| - chomwitt1 (QUIT: Ping timeout: 265 seconds) (~chomwitt@2a02:587:7a13:5400:1ac0:4dff:fedb:a3f1) | 00:27 | |
| + LainExperiments (~LainExper@user/LainExperiments) | 00:31 | |
| minute | josch: cool mod with the potentiometer. unsolicited advice: if you use magnet wire instead of stranded copper cable, you have less risk of ripping off smd pads, and also loose strands touching and shorting other stuff. | 01:00 |
|---|---|---|
| - mtm (QUIT: Ping timeout: 248 seconds) (~textual@47.202.75.129) | 01:02 | |
| - jackhill (QUIT: Ping timeout: 272 seconds) (~jackhill@kalessin.dragonsnail.net) | 01:04 | |
| + mtm (~textual@47.202.75.129) | 01:05 | |
| - vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:50) | 01:39 | |
| - LainExperiments (QUIT: Ping timeout: 240 seconds) (~LainExper@user/LainExperiments) | 01:40 | |
| + LainExperiments (~LainExper@user/LainExperiments) | 01:52 | |
| - jn (QUIT: Ping timeout: 252 seconds) (~quassel@user/jn/x-3390946) | 02:09 | |
| + jn (~quassel@user/jn/x-3390946) | 02:10 | |
| - mjw (QUIT: Ping timeout: 252 seconds) (~mjw@gnu.wildebeest.org) | 02:21 | |
| - LainExperiments (QUIT: Quit: Client closed) (~LainExper@user/LainExperiments) | 02:51 | |
| - nsc (QUIT: Ping timeout: 252 seconds) (~nicolas@190-98-142-46.pool.kielnet.net) | 03:23 | |
| + nsc (~nicolas@164-96-142-46.pool.kielnet.net) | 03:24 | |
| - paperManu (QUIT: Ping timeout: 245 seconds) (~paperManu@107.159.71.33) | 03:27 | |
| + LainExperiments (~LainExper@user/LainExperiments) | 03:56 | |
| - LainExperiments (QUIT: Ping timeout: 240 seconds) (~LainExper@user/LainExperiments) | 04:05 | |
| + jackhill (~jackhill@kalessin.dragonsnail.net) | 05:18 | |
| - Ar|stote|is (QUIT: Quit: https://quassel-irc.org - Chat comfortably. Anywhere.) (~linx@149.210.32.133) | 05:33 | |
| + Ar|stote|is (~linx@149.210.32.133) | 05:33 | |
| - n_to (QUIT: Quit: quitidiquit) (~n_to@2a03:4000:6:3662:24b1:57ff:fec6:76c1) | 06:21 | |
| + n_to (~n_to@2a03:4000:6:3662:24b1:57ff:fec6:76c1) | 06:21 | |
| - Ar|stote|is (QUIT: Ping timeout: 245 seconds) (~linx@149.210.32.133) | 06:27 | |
| + Ar|stote|is (~linx@149.210.32.196) | 06:31 | |
| - Ar|stote|is (QUIT: Ping timeout: 252 seconds) (~linx@149.210.32.196) | 06:50 | |
| + chomwitt1 (~chomwitt@2a02:587:7a13:5400:1ac0:4dff:fedb:a3f1) | 06:56 | |
| josch | minute: duh i was not thinking of using magnet wire but that's a great idea, thank you! | 07:03 |
| josch | it needed a fix in our infrastructure (piuparts didn't like packages with the Protected:yes field) but reform-tools finally transitioned to testing and we're a step closer to having it in a stable release \o/ | 07:06 |
| + Ar|stote|is (~linx@149.210.32.247) | 07:39 | |
| + mlarkin_ (~mlarkin@47.158.172.62) | 07:43 | |
| - mlarkin (QUIT: Ping timeout: 244 seconds) (~mlarkin@47.158.172.62) | 07:44 | |
| - chomwitt1 (QUIT: Quit: WeeChat 3.8) (~chomwitt@2a02:587:7a13:5400:1ac0:4dff:fedb:a3f1) | 08:00 | |
| - n_to (QUIT: Quit: quitidiquit) (~n_to@2a03:4000:6:3662:24b1:57ff:fec6:76c1) | 08:56 | |
| + n_to (~n_to@2a03:4000:6:3662:24b1:57ff:fec6:76c1) | 08:57 | |
| - n_to (QUIT: Quit: quitidiquit) (~n_to@2a03:4000:6:3662:24b1:57ff:fec6:76c1) | 09:18 | |
| + n_to (~n_to@2a03:4000:6:3662:24b1:57ff:fec6:76c1) | 09:19 | |
| - n_to (QUIT: Client Quit) (~n_to@2a03:4000:6:3662:24b1:57ff:fec6:76c1) | 09:19 | |
| + n_to (~n_to@2a03:4000:6:3662:24b1:57ff:fec6:76c1) | 09:20 | |
| frickler | josch: two comments from testing your new system-image so far: a) the greeter font is still the micro version (I patched that locally for me, not sure if there was an upstream patch yet) | 09:34 |
| frickler | b) reform-check errors out when "apt update" hasn't been run, why doesn't it just run it itself? | 09:34 |
| josch | frickler: thank you for your feedback! You are on a MNT Pocket Reform with RCORE RK3588 Module, right? | 09:35 |
| frickler | josch: yes. and with display v2 afaict | 09:35 |
| josch | frickler: about the console font: there was an attempt to fix this via a postinst snippet which changes FONTFACE and FONTSIZE in /etc/default/console-setup | 09:37 |
| josch | in reform-system-image we have a sed snippet that does this: | 09:38 |
| josch | sed -i -e "s/^FONTFACE=.*/FONTFACE=/" -e "s/^FONTSIZE=.*/FONTSIZE=/" /etc/default/console-setup | 09:38 |
| josch | but somehow this does not seem to have been effective? | 09:38 |
