| minute | grimmware: yep, this fix works https://source.mnt.re/reform/reform-debian-packages/-/merge_requests/140 | 00:07 |
|---|---|---|
| minute | ch: whenever you have time, could you confirm that your display still works after reboot with this kernel? https://source.mnt.re/reform/reform-debian-packages/-/merge_requests/140 | 00:07 |
| elb | minute: I'm pretty sure my adapter just doesn't go that fast, I ordered one with some other stuff I was getting from Digikey anyway | 00:09 |
| elb | minute: does that console line work on rk3588? if so, why doesn't it normally have a serial console? | 00:09 |
| minute | elb: that line is specifically for rk3588, and it was something missing in our uboot, that's why it wasn't there by default. also, with that command you can change the baud rate as well, that's why i looked it up for you | 00:10 |
| josch | elb: the reason it does not normally have a serial console is that u-boot does not add this to $bootargs | 00:10 |
| josch | elb: there is a MR fixing this which needs more testing | 00:10 |
| + lidstah (~lidstah@gateway/tor-sasl/lidstah) | 00:20 | |
| elb | oh, yeah, I know how to set up kernel serial console, I guess I just assumed there was a _reason_ it wasn't doing it | 00:21 |
| josch | nah just accidental omission, sorry XD | 00:21 |
| elb | personally, I feel like u-boot itself should run at a more ... pedestrian baud rate, as well | 00:22 |
| elb | but maybe that's just because my adapter won't do 1.5 Mbps ;-) | 00:22 |
| josch | elb: once you have an adapter that does 1.5 Mbps you'll definitely feel the difference | 00:24 |
| elb | I've used a lot of serial consoles in my day, I know that even 115200 is not lag-free | 00:24 |
| josch | using tmux with vim inside is much more pleasant at 1.5 Mbps | 00:24 |
| elb | it looks like u-boot is not in the pocket reform repo, is that correct? | 00:25 |
| elb | oh this makes sense, that's because it's SoM specific, not motherboard specific | 00:25 |
| elb | (I found it) | 00:25 |
| josch | elb: https://source.mnt.re/reform/reform-rk3588-uboot/ | 00:26 |
| josch | oh sry :D | 00:26 |
| elb | I guess I should look into recovery from a bad uboot flash before I go too far, here ;-) | 00:29 |
| elb | uh it'll use an sd card, right? | 00:29 |
| josch | elb: if you zeroed emmc, it will fall back to looking on sd-card, yes | 00:30 |
| - elb (QUIT: Remote host closed the connection) (~elb@68.133.31.194) | 00:31 | |
| + elb (~elb@68.133.31.194) | 00:33 | |
| elb | well that was unpleasant, my Emacs crashed | 00:33 |
| elb | (not on any reform, on my regular desktop) | 00:34 |
| minute | aw | 00:37 |
| lidstah | I was about to start asking some questions (before I purchase the reform), and you already answered one :) Hi all! | 00:37 |
| josch | _o/ | 00:38 |
| lidstah | I do have another question: I read that the official debian image for the reform is based on Debian unstable - And I'm the kind of guy who likes its Debian the boring way (aka stable). I've seen that there's a community maintained trixie image/iso/etc, but, should I use the official image, is it stable for everyday usage, or is the community trixie version best suited for me? (typical use here: | 00:53 |
| lidstah | tmux, ssh, vim, git, light webbrowsing, listening music, some emulation (fs-uae, dosbox and old consoles, the more demanding stuff being deferred to my proxmox cluster) | 00:53 |
| josch | lidstah: both have upsides and downsides -- it's hard to give a general recommendation | 00:53 |
| lidstah | and sorry for any mistakes, my english is far from perfect | 00:54 |
| josch | Debian unstable will sometimes break and you might be the one discovering this and then need to spend the time to fix things | 00:54 |
| josch | Debian stable is not necessarily more stable because the community maintained image/iso is maintained by only one person who also messes up badly from time to time | 00:55 |
| lidstah | got it - shit happens, I understand that | 00:57 |
| lidstah | mmh, might run unstable on the emmc, check how's stable by booting on sdcard sometimes/as a backup, and keep data on a nvme drive then | 00:59 |
