| + chrcav (~chrcav@user/chrcav) | 01:00 | |
| - vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:50) | 01:46 | |
| - wiedi (QUIT: Quit: ^C) (~wiedi@ip5f5908ad.dynamic.kabel-deutschland.de) | 02:27 | |
| + robin_ (~robin@user/terpri) | 02:33 | |
| - robin (QUIT: Ping timeout: 252 seconds) (~robin@user/terpri) | 02:35 | |
| - paperManu (QUIT: Ping timeout: 276 seconds) (~paperManu@107.159.213.145) | 03:10 | |
| + _justin_kelly71 (~justinkel@user/justin-kelly/x-6011154) | 06:35 | |
| + chomwitt (~chomwitt@2a02:85f:9a00:8300:1ac0:4dff:fedb:a3f1) | 07:05 | |
| josch | minute: the default sway config does not enable swayidle (with swaylock and brightnessctl) by default -- is that intentional | 08:13 |
|---|---|---|
| josch | (under gnome the problem is kinda the opposite: the default suspends the device while idle and then resume fails but that problem is already tracked in the community thread) | 08:14 |
| josch | oh bug: on fresh system images under sway, ~/.local/bin is not in PATH under sway -- but it *is* under gnome (both started with gdm) | 08:30 |
| josch | zeha: any opinion where you would hack that in? | 08:31 |
| + wiedi (~wiedi@ip5f5908ad.dynamic.kabel-deutschland.de) | 08:36 | |
| zeha | sway config, i think i even have it there | 09:10 |
| josch | yay i *finally* have an actual problem with my pocket \o/ | 09:21 |
| josch | so i had sway open, watched a video in mpv in one workspace, then pulled up another and opened htop in a terminal | 09:22 |
| josch | usb-c cable says that 10 W were pulled | 09:22 |
| josch | suddenly the unit switches off and 3 W are pulled | 09:22 |
| josch | pressing the buttons for the oled menu does nothing | 09:22 |
| josch | how to proceed? :) | 09:22 |
| * Guest9573 -> mjw | 09:32 | |
| - svp (QUIT: Read error: Connection reset by peer) (~svp@host-79-7-240-189.business.telecomitalia.it) | 10:04 | |
| + svp_ (~svp@host-79-7-240-189.business.telecomitalia.it) | 10:04 | |
| * svp_ -> svp | 10:04 | |
| - sterni (QUIT: ) (~quassel@user/sterni) | 10:15 | |
| + sterni (~quassel@user/sterni) | 10:15 | |
| josch | I measured both pouch cells -- both are at 4.0 V | 10:25 |
| josch | which makes no sense... | 10:25 |
| [tj] | is the charger board ok? | 10:26 |
| josch | huh but the other multimeter agrees -- i guess they are just very full | 10:28 |
| josch | [tj]: i see nothing odd with it | 10:28 |
| [tj] | 4.2 is full no? | 10:29 |
| josch | no clue -- i am used to lifepo4 :D | 10:30 |
| [tj] | :D | 10:31 |
| [tj] | I've no confidence is anything I might know these days, maybe I need more sleep | 10:31 |
| [tj] | so it doesn't power up at all? | 10:32 |
| josch | [tj]: if i press the diagonal arrow key in the bottom-left and enter and hold them, the oled does not power up | 10:33 |
| josch | photo of charger board: https://mister-muffin.de/p/9BOW.jpg | 10:34 |
| frickler | josch: did you try a hard power cycle via the switch? | 10:34 |
| [tj] | any chance the battery disconnect switch has moved? | 10:34 |
| josch | frickler: you mean the standby switch? yes i tried moving that switch | 10:35 |
| josch | [tj]: what is the battery disconnect switch? you mean the standby power switch? yes, i have toggled that and it's currently set to "away from the edge, towards the center" | 10:56 |
| [tj] | that is where mine is | 10:56 |
| [tj] | you can by pass the battery/charge controller, but I don't have a link for how to do it | 10:57 |
| minute | just yesterday we worked on a structured guide for this, but it isn't done yet (missing images) | 11:01 |
| minute | josch: charger board bypass works by unplugging the two cables from the wide 9pin connector on the bottom left of motherboard | 11:02 |
| minute | josch: then you borrow a 2mm jumper/shunt from the internal usb header (just below internal usb-c) | 11:02 |
| minute | josch: this you plug on pins 2+3 of the CHARGER connector (where you unplugged the two cables from) | 11:03 |
| minute | josch: pins are counted from the left | 11:03 |
| minute | josch: this way power to charger and power from charger lines are bridged and you can directly power the board from usb-c pd | 11:03 |
| josch | minute: thank you! that worked perfectly! | 11:07 |
| josch | pocket booted into sway again :) | 11:08 |
| - chomwitt (QUIT: Ping timeout: 272 seconds) (~chomwitt@2a02:85f:9a00:8300:1ac0:4dff:fedb:a3f1) | 11:08 | |
| - svp (QUIT: Remote host closed the connection) (~svp@host-79-7-240-189.business.telecomitalia.it) | 11:15 | |
| + svp (~svp@host-79-7-240-189.business.telecomitalia.it) | 11:15 | |
| minute | josch: then you need a new charger board. sorry for this! it looks like this problem, if it exists, manifests very early in use, hmm. we could probably stress test the remaining boards | 11:15 |
| josch | minute: i remember that you made changes to the charging board? i hope my unit did not have the revised version in it? | 11:17 |