| frickler | ah, right, now I remember, that somehow didn't work out, I explicitly did put the font and size I want into my local setup | 09:38 |
| josch | frickler: into uboot? | 09:38 |
| frickler | no, just into /etc/default/console-setup on my nvme | 09:39 |
| josch | frickler: okay but this is from a fresh system image | 09:40 |
| josch | and if the font is still tiny on that one, then something is amiss | 09:41 |
| josch | frickler: can you check what setting FONTFACE and FONTSIZE are in your /etc/default/console-setup on the system image with the tiny font? | 09:41 |
| frickler | hmm, let me double check, but I did flash a new system image to the sdcard | 09:41 |
| josch | and your /proc/cmdline should have fbcon=font:TER16x32 in it | 09:42 |
| josch | if it does not, reform-check should've complained | 09:42 |
| frickler | yes, I have both the cmdline entry and FONTFACE/SIZE empty | 09:42 |
| josch | huh | 09:43 |
| frickler | next thing: after running apt update reform-check complains about a lot of files differing, like: "??5?????? /usr/bin/reform-chat" and most other tools | 09:43 |
| frickler | also some tools still are in /usr/sbin, didn't you say that those moved? | 09:43 |
| josch | ch, hramrach: we have the fontsize problem again and even though u-boot sets fbcon=font:TER16x32 and even though FONTFACE and FONTSIZE are empty it's still present. See above. | 09:43 |
| josch | frickler: the differences dpkg shows are because you installed reform-tools from git over the version that was installed from the package, so files naturally differ and dpkg doesn't like that | 09:44 |
| frickler | hmm, actually apt policy shows reform-tools 1.68-1+reform from the mntre.com repo, maybe the image build didn't pick up your git version? | 09:45 |
| josch | frickler: it did but how should apt know that you manually installed the version from git? | 09:46 |
| frickler | shouldn't that version have a different version string? | 09:46 |
| josch | frickler: the system you are using first installed reform-tools 1.68-1+reform and *then* did a "sudo make install" for the version from git. So as far as apt is concerned, you have 1.68-1+reform installed. | 09:47 |
| josch | frickler: the only way to tell apt "no i have another version" is to first build a debian package from the version in git | 09:47 |
| frickler | josch: ah, kind of a hacky build process, then, fine ;) | 09:49 |
| josch | frickler: do you have an idea of how to improve this? | 09:50 |
| josch | frickler: also, above you said that some tools are still in /usr/sbin -- which ones? | 09:51 |
| frickler | josch: well as you said, build a dpkg with a custom version that likely should include the git sha and install that, but I know that's easier said than done | 09:52 |
| josch | the problem with that approach is, that we then duplicate the packaging again | 09:52 |
| frickler | josch: the only ones in /usr/bin are chat,check,compstat,handbook,help,pavucontrol | 09:53 |
| josch | we cannot just copypaste the packaging scripts from Debian because new reform-tools versions will often (as it does now) require changes to the packaging | 09:53 |
| josch | frickler: no, i mean which ones are in /usr/sbin. You said some are still in /usr/sbin | 09:54 |
| frickler | josch: yes, all others are in sbin, boot-config,config,display-config,emmc-bootstrap,flash-rescue,flash-uboot,hw-setup,mcu-tool,migrate,setup-encrypted-nvme,standby | 09:55 |
| josch | oh wow, that should not be happening | 09:56 |
| josch | wait, you did not list reform-setup-encrypted-disk -- is that by accident or is it actually there? | 09:57 |
| josch | *not there | 09:57 |
| frickler | no, only reform-setup-encrypted-nvme | 09:58 |
| josch | aha! | 09:58 |
| josch | frickler: do you recall from where you have the link to the artifacts you flashed? | 09:58 |
| frickler | from the link you posted the other day, https://source.mnt.re/reform/reform-system-image/-/jobs/7817 | 09:59 |
| josch | whoops | 09:59 |
| josch | it should've been job 7823 | 09:59 |
| josch | let me find where the old link is from | 09:59 |
| josch | frickler: i can only find https://source.mnt.re/reform/reform-system-image/-/jobs/7823 | 10:00 |
| frickler | ah, ok, let me retest then | 10:00 |
