | |
| + antti (~antti@user/antti) | 07:54 | |
| - antti (QUIT: Quit: auf Wiedersehen) (~antti@user/antti) | 08:23 | |
| + antti (~antti@user/antti) | 08:24 | |
| + hlep (~hlep@user/hlep) | 09:33 | |
| + gustav287 (~gustav@c-78-82-54-94.bbcust.telenor.se) | 10:02 | |
| - sheepyboi (QUIT: Ping timeout: 256 seconds) (~sheepyboi@c-73-233-94-233.hsd1.pa.comcast.net) | 10:30 | |
| + sheepyboi (~sheepyboi@c-73-233-94-233.hsd1.pa.comcast.net) | 10:31 | |
| - hlep (QUIT: Ping timeout: 265 seconds) (~hlep@user/hlep) | 10:53 | |
| - mjw (QUIT: Ping timeout: 255 seconds) (~mjw@2a02-a472-ffa1-1-f2e6-4728-2974-38b8.fixed6.kpn.net) | 11:00 | |
| + chomwitt (~chomwitt@2a02:85f:9a74:ce00:1ac0:4dff:fedb:a3f1) | 11:17 | |
| + aperezdc (~aperezdc@2a03:6000:6e61:633::43) | 12:36 | |
| + mjw (~mjw@2a02-a472-ffa1-1-f2e6-4728-2974-38b8.fixed6.kpn.net) | 12:50 | |
| - chomwitt (QUIT: Ping timeout: 256 seconds) (~chomwitt@2a02:85f:9a74:ce00:1ac0:4dff:fedb:a3f1) | 13:07 | |
| * dodo -> librehans | 13:20 | |
| - jjbliss (QUIT: Quit: nyaa~) (~jjbliss@2601:404:cb00:48e0:730:cb49:692d:8e3) | 13:21 | |
| - ch (QUIT: Quit: ch) (~ch@user/meow/ch) | 13:34 | |
| + ch (~ch@user/meow/ch) | 13:38 | |
| + jjbliss (~jjbliss@2601:404:cb00:48e0:730:cb49:692d:8e3) | 14:05 | |
| + paperManu (~paperManu@198.16.179.107) | 14:21 | |
| - sheepyboi (QUIT: Ping timeout: 258 seconds) (~sheepyboi@c-73-233-94-233.hsd1.pa.comcast.net) | 14:37 | |
| + sheepyboi2 (~sheepyboi@c-73-233-94-233.hsd1.pa.comcast.net) | 14:41 | |
| + manis (01a66df340@185.72.67.185) | 14:44 | |
| ch | ok my not-charging-from-empty issue is a firmware issue i think | 15:04 |
|---|---|---|
| ch | with everything in place vcc on the charger connector is 0.6V | 15:04 |
| ch | with the PROG sw set to ON i get 5V on there | 15:05 |
| ch | so my guess is that sysctl boots and immediately turns off the charger input | 15:05 |
| ch | flashed a new firmware, now to actually make it discharge | 15:58 |
| ch | animone: which voltages are you seeing? | 16:00 |
| minute | ch: thanks for the MR! | 16:31 |
| - emdevt (QUIT: Ping timeout: 248 seconds) (~emdevt@2001:7e8:faf6:f101:70f9:10a4:805e:9fe1) | 16:32 | |
| minute | animone: you said isolated. but with bypass jumper on the charger connector on mobo? | 16:32 |
| - mjw (QUIT: Ping timeout: 250 seconds) (~mjw@2a02-a472-ffa1-1-f2e6-4728-2974-38b8.fixed6.kpn.net) | 16:45 | |
| + chomwitt (~chomwitt@2a02:85f:9a74:ce00:1ac0:4dff:fedb:a3f1) | 16:49 | |
| - jjbliss (QUIT: Quit: nyaa~) (~jjbliss@2601:404:cb00:48e0:730:cb49:692d:8e3) | 17:01 | |
| ch | minute: sure thing. also i guess its my fault in the first place | 17:04 |
| ch | wondering now if i should change my install to trixie | 17:08 |
| - svp (QUIT: Ping timeout: 260 seconds) (~svp@2002:4f07:f0bd:0:69b1:b463:d245:e861) | 17:12 | |
| + mjw (~mjw@178.230.61.124) | 17:15 | |
| minute | i'm currently exploring kde plasma on pocket! (incl with external screen) | 17:19 |
| + svp (~svp@host-79-7-240-189.business.telecomitalia.it) | 17:19 | |
| animone | yes, I jumpered pins 2-3, so pin 1-4 all read 3.8V with the standby switch on | 17:23 |
| animone | and pin 5 reads.. 0-1.7V | 17:23 |
| animone | 1-4 are 5V ith the swich off | 17:23 |
| ch | maybe turn ON the prog switch, and see if that makes a difference | 17:28 |
| ch | (with standby switch on) | 17:28 |
| animone | ok, with PROG ON, 4.8V VCC and 3.3V standby | 17:30 |
| animone | no LED though | 17:32 |
| ch | well the led is controlled by the sysctl, which is now not starting | 17:32 |
| animone | I see, at least this tells me it's probably not a hardware issue | 17:33 |
| ch | if the batts are 'just empty', then you can try what i did earlier today. put the batts and charger connections back, standby off, prog on, plug in a charger, standby on, wait a bit | 17:33 |
| ch | like 5-10min | 17:33 |
| ch | then standby off, prog off, standby on and see if the oled responds | 17:33 |
| minute | makes sense! | 17:37 |