| josch | no need to be sorry -- this is much more frustrating for you than it is for me | 11:17 |
| minute | josch: no, the revised version is not done yet | 11:17 |
| josch | ah okay good | 11:17 |
| josch | i'm just happy that this was just the charging board problem | 11:17 |
| minute | right now we just replace people's charger boards for free if this happens | 11:17 |
| josch | okay, i'll just write to support@mntre.com to attach a new board to one of my remaining shop orders :) | 11:18 |
| asjackson | are there many mnt pockets still to be shipped? think i ordered jan 1st so hopefully soon :) no worries if not, just excited | 11:19 |
| minute | asjackson: yes :3 currently waiting for rcore v2 batch, should come in 1-2 wks | 11:21 |
| asjackson | exciting :3 | 11:22 |
| asjackson | thanks! | 11:22 |
| josch | minute: would it in any way useful to you to have one more of the broken boards back? or should i put it into the bin? | 11:23 |
| minute | josch: thanks, we have enough samples now, it can be recycled | 11:23 |
| josch | okay! | 11:24 |
| BoostisBetter | minute: so it is just assumed that the charging board failed. | 11:24 |
| minute | BoostisBetter: it is verified | 11:30 |
| BoostisBetter | minute: ah ok. Well in other news, I am going on 10 days of uptime on my pocket. It is great having it back in use. | 11:31 |
| minute | BoostisBetter: if the computer is dead with the charging board connected, but fine with the charging board disconnected, the charging board is logically the issue. well, ok, it could also be one of the cables. that hasn't happened yet afaik | 11:31 |
| josch | minute: is there something on the charging board one could measure with batteries attached to easily "make sure"? | 11:34 |
| - GNUmoon (QUIT: Remote host closed the connection) (~GNUmoon@gateway/tor-sasl/gnumoon) | 12:23 | |
| + GNUmoon (~GNUmoon@gateway/tor-sasl/gnumoon) | 12:24 | |
| + paperManu (~paperManu@107.159.213.145) | 12:31 | |
| + chomwitt (~chomwitt@2a02:85f:9a00:8300:1ac0:4dff:fedb:a3f1) | 12:46 | |
| + mark_ (~mjw@gnu.wildebeest.org) | 12:47 | |
| + mmmm (~mmmm@ip2504e528.dynamic.kabel-deutschland.de) | 13:02 | |
| + gustav28 (~gustav@c-78-82-53-72.bbcust.telenor.se) | 13:03 | |
| * mmmm -> josch-pocket | 13:03 | |
| josch | page 75 in https://mntre.com/documentation/pocket-reform-handbook.pdf says: "So only do this when the flashing script asks you to do so." | 13:24 |
| josch | i do not see any script doing such a thing in there | 13:25 |
| josch | pocket-reform-keyboard-fw/pocket-hid/flash.sh just calls "picotool load" | 13:25 |
| - josch-pocket (QUIT: Read error: Connection reset by peer) (~mmmm@ip2504e528.dynamic.kabel-deutschland.de) | 13:40 | |
| josch | fixed here: https://source.mnt.re/reform/pocket-reform/-/merge_requests/41 | 13:44 |
| + mmmm (~mmmm@ip2504e528.dynamic.kabel-deutschland.de) | 13:45 | |
| * mmmm -> josch-pocket | 13:45 | |
| josch-pocket | i now patched my keyboary firmware to have scroll and middle mouse button swapped | 13:46 |
| josch-pocket | so it's scroll, left, middle, right | 13:46 |
| josch-pocket | this was driving me insane and d | 13:47 |
| josch-pocket | and did not get better | 13:47 |
| josch-pocket | i would have my fingers hovering over one of the mouse keys for agonizing seconds before ultimately hitting the wrong button... | 13:48 |
| josch | f_: mmmm is the username that i give throw-away installations because it's easy to type and the same key on the german, english and neo2 layout because there is this bug with reform-setup-wizard: https://source.mnt.re/reform/mnt-reform-setup-wizard/-/issues/16 | 13:50 |
| josch | ooooh i was looking at the wrong script! | 13:58 |
| josch | it's scripts/update-keyboard-firmware.sh | 13:58 |
| josch | f_: wow you are indeed a ghost as far as the logging is concerned | 14:01 |
| minute | josch-pocket: we now have many competing flashing scripts | 14:27 |
| josch | i should sit together with zeha and think of a better solution https://xkcd.com/927/ | 14:27 |
| josch | on the other hand, if fwupd works, these scripts are only of limited value and that also only to those who are not afraid to compile their own stuff | 14:28 |
| minute | josch: yes. the universal update script that i made is just to get all kind of weird versions on the latest version that is fwupd capable | 14:34 |
| minute | josch: another thing, TIL about .dmrc | 14:34 |
| minute | josch: i'm thinking that the quickest/MVP way to switch from greetd to gdm might be writing a ~/.dmrc with the default desktop choice from the installer (currently it writes a greetd config file) | 14:35 |
| minute | (by installer i mean reform setup wizard) | 14:35 |
| minute | as in, switching to gdm would be the first step towards shipping gnome | 14:35 |
| minute | we can then also bring the more noisy log level back | 14:35 |
