| + mrdaught (~mrdaught@172.58.111.135) | 00:05 | |
| - mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@172.58.111.135) | 00:07 | |
| josch | okay, now building src:linux twice, each with -j2 -- should be done in the morning | 00:13 |
|---|---|---|
| josch | one build is 6.8-rc1 with the cpufreq patch stack | 00:14 |
| josch | one build is 6.8-rc7 with the normal patch stack | 00:14 |
| josch | currently broken is 6.8-rc7 with the cpufreq patch stack | 00:14 |
| josch | by trying out these two kernel, we can find out whether the problem is in the collabora patch stacks (normal vs cpufreq) or in what happened between linux 6.8-rc1 and 6.8-rc7 | 00:15 |
| + jacobk (~quassel@utdpat241106.utdallas.edu) | 00:17 | |
| + mrdaught (~mrdaught@172.58.111.135) | 00:21 | |
| - mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@172.58.111.135) | 00:21 | |
| minute | josch: very nice | 00:27 |
| ehenter | Hmm, I just tried chromium for fun and it really seems to be a bit of faster compared to firefox. I'm still unsure if I want to switch because of google. Debian apt seems to handle upgrades and new installations well enough so stuff do not break when using it from already usable system | 00:44 |
| + mrdaught (~mrdaught@bras-base-toroon01zb3-grc-78-184-147-249-228.dsl.bell.ca) | 00:52 | |
| ehenter | Now tried Covidence and other academic PDF sources, and perf increase is very noticeable at least for me... | 00:54 |
| - mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@bras-base-toroon01zb3-grc-78-184-147-249-228.dsl.bell.ca) | 00:54 | |
| ehenter | My question now is: is there any chromium fork available in debian repo which is ungoogled? | 00:58 |
| ehenter | Or if anyaone knows if even optimized firefox build helps with real world performance? | 01:00 |
| minute | firefox only performs well on desktop gpus i think | 01:00 |
| ehenter | Oh dang then. Is anyone tried build with pgo and other possible optimizations? I remember that in gentoo those helped on low-end x86 systems | 01:02 |
| minute | i haven't, personally | 01:08 |
| + mrdaught (~mrdaught@bras-base-toroon01zb3-grc-78-184-147-249-228.dsl.bell.ca) | 01:14 | |
| - mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@bras-base-toroon01zb3-grc-78-184-147-249-228.dsl.bell.ca) | 01:14 | |
| ehenter | Okay, hopefully someone else has been and reports back. But do know, minute, that I'm actually happy with current state anyway. I just look for extra perf if someone has some tweaks in their knowlege :) | 01:14 |
| minute | ehenter: cool! | 01:26 |
| ehenter | You know what's cooler? This is my first system in 5 years that do not crash every day and that I can rely on... | 01:32 |
| ehenter | This is a bit of rambling but I do have 1st gen AMD Ryzen together with AMD R9 Nano on Asrock board that likes to crash every time that I'd like to do something with it | 01:33 |
| ehenter | And I'm impressed that small studio in Europe (without forgetting RBZ) can make actually functioning computer which I can use for real work | 01:35 |
| - jacobk (QUIT: Ping timeout: 260 seconds) (~quassel@utdpat241106.utdallas.edu) | 01:47 | |
| jn | i think one of the fundamental things mnt got right is that the Reform itself is kind of "small" too (conceptually, not physically). i've seen certain other projects in the open source hardware field that have incredible feature creep and focus drift, and consequently don't achieve their goals | 01:49 |
| ehenter | Got to agree. They seem to deliver systems that people can use and that aligns with their specified goals. That's one reason why it feels so bad to complain about some missing "features" when all of the specified ones are fulfilled :D | 01:58 |
| - Gooberpatrol66 (QUIT: Ping timeout: 260 seconds) (~Gooberpat@user/gooberpatrol66) | 02:10 | |
| + mrdaught (~mrdaught@184.147.249.228) | 02:19 | |
| - mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@184.147.249.228) | 02:20 | |
| + jacobk (~quassel@utdpat242024.utdallas.edu) | 02:28 | |
| - mjw (QUIT: Ping timeout: 246 seconds) (~mjw@gnu.wildebeest.org) | 02:33 | |