| ch | alternatively try flashing https://source.mnt.re/reform/pocket-reform/-/merge_requests/47 but thats more work and also super fresh | 17:38 |
| animone | nice! I have an OLED | 17:40 |
| animone | when flashing, should I have the usb-pd plugged in or no? | 17:41 |
| animone | the instructions are a little vague about providing power | 17:42 |
| ch | tbh i would let it continue charging for a while and then boot normally and with usb-pd plugged, then flash (still plugged) | 17:43 |
| animone | ok, I'll wait a bit and see if the batteries charge | 17:43 |
| animone | batteries are charging, so at least the charger board is working, yay | 17:47 |
| ch | minute: thanks! i was about to click rebase xD | 17:51 |
| minute | ch: haha sorry, trigger fingers | 17:51 |
| ch | nah its great :) | 17:53 |
| - mjw (QUIT: Remote host closed the connection) (~mjw@178.230.61.124) | 17:54 | |
| + mjw (~mjw@178.230.61.124) | 17:54 | |
| animone | I'm going to let this charge up for a couple hours to get to something reasonable first | 17:57 |
| animone | should I then try booting with PROG ON, and flash from the OS, or go through the 'unbricking' flashing? | 17:58 |
| minute | animone: i wouldn't boot with prog on | 17:58 |
| animone | ok | 17:59 |
| minute | animone: i also think that wouldn't be easily possible because without sysctl firmware you can't turn on the power rails | 17:59 |
| animone | oh, right | 17:59 |
| minute | animone: and a bunch of signals would be in undefined states, especially for the display | 17:59 |
| minute | ok KDE is really awesome and i'm gonna switch to it from gnome at least for a while, but... the overlay plane used for mouse cursor on the internal display (DSI) instead of HDMI is buggy and throws kernel errors as soon as that cursor is being cropped at the screen edges, and has no alpha | 18:01 |
| minute | i'm gonna try KWIN_FORCE_SW_CURSOR to see if that's still fast enough | 18:01 |
| animone | I'll see if everything works normally (PROG off) in a couple hours and poke back here | 18:02 |
| minute | ok wow, with KWIN_FORCE_SW_CURSOR=1 the cursor is fixed but still fast and doesn't stutter when the shell draws unlike in GNOME... so KDE/kwin have proper multithreading for that it seems | 18:04 |
| minute | josch: we can ship plasma-desktop if we setup KWIN_FORCE_SW_CURSOR=1, and we should do it in a way that we can remove it later if this problem is fixed (not sure if necessary though). we also might need to switch to sddm, as gdm failed to correctly start plasma earlier... | 18:08 |
| minute | also, we might need to remove konqueror and make sure firefox is the default browser, because konqueror causes extreme graphics glitches + gpu faults on panfrost | 18:11 |
| - mjw (QUIT: Remote host closed the connection) (~mjw@178.230.61.124) | 18:13 | |
| + mjw (~mjw@178.230.61.124) | 18:13 | |
| josch | minute: nice that you found the KWIN_FORCE_SW_CURSOR=1 workaround. I also saw the black-box-cursor in the past in kde but didn't know how to fix it. The problem with shipping environment variables is, that those are stored in /etc and can thus be nasty when upgrading reform-tools with a config file changed by the user. Thus, "undoing" that workaround might not be too trivial. I'll have a look into the | 18:20 |
| josch | code whether there is another knob that one can turn to enable the software cursor. | 18:20 |
| josch | nope, there seems to be no other way than via the environment variable: https://sources.debian.org/src/kwin/4:6.3.6-1/src/compositor_wayland.cpp?hl=495#L515 | 18:22 |
| minute | josch: is there no place in /usr that can play the role of | 18:28 |
| minute | josch: /etc/environment...? | 18:29 |
| josch | not as far as i know, no | 18:29 |
| josch | but i wish there was | 18:29 |
| minute | ok hmm... | 18:29 |
| josch | on the other hand: is it that important to have it easily revertible via reform-tools upgrade? | 18:29 |
| - mjw (QUIT: Ping timeout: 256 seconds) (~mjw@178.230.61.124) | 18:30 | |