| josch | okay, i didn't know about ~/.dmrc -- i can read up about that and then propose a MR to reform-system-image which replaces greetd with gdm | 14:37 |
| minute | > The user's default session and language choices are stored in the ~/.dmrc file. When a user logs in for the first time, this file is created with the user's initial choices. The user can change these default values by simply changing to a different value when logging in. GDM will remember this change for subsequent logins | 14:37 |
| minute | apparently it does not use this file again though (i just tested) | 14:37 |
| minute | so let me erase my gdm config files | 14:37 |
| minute | this is a bit of a pandora's box https://wiki.archlinux.org/title/GDM | 14:39 |
| minute | > Some GDM settings are stored in a dconf database | 14:41 |
| minute | > Some | 14:41 |
| minute | why not all in one place ;__; | 14:41 |
| minute | uff uff uff | 14:47 |
| minute | the session choice is stored in /var/lib/AccountsService/users/<username> | 14:47 |
| minute | and indeed, changing the Session=gnome to Session=sway makes gdm login to sway | 14:49 |
| minute | in this file | 14:49 |
| minute | the .dmrc is seemingly ignored | 14:49 |
| minute | ah, it's complicated https://gitlab.gnome.org/GNOME/gdm/-/commit/6e2429a11efa64370386f85c79dc5995375d9ae2#21330aa240d761148218157a4a83dd3cdac67465_1587_1669 | 14:53 |
| minute | accountsservice has some templates for these files, /usr/share/accountsservice/user-templates | 14:56 |
| - josch-pocket (QUIT: Quit: josch-pocket) (~mmmm@ip2504e528.dynamic.kabel-deutschland.de) | 15:01 | |
| minute | zeha: any chance you could try the kernel from this build on your pocket? https://source.mnt.re/reform/reform-debian-packages/-/jobs/10184/artifacts/browse/repo/pool/main/l/linux/ | 15:03 |
| minute | it works for me (disp v1) and josch (disp v2) but not clear if after reboot display works with this version for your disp v1 | 15:04 |
| minute | (and it has the panel orientation in-kernel) | 15:04 |
| - bkeys (QUIT: Ping timeout: 260 seconds) (~Thunderbi@h211.16.186.173.dynamic.ip.windstream.net) | 15:15 | |
| + bkeys (~Thunderbi@66.110.201.50) | 15:21 | |
| - bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@66.110.201.50) | 15:32 | |
| + bkeys (~Thunderbi@66.110.201.50) | 15:32 | |
| - bkeys (QUIT: Client Quit) (~Thunderbi@66.110.201.50) | 15:33 | |
| + bkeys (~Thunderbi@66.110.201.50) | 15:34 | |
| - bkeys (QUIT: Client Quit) (~Thunderbi@66.110.201.50) | 15:35 | |
| + bkeys1 (~Thunderbi@66.110.201.50) | 15:35 | |
| - bkeys1 (QUIT: Client Quit) (~Thunderbi@66.110.201.50) | 15:37 | |
| + bkeys (~Thunderbi@66.110.201.50) | 15:38 | |
| - bkeys (QUIT: Client Quit) (~Thunderbi@66.110.201.50) | 15:41 | |
| + bkeys (~Thunderbi@66.110.201.50) | 15:41 | |
| - bkeys (QUIT: Client Quit) (~Thunderbi@66.110.201.50) | 15:43 | |
| + bkeys (~Thunderbi@66.110.201.50) | 15:43 | |
| - bkeys (QUIT: Read error: Connection reset by peer) (~Thunderbi@66.110.201.50) | 15:46 | |
| + bkeys1 (~Thunderbi@66.110.201.50) | 15:46 | |
| - bkeys1 (QUIT: Client Quit) (~Thunderbi@66.110.201.50) | 15:46 | |
| + bkeys (~Thunderbi@66.110.201.50) | 15:47 | |
| - bkeys (QUIT: Client Quit) (~Thunderbi@66.110.201.50) | 15:50 | |
| + bkeys1 (~Thunderbi@66.110.201.50) | 15:51 | |
| - chomwitt (QUIT: Ping timeout: 265 seconds) (~chomwitt@2a02:85f:9a00:8300:1ac0:4dff:fedb:a3f1) | 15:51 | |
| - bkeys1 (QUIT: Client Quit) (~Thunderbi@66.110.201.50) | 15:52 | |
| + bkeys (~Thunderbi@66.110.201.50) | 15:52 | |
| - mark_ (QUIT: Ping timeout: 265 seconds) (~mjw@gnu.wildebeest.org) | 15:54 | |
| - bkeys (QUIT: Client Quit) (~Thunderbi@66.110.201.50) | 15:55 | |
| + bkeys (~Thunderbi@66.110.201.50) | 15:55 | |
| - bkeys (QUIT: Client Quit) (~Thunderbi@66.110.201.50) | 15:58 | |
| + bkeys (~Thunderbi@66.110.201.50) | 15:58 | |
| - bkeys (QUIT: Client Quit) (~Thunderbi@66.110.201.50) | 16:01 | |
| + bkeys1 (~Thunderbi@66.110.201.50) | 16:02 | |
| * bkeys1 -> bkeys | 16:04 | |
| minute | josch: where does reform-desktop-full and reform-desktop-minimal come from? i'm trying to find the origin of those packages | 16:06 |
| - bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@66.110.201.50) | 16:08 | |
| minute | josch: where is the source repo for http://deb.debian.org/debian/pool/main/r/reform-tools/reform-tools_1.71-2.debian.tar.xz ? | 16:12 |
| + bkeys (~Thunderbi@66.110.201.50) | 16:13 | |
| minute | josch: i want to remove wayfire and reform-firedecor from the deps and add gnome, and have been searching for a long time through repos and scripts now and i just don't understand where the debian control files are for reform-tools | 16:14 |
| minute | ahh it's here https://salsa.debian.org/reform-team/reform-tools | 16:16 |
| + mark_ (~mjw@gnu.wildebeest.org) | 16:16 | |
