| + paperManu_ (~paperManu@198.58.139.163) | 00:11 | |
| - kensanata (QUIT: Quit: Ping timeout (120 seconds)) (~alex@user/kensanata) | 00:26 | |
| + kensanata (~alex@user/kensanata) | 00:27 | |
| - kensanata (QUIT: Ping timeout: 248 seconds) (~alex@user/kensanata) | 00:40 | |
| + kensanata (~alex@user/kensanata) | 00:43 | |
| bremner | minute: as promised, a writeup of what changed with SR's new patch for hibernation: https://www.cs.unb.ca/~bremner/blog/posts/hibernate-pocket-13/ | 00:56 |
|---|---|---|
| bremner | I don't really know who to bug about the PCI reset series not applying anymore. I guess I should see if it applies on a tagged upstream version. | 00:58 |
| - kensanata (QUIT: Ping timeout: 244 seconds) (~alex@user/kensanata) | 00:59 | |
| + kensanata (~alex@user/kensanata) | 01:02 | |
| minute | bremner: you could write to sebastian.reichel at collabora.com | 01:07 |
| minute | bremner: after all he already read your blog posts | 01:08 |
| bremner | minute: ack. I just wanted to see where the conflicts are happening, but it does seem to be with the active work at collabora more than what is in released linux | 01:14 |
| bremner | but the patch series is by someone else, so... | 01:14 |
| elb | tuigreet also doesn't start a full session | 01:14 |
| - kensanata (QUIT: Ping timeout: 248 seconds) (~alex@user/kensanata) | 01:14 | |
| elb | I'm going to have to learn more about how modern ridiculously overcomplicated desktop environments work than I had wanted to | 01:14 |
| + kensanata (~alex@user/kensanata) | 01:21 | |
| - kensanata (QUIT: Ping timeout: 256 seconds) (~alex@user/kensanata) | 01:30 | |
| elb | josch: it's not just gdm that won't run with the bad firmware-misc-nonfree, gnome-session also fails | 01:31 |
| + kensanata (~alex@user/kensanata) | 01:32 | |
| - kensanata (QUIT: Ping timeout: 260 seconds) (~alex@user/kensanata) | 01:37 | |
| + kensanata (~alex@user/kensanata) | 01:39 | |
| elb | ok I dunno if I have all the stuff a fancy session wants, but i have ssh-agent, so I'm happy ;-) | 01:43 |
| - kensanata (QUIT: Ping timeout: 260 seconds) (~alex@user/kensanata) | 01:45 | |
| + kensanata (~alex@user/kensanata) | 01:47 | |
| josch | if i didn't mess up, this pipeline will produce fixed firmware-misc-nonfree packages: https://source.mnt.re/josch/reform-debian-packages/-/pipelines/4885 | 01:57 |
| josch | It's too late over here to wait for the pipeline to finish and test. I'll do that tomorrow but maybe somebody in a different timezone feels motivated to do this before I get to do it in about 8 hours. :) | 01:57 |
| - kensanata (QUIT: Ping timeout: 244 seconds) (~alex@user/kensanata) | 02:00 | |
| minute | josch: uff! what was the bug? | 02:00 |
| + kensanata (~alex@user/kensanata) | 02:01 | |
| minute | bremner: btw did you try removing that cqe flag from emmc node in dts? | 02:02 |
| minute | anyways, really really need to slee | 02:02 |
| minute | sleep | 02:02 |
| minute | (sent from my MNT Reform Next) | 02:02 |
| minute | nighty night | 02:02 |
| - kensanata (QUIT: Quit: Ping timeout (120 seconds)) (~alex@user/kensanata) | 02:21 | |
| + kensanata (~alex@user/kensanata) | 02:21 | |
| - kensanata (QUIT: Client Quit) (~alex@user/kensanata) | 02:25 | |
| + kensanata (~alex@user/kensanata) | 02:25 | |
| - kensanata (QUIT: Quit: Ping timeout (120 seconds)) (~alex@user/kensanata) | 02:34 | |
| + kensanata (~alex@user/kensanata) | 02:34 | |
| - kensanata (QUIT: Ping timeout: 246 seconds) (~alex@user/kensanata) | 02:41 | |
| + kensanata (~alex@user/kensanata) | 02:46 | |
| - kensanata (QUIT: Ping timeout: 248 seconds) (~alex@user/kensanata) | 03:08 | |
| + kensanata (~alex@user/kensanata) | 03:11 | |
| - kensanata (QUIT: Client Quit) (~alex@user/kensanata) | 03:15 | |
| + kensanata (~alex@user/kensanata) | 03:15 | |
| - kensanata (QUIT: Ping timeout: 246 seconds) (~alex@user/kensanata) | 03:20 | |