| lidstah | thank you :) | 01:00 |
| josch | lidstah: if you run into problems, please shout! | 01:01 |
| lidstah | of course | 01:02 |
| lidstah | plus, tinkering with unstable will remember me the good ol' days :) | 01:04 |
| minute | huh, what https://xcancel.com/mariobalanica/status/1977826099104243980 | 01:05 |
| minute | this person managed to get amd rx580 working on rk3588 | 01:06 |
| Tramtrist | glad to see im not the only one using xcancel :) | 01:06 |
| minute | Tramtrist: it's quite useful :D | 01:06 |
| Tramtrist | Indeed... deleted my X account years ago but its very useful | 01:07 |
| minute | https://github.com/mariobalanica/arm-pcie-gpu-patches/tree/main/linux/6.16 | 01:17 |
| - pomel0 (QUIT: Ping timeout: 246 seconds) (~pomel0@user/pomel0) | 01:17 | |
| + pomel0 (~pomel0@user/pomel0) | 01:17 | |
| - pomel0 (QUIT: Ping timeout: 240 seconds) (~pomel0@user/pomel0) | 01:24 | |
| + pomel0 (~pomel0@user/pomel0) | 01:24 | |
| ch | minute: will test after athens (so hopefully tue evening) | 01:24 |
| minute | ch: nice! | 01:26 |
| + Guest57 (~Guest83@2603:800c:3d00:cf33:5141:e882:2da:2a) | 01:40 | |
| - Guest57 (QUIT: Client Quit) (~Guest83@2603:800c:3d00:cf33:5141:e882:2da:2a) | 01:41 | |
| - chrcav (QUIT: Ping timeout: 245 seconds) (~chrcav@user/chrcav) | 01:41 | |
| + antti- (~antti@user/antti) | 01:44 | |
| * antti- -> antti` | 01:44 | |
| - yqshao (QUIT: Quit: Ping timeout (120 seconds)) (~yqshao@user/yqshao) | 01:44 | |
| + yqshao (~yqshao@user/yqshao) | 01:44 | |
| - antti (QUIT: Ping timeout: 248 seconds) (~antti@user/antti) | 01:44 | |
| * antti` -> antti | 01:44 | |
| josch | huh, somehow this pipeline got stuck... https://source.mnt.re/reform/reform-system-image/-/pipelines/4904 | 01:46 |
| - kensanata (QUIT: Ping timeout: 244 seconds) (~alex@user/kensanata) | 01:46 | |
| + kensanata (~alex@user/kensanata) | 01:47 | |
| - kensanata (QUIT: Ping timeout: 248 seconds) (~alex@user/kensanata) | 02:27 | |
| + chrcav (~chrcav@user/chrcav) | 02:28 | |
| + murph_nj (~murph@2600:4040:af92:e600:63fd:ded8:1e89:d7b5) | 02:28 | |
| + kensanata (~alex@user/kensanata) | 02:29 | |
| - murphnj (QUIT: Ping timeout: 260 seconds) (~murph@user/murphnj) | 02:32 | |
| - chrcav (QUIT: Ping timeout: 248 seconds) (~chrcav@user/chrcav) | 02:44 | |
| + chrcav (~chrcav@user/chrcav) | 02:51 | |
| - mjw (QUIT: Ping timeout: 240 seconds) (~mjw@gnu.wildebeest.org) | 02:58 | |
| - paperManu (QUIT: Ping timeout: 264 seconds) (~paperManu@198.58.139.163) | 03:00 | |
| - paperManu_ (QUIT: Ping timeout: 256 seconds) (~paperManu@198.58.139.163) | 03:04 | |
| + paperManu (~paperManu@198.58.139.163) | 03:26 | |
| - lidstah (QUIT: Ping timeout: 272 seconds) (~lidstah@gateway/tor-sasl/lidstah) | 03:28 | |
| + lidstah (~lidstah@gateway/tor-sasl/lidstah) | 03:28 | |
| - paperManu (QUIT: Ping timeout: 240 seconds) (~paperManu@198.58.139.163) | 03:37 | |
| - op_4 (QUIT: Remote host closed the connection) (~tslil@user/op-4/x-9116473) | 04:05 | |
| + op_4 (~tslil@user/op-4/x-9116473) | 04:05 | |
| + ZetaR (~user@c-76-148-139-78.hsd1.fl.comcast.net) | 04:20 | |
| - pomel0 (QUIT: Ping timeout: 240 seconds) (~pomel0@user/pomel0) | 05:09 | |
| + pomel0 (~pomel0@user/pomel0) | 05:09 | |
| - pomel0 (QUIT: Ping timeout: 246 seconds) (~pomel0@user/pomel0) | 05:28 | |
| + pomel0 (~pomel0@user/pomel0) | 05:28 | |
| - pomel0 (QUIT: Ping timeout: 260 seconds) (~pomel0@user/pomel0) | 05:52 | |
| - aloo_shu (QUIT: Ping timeout: 244 seconds) (~aloo_shu@85.51.16.22) | 06:11 | |
| - lidstah (QUIT: Remote host closed the connection) (~lidstah@gateway/tor-sasl/lidstah) | 06:16 | |
| + lidstah (~lidstah@gateway/tor-sasl/lidstah) | 06:16 | |
| + pomel0 (~pomel0@user/pomel0) | 06:43 | |
| - deuill (QUIT: Quit: Gateway shutdown) (~deuill@mail.deuill.org) | 07:03 | |
| - kensanata (QUIT: Ping timeout: 256 seconds) (~alex@user/kensanata) | 08:37 | |
| + kensanata (~alex@user/kensanata) | 08:38 | |
| arjan | reform2_lpc kernel module is now available in guix https://codeberg.org/guix/guix/pulls/3095 | 09:54 |
| josch | arjan: nice! Tell me if there is anything we can do better in reform-tools to make your packaging work easier | 09:58 |