| josch | i suppose only few users will install ssdm+plasma and then there could be a reform-check in the future which checks whether whatever custom setup reform-setup-wizard performed to set up kde plasma is adjusted, no? | 18:30 |
| minute | interesting, KDE have a way to fix/improve things for cash https://discuss.kde.org/c/development/sponsored-work/31 | 18:30 |
| minute | josch: i currently think that we might, at some point, switch from gnome to kde as the default and make gnome one of the secondary options :3 | 18:31 |
| josch | oh interesting | 18:32 |
| minute | josch: because i don't see a solution for the cursor issue on gnome, and the shell's architecture prevents it from working smoothly without hardware sprites | 18:32 |
| minute | josch: the issue is open for months and isn't getting any attention | 18:32 |
| minute | josch: gnome has 120 open bugs tagged with "performance"... | 18:33 |
| josch | heh | 18:33 |
| minute | so it might not be the ideal choice for low-power systems after all | 18:33 |
| josch | hrm... i benchmarked gnome vs. sway when idling, let me look up the numbers... | 18:34 |
| minute | it's more minimalist and thus easier to manage, but i'm concerned with the usability and the perception of our devices being impacted by these issues | 18:34 |
| josch | https://community.mnt.re/t/battery-life-tests-with-rk3588/3397/15 says only 3 minutes (thus negligible) of battery life difference between sway and gnome | 18:36 |
| josch | tested workload was: "So I reduced the display backlight to minimum and turned off the keyboard backlight. Then I opened a terminal and ran htop" | 18:36 |
| minute | josch: that's cool, but compares only battery runtime when not doing typical work :D | 18:37 |
| josch | yes | 18:37 |
| josch | moving cursor and windows around might eat the battery -- but it's very hard to reproducibly test | 18:37 |
| minute | josch: i haven't compared energy use yet, but what i mean is responsiveness of the UI. i think it's really very bad when the cursor can stop moving for 100+ ms, and it happens all the time when using gnome shell functions (calendar, control panel etc) | 18:38 |
| josch | indeed, that sounds very bad | 18:38 |
| minute | josch: AFAIK it's because gnome-shell's rendering, including the cursor, happens in the same thread that manages all this UI (and those are written in javascript) | 18:39 |
| josch | minute: how about we just stick KWIN_FORCE_SW_CURSOR=1 into /etc/profile.d/reform.sh because that's the only way we know how to do it and deal with the problem when they come up? | 18:39 |
| minute | josch: and their medicine for that is to just bypass all GL rendering by using a hardware sprite for the cursor. which makes it rely in special, outside-of-GL hardware features | 18:39 |
| josch | unless you want to try and fix it properly first? | 18:39 |
| minute | josch: well, KDE is not really usable on pocket without this variable, so we need to make sure to ship it to existing users | 18:41 |
| minute | josch: when you move the cursor to the screen edge, the display blanks, sound drops out, kernel errors appear | 18:41 |
| minute | josch: because the overlay plane doesn't support what they're trying to do (arbitrarily cropping the "sprite") | 18:41 |
| minute | josch: sorry, i mean on RK3588 pocket (and DSI RK3588 reform) | 18:42 |
| minute | josch: maybe we could have some sort of management tool/script for /etc/environment that becomes part of reform-tools and is called also by a post install step? which checks if the env var is there, and if not, adds it? | 18:43 |
| josch | i'll have to run some upgrade tests to see what the user experience is in different scenarios | 18:43 |
| minute | another more heavyweight option would be to patch kwin. but it feels like a bigger burden | 18:44 |
| josch | there is also the option to ship a file /etc/profile.d/reform-kwin.sh which only includes KWIN_FORCE_SW_CURSOR=1 and which can be simply removed later | 18:44 |
| josch | removing a file is easier than parsing and editing one | 18:44 |