| + kensanata (~alex@user/kensanata) | 03:20 | |
| - kensanata (QUIT: Ping timeout: 256 seconds) (~alex@user/kensanata) | 03:36 | |
| - paperManu_ (QUIT: Quit: WeeChat 4.7.1) (~paperManu@198.58.139.163) | 03:37 | |
| - paperManu (QUIT: Ping timeout: 248 seconds) (~paperManu@198.58.139.163) | 03:40 | |
| + paperManu (~paperManu@198.58.139.163) | 03:43 | |
| + kensanata (~alex@user/kensanata) | 03:48 | |
| - kensanata (QUIT: Ping timeout: 246 seconds) (~alex@user/kensanata) | 03:53 | |
| + kensanata (~alex@user/kensanata) | 03:54 | |
| - kensanata (QUIT: Ping timeout: 248 seconds) (~alex@user/kensanata) | 03:58 | |
| + kensanata (~alex@user/kensanata) | 03:59 | |
| - kensanata (QUIT: Client Quit) (~alex@user/kensanata) | 04:04 | |
| + kensanata (~alex@user/kensanata) | 04:04 | |
| - kensanata (QUIT: Ping timeout: 244 seconds) (~alex@user/kensanata) | 04:09 | |
| + kensanata (~alex@user/kensanata) | 04:10 | |
| - kensanata (QUIT: Ping timeout: 260 seconds) (~alex@user/kensanata) | 04:15 | |
| + kensanata (~alex@user/kensanata) | 04:22 | |
| - paperManu (QUIT: Ping timeout: 246 seconds) (~paperManu@198.58.139.163) | 04:29 | |
| - kensanata (QUIT: Ping timeout: 248 seconds) (~alex@user/kensanata) | 04:48 | |
| + kensanata (~alex@user/kensanata) | 04:49 | |
| - kensanata (QUIT: Quit: Ping timeout (120 seconds)) (~alex@user/kensanata) | 04:56 | |
| + kensanata (~alex@user/kensanata) | 04:56 | |
| + sevarat (~sevarat@2601:647:6511:f9b0:2354:2d68:5792:891c) | 05:21 | |
| - kensanata (QUIT: Quit: Ping timeout (120 seconds)) (~alex@user/kensanata) | 05:24 | |
| + kensanata (~alex@user/kensanata) | 05:24 | |
| - sevarat (QUIT: Quit: nyaa~) (~sevarat@2601:647:6511:f9b0:2354:2d68:5792:891c) | 05:26 | |
| + sevarat (~sevarat@2601:647:6511:f9b0:2354:2d68:5792:891c) | 05:30 | |
| - mjw (QUIT: Ping timeout: 264 seconds) (~mjw@gnu.wildebeest.org) | 05:30 | |
| - kensanata (QUIT: Ping timeout: 256 seconds) (~alex@user/kensanata) | 05:36 | |
| + kensanata (~alex@user/kensanata) | 05:37 | |
| - kensanata (QUIT: Ping timeout: 240 seconds) (~alex@user/kensanata) | 05:55 | |
| + kensanata (~alex@user/kensanata) | 05:59 | |
| josch | oh, of course qcacld fails.. XD | 08:22 |
| + rwa_ (~rene@2001:9e8:33b8:6300:ac4d:8aa0:b7ef:664a) | 08:30 | |
| - rwa_ (QUIT: Remote host closed the connection) (~rene@2001:9e8:33b8:6300:ac4d:8aa0:b7ef:664a) | 09:11 | |
| + pomel0 (~pomel0@user/pomel0) | 09:16 | |
| - pomel0 (QUIT: Remote host closed the connection) (~pomel0@user/pomel0) | 09:24 | |
| + gustav25 (~gustav@c-78-82-53-102.bbcust.telenor.se) | 10:02 | |
| - wakest (QUIT: Remote host closed the connection) (m-a7d6fe@45.77.48.108) | 10:51 | |
| f_ | Is Reform Next available for purchase btw? | 11:13 |
| josch | f_: it's available for pre-order at crowdsupply: https://www.crowdsupply.com/mnt/mnt-reform-next | 11:17 |
| josch | minute: could you please merge this: https://source.mnt.re/reform/qcacld2/-/merge_requests/13 | 11:18 |
| f_ | okay cool! | 11:18 |
| f_ | compatible with every MNT Reform CPU module I suppose? | 11:18 |
| josch | f_: theoretically, probably yes but so far we only have the dtb for rk3588 | 11:19 |
| josch | f_: in practice, there are probably issues with SoMs like ls1028a which need an eDP adapter for the display, for example | 11:24 |
| f_ | > Dual screen (tested with 1440P, 120 Hz display) | 11:28 |
| f_ | \o/ so there's a non-zero chance of me being able to use my ultrawide screen at its max resolution of 3440x1440@165Hz :D | 11:29 |
| f_ | I wasn't sure whenever it could output 1440p 165Hz .. my laptop can output 1080p 165Hz but can't do 3440x1440 165Hz | 11:29 |
| f_ | josch: ok. rk3588 is what I will probably go with | 11:31 |
| f_ | so that's good | 11:31 |
| josch | go for rk3588 -- it's *really* good :) | 11:32 |
| josch | i'm only still sticking with a311d because i don't like change XD | 11:32 |