| + sts-q (~sts-q@89.58.73.149) | 09:58 | |
| arjan | thanks! I might try packaging the rest of reform-tools later, will keep that it mind | 09:59 |
| + gustav25 (~gustav@c-78-82-53-102.bbcust.telenor.se) | 10:02 | |
| - plomtest (QUIT: Remote host closed the connection) (~plom@user/plomtest) | 10:14 | |
| + plomtest (~plom@user/plomtest) | 10:14 | |
| + MartiniMoe (~moe@2001-4dd4-89cb-0-cf33-d0b9-e272-a563.ipv6dyn.netcologne.de) | 10:15 | |
| - schalken (QUIT: Ping timeout: 256 seconds) (~schalken@117-118-178-69.gci.net) | 10:26 | |
| + schalken (~schalken@117-118-178-69.gci.net) | 10:28 | |
| + aelius (~aelius@user/aelius) | 10:34 | |
| josch | arjan: the license is wrong -- you list it as gpl3+ but the lpc module is distributed under the terms of the gpl2 | 10:50 |
| arjan | oops good catch | 11:04 |
| f_ | \o/ | 11:05 |
| + rwa_ (~rene@2001:9e8:3394:5100:1b3a:6:8a18:abd5) | 11:06 | |
| - MartiniMoe (QUIT: Ping timeout: 246 seconds) (~moe@2001-4dd4-89cb-0-cf33-d0b9-e272-a563.ipv6dyn.netcologne.de) | 11:20 | |
| grimmware | minute: sick, just installed it! Hilariously it’s causing niri to not render something but my track record with that is just to build main and whatever problem I’ve discovered has already been fixed so fingers crossed. | 11:43 |
| - rwa_ (QUIT: Remote host closed the connection) (~rene@2001:9e8:3394:5100:1b3a:6:8a18:abd5) | 11:43 | |
| grimmware | minute: http://paste.debian.net/1401545/ that's new, that'll be why my screen is no longer starting up in nir | 12:00 |
| grimmware | *niri | 12:00 |
| grimmware | I'll come back and have a closer look at this later | 12:00 |
| - lidstah (QUIT: Ping timeout: 272 seconds) (~lidstah@gateway/tor-sasl/lidstah) | 12:06 | |
| + lidstah (~lidstah@gateway/tor-sasl/lidstah) | 12:08 | |
| grimmware | yeah looks like this has broken something :/ | 12:18 |
| minute | grimmware: that must be unrelated | 12:19 |
| minute | grimmware: i.e. other new kernel stuff | 12:19 |
| grimmware | yeah, makes sense | 12:19 |
| minute | grimmware: at least i have a hard time imagining how this patch would affect drm on that level hmm... | 12:19 |
| grimmware | is that branch ahead of main? | 12:19 |
| minute | or did i change some return value? | 12:19 |
| minute | grimmware: no it's rebased on main | 12:20 |
| grimmware | yeah I booted off of a branch rebased off of main a few times yesterday and didn't have this issue | 12:20 |
| minute | anyway i can do a quick test with sway, niri uses wlroots right? | 12:20 |
| grimmware | sway works fine | 12:20 |
| minute | huh | 12:20 |
| minute | did you rebuild niri in parallel or sth? | 12:21 |
| josch | grimmware: are you on debian and are maybe affected by the missing firmware issue? | 12:21 |
| grimmware | oh? | 12:21 |
| grimmware | minute: I just rebuilt niri to see if it eliminated the bug which it didn't | 12:21 |
| minute | grimmware: do you still have gpu accel? in sway, what do you see in glxinfo? | 12:23 |
| minute | maybe josch has the right hunch | 12:23 |
| grimmware | gimme a sec to switch back over | 12:24 |
| minute | josch: did the mali gpu firmware get mangled in the firmware misc package, was that the problem? | 12:24 |
| grimmware | minute: what am I looking for in glxinfo? | 12:25 |
| minute | grimmware: llvmpipe vs panfrost | 12:26 |
| minute | grimmware: i.e. the opengl / gpu driver | 12:26 |
| minute | llvmpipe = software rendering | 12:26 |
| grimmware | OpenGL renderer string: Mali-G610 (Panfrost) | 12:26 |
| grimmware | that what you're looking for? | 12:26 |
| josch | minute: the problem was that the tooling which de-duplicates the firmware files created the symlink but failed to install the file that the symlink is pointing to | 12:27 |
| minute | ok hm looks like gpu ist still ok grimmware | 12:27 |
| minute | josch: wow ok | 12:27 |
| josch | (the MNT repo has the patched package in it) | 12:27 |
| grimmware | hmm, I disabled the mnt repos whilst I was fiddling with reprepo so I think I'm going to set that back to normal and do an apt upgrade to see if I'm just missing something that should have been kept in step | 12:29 |
| josch | grimmware: what is your version of firmware-misc-nonfree? | 12:29 |