| minute | josch: wouldn't it make sense to "upstream" these back into the original reform-tools repo as well? | 16:17 |
| - bkeys (QUIT: Read error: Connection reset by peer) (~Thunderbi@66.110.201.50) | 16:29 | |
| + bkeys1 (~Thunderbi@66.110.201.50) | 16:29 | |
| * bkeys1 -> bkeys | 16:31 | |
| - bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@66.110.201.50) | 16:37 | |
| + bkeys (~Thunderbi@66.110.201.50) | 16:37 | |
| - bkeys (QUIT: Client Quit) (~Thunderbi@66.110.201.50) | 16:40 | |
| + bkeys (~Thunderbi@66.110.201.50) | 16:40 | |
| zeha | reading | 16:46 |
| zeha | minute: josch: the fwupd update mechanism should also be safer as (iirc) it asks the sysctl to reboot, which means it could finish whatever tasks it was doing | 16:47 |
| zeha | not saying it does that today | 16:48 |
| zeha | hmm, PreDepends linux-base (>= 4.11+reform20250520) | 16:49 |
| zeha | minute: josch: from which pipeline should i get that .deb? | 16:49 |
| minute | zeha: hehe, i had the same question, and it's in the same pipeline | 16:51 |
| minute | zeha: https://source.mnt.re/reform/reform-debian-packages/-/jobs/10184/artifacts/browse/repo/pool/main/l/linux-base/ | 16:51 |
| zeha | seems like an older linux-base from 20250520 is in the apt sources | 16:56 |
| zeha | initial reboot looks bad | 16:56 |
| zeha | so, initial reboot: display stays off | 16:57 |
| zeha | after power cycle it works | 16:57 |
| zeha | i changed nothing else, and my console was still correct and sway is also correct | 16:57 |
| + mmmm (~mmmm@195.12.232.251) | 16:59 | |
| zeha | ugh | 16:59 |
| zeha | after dpms off/on the panel flickers | 16:59 |
| * mmmm -> josch-pocket | 16:59 | |
| minute | zeha: ah yeah, i had this effect with the previous sleep fix | 17:03 |
| minute | zeha: well... what kind of flicker are you getting? like, oscillating in grey color, or a temporary glitch? | 17:03 |
| minute | zeha: anyway, sounds like i didn't improve anything on that front, so will just take those sleep changes out and only keep orientation and merge it | 17:04 |
| zeha | it feels like all colors are oscillating | 17:07 |
| minute | zeha: ok | 17:07 |
| minute | zeha: hmm it is strange, in this v2 i basically just removed mipi_dsi_dcs_enter_sleep_mode() for the disp v1 case... | 17:08 |
| - bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@66.110.201.50) | 17:08 | |
| zeha | and on the wakeup from dpms off there is some ghosting from the previous image | 17:08 |
| + bkeys (~Thunderbi@66.110.201.50) | 17:08 | |
| minute | zeha: yeah, that i've seen too | 17:08 |
| minute | zeha: after reboot it still stays off yes? | 17:08 |
| zeha | so | 17:09 |
| zeha | when the old kernel shuts down, the display starts fading out, retaining the old contents for a while | 17:09 |
| minute | yep ok | 17:09 |
| zeha | after the new kernel boots i have a working display | 17:09 |
| minute | ah! | 17:10 |
| minute | i think the fade out is from my removal of mipi_dsi_dcs_set_display_off(jdi->dsi); for disp v1, which i now realize | 17:10 |
| zeha | i rebooted/powercycled a bit, and the flickering is now always there in sway | 17:12 |
| zeha | http://per.namespace.at/~ch/IMG_6176.MOV | 17:13 |
| zeha | dunno if my internet is broken now or hetzner | 17:13 |
| minute | zeha: ok, the flickering comes from not turning the display off properly but then stopping the video stream | 17:14 |
| minute | i removed the sleep, but i _also_ removed the set_display_off which was probably wrong, so let me fix that | 17:15 |
| minute | aw, i guess the image got too big with gnome https://source.mnt.re/reform/reform-system-image/-/jobs/10197 | 17:21 |
| - bkeys (QUIT: Read error: Connection reset by peer) (~Thunderbi@66.110.201.50) | 17:21 | |
| + bkeys1 (~Thunderbi@66.110.201.50) | 17:21 | |
| josch-pocket | minute: when i built gnome images i bumped the variable at the top to 10G | 17:21 |
| minute | josch-pocket: ok thx, let me try ^^ | 17:22 |
| josch-pocket | minute: ROOTSIZE | 17:22 |
| minute | ah yes | 17:23 |
| josch-pocket | wow browsing gitlab with rk3588 is actually okay | 17:23 |
| minute | josch-pocket: yeah, on firefox it is still a bit slow but with webkit/chromium based browsers it's fast | 17:23 |
| * bkeys1 -> bkeys | 17:24 | |
| minute | josch-pocket: did you see my messages above about reform-desktop-full and reform-desktop-minimal? | 17:25 |
| - bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@66.110.201.50) | 17:26 | |
| + bkeys1 (~Thunderbi@66.110.201.50) | 17:27 | |
| - josch-pocket (QUIT: Ping timeout: 272 seconds) (~mmmm@195.12.232.251) | 17:27 | |
| * bkeys1 -> bkeys | 17:29 | |
| + josch-pocket (~josch-poc@195.12.232.247) | 17:35 | |
| - wickedshell (QUIT: Ping timeout: 245 seconds) (~wickedshe@c-174-56-48-20.hsd1.nm.comcast.net) | 17:35 | |
| josch-pocket | minute: i'm on my way to a camping trip so i only read your messages from earlier via the logs on mntre.com now sorry | 17:38 |