| josch | but at some point the performance of firefox/chromium on that cpu will make me switch :D | 11:33 |
| f_ | josch: performance isn't really an issue heh | 11:34 |
| f_ | actually I have an S905D3 which handles firefox just fine | 11:35 |
| josch | oh a311d also handles firefox just fine | 11:35 |
| f_ | and youtube tv (it's a webapp) and stuff | 11:35 |
| josch | the problem is not firefox but websites with too much javascript | 11:35 |
| f_ | youtube tv is a huge blob of javascripting | 11:35 |
| f_ | but yes this is an issue here too | 11:36 |
| josch | did you try gitlab or twitch? :) | 11:36 |
| f_ | twitch no, gitlab I use it everyday | 11:36 |
| f_ | laptop barely handles it | 11:36 |
| f_ | inside tor browser it works .. ish | 11:36 |
| josch | another issue is only having 3.6 gb ram which means that the memory used by firefox is constantly getting swapped out when i do actual work and needs to be loaded back in when i switch back to my firefox workspace and that is sloooow | 11:37 |
| f_ | aahh | 11:37 |
| f_ | even 8 GB isn't enough these days | 11:37 |
| f_ | (laptop has 8 GB of RAM, upgradable up to 32 GB of RAM .. but I didn't really feel the need to upgrade it) | 11:38 |
| josch | i don't need much. Before i switched to the reform my intel laptop had a slower processor and less ram than rk3588. So rk3588 is actually overkill for me. XD | 11:41 |
| f_ | Kinda overkill for me too lol | 11:44 |
| f_ | but it's the one that lets me use 2 displays with one of them maxed out at 1440p@165Hz :p | 11:44 |
| f_ | Has anyone tried to compile a whole linux kernel on it btw? | 11:45 |
| josch | multiple times on all mnt SoMs, yes | 11:45 |
| f_ | On my laptop, compiling the linux-postmarketos-amlogic kernel takes ~4 hours | 11:45 |
| josch | it heavily depends on the kernel config | 11:45 |
| f_ | (kernel package in postmarketOS with amlogic kernel config) | 11:45 |
| josch | a minimal config created with localmodconfig takes 15 minutes | 11:46 |
| josch | a full debian kernel takes 1:15 hours | 11:46 |
| f_ | and an allyesconfig? | 11:46 |
| f_ | oh, not bad at all | 11:46 |
| josch | i didn't try allyesconfig but i can give it a go if you're interested :) | 11:46 |
| minute | josch: i'll do MR, CI etc after coffee! | 11:53 |
| minute | btw i just turned on my pocket and my next and the next arrives at desktop much faster | 11:54 |
| minute | my pocket has experimental plymouth still, need to turn it off to see what's going on maybe | 11:54 |
| josch | minute: cool, no rush -- i'm using my own brances on gitlab in the meantime. I'm currently waiting for the next pipeline results to build and test new system images. | 11:55 |
| minute | after entering decryption passphrase, there is a long delay before it arrives at gdm. on next, there's no delay _but_ it also doesn't use an encrypted ssd at the moment (uses emmc) | 11:55 |
| * buckket1 -> buckket | 12:05 | |
| josch | thank you for merging qcacld2!13 | 12:07 |
| - buckket (QUIT: Quit: buckket) (~buckket@vps.buckket.org) | 12:08 | |
| josch | fixing that driver is not much fun... took me 3 hours doing that today which is the same time that i spent yesterday to do something actually productive: fixing the firmware-nonfree issue... | 12:09 |
| + buckket (~buckket@vps.buckket.org) | 12:09 | |
| minute | josch: oh noo :/ i'm so sorry | 12:12 |
| josch | not your fault -- it's the the result of vendors not upstreaming their drivers :( | 12:15 |
| josch | i'm just surprised i didn't break the thing yet because i cannot test anything | 12:16 |
| josch | but i heard grimmware is working on changing that and increasing my motivation for qcacld2 :P | 12:16 |
| grimmware | I have an anti static bag, some padding and a box, I just need to remember to send it:) | 12:23 |
| minute | elb: josch: CI runner assigned to https://source.mnt.re/elb/pocket-reform | 12:24 |