| josch | see https://community.mnt.re/t/request-for-help-reform-tools-1-69-release/3154 and https://community.mnt.re/t/migration-from-i-mx8mplus-to-a311d/2317/42 | 10:00 |
| frickler | hmm, https://source.mnt.re/reform/reform-system-image/-/jobs/7817/artifacts/file/pocket-reform-system-rk3588.img.gz still works it seems | 10:01 |
| josch | frickler: what do you mean by "still works"? | 10:01 |
| frickler | josch: I could still download it. https://mntre.com/reform-irc-logs/2025-03-03.log.html#t15:22:38 is where I got this from | 10:02 |
| frickler | but anyway, let me go for the newer one now | 10:02 |
| josch | frickler: yes, gitlab will only delete the artifacts on march 6 14:25 UTC | 10:03 |
| josch | frickler: sorry, i forgot to post the updated link to IRC as well | 10:04 |
| - jacobk (QUIT: Ping timeout: 260 seconds) (~quassel@47-186-65-73.dlls.tx.frontiernet.net) | 10:04 | |
| josch | frickler: depending on how much time you have, you could also wait until i fixed the "apt update" problem and the fontsize problem and only test then | 10:05 |
| frickler | josch: btw. I use FONTFACE="DejaVu" and FONTSIZE="24x43", which may be a bit extreme for general use | 10:05 |
| + jacobk (~quassel@47-186-65-73.dlls.tx.frontiernet.net) | 10:05 | |
| frickler | josch: ok, then I'll do some dayworking first and wait for a new build | 10:05 |
| josch | frickler: the problem with picking a FONTSIZE in /etc/default/console-setup for the system image is, that those are also run on machines that are *not* the pocket reform and where the default font size is thus fine | 10:06 |
| josch | this is also why i have a hard time testing these things :D | 10:06 |
| frickler | someone should sponsor you a pocket I guess :-) | 10:07 |
| josch | frickler: too late, wickedshell already did exactly that XD | 10:08 |
| josch | (just waiting for MNT to ship it) | 10:09 |
| josch | <= afk | 10:10 |
| frickler | gotta tell minute to do some proper prioritization :D | 10:27 |
| minute | well, the issue is we're waiting for rcore pcb restock. | 10:28 |
| minute | we finally got pcba confirmation photos yesterday so then it's normally a matter of days... | 10:29 |
| + chomwitt (~chomwitt@2a02:587:7a13:5400:1ac0:4dff:fedb:a3f1) | 10:30 | |
| - ericsfraga (QUIT: Remote host closed the connection) (~user@2a00:23cc:b452:2900::99e) | 10:38 | |
| frickler | ah, ok. are these the same boards for pocket and classic btw? | 10:42 |
| + andreas-e (~Andreas@2001:861:c4:f2f0::c64) | 10:43 | |
| [tj] | minute: do you want my dead charger board for testing? my sister in law is visiting and can post from germany when she gets home which makes returning it really easy for me | 11:05 |
| minute | [tj]: when would that be approx? | 11:06 |
| minute | today we got the opportunity from mps to actually send in samples. i can wait a few days ofc | 11:06 |
| minute | frickler: yes. and they're really popular :0 | 11:06 |
| [tj] | minute: she is here until next week, I think she would be able to post on the 14th | 11:07 |
| minute | [tj]: that's a bit late i think, alternatively we could make a ups label for you | 11:08 |
| [tj] | sure I can post it today | 11:09 |
| gordon1 | minute: i noticed your poll a week ago mentioned ampere altra, did you take a look at it already? is it possible to do free schematics board with it or do you need to sell your soul and/or pull rk3588 option with adapter board? | 11:13 |
| minute | gordon1: i'm in contact with ampere, full custom oshw board seems possible but would be a longer r&d project. com-hpc with adlink module would be easier but apparently com-hpc might be incompatible with oshw (licensing stuff) | 11:17 |
| gordon1 | righty, cool | 11:17 |
| gordon1 | that's weird that com-hpc is what incompatible than the actual chip ifself | 11:18 |
| - chomwitt (QUIT: Ping timeout: 260 seconds) (~chomwitt@2a02:587:7a13:5400:1ac0:4dff:fedb:a3f1) | 11:21 | |
| josch | frickler: i'd not like other customers to wait longer for their units because of me though :) | 11:22 |
| hramrach | frickler: for me when the fon is not set console-setup does not change the font. Did that change with recent console-setup, or is something else changing it? | 11:23 |
| [tj] | minute I can post whenever I get a label, I've pulled it and packed it up. Its nice to get to recycle the mnt packing peanuts | 11:25 |