| - jacobk (QUIT: Ping timeout: 268 seconds) (~quassel@utdpat242024.utdallas.edu) | 02:40 | |
| + Gooberpatrol66 (~Gooberpat@user/gooberpatrol66) | 03:12 | |
| - digitalrayne (QUIT: Quit: じゃまたね) (~digitalra@vps-446f4f39.vps.ovh.ca) | 03:53 | |
| + digitalrayne (~digitalra@vps-446f4f39.vps.ovh.ca) | 03:53 | |
| + jacobk (~quassel@utdpat242024.utdallas.edu) | 04:01 | |
| erle | http://news.dieweltistgarnichtso.net/notes/lego-mnt-reform-keyboard-holder-2.html | 04:25 |
| + mrdaught (~mrdaught@184.147.249.228) | 04:33 | |
| - mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@184.147.249.228) | 04:34 | |
| - Ar|stote|is (QUIT: Quit: https://quassel-irc.org - Chat comfortably. Anywhere.) (~linx@149.210.12.179) | 05:34 | |
| + Ar|stote|is (~linx@149.210.12.179) | 05:34 | |
| josch | lesson learned: with -j2, it takes longer than six hours for the kernel to build :D | 06:50 |
| + chomwitt (~chomwitt@2a02:587:7a18:fc00:1ac0:4dff:fedb:a3f1) | 07:02 | |
| - chomwitt (QUIT: Ping timeout: 260 seconds) (~chomwitt@2a02:587:7a18:fc00:1ac0:4dff:fedb:a3f1) | 07:09 | |
| - jacobk (QUIT: Ping timeout: 268 seconds) (~quassel@utdpat242024.utdallas.edu) | 08:10 | |
| josch | Build needed 07:56:22 uff... | 08:18 |
| + mrdaught (~mrdaught@bras-base-toroon01zb3-grc-78-184-147-249-228.dsl.bell.ca) | 08:20 | |
| - mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@bras-base-toroon01zb3-grc-78-184-147-249-228.dsl.bell.ca) | 08:24 | |
| josch | 6.8-rc1 with the experimental collabora cpufreq patch stack: https://mister-muffin.de/reform/linux-6.8-rc1-rk3588-test-cpufreq/ | 08:29 |
| josch | 6.8-rc7 with the collabora patch stack for 6.8-rc1: https://mister-muffin.de/reform/linux-6.8-rc7-rk3588 | 08:29 |
| + jacobk (~quassel@64.189.201.150) | 08:50 | |
| - klardotsh (QUIT: Ping timeout: 246 seconds) (~klardotsh@c-67-170-115-80.hsd1.wa.comcast.net) | 10:36 | |
| - murdock (QUIT: Ping timeout: 252 seconds) (~murdock@dhcp-209-99-207-141.greenmountainaccess.net) | 11:35 | |
| + murdock (~murdock@dhcp-209-99-207-141.greenmountainaccess.net) | 11:35 | |
| + S0rin (~S0rin@user/s0rin) | 11:50 | |
| + pandora (uid585533@id-585533.ilkley.irccloud.com) | 11:55 | |
| - buckket (QUIT: Quit: buckket) (~buckket@vps.buckket.org) | 11:59 | |
| + buckket (~buckket@vps.buckket.org) | 12:00 | |
| + mjw (~mjw@gnu.wildebeest.org) | 12:02 | |
| grimmware | minute: so here’s a totally unsolicited opinion - when the rack reform starts to be more of a thing, would you consider offering early units *without* a SoM? The reason I ask is because I think you’ll find that a bunch of people would order one to put an older module to good use whereas they might not if it just means they’re going to end up with another module. | 12:19 |
| grimmware | I’m already agonizing over the RCORE on the basis that I don’t even have my Pocket yet and I’m already considering preordering a replacement SoM | 12:20 |
| minute | grimmware: yeah that was one of the plans. i'm still not sure about the rack for that though because it feels overkill in size and material amount. a smaller solution might be more adequate | 12:29 |
| grimmware | ++ even better IMO | 12:30 |
| grimmware | I can see how the pocket parts for the lid half without the screen and a cheaper chassis would make for good reuse of the existing design. | 12:33 |
| jn | a 3U eurorack cluster of reform carriers would be fun, but also very niche | 12:36 |
| - mjw (QUIT: Ping timeout: 260 seconds) (~mjw@gnu.wildebeest.org) | 12:39 | |
| minute | grimmware: yep, was thinking in pretty much that direction as well | 12:52 |
| josch | or the reform next parts -- unless of course the reform next itself re-uses the pocket parts... | 12:59 |
| josch | one idea for the reform next was to have the I/O pcbs replacable and that might be a super nice feature for a "desktop" box as well | 12:59 |
| * Guest114 -> mjw | 13:06 | |
| - jacobk (QUIT: Quit: No Ping reply in 210 seconds.) (~quassel@64.189.201.150) | 13:23 | |
| + jacobk (~quassel@64.189.201.150) | 13:24 | |