| grimmware | I should be safe from the current gdm situation cos I'm still using a terminal greeter | 12:29 |
| josch | if you don't have version 20250917-1 then that's not the problem | 12:29 |
| grimmware | no, in fact I'm behind that | 12:30 |
| josch | good | 12:30 |
| + MartiniMoe (~moe@2001-4dd4-89cb-0-d0e4-da8c-8150-98a9.ipv6dyn.netcologne.de) | 12:30 | |
| grimmware | welp this is still looking like it could be a me problem given we're not at any other conclusion yet so I'm gonna step away from it for a bit and do some chores and then come back when I can debug in a clean house | 12:31 |
| grimmware | if anyone has any more ideas leave them here and I'll come back round to them but I wouldn't sweat it too much until I can eliminate some other possibilities | 12:31 |
| grimmware | thanks for your help :) | 12:32 |
| - MartiniMoe (QUIT: Ping timeout: 248 seconds) (~moe@2001-4dd4-89cb-0-d0e4-da8c-8150-98a9.ipv6dyn.netcologne.de) | 12:37 | |
| minute | grimmware: i'll try to build niri and see if it works for me | 12:39 |
| grimmware | minute: ty! I think you need to set render-drm-device in the config for it to work correctly in the first place, I've just tried fiddling with that setting and no dice | 12:41 |
| minute | grimmware: ah sorry, niri actually uses smithay, not wlroots | 12:46 |
| minute | grimmware: you need to manually set the render device? | 12:46 |
| grimmware | I've had to in the past hitting this exact same problem | 12:47 |
| minute | sounds weird | 12:47 |
| minute | (it's building here) | 12:48 |
| grimmware | https://github.com/YaLTeR/niri/issues/1199 has some relevant context | 12:49 |
| minute | done building, lets see if i can start it | 12:56 |
| minute | grimmware: it starts fine on top of gnome (nested) | 12:57 |
| + MartiniMoe (~moe@2001-4dd4-89cb-0-9e46-2680-ac3e-2445.ipv6dyn.netcologne.de) | 12:57 | |
| - MartiniMoe (QUIT: Client Quit) (~moe@2001-4dd4-89cb-0-9e46-2680-ac3e-2445.ipv6dyn.netcologne.de) | 12:57 | |
| minute | ok, quitting gnome | 12:57 |
| minute | grimmware: elb: josch: ch: there's a bug in the pocket keyboard fw. ctrl+alt+f2 -> ctrl+alt+hyper+2 (and alt+hyper+2 if on tty) works _only_ if you press alt and the 2 first, then hyper at the end. i think the hyper modifies the alt or sth | 13:01 |
| amk | https://soju.dadd.ie/uploads/akraak/38e5d421-1000018898.jpg half life 2 running via box64+zink :) pretty smooth too | 13:01 |
| grimmware | hahah, I knew there was something up with terminal switching but I only ever ran into it when I was desperately trying to debug something else | 13:02 |
| grimmware | amk: sweet. | 13:02 |
| minute | amk: very nice! is zink faster than bumping the OpenGL version to 3.x? | 13:04 |
| minute | grimmware: typing this from niri running on my pocket reform without any render device config | 13:04 |
| amk | minute: the opengl renderer seemed buggy the screen was all garbled and stuff so I gave zink a try which fixed the issues | 13:05 |
| minute | amk: wow ok | 13:05 |
| + chomwitt (~chomwitt@2a02:85f:9a5f:900:1ac0:4dff:fedb:a3f1) | 13:05 | |
| grimmware | minute: legit, thank you so much for looking. I'll finish this apt upgrade and then I'll give it all another go, I'll also try pulling the default config but I think that pretty firmly makes it a me problem which is, ultimately, optimal | 13:06 |
| grimmware | minute: also be sure to have a good play around with niri, it's lovely | 13:06 |
| minute | grimmware: yeah it's very interesting indeed | 13:08 |
| minute | grimmware: is there something like alt-tab app switcher, i.e. cycling through apps/windows/overview? | 13:08 |
| minute | niri doesn't consider by default the rotation of the panel btw | 13:09 |
| minute | i had to set transform "90" for DSI-1 | 13:09 |
| minute | oh, there's a hot corner for some kind of overview | 13:10 |
| grimmware | Yeah I think super+o as well | 13:10 |
| minute | nice, niri is like sway but with really nice mouse support and less jank (some smooth animations) | 13:10 |
| minute | grimmware: do you use a bar? | 13:11 |
| grimmware | It’s made me ditch local tmux | 13:11 |