| minute | [tj]: awesome, i'm on the way to mnt hq and will make a label there | 11:26 |
| frickler | hramrach: I need to check again, but iirc it is just unreadably small right from the start. so maybe the cmdline setting isn't really working. also TER16x32 is still too small for my eyes, but that may be a matter of taste or age | 11:26 |
| minute | [tj]: actually i need your sender address and email, maybe you can send me a quick mail to lukas@mntre.com | 11:27 |
| josch | hramrach: if you find some time, maybe you can flash the image from https://source.mnt.re/reform/reform-system-image/-/jobs/7823 and check what fontsize you get. That image should have an empty FONTSIZE in /etc/default/console-setup | 11:27 |
| [tj] | minute: sent! | 11:28 |
| grimmware | ugh, great, just picked up my pocket and turned the screen back on and it's flickering | 11:29 |
| josch | grimmware: with your change?? | 11:29 |
| grimmware | josch: I also built with the more recent mesa patch that was in main | 11:30 |
| josch | oh wait, sorry i mixed up issues | 11:30 |
| grimmware | I vaguely recall that was supposed to *stop* flickering which I never had before | 11:30 |
| josch | grimmware: you updated to mesa from the MNT repos? | 11:30 |
| grimmware | oh hang on, no | 11:31 |
| josch | grimmware: what is your current mesa version? | 11:31 |
| josch | grimmware: would be interesting to know if it also stops flickering for you | 11:31 |
| grimmware | yeah that's a good point, let me check version before updating | 11:31 |
| grimmware | I need to do a full-upgrade now I'm not worried about losing old kernel versions | 11:32 |
| grimmware | https://paste.debian.net/1361444/ | 11:34 |
| grimmware | that's just a dpkg -l | grep mesa | 11:34 |
| grimmware | gonna do a full upgrade | 11:34 |
| josch | minute: for the flickering problem which we fixed by reverting a commit in mesa, we also got feedback from Diederik de Haas in Debian bug #1099032 who cannot reproduce this issue on Pine64 PineTab2 (rk3566) which also has an Arm Mali-G52 GPU and who suggests that the problem might come from an issue with our device tree which might be missing the "mali-supply" entry as recommended in | 11:35 |
| josch | Documentation/devicetree/bindings/gpu/arm,mali-bifrost.yaml. The fix might look something like https://paste.debian.net/hidden/99e8c06b/ but since I have no clue what I'm doing i have no idea which regulator to put there... | 11:35 |
| josch | grimmware: this is odd | 11:35 |
| josch | grimmware: with mesa 24.3.4-3 you should not see any flickering | 11:36 |
| josch | grimmware: the flickering issue that we recently fixed by reverting a mesa commit (found by delanara) was found in mesa 25.0.0 | 11:36 |
| josch | grimmware: there is a video in https://community.mnt.re/t/display-flickering-on-pocket-reform-related-to-mesa-25-0-0/3133/15 -- is that how your flickering looks like? | 11:37 |
| minute | grimmware: josch: afaik flickering happens only on a311d with gnome? | 11:38 |
| josch | i had flickering on classic reform a311d also with sway | 11:38 |
| minute | oh wow ok | 11:38 |
| josch | (just not as much as delanara on their pocket) | 11:38 |
| minute | right, i forgot about that | 11:38 |
| josch | (maybe twice per minute) | 11:38 |
| josch | right, i forgot that grimmware has a rk3588 pocket reform | 11:39 |
| josch | so different issue | 11:39 |
| grimmware | it's very gentle and only on grey backgrounds by the looks of things right now - I'll see how I get on after things like reboot and whether it recurs after upgrade. I'll report back | 11:42 |
| minute | grimmware: it could be an oscillation in the panel | 11:45 |
| minute | grimmware: these go away after like 30 minutes | 11:45 |
| minute | this happens if panel is driven with wrong or no mode for a while | 11:46 |
| frickler | hramrach: josch: ok, did some setfont checking, so the image is actually using the Ter32x16 font, I've just gotten so used to my larger font that it seemed still incredibly small to me. but the default font is even tinier, so the current state sure is some improvement | 11:54 |
| josch | okay i see, thank you for checking! | 11:55 |
| ch | so is there an issue or not? | 11:55 |
| josch | ch, hramrach: probably false alarm since frickler was used to having an even bigger font than what is now the default via u-boot | 11:56 |