| minute | josch: yep, this requires a bunch of cables then, though | 13:31 |
| + mark_ (~mjw@gnu.wildebeest.org) | 13:39 | |
| + Nixkernal (~quassel@2a02:1210:1613:e600:517b:f380:8a4b:cb38) | 13:49 | |
| - mark_ (QUIT: Ping timeout: 240 seconds) (~mjw@gnu.wildebeest.org) | 14:08 | |
| + mrdaught (~mrdaught@bras-base-toroon01zb3-grc-78-184-147-249-228.dsl.bell.ca) | 14:22 | |
| minute | josch: sorry if you already answered this and i didn't see your reply, did you have some thoughts about packaging ly? or should i rather give agreetd another try? | 14:22 |
| - mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@bras-base-toroon01zb3-grc-78-184-147-249-228.dsl.bell.ca) | 14:24 | |
| josch | minute: i have only had a superficial look at ly and the README says that wayland support is in its early stages | 14:33 |
| josch | it seems to require a bunch of xorg stuff and the issue tracker lists things like always including startx in the session options | 14:33 |
| josch | did you try ly and does it work for you? | 14:34 |
| josch | about greetd: i have been using the agreety greeter for several months but it is very bare-bones and does not look pretty. But there are alternatives like https://git.sr.ht/~kennylevinsen/gtkgreet or https://git.sr.ht/~kennylevinsen/wlgreet which are more pretty. This one even has a very similar style as ly if that's what you were looking for: https://github.com/apognu/tuigreet | 14:39 |
| minute | josch: thanks! i remember there were some issues with greetd before, i will try greetd+tuigreet and the others today to see if we can use them | 14:46 |
| minute | funny, gtkgreet uses cage too | 14:47 |
| minute | (as does our setup wizard) | 14:47 |
| josch | minute: yes, there was this bug: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1025872 | 14:47 |
| josch | but that one got fixed last december | 14:47 |
| grimmware | I use ly with sway, mostly for the ascii fire animation | 14:51 |
| + mrdaught (~mrdaught@172.58.139.141) | 14:55 | |
| - jacobk (QUIT: Quit: No Ping reply in 180 seconds.) (~quassel@64.189.201.150) | 14:56 | |
| - mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@172.58.139.141) | 14:56 | |
| + jacobk (~quassel@64.189.201.150) | 14:57 | |
| grimmware | just had a look at the setup, afaicr it's giving options based on what's in /usr/share/wayland-sessions and it has separate wrapper scripts for loading up both x and wayland sessions which is incidentally a really nice place to put e.g. `export MOZ_ENABLE_WAYLAND=1 | 15:00 |
| + mrdaught (~mrdaught@172.58.139.141) | 15:01 | |
| grimmware | I set it up once and it's worked ever since but I've not changed much wrt desktop environment, I've been on sway for a while now | 15:01 |
| - mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@172.58.139.141) | 15:01 | |
| grimmware | so like, no strong feelings about it other than I like the ascii fire animation and the ascii fire animation makes my fans spin up if I don't log in fast enough | 15:02 |
| grimmware | oh, actually sway broke once and it made it easy to just launch a shell so that was nice | 15:02 |
| minute | heh, the wlgreet package ships a broken sway config | 15:05 |
| minute | "Unable to access background file /etc/greetd/background" | 15:05 |
| josch | minute: you found this bug: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1053450 | 15:10 |
| minute | ah :D | 15:10 |
| minute | ok, tuigreet doesn't build with debian's rustc | 15:10 |
| minute | > error[E0658]: use of unstable library feature 'build_hasher_simple_hash_one' | 15:11 |
| minute | also wlgreet is buggy, when switching away from its VT and then switching back, it can no longer receive input | 15:12 |
| josch | ups | 15:12 |
| minute | but instead there is a mouse cursor :D | 15:12 |
| minute | also i don't see a way to set the font and also it can't select the session, sigh | 15:13 |
| minute | trying cargo update on tuigreet... | 15:16 |
| minute | tehe, wlgreet includes the font in the binary https://git.sr.ht/~kennylevinsen/wlgreet/tree/master/item/src/draw.rs#L9 | 15:18 |