| grimmware | Just use alacrity with vim copy mode keybindings to emulate tmux | 13:11 |
| minute | oh nice | 13:12 |
| minute | `niri validate` is also really nice | 13:12 |
| grimmware | I like that you can always launch a new terminal and it’s not hilariously small | 13:12 |
| minute | haha right!! | 13:16 |
| minute | the workspace can just grow | 13:16 |
| minute | that's actually the most annoying thing about classic tiling, you're so right | 13:16 |
| minute | or some apps that are just squeezed in very unnatural ways because they expect floating | 13:16 |
| minute | grimmware: btw i'd just recommend to remove the render-device explicit config. it just works out of the box for me | 13:24 |
| grimmware | Yeah I’ve just tried that but no dice! The dumb thing as well is if I quit nori my screen stays off :/ | 13:25 |
| grimmware | Which would be less of a problem if i wasn’t trying to clean the house at the same time | 13:25 |
| grimmware | I mean this should be self evident from my presence here but I am doing a *terrible* job of cleaning the house | 13:28 |
| minute | grimmware: hah. i also should do some chores. btw i built latest niri from today's git | 13:29 |
| minute | grimmware: and started directly from tty where i logged in as my user | 13:29 |
| minute | (i.e. no greeter) | 13:29 |
| minute | i'm trying to find out how to start niri from gdm... | 13:30 |
| minute | btw cursor is more laggy for me on 120hz screen than on gnome, interesting | 13:32 |
| grimmware | hmm, I launched from greeter... perhaps I should try from a tty. | 13:32 |
| grimmware | I'm actually using their systemd setup which could be a factor | 13:33 |
| grimmware | I also built from main today' | 13:34 |
| minute | josch: about steam, i have `export PAN_MESA_DEBUG=gl3` in my /etc/profile and forgot about that! i get opengl 3.3 by default | 13:35 |
| minute | hmm maybe that's also why niri works for me? :D | 13:35 |
| minute | grimmware: ah, systemd setup sounds like it could be the issue, maybe unexpected permissions etc | 13:36 |
| minute | grimmware: try just starting the niri executable from tty | 13:36 |
| + paperManu (~paperManu@198.58.139.163) | 13:38 | |
| minute | ah i should have used niri --session or niri-session | 13:44 |
| josch | minute: PAN_MESA_DEBUG=gl3 was the missing piece! The game runs now o0 | 13:46 |
| minute | josch: neat. i should add that to my fedi post | 13:46 |
| + wakest (m-a7d6fe@67.219.99.225) | 13:46 | |
| josch | awesome, great you remembered that :) | 13:50 |
| josch | but it's incredible how "out of the box" this is working now o0 | 13:51 |
| minute | josch: yesss | 13:51 |
| minute | > /home/minute/src/niri/target/debian/niri_25.8.0-1_arm64.deb | 13:51 |
| grimmware | the PAN_MESA_DEBUG did not save me | 13:52 |
| minute | grimmware: yeah thought so. did you try launching from tty? | 13:52 |
| grimmware | the annoying thing is it essentially turns the screen off and I then can't turn it back on again | 13:52 |
| minute | grimmware: do you have an external screen perhaps? | 13:53 |
| grimmware | minute: yeah, I'd need to check if there's anything that's a little more intelligent going on with the session service, I might have to temporarily disable it | 13:53 |
| grimmware | cos it could be that the binary is checking for a socket or some shit | 13:53 |
| grimmware | minute: I do, I'm supopsed to be cleaning | 13:53 |
| grimmware | I just can't leave well enough alone | 13:54 |
| minute | ok | 13:54 |
| grimmware | too excited about having the screen flicker fixed | 13:54 |
| minute | it's weird though that the display doesn't turn on again | 13:54 |
| minute | this could be the behavior that prompted me to introduce a fix that introduced the flicker... because on ch's pocket the screen wouldn't come up after reboot | 13:55 |
| minute | grimmware: ok i built the deb (using cargo-deb) and installed it and now i was able to launch niri from gdm just like that. it works | 14:01 |
| minute | and now it's much faster for some reason... | 14:03 |
| grimmware | minute: yeah I think it mighht be that issue again because I just disabled and then attempted to re-enable my display under sway and it didn't come back on | 14:03 |