| - sevarat (QUIT: Ping timeout: 248 seconds) (~sevarat@2601:647:6511:f9b0:2354:2d68:5792:891c) | 12:30 | |
| josch | oh great, the emmc problems on ls1028a are back... https://mister-muffin.de/p/Fo4A.txt | 12:46 |
| minute | josch: ugh, that looks annoying | 12:55 |
| josch | root@reform:~# cat /proc/device-tree/model | 13:03 |
| josch | MNT Reform 2 with LS1028A Module | 13:03 |
| josch | Linux reform 6.17.2-mnt-reform-arm64 #1 SMP PREEMPT Debian 6.17.2-1~exp1+reform20251015T205418Z (2025 aarch64 GNU/Linux | 13:03 |
| josch | reform-setup-wizard also started successfully (just heard the screen reader voice) | 13:04 |
| josch | so the problem might be a very funny one | 13:04 |
| josch | the kernel which currently booted just fine on LS1028A was built from the *exact same sources* locally on my a311d | 13:04 |
| josch | the kernel that does not work with above output on dmesg was built my MNT gitlab CI | 13:04 |
| josch | maybe this problem is a random one | 13:04 |
| josch | or maybe it's a problem which is related to what is installed in the build environment | 13:05 |
| minute | huh | 13:05 |
| minute | could also be a driver module loading timing thing | 13:06 |
| minute | remember how usb3 on ls1028a only works if kernel/dmesg logging is active ;D | 13:06 |
| josch | what i don't *want* to remember?? XD | 13:11 |
| josch | i'll put this aside for now -- futer me has the irc history to pick this up again | 13:12 |
| josch | rk3588 is a bit more funny :) | 13:13 |
| josch | and it works out-of-the-box | 13:17 |
| josch | hats off to collabora | 13:17 |
| josch | my firmware-misc-nonfree fix also works | 13:17 |
| josch | i'll cherry-pick that one into reform-debian-packages main until it gets fixed in Debian | 13:17 |
| minute | nice | 13:17 |
| minute | ah btw still curious what caused the breakage in firmware-misc-nonfree | 13:18 |
| minute | i'm currently building box64/32 (not 86) on next | 13:18 |
| josch | minute: the version of box64 is debian is too old? | 13:18 |
| minute | josch: don't know, i just wanna try the latest git stuff | 13:19 |
| minute | josch: does the debian build build box32? | 13:19 |
| minute | because it's still experimental according to the build options | 13:19 |
| minute | (so i didn't expect it to be there) | 13:19 |
| josch | ups, yes it builds it but i forgot to add the file into the binary packages XD | 13:21 |
| minute | also i'm testing latest reform next build at home with various things to see how it behaves :D | 13:21 |
| josch | 0.3.8 got released 5 days ago, let me do another upload | 13:24 |
| minute | now installing steam using the script included with box64 | 13:24 |
| josch | tell me if that script works, then i can stuff it into the Debian package | 13:27 |
| minute | wow, steam just works | 13:27 |
| minute | josch: yes, the script works | 13:28 |
| minute | this is astounding, steam just works and it works without armhf arch | 13:28 |
| grimmware | wait what | 13:29 |
| josch | \o/ | 13:29 |
| grimmware | btw my CI runner for reform-debian-packages times out before it can get to the reprepo stage due to the intel builds which I don't need - is there a way to disable this? | 13:31 |
| josch | grimmware: two options: edit .gitlab-ci.yml or ask minute to enable the runner | 13:32 |
| grimmware | well I don't know how the fuck I missed that last time I looked | 13:34 |
| gordon2 | wat | 13:35 |
| minute | so far steam runs very well, even the desktop notifications incl sound | 13:36 |
| gordon2 | i managed to make steamcmd work but that required some trickery with box64 | 13:36 |
| minute | and i updated my profile inside of it | 13:36 |
| gordon2 | maybe it's gentoo thing | 13:36 |
| minute | i did zero trickery! | 13:36 |
| minute | except enabling box32 in box64 build | 13:36 |
| gordon2 | oh that didn't work for me at all, it wasn't able to run any 32-bit stuff | 13:37 |