| minute | josch: oh yeah, totally. but we can't put that in a .deb, if i understood the rules correctly, yes? | 18:47 |
| minute | josch: but a script managing this file would be ok? | 18:47 |
| minute | hmm actually packages put default configs in /etc all the time, they just don't get removed when uninstalling, right? (unless --purge ?) | 18:48 |
| josch | minute: yes, we can easily put /etc/profile.d/reform-kwin.sh in reform-tools.deb but upgrading reform-tools would *not* remove the file without some magic. Doing so is not too difficult though and we did that in the past with /etc/flash-kernel/ubootenv.d/00reform2_ubootenv and /etc/modprobe.d/reform.conf | 18:51 |
| josch | but remove-on-upgrade would only work with a separate file for the setting, making that option attractive | 18:52 |
| josch | this is the packaging file handling the removal: https://source.mnt.re/reform/reform-debian-packages/-/blob/main/reform-tools/debian/reform-tools.conffiles?ref_type=heads | 18:52 |
| + kensanata (~alex@user/kensanata) | 18:57 | |
| - ch (QUIT: *.net *.split) (~ch@user/meow/ch) | 18:59 | |
| - gustav287 (QUIT: *.net *.split) (~gustav@c-78-82-54-94.bbcust.telenor.se) | 18:59 | |
| - skipwich (QUIT: *.net *.split) (~skipwich@user/skipwich) | 18:59 | |
| - op_4 (QUIT: *.net *.split) (~tslil@user/op-4/x-9116473) | 18:59 | |
| - gsora (QUIT: *.net *.split) (~gsora@user/gsora) | 18:59 | |
| - vkoskiv_ (QUIT: *.net *.split) (~vkoskiv@109-204-186-7.cust.valoonet.fi) | 18:59 | |
| - amk (QUIT: *.net *.split) (~amk@user/amk) | 18:59 | |
| - kensanata6 (QUIT: *.net *.split) (~alex@user/kensanata) | 18:59 | |
| - schalken (QUIT: *.net *.split) (~schalken@117-118-178-69.gci.net) | 18:59 | |
| + op_4 (~tslil@user/op-4/x-9116473) | 18:59 | |
| minute | josch: ahh very interesting @ remove-on-upgrade, thanks! so this seems like a good way forward then | 19:00 |
| + ch (~ch@user/meow/ch) | 19:00 | |
| minute | besides this, the only hard hurdle is the failing start from gdm (will look into it) and, similar to gnome, defaulting to bad non integer scaling | 19:00 |
| minute | (which also causes gfx bugs) | 19:01 |
| + mjw (~mjw@gnu.wildebeest.org) | 19:02 | |
| josch | in the meantime i found your fedi post about the "about this system" dialog in kde and am reading up *how* they acquire this information | 19:02 |
| josch | for example, i was not aware of /proc/device-tree/chosen/u-boot,version | 19:02 |
| minute | oh same | 19:03 |
| minute | ok, starting plasma from GDM worked now | 19:03 |
| minute | not sure why, maybe because of some initial setup that had to be done... | 19:03 |
| + vkoskiv (~vkoskiv@109-204-186-7.cust.valoonet.fi) | 19:04 | |
| + gustav287 (~gustav@c-78-82-54-94.bbcust.telenor.se) | 19:04 | |
| + skipwich (~skipwich@user/skipwich) | 19:04 | |
| + amk (~amk@user/amk) | 19:04 | |
| + schalken (~schalken@117-118-178-69.gci.net) | 19:04 | |
| minute | but now there are new strange graphics problems: overview makes all screens completely black, and text rendering in window switcher is broken (only shows rectangles instead of glyphs) | 19:05 |
| minute | i wonder if that's depending on starting from gdm. will try starting from console again | 19:06 |
| minute | indeed | 19:07 |
| + gsora (~gsora@user/gsora) | 19:07 | |
| minute | josch: starting plasma from gdm yields in broken rendering. starting from console with just `startplasma-wayland` and everything is fine. not good | 19:07 |
| minute | i'll try to switch to sddm... | 19:08 |
| minute | `sudo dpkg-reconfigure sddm` | 19:08 |
| minute | hmm, plymouth complains about label-pango.so missing | 19:15 |
| minute | (it's the package plymouth-label) | 19:17 |
| josch | now that i learned how KDE gets the data in its "about this system" dialog i wonder if reform-check should be running "solid-hardware6 list details" and show something similar... | 19:18 |
| minute | josch: why not :3 | 19:19 |
| josch | i know, right? :D | 19:20 |
| ch | otoh theres a lot of stuff in there thats kinda meh? | 19:20 |