| minute | grimmware: damn | 14:03 |
| minute | grimmware: ok thanks for testing that, it's what i feared and why i didn't directly merge it | 14:03 |
| minute | why is my display different than yours and ch's in that regard... | 14:04 |
| minute | anyway, i won't give up on that issue | 14:04 |
| grimmware | minute: do you have a workflow for just applying the patches to the kernel tree so that you can diff them without having to do a diff of a diff? | 14:04 |
| minute | grimmware: yeah | 14:05 |
| minute | grimmware: well, the 2 patches are merged now in my MR | 14:06 |
| minute | grimmware: so, no diff of diff needed | 14:06 |
| minute | grimmware: btw because i was lazy i just created that patch in emacs. emacs can directly edit a diff and adjust it if you're careful enough | 14:07 |
| minute | grimmware: so i just edited the diff and pushed that and i was lucky it built | 14:07 |
| minute | grimmware: but normally i would git clone the same linux version that debian uses, but from kernel.org, then apply any needed patches from reform-debian-packages (that can be easily done with bash for loop), then commit that as a baseline. | 14:08 |
| minute | grimmware: from there, any changes will show up in local git as diffs to what is in reform-debian-packages, so i can commit new things and "git show xyz > foo.patch" for example | 14:08 |
| grimmware | minute: I'm really just after being able to directly compare the different patched versions of the file directly myself between commits because dealing with the diffs of diffs is a bit of a mindfuck | 14:09 |
| minute | grimmware: i wouldn't bother | 14:09 |
| grimmware | like I've hacked this together before by just adding an exit to one of the build scripts | 14:09 |
| minute | btw the following works for me: niri msg output DSI-1 off; and then: niri msg output DSI-1 on | 14:13 |
| minute | i understand that doesn't work for you so there must be some subtle difference for those display panels | 14:13 |
| minute | grimmware: btw so far i thought you have display v1. maybe you have v2? because you were talking about the "fade out" effect that happens only on v2 i think | 14:14 |
| minute | grimmware: what's the output of `cat /sys/devices/platform/fde30000.dsi/fde30000.dsi.0/mnt_pocket_reform_panel_version` ? | 14:15 |
| minute | (for me it's 1) | 14:15 |
| grimmware | 1 | 14:16 |
| grimmware | I've had this thing since before there *was* a v2 :) | 14:16 |
| minute | grimmware: ok, gotcha! | 14:17 |
| grimmware | as far as I'm aware we have the same configuration, v1 rk3588 16GB | 14:17 |
| minute | yeah. but i guess displays are from different batches | 14:19 |
| minute | anyway, i have to do more display hacking soon to help bringup the edk2 display stuff as well | 14:19 |
| minute | i think i'll rework the driver so it's more easy to read/follow/debug | 14:19 |
| minute | and that more of the sequencing can be adjusted at runtime so we can test this out faster with scripts | 14:20 |
| + mjw (~mjw@gnu.wildebeest.org) | 14:20 | |
| minute | rather than rebuilds + reboots | 14:20 |
| grimmware | yeah that sounds good | 14:21 |
| grimmware | minute: do you still get ghosting? | 14:23 |
| minute | grimmware: don't know, maybe? i didn't pay attention recently | 14:26 |
| minute | my panel is quite crusty so i wanna replace it with v2 soon | 14:26 |
| minute | after i fix those issues for v1 users... | 14:26 |
| grimmware | like, if my panel turns out to be kind of unique I'm not against also replacing but I would like to know that I'm not just walking away from a problem that I could help solve if it's actually useful to other people | 14:28 |
| minute | grimmware: i think ch has the same kind of panel | 14:28 |
| minute | i'll check on monday or tuesday if i have a panel in old stock that behaves like this | 14:29 |
| minute | grimmware: btw yes i still get ghosting, esp on gray (just saw in gdm) | 14:30 |
| grimmware | yeah same | 14:31 |
| minute | we should probably offer disp v2 in the shop soon :3 | 14:31 |
| - wakest (QUIT: Remote host closed the connection) (m-a7d6fe@67.219.99.225) | 14:39 | |