| gordon2 | interesting | 13:37 |
| minute | i think they fixed it quite recently... when did you build it? | 13:37 |
| minute | btw i'm also on gnome | 13:37 |
| gordon2 | overall i had very bad experience with box64, wonder what am i doing wrong | 13:37 |
| gordon2 | except using gentoo ofc | 13:37 |
| minute | i remember sway was more troubling because steam doesn't expect tiled window manager perhaps | 13:37 |
| gordon2 | i built it a while ago, let me check | 13:38 |
| minute | gordon2: weird. box64 worked quite well for me for lots of things | 13:38 |
| minute | haha but axiom verge doesn't launch, at least not in vulkan mode | 13:38 |
| gordon2 | and it does not work in wine for me at all | 13:38 |
| gordon2 | yeah, it was in august i last compiled it | 13:38 |
| gordon2 | maybe need to try again | 13:39 |
| minute | oh but it launches in opengl mode | 13:39 |
| gordon2 | oh, i use that thing for it | 13:39 |
| gordon2 | one sec | 13:39 |
| gordon2 | the thing that translates opengl to es or something | 13:40 |
| josch | minute: you were wondering above about firmware-misc-nonfree. The problem was, that the package included a broken symlink for the mali_csffw.bin firmware blob. It should be fixed by this change: https://salsa.debian.org/kernel-team/firmware-nonfree/-/merge_requests/132 | 13:41 |
| grimmware | I love using steam on my pocket. Absolutely marvelling at playing amd64 Quadrilateral Cowboy on my arm64 open source linux machine without skipping a beat is incredible | 13:43 |
| gordon2 | minute: gl4es | 13:43 |
| grimmware | 18 year old me would lose his tiny mind | 13:43 |
| gordon2 | here we go, that took me a while | 13:43 |
| gordon2 | https://ptitseb.github.io/gl4es/ | 13:43 |
| minute | gordon2: gl4es is not needed on rk3588 i think... what was your usecase again? | 13:44 |
| gordon2 | ... i don't quite remember, i think it just was suggested by box64 ebuild | 13:44 |
| minute | grimmware: yeah i also feel kind of vindicated r/n because gaming is an integral part of computering for me / how i grew up with c64, amiga etc | 13:45 |
| gordon2 | tbh i think it was that nothing worked and i was desperate to try anything | 13:45 |
| gordon2 | duskers still refuse to run :( | 13:45 |
| minute | grimmware: ah you're using steam for longer on pocket already? :3 with full UI? | 13:45 |
| gordon2 | with new box64 | 13:45 |
| minute | aw. is that a game? | 13:46 |
| gordon2 | yep https://www.gog.com/en/game/duskers , native linux | 13:46 |
| gordon2 | but x86_64 ofc | 13:46 |
| grimmware | minute: yeah, I fucked around with it a whole bunch installing armhf libraries and stuff a while ago and got kinda close and then at some point in the intervening time (probably a package upgrade) it just started working | 13:46 |
| grimmware | I did have some problems with the mouse on fullscreen games for a bit but that turned out to be a bug in niri which was fixed in main by the time I thought to look | 13:47 |
| gordon2 | ACTION meanwhile is trying few more trackball cup designs | 13:47 |
| grimmware | niri is fantastic by the way, it matches the pocket reform so well | 13:47 |
| minute | grimmware: ah nice, need to look into it and find a way to configure it with the same shortcuts etc | 13:48 |
| minute | but first, more battery system fixes... | 13:49 |
| grimmware | <3 battery system fixes are top of my wishlist | 13:49 |
| grimmware | the movement on that front has really enthused me about using my pocket again | 13:50 |
| minute | grimmware: on my test with charging board v2 with fresh cells i didn't have any more reboot issuse | 13:50 |
| minute | grimmware: would you like to betatest one? (charging board v2 that is) | 13:50 |
| grimmware | minute: I would absolutely love to betatest one | 13:51 |
| minute | ok cool, gonna put that in my todo for monday | 13:51 |