| frickler | well it depends on whether you want to cater for old people with bad eyes by default, then you'd switch to an even larger font. but the setup does seem to be working as intended currently | 11:57 |
| ch | there's no larger font in the kernel, so excluding other points, its not possible | 11:58 |
| frickler | yes, you could only switch after booting by setting an even larger default in /etc/default/console-setup once again | 11:59 |
| ch | sure, but switching mid-boot is terrible | 12:00 |
| abortretryfail | minute: re: magnet wire. What ho you use to strip the varnish? I've found the stuff is horrible to work with after winding a bunch of transformers for radio kits. | 12:07 |
| abortretryfail | s/ho/do/ | 12:07 |
| grimmware | ugh something fucky is happening, now it's not booting | 12:26 |
| grimmware | I need to hook up serial again | 12:26 |
| hramrach | frickler: We are limited to the fonts included in the kernel for the defaults. The setup was changed so that you *can* set custom font now, and it's not reverted to 32x16 by reform-tools. | 12:35 |
| * Guest6496 -> mjw | 12:36 | |
| hramrach | Amybe reform-setup could offer font customization or something but the standard tools can be used now as well | 12:38 |
| josch | related is https://source.mnt.re/reform/mnt-reform-setup-wizard/-/issues/18 | 12:40 |
| - Ar|stote|is (QUIT: Ping timeout: 272 seconds) (~linx@149.210.32.247) | 12:43 | |
| + glu_ (~glu@user/glu) | 12:44 | |
| + chomwitt (~chomwitt@2a02:587:7a13:5400:1ac0:4dff:fedb:a3f1) | 12:46 | |
| + Ar|stote|is (~linx@149.210.14.170) | 12:47 | |
| + paperManu (~paperManu@107.159.71.33) | 12:50 | |
| - glu_ (QUIT: Ping timeout: 272 seconds) (~glu@user/glu) | 12:51 | |
| grimmware | okay I've got boot again, I've got a sneaking suspicion extlinux fucked me again | 12:58 |
| grimmware | yeah what do you know it's dropped my device tree again | 12:58 |
| grimmware | fuck sake | 12:59 |
| + gustav28 (~gustav@c-78-82-52-230.bbcust.telenor.se) | 13:02 | |
| + glu_ (~glu@user/glu) | 13:03 | |
| - mtm (QUIT: Ping timeout: 268 seconds) (~textual@47.202.75.129) | 13:03 | |
| grimmware | yeah I left this landmine for myself | 13:03 |
| - bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@38-146-94-247.echocast.zone) | 13:03 | |
| + bkeys1 (~Thunderbi@38-146-94-247.echocast.zone) | 13:03 | |
| + mtm (~textual@47.202.75.129) | 13:04 | |
| - glu_ (QUIT: Remote host closed the connection) (~glu@user/glu) | 13:06 | |
| * bkeys1 -> bkeys | 13:06 | |
| - bkeys (QUIT: Client Quit) (~Thunderbi@38-146-94-247.echocast.zone) | 13:07 | |
| + bkeys (~Thunderbi@38-146-94-247.echocast.zone) | 13:07 | |
| minute | [tj]: you should have email with label now | 13:13 |
| grimmware | josch: https://source.mnt.re/reform/reform-tools/-/merge_requests/114 I forgot to actually fix this in the repos so my change in /usr/share got clobbered on package update | 13:15 |
| - bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@38-146-94-247.echocast.zone) | 13:16 | |
| + bkeys (~Thunderbi@38-146-94-247.echocast.zone) | 13:16 | |
| - chomwitt (QUIT: Ping timeout: 265 seconds) (~chomwitt@2a02:587:7a13:5400:1ac0:4dff:fedb:a3f1) | 13:17 | |
| grimmware | minute: either update or reboot has fixed the flicker, I think your theory is probably correct so there might be a subsequent kernel patch update that needs doing to ensure that the right mode gets set. The behaviour of my patch is not *quite* the same as what the screen would do on return before (the backlight powers on blank and then the image returns) so I wouldn't be surprised if it is indeed running on the wrong mode | 13:17 |
| grimmware | for a moment or something | 13:17 |
| grimmware | at any rate I'll keep an eye on it | 13:17 |
| - bkeys (QUIT: Read error: Connection reset by peer) (~Thunderbi@38-146-94-247.echocast.zone) | 13:18 | |
| + bkeys (~Thunderbi@38-146-94-247.echocast.zone) | 13:19 | |
| - bkeys (QUIT: Client Quit) (~Thunderbi@38-146-94-247.echocast.zone) | 13:22 | |
| + bkeys1 (~Thunderbi@38-146-94-247.echocast.zone) | 13:22 | |
| - bkeys1 (QUIT: Client Quit) (~Thunderbi@38-146-94-247.echocast.zone) | 13:23 | |
| + bkeys (~Thunderbi@38-146-94-247.echocast.zone) | 13:23 | |
| + bkeys1 (~Thunderbi@38-146-94-247.echocast.zone) | 13:27 | |
| - bkeys (QUIT: Read error: Connection reset by peer) (~Thunderbi@38-146-94-247.echocast.zone) | 13:27 | |