| minute | yeah, tuigreet is not compilable using debian's rust, giving up on that | 15:21 |
| minute | lets try gtkgreet | 15:21 |
| + mrdaught (~mrdaught@172.58.139.141) | 15:29 | |
| - mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@172.58.139.141) | 15:29 | |
| minute | now trying ReGreet | 15:33 |
| minute | ReGreet cleanly builds with debian rustc, so that's good. but now the sway started by greetd is broken | 15:52 |
| + mrdaught (~mrdaught@172.58.139.141) | 15:52 | |
| - cobra (QUIT: Ping timeout: 268 seconds) (~cobra@user/Cobra) | 15:54 | |
| - mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@172.58.139.141) | 15:55 | |
| + cobra (~cobra@user/Cobra) | 16:03 | |
| minute | sway started by greetd is just completely broken for me... it only shows a background and a cursor but can't show any windows, but there are no visible problems in sway log | 16:07 |
| grimmware | Display managers are such a shitshow since Wayland. | 16:13 |
| minute | yeah i give up on this, we just have to ship ly | 16:13 |
| grimmware | It’s certainly something that community members are able to iterate on later if there are rough edges | 16:15 |
| aperezdc | minute: I've been using greetd+tuigreeter with labwc (also wlroots-based) and works here, may I help somehow troubleshoot the issue? | 16:18 |
| aperezdc | ACTION used ly after it was first mentioned here, moved over to greetd after discovering it a couple of weeks later | 16:19 |
| aperezdc | but yeah, wayland and display managers is a bit hit and miss, I had some trouble with emptty and that's why I had been running labwc by hand after logging in using a regular console | 16:20 |
| aperezdc | but then ly and greetd had worked okay | 16:20 |
| sigrid | I don't even run any display managers. my linux laptop boots directly into sway after linux kernel password prompt to mount encrypted rootfs | 16:24 |
| grimmware | Oh I should do that. | 16:25 |
| grimmware | Momentum is a hell of a drug | 16:26 |
| grimmware | sigrid: did you end up getting any of the other modules in the end? | 16:27 |
| - murdock (QUIT: Ping timeout: 240 seconds) (~murdock@dhcp-209-99-207-141.greenmountainaccess.net) | 16:28 | |
| + murdock (~murdock@dhcp-209-99-207-141.greenmountainaccess.net) | 16:28 | |
| sigrid | nope. I am considering getting rk3588 | 16:30 |
| sigrid | which would involve porting 9front to it before it's usable in any way for me | 16:31 |
| sigrid | which is a bit weird because I'd love to use the reform to do it. chicken and egg situation | 16:31 |
| grimmware | heh, yeah I know that one | 16:34 |
| grimmware | not the porting so much as "I don't want to do X unless I can do it with Y" | 16:34 |
| minute | aperezdc: how did you build tuigreet? | 16:42 |
| minute | it doesn't build with debian rustc so probably i would need a toolchain with rustup, feels weird to do so much setup for a tui | 16:44 |
| grimmware | "just rewrite it in Rust" | 16:45 |
| minute | :S | 16:54 |
| + mrdaught (~mrdaught@172.58.139.141) | 16:54 | |
| minute | ok, got some lunch and feeling a bit better, now checking what else is available in terms of TUI display managers | 16:59 |
| aperezdc | minute: ah, Arch has it packaged | 17:00 |
| aperezdc | so I did't really build it myself | 17:01 |
| aperezdc | it seems that the rustc/cargo versions in Arch are recent enough | 17:01 |
| aperezdc | that'd be 1.76.0 | 17:03 |
| - mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@172.58.139.141) | 17:04 | |
| minute | now testing https://github.com/coastalwhite/lemurs | 17:04 |
| minute | aperezdc: aha, i see | 17:04 |
| minute | aperezdc: in debian i have rustc 1.70.0 | 17:04 |
| minute | sigrid: yay for 3588front | 17:05 |
| + mrdaught (~mrdaught@172.58.139.141) | 17:08 | |
| - mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@172.58.139.141) | 17:11 | |
| josch | minute: so apparently i have too much free time, so i tried building an initramfs without /usr/share/initramfs-tools/scripts/init-top/reform (which does the early module loading) and I *do* get the display up with a fresh system image on a311d | 17:15 |