| minute | sddm works, but looks kinda garish by default... i think it also does some weird non integer scaling. but the KDE plasma started from it does not have rendering issues like when it's started from gdm. that smells like gdm is adding some env variables maybe | 19:20 |
| ch | :) | 19:20 |
| josch | ch: in where? | 19:20 |
| ch | `solid-hardware6 list details` | 19:20 |
| josch | ch: i was not talking about just cat-ing it unfiltered :) | 19:21 |
| ch | right | 19:21 |
| ch | but then the interesting things look like find /sys :D | 19:21 |
| ch | josch: did you say the other day the pocket hid firmware is not on fwupd yet? | 19:21 |
| josch | yes, but the advantage is that *somebody else* implemented it :) | 19:21 |
| josch | ch: yes, and i only repeated what minute said | 19:21 |
| minute | ch: the pocket hid firmware is not on fwupd yet. the pocket sysctl is. | 19:21 |
| ch | is there something i can do to make that happen? | 19:22 |
| minute | ch: totally | 19:22 |
| minute | ch: if you feel adventurous, you can port over the same things that were needed for the sysctl-fw to the hid fw | 19:22 |
| + AnimaInvicta (~AnimaInvi@88-120-179-216.subs.proxad.net) | 19:23 | |
| ch | josch: lshw also exists btw | 19:23 |
| josch | ACTION whispers *do it*... | 19:23 |
| ch | i thought thats done | 19:23 |
| josch | https://fwupd.org/lvfs/search?value=mnt says no | 19:23 |
| minute | josch: well that part is my responsibility | 19:24 |
| ch | `fwupdmgr get-devices` gives me `Pocket Reform Input 1.0` and flags=Updatable | 19:24 |
| minute | ok let me check. indeed there's more here than i remembered https://source.mnt.re/reform/pocket-reform/-/tree/main/pocket-reform-keyboard-fw/pocket-hid?ref_type=heads | 19:24 |
| minute | ch: looks like you did the required work already in january. sorry! | 19:25 |
| ch | np | 19:25 |
| minute | there's even a .cab in CI already | 19:26 |
| bremner | josch: btw, there _is_ a discourse thread about hibernate on the pocket rk3588. Not sure how helpful it is for onboarding, but maybe better than IRC? | 19:28 |
| josch | oh wow, i just learned/remembered that we already kicked everything in /etc out of reform-tools except for /etc/kernel/postinst.d/zz-reform-tools which we can kick out now that the kernel supports /usr/share/kernel/postinst.d | 19:28 |
| josch | bremner: please mentioned it in the thread if you have the link at hand :) | 19:29 |
| minute | > pocket-hid.cab was uploaded 13 seconds ago | 19:29 |
| ch | maybe longer text for FWCAB_FIRMWARE_DESCRIPTION might be needed or something | 19:29 |
| minute | ch: yes, > Not enough useful words, minimum is 5 and was 1 | 19:30 |
| minute | ch: and the other blocker is that someone has to install it via fwupd and upload a success report :D | 19:30 |
| minute | ch: but maybe you can see it if you have the embargo repo active still > The firmware is in the embargo-mntre remote. It was moved 2 minutes ago | 19:31 |
| minute | ok, now trying to remember how to activate that "splash" commandline thing | 19:31 |
| josch | minute: i also aways forget and when i do, i usually run: | 19:32 |
| josch | dpkg -L reform-tools | grep u-boot-m | 19:32 |
| minute | :0 | 19:32 |
| josch | and that shows me /usr/share/u-boot-menu/conf.d/reform.conf and there its "documented" | 19:32 |
| minute | many thanks, i will test this finally now | 19:33 |
| minute | huh, looks like i again didn't have u-boot-menu | 19:34 |
| ch | https://paste.debian.net/plain/1395443 "you're drunk, fwupd!" | 19:39 |
| minute | josch: looks like u-boot-menu already adds alternative entries with "splash" in them, i guess that mirrors the "graphical" vs non-graphical boot options i faintly remember from debian with grub on x86? (but they're not the default) | 19:42 |
| minute | ch: heh! | 19:43 |
| ch | i think its really not visible in the embargo remote | 19:44 |
| minute | ok, maybe it's still propagating to CDNs or something | 19:47 |