| josch-pocket | minute: do you actually need to remove wayfire and reform-firedecor from the deps before you can proceed | 17:39 |
| josch-pocket | ? | 17:39 |
| josch-pocket | sorry typing is still not easy | 17:39 |
| minute | josch-pocket: no, i did a workaround | 17:39 |
| josch-pocket | i can send you a patch later today but i do not have my ssh key with me to push things myself im afraiy | 17:40 |
| minute | josch-pocket: i did it like this now https://source.mnt.re/reform/reform-system-image/-/commit/1bf54df850b528128ca3d3a9eab70cdd6c1c892d | 17:40 |
| josch-pocket | that's good! | 17:41 |
| minute | josch-pocket: it's not so much about patching etc but more about the fact that the reform-desktop-... dependencies are not in MNT repos, which is very confusing when trying to understand reform-system-images | 17:41 |
| minute | reform-system-image | 17:41 |
| - bkeys (QUIT: Read error: Connection reset by peer) (~Thunderbi@66.110.201.50) | 17:41 | |
| + bkeys (~Thunderbi@66.110.201.50) | 17:42 | |
| josch-pocket | they... should be | 17:42 |
| josch-pocket | one sec train | 17:42 |
| minute | josch-pocket: no rush, just something i'd like to understand or clear up later ^^ | 17:42 |
| - bkeys (QUIT: Client Quit) (~Thunderbi@66.110.201.50) | 17:46 | |
| + bkeys1 (~Thunderbi@66.110.201.50) | 17:47 | |
| + josch-pocket_ (~josch-poc@ip-185-104-138-79.ptr.icomera.net) | 17:47 | |
| * bkeys1 -> bkeys | 17:49 | |
| - josch-pocket (QUIT: Ping timeout: 260 seconds) (~josch-poc@195.12.232.247) | 17:50 | |
| josch-pocket_ | the summary is that reform-tools is now a distro agnostic package without debian integration. to hot patch things you either do what you did or ask me or edit this patch | 17:50 |
| * josch-pocket_ -> josch-pocket | 17:50 | |
| josch-pocket | https://source.mnt.re/reform/reform-debian-packages/-/blob/main/patches/reform-tools?ref_type=heads#L166 | 17:50 |
| - bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@66.110.201.50) | 17:52 | |
| + bkeys (~Thunderbi@66.110.201.50) | 17:52 | |
| josch-pocket | also reform-desktop packages are in the mnt repo https://mntre.com/reform-debian-repo/pool/main/r/reform-tools/ | 17:52 |
| minute | josch-pocket: what i meant is that if i "rg wayfire" in reform-tools or reform-system-image i don't get any results, so i don't know why it is installed, where it comes from | 17:53 |
| minute | josch-pocket: because reform-tools on source.mnt.re does not have the "debian" folder that you have in your salsa git version | 17:54 |
| - bkeys (QUIT: Client Quit) (~Thunderbi@66.110.201.50) | 17:54 | |
| + bkeys (~Thunderbi@66.110.201.50) | 17:54 | |
| minute | josch-pocket: this feels odd because the information that is needed in terms of what are the ingredients of a reform-system-image are at the moment not to be found in the mnt repos | 17:55 |
| minute | josch-pocket: s/mnt repos/mnt git repos/ | 17:55 |
| - bkeys (QUIT: Client Quit) (~Thunderbi@66.110.201.50) | 17:57 | |
| josch-pocket | if you rather work with a debian directory of reform-tools than the patch in reform-debian-packages then i can make it so that it works like that again no problem | 17:57 |
| + bkeys (~Thunderbi@66.110.201.50) | 17:57 | |
| josch-pocket | it would be the same information just expressed differently | 17:58 |
| minute | josch-pocket: the list in that patch looks quite different to what i mean | 17:58 |
| minute | josch-pocket: i mean this https://salsa.debian.org/reform-team/reform-tools/-/blob/master/debian/control?ref_type=heads | 17:58 |
| josch-pocket | the patch is against that file | 17:59 |
| minute | josch-pocket: ah ok. that is a lot of indirection to wrap my head around, sorry ^^ | 17:59 |
| josch-pocket | if you rather have the full file i can implement that too :) | 17:59 |
| josch-pocket | i agree | 18:00 |
| josch-pocket | my idea was to have motivation to keep the diff small | 18:00 |
| minute | josch-pocket: that would be great, because i only found the debian control files by googling, afaik this dependency is not mentioned in the scripts or docs | 18:00 |
| josch-pocket | upstream everything to debian etc | 18:00 |
| josch-pocket | minute: also, i can give you git push access on salsa | 18:00 |
| minute | josch-pocket: yeah, but the memory in my head is too small to keep track of what lives in external repos and where to find them | 18:00 |
| josch-pocket | right makes sense | 18:01 |
| minute | like, if i have to work on this again in a few months maybe i won't remember | 18:01 |
| - bkeys (QUIT: Ping timeout: 252 seconds) (~Thunderbi@66.110.201.50) | 18:02 | |
| josch-pocket | understood. i'll turn the patch into full files you can edit in reform-debian-packages, okay? | 18:02 |
| minute | josch-pocket: a list in the readme about the git repos that end up being the main ingredients for the system image would also help | 18:02 |