| minute | josch: nice | 17:15 |
| minute | meh, 1. lemurs authentication doesn't work 2. journalctl scribbles over the UI lol | 17:16 |
| minute | i'm thinking to just not use any display manager and instead use the normal login mechanism and then, if we're on tty1, autostart either sway or wayfire, and if you need a shell you can switch to tty2 | 17:17 |
| minute | (and we explain how you change all that) | 17:17 |
| + mrdaught (~mrdaught@172.58.139.141) | 17:20 | |
| - mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@172.58.139.141) | 17:20 | |
| minute | reform-system-image already customizes /etc/skel/.profile so i'll do that there | 17:24 |
| minute | yeah, that works for now | 17:31 |
| josch | minute: a few lines above /etc/skel/.profile in mkimage.sh there is a comment for something that should be moved to the first-boot script | 17:31 |
| aperezdc | minute: seems good enough, and likely more reliable right now | 17:31 |
| minute | josch: i see, it's good to have locales already generated though i think | 17:32 |
| minute | josch: reform-setup-wizard lets one select the timezone | 17:32 |
| josch | cool! | 17:33 |
| + mark_ (~mjw@gnu.wildebeest.org) | 17:33 | |
| - Nixkernal (QUIT: Quit: https://quassel-irc.org - Chat comfortably. Anywhere.) (~quassel@2a02:1210:1613:e600:517b:f380:8a4b:cb38) | 17:34 | |
| minute | aha, sway launched in this way (as my cage replacement) after logging in as root works fine | 17:36 |
| minute | gtk4 apps don't wanna show up though... :D | 17:38 |
| minute | ah, just hung for a while hm | 17:39 |
| + mrdaught (~mrdaught@172.58.139.141) | 17:39 | |
| minute | ok this is some kind of .cache issue | 17:41 |
| - mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@172.58.139.141) | 17:42 | |
| josch | if you are reading this and are running ls1028a or imx8mq and feel adventurous and also have too much free time, consider running "sudo rm /usr/share/initramfs-tools/scripts/init-top/reform && sudo update-initramfs -u && systemctl reboot" and then reboot a few times to make sure that the display comes on reliably. Then tell me about your results. :) | 17:48 |
| - mark_ (QUIT: Ping timeout: 264 seconds) (~mjw@gnu.wildebeest.org) | 17:51 | |
| + mrdaught (~mrdaught@172.58.139.141) | 17:57 | |
| josch | i did it on my a311d reform and everything seems to work fine | 18:00 |
| josch | minute: i'll conditionalize the modprobe in initramfs to only run on imx8mq and ls1028a, okay? | 18:01 |
| minute | josch: ok! | 18:01 |
| - mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@172.58.139.141) | 18:01 | |
| - S0rin (QUIT: Ping timeout: 268 seconds) (~S0rin@user/s0rin) | 18:03 | |
| josch | minute: can writing a broken u-boot to emmc on rk3588 soft-brick it similar to the banana-pi? I'm asking because I'm currently addink rk3588 branches into the reform-tools staging branch and was wondering whether similar treatment as for a311d is needed for rk3588? | 18:05 |
| minute | josch: possibly yeah | 18:06 |
| minute | josch: what's your opinion about reducing kernel spam in the console? | 18:07 |
| minute | i.e. lower loglevel by default | 18:07 |
| minute | i've seen users confused by kernel spam writing over the login etc | 18:07 |
| josch | personally, when i'm on a tty, i expect the kernel messages when i do things like connect to wifi or attach a device | 18:15 |
| josch | but i see how it's confusing when you are at the login, for example | 18:15 |
| josch | if you just want to prevent this from happening at the login (do you?) maybe there exist other options? | 18:15 |
| josch | the kernel messages are definitely printed with agreetd, for example | 18:16 |
| josch | maybe it'd make sense to send them to ttyX and do the login on ttyY? | 18:16 |
| josch | so that one can still get to the messages if one wants to by switching the tty? | 18:16 |
| josch | i don't have a strong opinion on the topic | 18:17 |
| - murdock (QUIT: Ping timeout: 252 seconds) (~murdock@dhcp-209-99-207-141.greenmountainaccess.net) | 18:18 | |