| ch | https://source.mnt.re/reform/pocket-reform/-/merge_requests/49 this adds some words, just in case | 19:47 |
| minute | josch: oh wow, i had a very strange looking splash now, probably some debian default :D how do i get yours? | 19:49 |
| minute | ch: words are good, it likes words | 19:50 |
| minute | josch: also, sigh that a tiny bit of console stuff still flashes between plymouth disappearing and sddm appearing, no clean transition | 19:51 |
| - Batto (QUIT: Remote host closed the connection) (~joona@user/Batto) | 20:04 | |
| + Batto (~joona@user/Batto) | 20:10 | |
| ch | josch: https://paste.debian.net/plain/1395449 | 20:20 |
| minute | trying `sudo plymouth-set-default-theme reform-y2k` | 20:21 |
| minute | hmm, didn't we once ship some skel files to mask autostart of blueman.desktop, nm-applet.desktop, pasystray.desktop | 20:28 |
| minute | indeed we did, they're just missing in my account | 20:28 |
| minute | josch: your plymouth reform-y2k theme is much better than the default for sure, cool! only missing the inter font and a transparent png in place of the debian logo (not because i don't like the debian logo, it's just an aesthetic clash). otherwise awesome | 20:39 |
| - gustav287 (QUIT: Quit: Quit) (~gustav@c-78-82-54-94.bbcust.telenor.se) | 22:15 | |
| ch | ║ Downgrade Pocket Reform Input 1.0 from 20250429 to gbeaad31? ║ | 22:31 |
| ch | minute: i uploaded the upgrade report (i think) | 22:34 |
| ch | downgrade report i guess | 22:34 |
| - svp (QUIT: Quit: Gateway shutdown) (~svp@host-79-7-240-189.business.telecomitalia.it) | 22:57 | |
| + svp (~svp@2002:4f07:f0bd:0:69b1:b463:d245:e861) | 22:59 | |
| josch | minute: it does? For me it generates extra entries with "single" in them for rescue mode. | 23:06 |
| josch | minute: in reform-system-image i enable it like this: https://source.mnt.re/reform/reform-system-image/-/blob/main/mkimage.sh?ref_type=heads#L484 | 23:06 |
| josch | well, "set it as the default" that is | 23:07 |
| josch | ch: uuuuh where do i have to click to merge that? :) | 23:07 |
| ch | reform-tools | 23:08 |
| ch | :D | 23:08 |
| ch | would be good if i'd find my usb-pd debug hardware | 23:08 |
| * plomlompom -> plomlomp0m | 23:09 | |
| josch | ch: thank you! | 23:09 |
| josch | minute: the desktop files are shipped here: https://source.mnt.re/reform/reform-system-image/-/tree/main/etc/skel/.config/autostart?ref_type=heads | 23:11 |
| josch | minute: if you see the debian logo in the reform-y2k theme you might be missing the reform-branding package on your system? | 23:11 |
| ch | looks like sysctl has a crashbug that i can trigger with a power bank like device | 23:20 |
| ch | what the hell | 23:21 |
| josch | ch: see it this way: good that you run into this and not somebody who doesn't know how to fix it :) | 23:30 |
| ch | well i also don't know how to fix it :D | 23:30 |
| josch | ...yet :) | 23:31 |
| minute | looks like swd might be helpful at some point... | 23:31 |
| minute | i've also had some weird sysctl crashes after poweroff. it always recovers by watchdog though. still, interesting to figure out what happens there | 23:31 |
| minute | also, i wonder if static analysis might yield something | 23:32 |
| ch | yeah watchdog brings it back | 23:32 |
| josch | minute: if you want to play more with plymouth, edit /usr/share/plymouth/themes/reform-y2k/reform-y2k.plymouth and set Font and TitleFont to "Inter Medium 11" and then put this file https://paste.debian.net/hidden/20dace27/ into /usr/share//initramfs-tools/hooks/reform-plymouth and make it executable and then rebuild your initramfs | 23:33 |
| josch | background: if you want a different font (inter) then we somehow have to copy it to the initramfs and run fontconfig in it | 23:34 |
| minute | josch: ok, thanks :33 | 23:38 |
| animone | I'm still waiting on my batteries to charge up.. it's nearly there.. | 23:47 |
| animone | oh, this has really old keyboard firmware: HID20240416 | 23:47 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!