| * bkeys1 -> bkeys | 13:27 | |
| - bkeys (QUIT: Client Quit) (~Thunderbi@38-146-94-247.echocast.zone) | 13:28 | |
| + bkeys (~Thunderbi@38-146-94-247.echocast.zone) | 13:29 | |
| - bkeys (QUIT: Client Quit) (~Thunderbi@38-146-94-247.echocast.zone) | 13:32 | |
| + bkeys (~Thunderbi@38-146-94-247.echocast.zone) | 13:32 | |
| - bkeys (QUIT: Client Quit) (~Thunderbi@38-146-94-247.echocast.zone) | 13:34 | |
| + bkeys (~Thunderbi@38-146-94-247.echocast.zone) | 13:36 | |
| - bkeys (QUIT: Read error: Connection reset by peer) (~Thunderbi@38-146-94-247.echocast.zone) | 13:38 | |
| grimmware | heh, yeah there it is again after the screen coming back! | 13:39 |
| + bkeys (~Thunderbi@38-146-94-247.echocast.zone) | 13:39 | |
| - bkeys (QUIT: Client Quit) (~Thunderbi@38-146-94-247.echocast.zone) | 13:40 | |
| + bkeys (~Thunderbi@38-146-94-247.echocast.zone) | 13:40 | |
| minute | josch: some input by daniels and kusma (mesa) on a311d flicker: the patch introduces AFBC, which is arm framebuffer compression. this feeds into the DSI controller, so it changes the way that the display is passed from the GPU to the display engine. it might be this triggers a bug in the DSI driver or engine and we would need to adjust the driver or some timings to fix it | 13:41 |
| minute | josch: working AFBC would increase GPU performance on a311d. | 13:41 |
| - bkeys (QUIT: Read error: Connection reset by peer) (~Thunderbi@38-146-94-247.echocast.zone) | 13:43 | |
| + bkeys (~Thunderbi@38-146-94-247.echocast.zone) | 13:43 | |
| - bkeys (QUIT: Read error: Connection reset by peer) (~Thunderbi@38-146-94-247.echocast.zone) | 13:48 | |
| + bkeys1 (~Thunderbi@38-146-94-247.echocast.zone) | 13:48 | |
| - bkeys1 (QUIT: Client Quit) (~Thunderbi@38-146-94-247.echocast.zone) | 13:50 | |
| + bkeys (~Thunderbi@38-146-94-247.echocast.zone) | 13:50 | |
| - bkeys (QUIT: Client Quit) (~Thunderbi@38-146-94-247.echocast.zone) | 13:52 | |
| + bkeys1 (~Thunderbi@38-146-94-247.echocast.zone) | 13:52 | |
| - Ar|stote|is (QUIT: Ping timeout: 245 seconds) (~linx@149.210.14.170) | 13:54 | |
| - bkeys1 (QUIT: Client Quit) (~Thunderbi@38-146-94-247.echocast.zone) | 13:54 | |
| + bkeys (~Thunderbi@38-146-94-247.echocast.zone) | 13:55 | |
| - bkeys (QUIT: Client Quit) (~Thunderbi@38-146-94-247.echocast.zone) | 13:55 | |
| + bkeys (~Thunderbi@38-146-94-247.echocast.zone) | 13:56 | |
| - bkeys (QUIT: Client Quit) (~Thunderbi@38-146-94-247.echocast.zone) | 13:57 | |
| + bkeys (~Thunderbi@38-146-94-247.echocast.zone) | 13:57 | |
| + Ar|stote|is (~linx@149.210.32.12) | 13:58 | |
| - bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@38-146-94-247.echocast.zone) | 14:10 | |
| + bkeys (~Thunderbi@38-146-94-247.echocast.zone) | 14:10 | |
| - bkeys (QUIT: Ping timeout: 245 seconds) (~Thunderbi@38-146-94-247.echocast.zone) | 14:17 | |
| josch | minute: oh wow nice i just imagine i understood any of what you wrote XD | 14:17 |
| josch | but i guess bottom line is, that it's likely not a dts fix | 14:17 |
| josch | i wish daniels and kusma had left that comment in the mesa issue tracker XD | 14:18 |
| josch | the conversation happened on irc? | 14:18 |
| + bkeys (~Thunderbi@38-146-94-247.echocast.zone) | 14:22 | |
| minute | josch: yeah panfrost channel on oftc | 14:25 |
| minute | josch: narmstrong says > so, the way afbc has been integrated into A311D, means yes it can miss a vsync on load | 14:26 |
| minute | so that's exactly what happens | 14:26 |
| minute | it is _supposed_ to work at least > AFBC has been integrated and validated with the ARM mali blob, so I'm pretty sure it works fine :-) | 14:27 |
| minute | i guess to get afbc working we need to fiddle with dsi timings... nooo | 14:27 |
| minute | until then we need to carry this revert patch | 14:27 |
| abortretryfail | :( | 14:30 |
| [tj] | minute: posted | 14:32 |
| josch | minute: thanks a lot for investigating this :) | 14:34 |
| josch | grimmware: no, U_BOOT_FDT_DIR="/dtbs/" is no longer needed | 14:34 |
| josch | grimmware: in fact i added a commit going the opposite direction: https://source.mnt.re/reform/reform-tools/-/commit/2068a25bf78ca6497f65eec84b1e26a12cc89189 | 14:35 |
| josch | grimmware: reason is that i plan to press the merge button here: https://source.mnt.re/reform/reform-debian-packages/-/merge_requests/86 | 14:36 |