| grimmware | excellent :) | 13:51 |
| minute | grimmware: can you DM me your shipping address or previous MNT shop order number, then it'll be easiest for me | 13:52 |
| grimmware | yeah lemme dig out a shop order number | 13:54 |
| + paperManu (~paperManu@198.58.139.163) | 14:01 | |
| josch | the pocket reboot issues are really funky. The last reboot I had was with old firmware on battery while watching a video. Since then: nothing. It had a week uptime (on power) and recorded 1.5 hours of video without problems. So the reboots are really seldom for me for some reason. Maybe my cells being different has to do something with it? | 14:05 |
| grimmware | I can see that. It occurred to me that mine may be so consistent and problematic because of my extra i2c peripherals drawing that extra bit of power | 14:07 |
| grimmware | I think spiking load average is a huge factor | 14:08 |
| grimmware | I would imagine that if my pocket could make it past boot it would actually be able to run fine until I do something like launch a browser | 14:08 |
| grimmware | (when it's in the undervoltage state that is) | 14:09 |
| grimmware | also I've ridden this thing *hard* since it turned up, and I got mine pretty early on | 14:09 |
| + paperManu_ (~paperManu@198.58.139.163) | 14:13 | |
| josch | speaking of riding things hard, i need tips for how to take care of copper surfaces | 14:16 |
| josch | pure copper heatsinks look awesome when new | 14:17 |
| josch | but after some time: https://mister-muffin.de/p/VXw8.jpg | 14:17 |
| josch | i wonder if there are easy ways to protect the surface without compromising on the heat-transfer properties | 14:17 |
| grimmware | oh gosh is that actually oxidizing? | 14:18 |
| minute | josch: for me the reboots are gone now | 14:18 |
| minute | josch: omg grünspan | 14:18 |
| grimmware | I kinda love the tarnished look on the pocket's logo but I will agree that is a bit extra | 14:19 |
| josch | :D | 14:19 |
| minute | josch: we should send you the updated heatsink in any case | 14:25 |
| josch | oh? why? | 14:25 |
| minute | josch: it's a bit thinner, less warp | 14:30 |
| josch | i read about the warping from others but i somehow haven't seen this issue here so far -- i don't know what's different | 14:31 |
| minute | josch: yeah it's maybe differences in case parts that play into it also | 14:32 |
| josch | maybe -- but in any case, i'm happy with the heatsink and... | 14:32 |
| josch | i actually think the one with the fins looks a bit cooler XD | 14:32 |
| josch | f_: i'm currently timing a "allyesconfig" on rk3588 for you -- temperatures stay below 74 °C, so no throttling of the cpu due to overheating | 14:33 |
| josch | i have the acrylic back open to help with the temperatures | 14:34 |
| josch | building firmware-nonfree increases the time that reform-debian-package pipeline takes from 40 minutes to over 1 hour. It also increases the artifact size by 400 MiB. | 14:43 |
| josch | a fived version of firmware-misc-nonfree is now in the MNT repos: https://mntre.com/reform-debian-repo/pool/main/f/firmware-nonfree/ | 15:05 |
| josch | elb, cararemixed: in case you want to test ^ | 15:06 |
| elb | is there a way to slow down the rk3588 console | 16:15 |
| elb | I get terrible corruption | 16:15 |
| elb | josch: confirmed works here | 16:16 |
| josch | \o/ | 16:19 |
| grimmware | oh my god how is it the case that every time I want to test out a change I get in a fight with a fucking build system | 16:20 |
| + mjw (~mjw@gnu.wildebeest.org) | 16:27 | |
| grimmware | Hilarious side effect of the font patching used for the MNT logo in waybar if you use the window title module https://usercontent.irccloud-cdn.com/file/qO8cSUC5/1760797729.JPG | 16:29 |
| grimmware | Kinda love it. | 16:29 |
| + pomel0 (~pomel0@user/pomel0) | 16:40 | |