| + S0rin (~S0rin@user/s0rin) | 18:20 | |
| minute | hm :| | 18:20 |
| minute | i usually use dmesg -w if i want to see messages | 18:21 |
| josch | i use that too, yes | 18:21 |
| josch | minute: i say just go for it | 18:21 |
| minute | ok cool | 18:21 |
| josch | if it turns out to be bad, you can always revert :) | 18:21 |
| minute | true true :3 | 18:21 |
| josch | minute: if you need a package that is not yet in debian like ly, tell me early so that i can get it through NEW :) | 18:22 |
| josch | minute: also, just making sure, you saw the new rk3588 kernel images i posted links to this morning? | 18:23 |
| minute | josch: oh yeah, i will try them after finishing my setup integration work foo | 18:24 |
| josch | i'm very much looking forward to that first-boot wizard :) | 18:24 |
| + murdock (~murdock@2600:380:8c1c:7563:cc0d:8cf3:991f:82bd) | 18:24 | |
| josch | that'll be a very nice solution for the things that are currently bad hacks XD | 18:25 |
| minute | yep | 18:25 |
| josch | if you make the wizard ready for being uploaded to crates.io i think there is little that stops it from getting uploaded into debian proper | 18:26 |
| josch | (because in contrast to reform-tools, just installing the first-boot wizard will not mess with the system) | 18:26 |
| minute | interesting | 18:31 |
| josch | Since it is rust i guess you can always just wget the statically linked binary blob and it will work. My own motivation to do the packaging for that would be that having it in Debian gives a better assurance that what you have running on your reform is indeed FOSS and not just $binary downloaded from $somewhere. | 18:34 |
| josch | The rust team auto-generates dependencies but for that to work, the software should be available as a crate, ideally from the official crates.io repo. | 18:35 |
| aperezdc | minute: GTK4 compiles shaders and those might take a bit of time on first runs, IIRC | 18:35 |
| josch | interesting -- can those be precompiled maybe? | 18:35 |
| josch | we could fill the respective cashes during the system image build if that is possible | 18:35 |
| josch | *caches | 18:35 |
| minute | josch: this is my plan https://source.mnt.re/reform/reform-system-image/-/issues/26 | 18:37 |
| aperezdc | the compiled result would be different for etnaviv/mali/etc, so... not sure if it's worth it | 18:37 |
| minute | aperezdc: yeah... i now have a nice background image in there that can bridge the few seconds of waiting maybe ^^ | 18:37 |
| minute | will see how well it works | 18:38 |
| - pandora (QUIT: Quit: Connection closed for inactivity) (uid585533@id-585533.ilkley.irccloud.com) | 18:44 | |
| - murdock (QUIT: Read error: Connection reset by peer) (~murdock@2600:380:8c1c:7563:cc0d:8cf3:991f:82bd) | 18:55 | |
| + murdock (~murdock@dhcp-209-99-207-141.greenmountainaccess.net) | 18:56 | |
| josch | cool! i read the issue and will leave comments i guess some time later tonight :) | 19:01 |
| + mrdaught (~mrdaught@172.58.139.141) | 19:02 | |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:50) | 19:02 | |
| - mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@172.58.139.141) | 19:02 | |
| minute | josch: any idea how i'm holding it wrong? https://source.mnt.re/reform/mnt-reform-setup-wizard/-/jobs/3347#L54 | 19:04 |
| - S0rin (QUIT: Ping timeout: 260 seconds) (~S0rin@user/s0rin) | 19:05 | |
| josch | minute: you forgot to install crossbuild-essential-arm64 | 19:06 |
| minute | oh :3 | 19:06 |
| + S0rin (~S0rin@user/s0rin) | 19:06 | |
| josch | also, try adding -oDebug::pkgProblemResolver=true -oDebug::pkgDepCache::Marker=1 -oDebug::pkgDepCache::AutoInstall=1 to the apt call to let it tell you why it cannot install things | 19:06 |
| josch | (i know, this is so intuitive compared to a --debug switch) | 19:07 |
| josch | minute: you need both: build-essential and crossbuild-essential-arm64 | 19:08 |
| josch | before you had build-essential:arm64 but you just need plain build-essential | 19:08 |
| minute | mhm mhm | 19:09 |
| minute | josch: thank you | 19:10 |
| minute | incredible how much of a struggle this is | 19:43 |