| minute | [tj]: awesome thanks | 14:37 |
| josch | grimmware: in #1091979, vagrantc put Arnaud into CC to fix this for good in Debian instead of reform-tools having to work around this | 14:38 |
| - amk (QUIT: Ping timeout: 248 seconds) (~amk@user/amk) | 14:53 | |
| + chomwitt (~chomwitt@2a02:587:7a13:5400:1ac0:4dff:fedb:a3f1) | 14:54 | |
| - bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@38-146-94-247.echocast.zone) | 15:00 | |
| + bkeys1 (~Thunderbi@38-146-94-247.echocast.zone) | 15:00 | |
| - chomwitt (QUIT: Ping timeout: 252 seconds) (~chomwitt@2a02:587:7a13:5400:1ac0:4dff:fedb:a3f1) | 15:02 | |
| - bkeys1 (QUIT: Ping timeout: 276 seconds) (~Thunderbi@38-146-94-247.echocast.zone) | 15:05 | |
| + bkeys (~Thunderbi@38-146-94-247.echocast.zone) | 15:06 | |
| abortretryfail | Does MB 3.0 use the same LPC firmware code as 2.5? | 15:06 |
| [tj] | which lines in a usb-c cable are relevant for power? I got a cable tester and the cable I used when the charge controller died has everything connected apart from cc1 (and D+/D-) in 1 orientation which I think just the mechanics of the connector | 15:09 |
| abortretryfail | [tj]: The schematics are good for questions like that. | 15:13 |
| abortretryfail | What machine is it? Pocket Reform? | 15:13 |
| [tj] | I don't think the pd power negotiation is reform dependant | 15:15 |
| - bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@38-146-94-247.echocast.zone) | 15:17 | |
| + bkeys (~Thunderbi@38-146-94-247.echocast.zone) | 15:17 | |
| abortretryfail | It's not, but the schematics are still useful for that. | 15:18 |
| abortretryfail | cc1 and cc2 look like they are involved. | 15:18 |
| minute | abortretryfail: there is updated fw for mb3 | 15:23 |
| abortretryfail | build.sh in the reform2-lpc-fw folder doesn't have anything about it. Is that on a separate branch? | 15:29 |
| minute | abortretryfail: oh wow, looks like i didn't push that | 15:30 |
| abortretryfail | Reasoning: I'm brainstorming an idea to use a FET and some extra resistors to have a switchable high/low supply current mode via the LTC4020's ILimit pin | 15:30 |
| minute | abortretryfail: pushed | 15:31 |
| abortretryfail | Thanks :) | 15:33 |
| - wickedshell (QUIT: Ping timeout: 260 seconds) (~wickedshe@2601:8c0:800:4baa:ebe7:4dfe:1b3d:cca4) | 15:37 | |
| abortretryfail | huh the only change is the #define CHARGE_SOFT_START 0? | 15:40 |
| minute | abortretryfail: this is also relevant https://source.mnt.re/reform/reform/-/commit/e783e55d2e9a4b73989f6a490b2af959bc29223e | 15:55 |
| + ericsfraga (~user@2a00:23cc:b452:2900::99e) | 15:57 | |
| + mtm_ (~textual@47.202.75.129) | 16:34 | |
| - mtm (QUIT: Ping timeout: 244 seconds) (~textual@47.202.75.129) | 16:35 | |
| grimmware | josch: ah okay legit. Do you know when you're gonna merge? | 16:39 |
| abortretryfail | Oh, I think I'm running a version that has deep sleep. It takes a while to wake up sometimes, but it's not that bad tbh. | 17:07 |
| + chomwitt (~chomwitt@2a02:587:7a13:5400:1ac0:4dff:fedb:a3f1) | 17:08 | |
| - chomwitt (QUIT: Ping timeout: 268 seconds) (~chomwitt@2a02:587:7a13:5400:1ac0:4dff:fedb:a3f1) | 17:15 | |
| ch | [tj]: cc1/cc2 are used for sensing and pd handshake | 17:16 |
| [tj] | yes, this I know | 17:16 |
| [tj] | I have read pd and found it very upsetting | 17:17 |
| ch | accurate | 17:17 |
| [tj] | but if we are looking for a root cause for the charger board ic exploding then its worth taking a holistic approach | 17:17 |
| ch | power is on vbus and gnd, imo | 17:17 |
| [tj] | sorry, was writing stuff for work | 17:17 |
| - bkeys (QUIT: Ping timeout: 252 seconds) (~Thunderbi@38-146-94-247.echocast.zone) | 17:19 | |
| - jacobk (QUIT: Ping timeout: 248 seconds) (~quassel@47-186-65-73.dlls.tx.frontiernet.net) | 17:20 | |
| + bkeys (~Thunderbi@38-146-94-247.echocast.zone) | 17:25 | |
| josch | grimmware: merged! | 17:25 |
| josch | grimmware: thus, it will be in the MNT repos in 1.5 hours | 17:25 |
| josch | minute: could you enable gitlab CI for https://source.mnt.re/grimmware/reform-tools/ please? Thank you! | 17:30 |
| ch | is the kernel change already merged or are there kernel packages to test? | 17:37 |
| ch | (for the display issue) | 17:37 |
| minute | ch: there are even sd images to test! | 17:41 |
| ch | :) | 17:42 |