| minute | josch-pocket: thank you! | 18:02 |
| josch-pocket | that is what you need? | 18:02 |
| josch-pocket | just making sure | 18:02 |
| + bkeys (~Thunderbi@66.110.201.50) | 18:03 | |
| minute | josch-pocket: like, the debian control files would be great to have in the repo (i don't think other distros will be sad about them), and if you have time, an "ingredient list" in the README would be great at some point. just so that people like me can reconstruct a mental image of what the system image is made from (reform-debian-packages, reform-tools, reform-desktop-full, ...) | 18:04 |
| minute | reform-setup-wizard | 18:04 |
| josch-pocket | thank you i copy pasted that to my todo list :) | 18:05 |
| minute | josch-pocket: awesome! | 18:05 |
| minute | josch-pocket: on the bright side, we can get rid of maintaining reform-firedecor | 18:05 |
| minute | uff, 500 Internal Server Error when uploading the system images... ;__; | 18:07 |
| - bkeys (QUIT: Client Quit) (~Thunderbi@66.110.201.50) | 18:07 | |
| minute | > /dev/sdb 295G 278G 2.7G 100% /mnt/gitlab-artifacts | 18:07 |
| minute | ah. | 18:07 |
| + bkeys (~Thunderbi@66.110.201.50) | 18:07 | |
| josch-pocket | minute: thanks to you as well for your quick help this morning because i just had to show my bahncard and without a working pocket i would have had no pdf to show just now XD | 18:09 |
| minute | josch-pocket: omg | 18:09 |
| josch-pocket | conductor was just here scanning the pocket display :) | 18:10 |
| minute | nice :DD | 18:11 |
| minute | hmm, 75 GiB of artifacts from 2025-05-20 | 18:11 |
| - bkeys (QUIT: Ping timeout: 245 seconds) (~Thunderbi@66.110.201.50) | 18:12 | |
| minute | 5x 15.1GB | 18:12 |
| minute | ah, it's a bunch of system image builds | 18:13 |
| josch-pocket | i tried limiting my system image building to reform-system-any but that messed up the official sysimage link yesterday :( | 18:13 |
| - aperezdc (QUIT: Remote host closed the connection) (~aperezdc@46.23.89.43) | 18:20 | |
| josch-pocket | zeha: wow my rk3588 cannot handle your video | 18:21 |
| + aperezdc (~aperezdc@46.23.89.43) | 18:21 | |
| josch-pocket | utilizes all cores with mpv and drops most frames | 18:21 |
| minute | gitlab artifact upload is really slow somehow | 18:29 |
| minute | nload says around 100mbit/s incoming on source.mnt.re | 18:30 |
| minute | not great for many gigabytes | 18:30 |
| minute | i did an iperf test between the build server and the gitlab server, it's almost 5 gbit/s | 18:36 |
| minute | so... wtf | 18:36 |
| * mjw -> Guest2919 | 18:36 | |
| * mark_ -> mjw | 18:36 | |
| minute | maybe slow compression | 18:36 |
| minute | gah > ERROR: Uploading artifacts as "archive" to coordinator... 413 Request Entity Too Large | 18:37 |
| minute | i guess i should rebase on the any image | 18:37 |
| - josch-pocket (QUIT: Ping timeout: 244 seconds) (~josch-poc@ip-185-104-138-79.ptr.icomera.net) | 18:37 | |
| minute | i also wonder what's the size of gnome vs gnome-core deps | 18:46 |
| minute | ah, now with fastzip featureflag the upload of only the "any" image took < 2 seconds (??) | 18:54 |
| minute | hmm it's still "only" 1.86GB | 18:55 |
| minute | flashing to microsd for pocket... lets see if we'll get gnome out of the box ;0 | 18:56 |
| + josch-pocket (~josch-poc@185.109.152.61) | 18:59 | |
| josch-pocket | it should work. at least it worked last we | 19:00 |
| josch-pocket | week | 19:00 |
| josch-pocket | i onl | 19:00 |
| josch-pocket | i only added gnome and disabled greetd via systemd | 19:00 |
| minute | i bumped the runner concurrency from 1 to 4 on the arm builder :3 | 19:02 |
| minute | it was idling a lot when a job is just installing deb packages... | 19:02 |
| josch-pocket | that should help a lot, given how much we split up the jobs | 19:04 |
| + bkeys (~Thunderbi@173.186.16.211) | 19:04 | |
| minute | i wonder if this is a good idea or not :0 https://github.com/ilikenwf/apt-fast | 19:05 |
| minute | josch-pocket: also, yay @ reform-system-image-any... maybe a good transition would be to only build platform specific images for some platforms where we know that they don't have usable uboot on emmc (you probably already had this in mind?) | 19:06 |
| - josch-pocket (QUIT: Ping timeout: 276 seconds) (~josch-poc@185.109.152.61) | 19:09 | |
| + josch-pocket (~josch-poc@195.12.232.246) | 19:09 | |
| josch-pocket | minute: you modified emulate.sh and removed the graphics -- why? | 19:13 |
| josch-pocket | -device virtio-gpu-pci is needed for automated testing via vnc | 19:17 |
| minute | josch-pocket: but there is -device virtio-gpu-pci,edid=on,xres=1024,yres=768 ? | 19:26 |
| minute | josch-pocket: why you need 2x virtio-gpu-pci? | 19:27 |
| josch-pocket | huh err... | 19:27 |
| josch-pocket | okay seems to be an oversight from me thank you for the cleanup | 19:27 |