| minute | locally it works except for the linking, because of broken unstable i can't currently install the arm64 libs of gtk4 | 19:43 |
| - erle (QUIT: Ping timeout: 256 seconds) (~erle@2a02:8109:da01:6400::311a) | 19:45 | |
| + erle (~erle@2a02:8109:da01:6400::311a) | 19:56 | |
| minute | ah i think the problem is that the gtk4 version in the debian that is used for rust is too old | 20:09 |
| minute | in the debian that is used for the rust docker image | 20:09 |
| + mark_ (~mjw@gnu.wildebeest.org) | 20:16 | |
| - mjw (QUIT: Killed (tantalum.libera.chat (Nickname regained by services))) (~mjw@2001:1c06:2488:1400:4fd:39a7:74ac:7bae) | 20:22 | |
| * mark_ -> mjw | 20:22 | |
| + Guest4862 (~mjw@2001:1c06:2488:1400:4fd:39a7:74ac:7bae) | 20:23 | |
| josch | minute: i have no clue how cargo does cross-compiling but the log suggests it is using pkg-config and in that case you might need to set PKGCONFIG_EXECUTABLE=/usr/bin/aarch64-linux-gnu-pkg-config | 20:25 |
| josch | and/or PKG_CONFIG_EXECUTABLE=/usr/bin/aarch64-linux-gnu-pkg-config | 20:25 |
| minute | josch: i already figured that part out, but the issue is the gtk4 version, it needs at least 4.11 | 20:25 |
| minute | so i can't use a bookworm container | 20:25 |
| josch | right, unstable has 4.12 | 20:26 |
| - S0rin (QUIT: Ping timeout: 264 seconds) (~S0rin@user/s0rin) | 20:26 | |
| minute | so i'm trying to construct something that uses unstable and also can add the arm64 cross build target to rust | 20:27 |
| + S0rin (~S0rin@user/s0rin) | 20:31 | |
| + klardotsh (~klardotsh@c-67-170-115-80.hsd1.wa.comcast.net) | 20:42 | |
| minute | ok i got it to the linking step in my docker container locally | 20:43 |
| minute | josch: "aarch64-linux-gnu-pkg-config" does not seem to be a thing in debian | 20:44 |
| josch | minute: it is. Install pkgconf:arm64 | 20:47 |
| minute | oh ok | 20:47 |
| josch | this installs a symlink which ends up at /usr/bin/pkgconf from pkgconf-bin | 20:48 |
| josch | and pkgconf-bin is from your native architecture | 20:48 |
| josch | this works because pkgconf-bin is marked as Multi-Arch:foreign | 20:48 |
| minute | ahhh i made a terrible mistake, but now i got it | 20:56 |
| josch | Do tell, because i have no clue how cargo cross-builds :) | 20:57 |
| minute | i got a finished build inside my test docker container, now retrying it cleanly | 21:02 |
| - S0rin (QUIT: Ping timeout: 264 seconds) (~S0rin@user/s0rin) | 21:13 | |
| minute | excuse me https://source.mnt.re/reform/mnt-reform-setup-wizard/-/jobs/3365#L270 | 21:16 |
| josch | managing to find out that you need dbus-update-activation-environment in your sway config must've been "fun" | 21:16 |
| minute | josch: ah well that's an old and well-known one for me :D | 21:17 |
| minute | josch: this is copied from debian's /etc/sway/config.d/... | 21:17 |
| josch | minute: the line you linked is the 64-bit time_t stuff | 21:17 |
| minute | josch: yeah ;/ it literally just worked here locally 2 minutes ago | 21:17 |
| josch | minute: instead of copy-ing those lines, you could just include /etc/sway/config.d/50-systemd-user.conf, no? | 21:18 |
| + S0rin (~S0rin@user/s0rin) | 21:18 | |
| minute | josch: i could, but if that file or directory contents changes in a weird way then the setup breaks | 21:19 |
| minute | i will try to build with testing-slim | 21:20 |
| minute | > Job succeeded | 21:24 |
| josch | i just tried the executable locally :) | 21:25 |
| minute | niice | 21:25 |
| minute | (it's missing 2 svgs) | 21:25 |
| minute | careful btw, it actually wants to change stuff | 21:25 |
| josch | but not if i don't click anything, right? | 21:26 |
| josch | i'm also not running it as root :) | 21:26 |
| minute | josch: right | 21:30 |
| minute | ok gonna merge in some of plomlompom's stuff | 21:30 |
| josch | what is the reason for limiting people to what they can use as their login password? | 21:35 |
| + mrdaught (~mrdaught@172.58.139.13) | 21:43 | |
| - mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@172.58.139.13) | 21:43 | |