| + aloo_shu (~aloo_shu@85.51.16.22) | 14:54 | |
| grimmware | Okay back at an external monitor and niri will actually fail to render anything even on the external monitor unless I have that render-drm-device set | 14:56 |
| grimmware | ah great qutebrowser no longer actually renders anything what the hell | 14:57 |
| minute | grimmware: started from tty? | 15:00 |
| grimmware | nah | 15:03 |
| minute | grimmware: then we can't really compare | 15:06 |
| grimmware | Yeah sorry just debugging what I can in a single boot first because I have to turn the whole machine off and on again for every test of the inbuilt display | 15:08 |
| minute | grimmware: ok gotcha. | 15:09 |
| minute | grimmware: so the oscillation/flicker comes from the display not being fully turned off but the dsi stream from the computer stopping. | 15:10 |
| - chomwitt (QUIT: Ping timeout: 246 seconds) (~chomwitt@2a02:85f:9a5f:900:1ac0:4dff:fedb:a3f1) | 15:20 | |
| + paperManu_ (~paperManu@198.58.139.163) | 15:29 | |
| - paperManu (QUIT: Ping timeout: 246 seconds) (~paperManu@198.58.139.163) | 15:32 | |
| grimmware | minute: I've just thrown together a build that's a combination of your patch and my patches (much easier with the patches merged together), there's a few things that are happening in init that are happening in the wrong order according to the datasheet and a missing soft_reset. I feel like this would be consistent with the behaviour that we're seeing and it would make sense if the batch that | 15:47 |
| grimmware | you're on was just a bit more forgiving. I mean it's still kind of a shot in the dark but I figure if it lands us back exactly where we are then that's fine too. | 15:47 |
| grimmware | just because we'd eliminate that difference as a source of where this could come from | 15:47 |
| + wakest (m-a7d6fe@67.219.99.225) | 15:53 | |
| elb | minute: that (re: hyper order) actually makes sense, I suppose the hyper should modify _every_ key that is currently pressed and right now it may not | 15:57 |
| elb | ctrl+alt+n then hold and hit hyper works for me, confirmed | 15:58 |
| minute | grimmware: sure, please show! | 16:07 |
| minute | elb: great. yeah it looks like hyper changes alt to altgr | 16:08 |
| elb | oh, so it probably _is_ recognizing all the keys, and it's just not the right keystroke | 16:11 |
| elb | I've got the firmware open now | 16:11 |
| grimmware | minute: https://source.mnt.re/grimmware/reform-debian-packages/-/pipelines/4907 is the pipeline, it's currently building | 16:11 |
| minute | elb: yeah in the matrix defintion, in the hyper layer, altgr / rightalt or however it's called should be changed to leftalt on that key | 16:14 |
| grimmware | am I right in thinking that I don't actually need qcacld on rk3588? | 16:18 |
| elb | I actually wondered about those keys when I created my layout, in general the swapping of left-for-right modifiers | 16:19 |
| grimmware | right here goes nothing | 16:24 |
| grimmware | lol no it hasn't come on at all this time | 16:27 |
| lidstah | Hi all, and just ordered the reform (spent half of the night reading forums and docs, got the printed manual too) :) | 16:31 |
| lidstah | s/got/took/ | 16:32 |
| grimmware | hi lidstah! | 16:33 |
| grimmware | I'm digging around in pocket reform display code | 16:33 |
| lidstah | Hi grimmware :) | 16:33 |
| lidstah | browsing the site and shop also made me want to recap my old A1200 (although it doesn't seem there's a ZZ9000 A1200 version, I'm wondering if the Zorro-II version would work in a towered A1200…) | 16:36 |
| lidstah | (and reading further, nope) | 16:38 |
| grimmware | aight that's another build kicked off with some more of the recommended delays added, imma go shower and then walk the dog and come back and test the new build | 16:44 |
| grimmware | so glad I got a suitable setup for testing this in CI <3 | 16:45 |
| - xktr (QUIT: Ping timeout: 256 seconds) (~xktr@user/xktr) | 16:56 | |
| + xktr (~xktr@user/xktr) | 16:57 | |
| josch | grimmware: no qcacld needed on rk3588 -- can you confirm that reform-check is telling you this information as well? | 16:58 |