| minute | josch-pocket: np! that was actually recently found by plomlompom | 19:32 |
| minute | we were trying to get a qemu server going with vnc | 19:32 |
| minute | except for a silly mistake i made the setup wizard with gnome worked well in my first system image test | 19:33 |
| minute | gnome has weird icons somehow though, let's see | 19:34 |
| minute | the next showstopper is disabling all kinds of suspend | 19:35 |
| + josch-pocket_ (~josch-poc@80.239.189.35) | 19:35 | |
| minute | lmao, after reboot, gdm doesn't come up, instead, normal login | 19:35 |
| josch-pocket_ | because we have that running on debian ci systems to test reform-setup-wizard | 19:35 |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:50) | 19:36 | |
| - josch-pocket (QUIT: Ping timeout: 252 seconds) (~josch-poc@195.12.232.246) | 19:36 | |
| minute | ah, `systemctl enable gdm` says The unit files have no installation config... | 19:37 |
| + josch-pocket (~josch-poc@80.239.189.33) | 19:38 | |
| - josch-pocket_ (QUIT: Ping timeout: 245 seconds) (~josch-poc@80.239.189.35) | 19:40 | |
| minute | gah, greetd is running, that's why | 19:42 |
| - josch-pocket (QUIT: Quit: josch-pocket) (~josch-poc@80.239.189.33) | 19:53 | |
| - bkeys (QUIT: Ping timeout: 268 seconds) (~Thunderbi@173.186.16.211) | 20:09 | |
| + bkeys (~Thunderbi@66.110.201.50) | 20:19 | |
| - bkeys (QUIT: Client Quit) (~Thunderbi@66.110.201.50) | 20:20 | |
| + bkeys (~Thunderbi@66.110.201.50) | 20:20 | |
| minute | ACTION is in the gnome customization rabbithole (especially keybindings) | 20:24 |
| + chomwitt (~chomwitt@2a02:85f:9a00:8300:1ac0:4dff:fedb:a3f1) | 20:24 | |
| grimmware | I am still very much loving Niri for the pocket | 20:24 |
| minute | waiting for niri to arrive in debian :D | 20:25 |
| grimmware | do you know if it's in the works? | 20:26 |
| minute | no idea | 20:26 |
| + bkeys1 (~Thunderbi@66.110.201.50) | 20:27 | |
| grimmware | doesn't look like it | 20:28 |
| - bkeys (QUIT: Ping timeout: 276 seconds) (~Thunderbi@66.110.201.50) | 20:28 | |
| grimmware | looks like it would need someone to step up and volunteer and I don't think I hate myself enough to do that without some knowledge that it's actually going to help people I care about heh | 20:30 |
| - bkeys1 (QUIT: Ping timeout: 248 seconds) (~Thunderbi@66.110.201.50) | 20:31 | |
| + bkeys (~Thunderbi@66.110.201.50) | 20:33 | |
| - bkeys (QUIT: Read error: Connection reset by peer) (~Thunderbi@66.110.201.50) | 20:35 | |
| + bkeys (~Thunderbi@66.110.201.50) | 20:38 | |
| - bkeys (QUIT: Read error: Connection reset by peer) (~Thunderbi@66.110.201.50) | 20:41 | |
| + bkeys1 (~Thunderbi@66.110.201.50) | 20:41 | |
| - bkeys1 (QUIT: Ping timeout: 244 seconds) (~Thunderbi@66.110.201.50) | 20:45 | |
| + bkeys (~Thunderbi@38-146-94-247.echocast.zone) | 20:55 | |
| - bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@38-146-94-247.echocast.zone) | 21:05 | |
| + bkeys (~Thunderbi@38-146-94-247.echocast.zone) | 21:05 | |
| sad_plan | minute: im just curious really. is there a specific reason for going with ARM arch to begin with? i know opensource hardware is.. scarse, but I wouldve thought that it its even more so with ARM. is it because of the modularity perhaps? | 21:08 |
| minute | sad_plan: arm socs are simpler than amd/intel chips, generally less power hungry (fanless), no NDA for reference schematics, parts can be bought from multiple sources, multiple diverse vendors of the chips, ... | 21:11 |
| - bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@38-146-94-247.echocast.zone) | 21:12 | |
| minute | sad_plan: intel/amd soms are generally rare or expensive | 21:12 |
| + bkeys (~Thunderbi@38-146-94-247.echocast.zone) | 21:12 | |
| sad_plan | I see. that all makes sense. thanks for clearifying C: | 21:13 |
| minute | sad_plan: also consider the situation was even more like this when we started, in around 2017 | 21:13 |
| minute | sad_plan: now the x86 world also has more power efficient chips, but the other issues still exist | 21:13 |
| minute | sad_plan: i was also very interested in risc-v, but competitive chips didn't materialize | 21:14 |
| minute | sad_plan: and surprisingly arm got very competitive in terms of performance, more so than i imagined, even apple migrated to arm on their laptops+desktops | 21:14 |
| minute | sad_plan: so in general i feel like i made a good choice | 21:15 |
| sad_plan | I see. risc-v is still rather early. even *performant* risc-v chips are still terribly slow. atleast last time I checked. it would be interesting to see what becomes of it though, when developement gets further along | 21:15 |
| sad_plan | I agree. I was supprised when apple made their choise to go with ARM, but they did so with a win imo | 21:15 |
| minute | sad_plan: parts of the reverse engineering / open source driver development scene around arm socs are also quite fun | 21:16 |