| vagrantc | huh. when i downgraded my cells to 1500mAh cells (because of a failure) ... the battery module still reports the design capacity as 18000mAh ... | 21:46 |
| vagrantc | might be why it suddently cut power the other day? | 21:46 |
| vagrantc | one cell dropped to 2.4v while the others were ~3.1+ | 21:48 |
| vagrantc | wouldn't surprise me if some of the cells are more worn than the others ... but maybe also not measuring the full capacity correctly? | 21:48 |
| minute | josch: the characters that can be safely piped to chpasswd | 21:54 |
| minute | josch: we don't have a better solution yet | 21:54 |
| minute | i wonder how debian-installer solves this | 21:57 |
| minute | on the positive side, i just figured out how to include all necessary svg graphics in the executable so we really only need 1 binary | 21:58 |
| minute | now learning how to make this executable smaller :D | 22:01 |
| minute | oh wow just building in release mode yields a 6.6M executable vs 102M in debug (on x64) | 22:02 |
| josch | minute: jusr run the crypt(3) algo over the password string and then pass that to usermod --password=$crypted | 22:03 |
| minute | josch: ah! | 22:04 |
| - S0rin (QUIT: Ping timeout: 260 seconds) (~S0rin@user/s0rin) | 22:05 | |
| minute | ok, size-optimized binary is now 1.7M | 22:05 |
| grimmware | Is crypt still good? | 22:06 |
| minute | with upx we're down to 600kB! | 22:06 |
| minute | the question is, can i cross-run upx | 22:06 |
| + S0rin (~S0rin@user/s0rin) | 22:08 | |
| minute | oh mega neat, upx can cross-compress an arm64 executable | 22:08 |
| josch | grimmware: i guess that depends on the hash, see crypt(5) | 22:08 |
| josch | according to that page, yescrypt is currently the best | 22:10 |
| minute | 589 kb :3 https://source.mnt.re/reform/mnt-reform-setup-wizard/-/jobs/3370/artifacts/browse/target/aarch64-unknown-linux-gnu/release/ | 22:17 |
| minute | (and it works on the target) | 22:19 |
| - S0rin (QUIT: Ping timeout: 264 seconds) (~S0rin@user/s0rin) | 22:28 | |
| - murdock (QUIT: Ping timeout: 272 seconds) (~murdock@dhcp-209-99-207-141.greenmountainaccess.net) | 22:28 | |
| - colinsane (QUIT: Quit: bye) (~colinunin@97-113-159-4.tukw.qwest.net) | 22:28 | |
| + chomwitt (~chomwitt@2a02:587:7a18:fc00:1ac0:4dff:fedb:a3f1) | 22:30 | |
| + colinsane (~colinunin@97-113-159-4.tukw.qwest.net) | 22:30 | |
| + murdock (~murdock@2600:380:8c1c:7563:cc0d:8cf3:991f:82bd) | 22:32 | |
| + mrdaught (~mrdaught@172.58.139.84) | 22:38 | |
| - robin_ (QUIT: Remote host closed the connection) (~robin@user/terpri) | 22:49 | |
| + robin (~robin@user/terpri) | 22:50 | |
| + S0rin (~S0rin@user/s0rin) | 23:08 | |
| - mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@172.58.139.84) | 23:21 | |
| + mrdaught (~mrdaught@172.58.139.84) | 23:22 | |
| grimmware | josch: gotcha, just been reading about the morris worm and the conflation between the program and the algorithm gave me pause :P | 23:24 |
| - mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@172.58.139.84) | 23:28 | |
| + mrdaught (~mrdaught@172.58.139.84) | 23:33 | |
| + sir-photch (~m-hy5poy@2a01:4f8:c2c:5963::1) | 23:33 | |
| sir-photch | hello world | 23:34 |
| + Guest50 (~Guest50@91.26.88.219) | 23:35 | |
| - Guest50 (PART: !!unknown attribute: msg!!) (~Guest50@91.26.88.219) | 23:36 | |
| - sir-photch (QUIT: Remote host closed the connection) (~m-hy5poy@2a01:4f8:c2c:5963::1) | 23:38 | |
| - mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@172.58.139.84) | 23:38 | |
| + sir-photch (~m-hy5poy@2a01:4f8:c2c:5963::1) | 23:38 | |
| - murdock (QUIT: Read error: Connection reset by peer) (~murdock@2600:380:8c1c:7563:cc0d:8cf3:991f:82bd) | 23:43 | |
| + murdock (~murdock@dhcp-209-99-207-141.greenmountainaccess.net) | 23:43 | |
| + mrdaught (~mrdaught@172.58.139.84) | 23:45 | |
| - mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@172.58.139.84) | 23:46 | |
| - S0rin (QUIT: Ping timeout: 264 seconds) (~S0rin@user/s0rin) | 23:50 | |
| + S0rin (~S0rin@user/s0rin) | 23:51 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!