| + chomwitt (~chomwitt@2a02:85f:9a5f:900:1ac0:4dff:fedb:a3f1) | 17:16 | |
| grimmware | reform-check is telling me... a lot | 17:59 |
| grimmware | I'll have a look at that tomorrow | 17:59 |
| grimmware | my adjusted timings didn't help, I think I'm going to have to come back to this during the week and add a bunch of debug so I can actually prove to myself that I know any of what's going on here | 18:00 |
| - Ar|stote|is (QUIT: Read error: Connection reset by peer) (~linx@149.210.6.130) | 18:03 | |
| + Ar|stote|is (~linx@149.210.6.130) | 18:09 | |
| - pomel0 (QUIT: Ping timeout: 240 seconds) (~pomel0@user/pomel0) | 18:18 | |
| minute | lidstah: cool, welcome! | 18:20 |
| - chomwitt (QUIT: Ping timeout: 244 seconds) (~chomwitt@2a02:85f:9a5f:900:1ac0:4dff:fedb:a3f1) | 18:20 | |
| minute | lidstah: and yeah, zz9k should work with towered a1200, i tested it once (i don't have that tower anymore though). it also works with zorro adapter on a500 | 18:20 |
| - timide (QUIT: Remote host closed the connection) (~timide@user/timide) | 18:24 | |
| + rwa_ (~rene@2001:9e8:3394:5100:1b3a:6:8a18:abd5) | 18:47 | |
| + chomwitt (~chomwitt@2a02:85f:9a5f:900:1ac0:4dff:fedb:a3f1) | 18:48 | |
| + timide (~timide@user/timide) | 18:57 | |
| lidstah | minute: ah thanks for the answer about the zz9000 and the A1200 | 19:20 |
| lidstah | might be a nice project for next summer (although I should recap it first and get a new power supply for it, it's been stored properly but I haven't powered it on since ~15 years) | 19:29 |
| + bkeys1 (~Thunderbi@98.19.131.193) | 19:47 | |
| - bkeys (QUIT: Ping timeout: 244 seconds) (~Thunderbi@98.19.131.193) | 19:49 | |
| * bkeys1 -> bkeys | 19:49 | |
| - mhoye (QUIT: Ping timeout: 244 seconds) (~mhoye@li319-32.members.linode.com) | 19:49 | |
| + mhoye (~mhoye@li319-32.members.linode.com) | 19:51 | |
| - rwa_ (QUIT: Quit: Leaving) (~rene@2001:9e8:3394:5100:1b3a:6:8a18:abd5) | 20:20 | |
| - chomwitt (QUIT: Ping timeout: 248 seconds) (~chomwitt@2a02:85f:9a5f:900:1ac0:4dff:fedb:a3f1) | 20:45 | |
| - kensanata (QUIT: Ping timeout: 248 seconds) (~alex@user/kensanata) | 20:47 | |
| + kensanata (~alex@user/kensanata) | 20:48 | |
| + chomwitt (~chomwitt@2a02:85f:9a5f:900:1ac0:4dff:fedb:a3f1) | 20:56 | |
| - chomwitt (QUIT: Ping timeout: 250 seconds) (~chomwitt@2a02:85f:9a5f:900:1ac0:4dff:fedb:a3f1) | 21:02 | |
| + paperManu (~paperManu@198.58.139.163) | 21:17 | |
| + wielaard (~mjw@gnu.wildebeest.org) | 21:31 | |
| - mjw (QUIT: Ping timeout: 256 seconds) (~mjw@gnu.wildebeest.org) | 21:33 | |
| + AnimaInvicta (~AnimaInvi@88-120-179-216.subs.proxad.net) | 21:36 | |
| + chomwitt (~chomwitt@2a02:85f:9a5f:900:1ac0:4dff:fedb:a3f1) | 21:39 | |
| * wielaard -> mjw | 21:41 | |
| - jacqueline (QUIT: Ping timeout: 264 seconds) (~jacquelin@user/jacqueline) | 21:49 | |
| - switchy (QUIT: Ping timeout: 264 seconds) (~switchy@mechboards/switchy) | 21:51 | |
| - se6astian (QUIT: Ping timeout: 264 seconds) (~quassel@46.101.193.235) | 21:52 | |
| + se6astian (~quassel@46.101.193.235) | 21:53 | |
| - frickler (QUIT: Ping timeout: 264 seconds) (~jens@user/frickler) | 21:55 | |
| - sknebel (QUIT: Ping timeout: 264 seconds) (~quassel@v22016013254630973.happysrv.de) | 21:55 | |
| + sknebel (~quassel@v22016013254630973.happysrv.de) | 21:56 | |
| + frickler (~jens@user/frickler) | 21:57 | |
| + switchy (~switchy@mechboards/switchy) | 22:03 | |
| + jacqueline (~jacquelin@user/jacqueline) | 22:04 | |
| - gustav25 (QUIT: Quit: Quit) (~gustav@c-78-82-53-102.bbcust.telenor.se) | 22:15 | |
| - digitalrane (QUIT: Ping timeout: 250 seconds) (~digitalra@user/digitalrane) | 22:31 | |
| - chomwitt (QUIT: Ping timeout: 244 seconds) (~chomwitt@2a02:85f:9a5f:900:1ac0:4dff:fedb:a3f1) | 23:12 | |
| - cobra (QUIT: Ping timeout: 265 seconds) (~cobra@user/Cobra) | 23:22 | |
| - AnimaInvicta (PART: !!unknown attribute: msg!!) (~AnimaInvi@88-120-179-216.subs.proxad.net) | 23:28 | |
| + deuill (~deuill@mail.deuill.org) | 23:29 | |
| + cobra (~cobra@user/Cobra) | 23:37 | |
| + pomel0 (~pomel0@user/pomel0) | 23:45 | |
| - cobra (QUIT: Quit: ZNC 1.8.2 - https://znc.in) (~cobra@user/Cobra) | 23:57 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