| minute | so, we're learning a lot about linux and chip internals all the time, and the project is also about learning | 21:16 |
| minute | on intel we probably wouldn't have to touch the kernel that much and also wouldn't learn so much about it | 21:16 |
| + bkeys1 (~Thunderbi@38-146-94-247.echocast.zone) | 21:17 | |
| sad_plan | thats just the iceing on the cake if its fun too, and not some terribly tedious task | 21:17 |
| sad_plan | learning is great. I love learning | 21:17 |
| minute | yeah it can be tough at times, but most of the time we found solutions for hairy problems | 21:18 |
| - bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@38-146-94-247.echocast.zone) | 21:18 | |
| * bkeys1 -> bkeys | 21:18 | |
| minute | i think the only big remaining issue is proper suspend/hibernate support | 21:18 |
| + bkeys1 (~Thunderbi@66.110.201.50) | 21:19 | |
| sad_plan | yeah, ive seen that being mentioned a few times. from what ive heard, this has been somewhat of an issue with linux in general, atleast in the past. maybe its better now than what it was when I started using linux ~6-7 years ago | 21:20 |
| - bkeys (QUIT: Ping timeout: 252 seconds) (~Thunderbi@38-146-94-247.echocast.zone) | 21:22 | |
| * bkeys1 -> bkeys | 21:22 | |
| - bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@66.110.201.50) | 21:24 | |
| + bkeys1 (~Thunderbi@66.110.201.50) | 21:24 | |
| - bkeys1 (QUIT: Client Quit) (~Thunderbi@66.110.201.50) | 21:26 | |
| + bkeys (~Thunderbi@66.110.201.50) | 21:27 | |
| + josch-pocket (~josch-poc@195.12.232.246) | 21:41 | |
| - bkeys (QUIT: Read error: Connection reset by peer) (~Thunderbi@66.110.201.50) | 21:43 | |
| + bkeys1 (~Thunderbi@66.110.201.50) | 21:43 | |
| * bkeys1 -> bkeys | 21:45 | |
| minute | current annoyance with fresh debian/gnome system https://gitlab.gnome.org/GNOME/gnome-software/-/issues/2815 | 21:48 |
| + bkeys1 (~Thunderbi@66.110.201.50) | 21:49 | |
| - bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@66.110.201.50) | 21:49 | |
| * bkeys1 -> bkeys | 21:49 | |
| minute | gnome-terminal is quite sluggish on gnome with default settings on rk3588 | 21:50 |
| sad_plan | I think we've come a long way, if thats the most annoying thing with gnome/debian currently :p | 21:50 |
| sad_plan | and how do you fix the slugginess? | 21:51 |
| minute | i guess we switch back to "foot" for now, until gtk4 gnome-terminal is released. or perhaps we're on the wrong renderer by default | 21:52 |
| + bkeys1 (~Thunderbi@66.110.201.50) | 21:54 | |
| sad_plan | I see. foot should be relativly fast. atleast thats what they claim. never really used it much myself. I mostly just use st. although ive been using konsole on my desktop, because I havent bothered to get everything working under my own setup yet | 21:54 |
| - bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@66.110.201.50) | 21:55 | |
| * bkeys1 -> bkeys | 21:55 | |
| vagrantc | minute: i've had no issues with sakura on rk3588 ... but ask 9 people a good terminal emulator and get 12 different answers | 22:00 |
| minute | vagrantc: ah, let me try | 22:00 |
| sad_plan | ^ foss in a nutshell | 22:00 |
| - bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@66.110.201.50) | 22:00 | |
| + bkeys1 (~Thunderbi@66.110.201.50) | 22:00 | |
| vagrantc | though i also mostly have been using console only lately, only occasionally starting up sway :) | 22:02 |
| * bkeys1 -> bkeys | 22:03 | |
| sigrid | I use https://tomscii.sig7.se/zutty/ on my system(s) in general, though it's not wayland-specific | 22:04 |
| minute | vagrantc: hmm sakura is also quite sluggish compared to foot, maybe it's pocket reform specific (rotated hidpi display) | 22:06 |
| minute | (or gnome/mutter specific) | 22:06 |
| vagrantc | curious | 22:07 |
| - bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@66.110.201.50) | 22:13 | |
| + bkeys1 (~Thunderbi@66.110.201.50) | 22:13 | |
| - gustav28 (QUIT: Quit: Quit) (~gustav@c-78-82-53-72.bbcust.telenor.se) | 22:15 | |
| * bkeys1 -> bkeys | 22:16 | |
| - bkeys (QUIT: Ping timeout: 252 seconds) (~Thunderbi@66.110.201.50) | 22:26 | |
| + bkeys (~Thunderbi@173.186.16.211) | 22:43 | |
| - josch-pocket (QUIT: Quit: josch-pocket) (~josch-poc@195.12.232.246) | 23:00 | |
| minute | bizarre, i was like, something is odd... so gnome picks "225%" as default scale on pocket reform | 23:10 |
| - chomwitt (QUIT: Ping timeout: 245 seconds) (~chomwitt@2a02:85f:9a00:8300:1ac0:4dff:fedb:a3f1) | 23:13 | |
| minute | setting to 200% everything is much faster m))) | 23:15 |
| vagrantc | yay! | 23:18 |
| minute | https://source.mnt.re/reform/reform-system-image/-/merge_requests/129 | 23:21 |
| minute | ooof a bit frustrating https://source.mnt.re/reform/reform-system-image/-/jobs/10226#L12183 | 23:50 |
| minute | ok, understood | 23:53 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