| ch | i think i'd like to live dangerously and try it on my normal install | 17:42 |
| + wickedshell (~wickedshe@2601:8c0:800:4baa:fd8a:dda7:847a:75e9) | 17:43 | |
| - wickedshell (QUIT: Ping timeout: 276 seconds) (~wickedshe@2601:8c0:800:4baa:fd8a:dda7:847a:75e9) | 17:50 | |
| minute | josch: done! | 17:52 |
| ch | ah yes, disabling sleep did work, i remember that from c3 | 17:56 |
| josch | minute: thank you! | 17:56 |
| josch | is a311d ethernet known to not come up after a fresh boot sometimes? | 17:57 |
| minute | josch: yep | 17:59 |
| josch | do we have an issue open for that? | 18:00 |
| minute | i don't believe so | 18:01 |
| josch | i only notice now that i use an a311d as a media center | 18:01 |
| josch | minute: do you have a keyword for me with which i can maybe find past discussion about that in irc logs? | 18:01 |
| minute | josch: not at the moment, sorry... | 18:01 |
| josch | no problem :) | 18:02 |
| josch | i'll search more later | 18:02 |
| - ericsfraga (QUIT: Read error: Connection reset by peer) (~user@2a00:23cc:b452:2900::99e) | 18:04 | |
| josch | but i see getting ethernet to work in the first place was a massive struggle XD | 18:06 |
| minute | oh yeah :D | 18:08 |
| + jacobk (~quassel@utdpat241060.utdallas.edu) | 18:11 | |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:50) | 18:14 | |
| - jacobk (QUIT: Ping timeout: 276 seconds) (~quassel@utdpat241060.utdallas.edu) | 18:58 | |
| + mark_ (~mjw@gnu.wildebeest.org) | 19:47 | |
| * mjw -> Guest4439 | 19:48 | |
| - Guest4439 (QUIT: Killed (mercury.libera.chat (Nickname regained by services))) (~mjw@2001:1c06:2486:a800:7602:5eff:dc71:a72c) | 19:48 | |
| * mark_ -> mjw | 19:48 | |
| + Guest4439 (~mjw@2001:1c06:2486:a800:7602:5eff:dc71:a72c) | 19:48 | |
| + wickedshell (~wickedshe@2601:8c0:800:4baa:314c:a978:edb5:94da) | 19:56 | |
| + jacobk (~quassel@utdpat241106.utdallas.edu) | 20:24 | |
| - jacobk (QUIT: Ping timeout: 248 seconds) (~quassel@utdpat241106.utdallas.edu) | 20:28 | |
| - andreas-e (QUIT: Quit: Leaving) (~Andreas@2001:861:c4:f2f0::c64) | 20:30 | |
| grimmware | minute: looks like you were right, the flickering goes away after a while | 20:42 |
| + jacobk (~quassel@utdpat241106.utdallas.edu) | 21:32 | |
| - bkeys (QUIT: Ping timeout: 248 seconds) (~Thunderbi@38-146-94-247.echocast.zone) | 21:42 | |
| + digitalrane_ (~digitalra@user/digitalrane) | 21:46 | |
| - digitalrane (QUIT: Ping timeout: 250 seconds) (~digitalra@user/digitalrane) | 21:46 | |
| * digitalrane_ -> digitalrane | 21:51 | |
| + chomwitt (~chomwitt@2a02:587:7a13:5400:1ac0:4dff:fedb:a3f1) | 21:59 | |
| - jacobk (QUIT: Ping timeout: 260 seconds) (~quassel@utdpat241106.utdallas.edu) | 22:03 | |
| - gustav28 (QUIT: Quit: Quit) (~gustav@c-78-82-52-230.bbcust.telenor.se) | 22:15 | |
| - wickedshell (QUIT: Read error: Connection reset by peer) (~wickedshe@2601:8c0:800:4baa:314c:a978:edb5:94da) | 22:17 | |
| minute | grimmware: yeah i kinda have a symbiotic relationship with these panels after all that time lol | 22:23 |
| grimmware | I'm like, 0 surprised | 22:24 |
| grimmware | I'll probably hit you up at some point for any intuition on where that might need to be fixed because it's a little bit jarring and it definitely didn't happen before | 22:25 |
| grimmware | I mean realistically I probably need to check out a commit from back in November and stare at it for a while next to the current code but I am not an expert so any hints are greatly appreciated | 22:26 |
| grimmware | Honestly so glad to have a platform that I love so much where I can contribute usefully, even if it's only through determination | 22:27 |
| + bkeys (~Thunderbi@38-146-94-247.echocast.zone) | 23:05 | |
| - bkeys (QUIT: Client Quit) (~Thunderbi@38-146-94-247.echocast.zone) | 23:07 | |
| + bkeys (~Thunderbi@38-146-94-247.echocast.zone) | 23:07 | |
| + bkeys1 (~Thunderbi@2607:fb90:e341:4305:160:59c6:662:393e) | 23:18 | |
| - bkeys (QUIT: Ping timeout: 252 seconds) (~Thunderbi@38-146-94-247.echocast.zone) | 23:19 | |
| * bkeys1 -> bkeys | 23:19 | |
| - bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@2607:fb90:e341:4305:160:59c6:662:393e) | 23:47 | |
| josch | frickler: reform-check now has an --offline option and asks interactively if it should run "apt update" for you | 23:47 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!