| gordon2 | yeah, there is too much variation, i printed 5 trackball cups and some of them work well with SiN and some of them just refuse to register any tracking, more research needed | 16:48 |
| f_ | josch: you're awesome | 16:49 |
| josch | f_: 151 minutes | 16:51 |
| f_ | Impressive! | 16:51 |
| josch | indeed, i recently built the Debian kernel on my A311D with -j2 and it took 14:20 hours :D | 16:53 |
| elb | this is what I get on my RK3588 console: https://lostbits.net/elb/private/unfortunate_messages.png | 16:58 |
| josch | ouch o0 | 16:59 |
| josch | elb: rk3588 pocket reform, right? | 16:59 |
| f_ | elb: did you connect ground? | 17:01 |
| f_ | also love the >.> in the logs | 17:01 |
| elb | I connected the ground, rk3588 | 17:06 |
| elb | I think my serial dongle is just not reliably fast enough | 17:06 |
| elb | 1.5 Mbps seems ... excessive | 17:06 |
| elb | oh, yes, sorry, I answered the wrong part of that question -- pocket reofrm | 17:09 |
| elb | I set up a getty on tty2 so that I don't get into this trap in the future, and I see the getty running, but neither alt+hyper+2 nor ctl+alt+hyper+2 is taking me to tty2; any ideas? | 17:14 |
| - swivel (QUIT: Ping timeout: 255 seconds) (~swivel@shells.gnugeneration.com) | 18:19 | |
| + swivel (~swivel@shells.gnugeneration.com) | 18:21 | |
| grimmware | minute: that additional soft reset did not fix the flicker :/ | 18:35 |
| grimmware | at least I've got my build pipeline working now though. I can go back to the other branch I was working on with a more speculative attempt at the fix. | 18:36 |
| + wielaard (~mjw@gnu.wildebeest.org) | 19:28 | |
| - mjw (QUIT: Killed (uranium.libera.chat (Nickname regained by services))) (~mjw@gnu.wildebeest.org) | 19:30 | |
| * wielaard -> mjw | 19:30 | |
| grimmware | I've just tested a branch where I've reordered init to map better to what it says on the v1 for the datasheet, but it has the annoying side effect of only succesfully initializing the display every other time (obviously not that helpful) and doesn't appear to have fixed the flicker, but it no longer does that thing where it'll look like it's browning out and has the screen fade | 19:53 |
| grimmware | I'll be testing out some iterations of the same over the coming week I guess | 19:53 |
| grimmware | minute: is there any rationale to the two patches on the pocket panel other than that's just what happened? I'm wondering if it's worth consolidating them. | 20:13 |
| minute | grimmware: yeah i already did | 22:09 |
| minute | grimmware: a week ago or so. but i couldn't test it because CI was broken and then i ran outa time | 22:09 |
| minute | grimmware: https://source.mnt.re/reform/reform-debian-packages/-/commit/a10e45e0001e9f3a29eb95f78cc6f63733805b96 | 22:11 |
| - gustav25 (QUIT: Quit: Quit) (~gustav@c-78-82-53-102.bbcust.telenor.se) | 22:15 | |
| + colinsane (~colinunin@97-113-71-58.tukw.qwest.net) | 22:23 | |
| minute | elb: sometimes resetting uart adapter helps, i.e. replug it | 22:50 |
| minute | grimmware: lol @ mnt logo font glitch | 22:50 |
| minute | elb: you can change the baud rate by kernel parameter i think, and/or DTS | 22:51 |
| minute | elb: i.e. console=ttyS2,115000, see https://www.kernel.org/doc/html/v4.14/admin-guide/kernel-parameters.html -> console= | 22:52 |
| minute | installing test kernel on my pocket reform... | 23:18 |
| minute | grimmware: i think my MR fixes the flicker issue. but i still have some residual flicker from last reboot, so need to wait a little more. | 23:24 |
| minute | grimmware: sometimes i also get color snow but it goes away after turning the display off and on again in gnome, so this is useful for tweaking the sleep intervals for fixing that issue too. i'm thinking to maybe make sysfs attrs for the delays so i can dynamically adjust them while testing, skipping rebuild loops | 23:25 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
