| - AnimaInvicta (PART: !!unknown attribute: msg!!) (~AnimaInvi@88-169-191-86.subs.proxad.net) | 00:08 | |
| - mjw (QUIT: Ping timeout: 260 seconds) (~mjw@206.12.14.167) | 00:53 | |
| - potash6 (QUIT: Quit: The Lounge - https://thelounge.chat) (~potash@user/foghorn) | 00:59 | |
| - voltaire28 (QUIT: Ping timeout: 260 seconds) (~jlafon@17.89.66.37.rev.sfr.net) | 01:14 | |
| + mjw (~mjw@206.12.14.167) | 01:25 | |
| - Gooberpatrol66 (QUIT: Ping timeout: 261 seconds) (~Gooberpat@user/gooberpatrol66) | 01:30 | |
| josch | minute: I noticed that 9100-debug-rk3588-dp-max-lanes.patch and 9101-debug-rk3588-dp-link-training.patch are only in linux/patches7.0 and not in linux/patches7.1 -- intentionally missing? | 01:31 |
|---|---|---|
| + oliverD (~Thunderbi@user/oliverd) | 01:36 | |
| - oliverD (QUIT: Remote host closed the connection) (~Thunderbi@user/oliverd) | 01:36 | |
| + paperManu_ (~paperManu@204.48.0.160) | 02:10 | |
| minute | josch: i think they're replaced by the all-in-one analogix dp patch | 02:15 |
| minute | josch: but also, the alt-mode functionality is disabled in device tree until i make it better | 02:16 |
| minute | josch: the size of my patches was already more than big enough so i wanted to release a new stable in-between point that already gives me most of what i need on the SC side | 02:17 |
| minute | as for user value, i focused on the various pocket panel driver scenarios and issues | 02:18 |
| - a3f (QUIT: Read error: Connection reset by peer) (~a.fatoum@flummi.grey.stw.pengutronix.de) | 02:18 | |
| + a3f (~a.fatoum@flummi.grey.stw.pengutronix.de) | 02:21 | |
| - mjw (QUIT: Ping timeout: 266 seconds) (~mjw@206.12.14.167) | 02:23 | |
| + mjw (~mjw@206.12.14.167) | 02:30 | |
| + Gooberpatrol66 (~Gooberpat@user/gooberpatrol66) | 02:44 | |
| - paperManu (QUIT: Ping timeout: 268 seconds) (~paperManu@204.48.0.160) | 03:09 | |
| josch | minute: I upgraded firmware and kernel. Things mostly work. I posted in the forum. | 03:23 |
| josch | minute: biggest regression is that pressing and holding super+enter to power on doesn't seem to work anymore? | 03:23 |
| - mjw (QUIT: Ping timeout: 243 seconds) (~mjw@206.12.14.167) | 03:29 | |
| - colinsane (QUIT: Ping timeout: 264 seconds) (~colinunin@97-113-85-96.tukw.qwest.net) | 03:32 | |
| - _justin_kelly4 (QUIT: Ping timeout: 246 seconds) (~justinkel@user/justin-kelly/x-6011154) | 03:36 | |
| + _justin_kelly4 (~justinkel@user/justin-kelly/x-6011154) | 03:39 | |
| + colinsane (~colinunin@97-113-85-96.tukw.qwest.net) | 03:46 | |
| - Svp (QUIT: Ping timeout: 248 seconds) (~svp@host-79-57-105-252.business.telecomitalia.it) | 04:03 | |
| + potash6 (~potash@user/foghorn) | 04:04 | |
| + Svp (~svp@host-79-57-105-252.business.telecomitalia.it) | 04:20 | |
| - lidstah (QUIT: Remote host closed the connection) (~lidstah@gateway/tor-sasl/lidstah) | 04:29 | |
| + lidstah (~lidstah@gateway/tor-sasl/lidstah) | 04:29 | |
| - lidstah (QUIT: Remote host closed the connection) (~lidstah@gateway/tor-sasl/lidstah) | 05:00 | |
| + lidstah (~lidstah@gateway/tor-sasl/lidstah) | 05:05 | |
| - lidstah (QUIT: Remote host closed the connection) (~lidstah@gateway/tor-sasl/lidstah) | 05:16 | |
| + lidstah2 (~lidstah@gateway/tor-sasl/lidstah) | 05:16 | |
| - lidstah2 (QUIT: Remote host closed the connection) (~lidstah@gateway/tor-sasl/lidstah) | 05:27 | |
| + lidstah (~lidstah@gateway/tor-sasl/lidstah) | 05:28 | |
| - RandyK (QUIT: Remote host closed the connection) (~RandyK@user/randyk) | 05:29 | |
| + RandyK (~RandyK@user/randyk) | 05:29 | |
| - lidstah (QUIT: Remote host closed the connection) (~lidstah@gateway/tor-sasl/lidstah) | 05:37 | |
| + lidstah (~lidstah@gateway/tor-sasl/lidstah) | 05:38 | |
| - lidstah (QUIT: Remote host closed the connection) (~lidstah@gateway/tor-sasl/lidstah) | 06:24 | |
| + lidstah (~lidstah@gateway/tor-sasl/lidstah) | 06:25 | |
| - lidstah (QUIT: Remote host closed the connection) (~lidstah@gateway/tor-sasl/lidstah) | 06:27 | |
| + lidstah (~lidstah@gateway/tor-sasl/lidstah) | 06:28 | |
| - potash6 (QUIT: Quit: The Lounge - https://thelounge.chat) (~potash@user/foghorn) | 06:51 | |
| + potash6 (~potash@user/foghorn) | 06:52 | |
| - potash6 (QUIT: Quit: The Lounge - https://thelounge.chat) (~potash@user/foghorn) | 07:31 | |
| + potash6 (~potash@user/foghorn) | 08:00 | |
| + potash64 (~potash@user/foghorn) | 08:07 | |
| - potash6 (QUIT: Ping timeout: 264 seconds) (~potash@user/foghorn) | 08:09 | |
| * potash64 -> potash6 | 08:09 | |
| + potash68 (~potash@user/foghorn) | 08:15 | |
| - potash6 (QUIT: Ping timeout: 264 seconds) (~potash@user/foghorn) | 08:17 | |
| * potash68 -> potash6 | 08:17 | |
| + mjw (~mjw@206.12.14.47) | 08:18 | |
| + oliverD (~Thunderbi@user/oliverd) | 09:14 | |
| - elb2 (QUIT: Ping timeout: 258 seconds) (~elb@user/elb) | 09:22 | |
| - oliverD (QUIT: Ping timeout: 272 seconds) (~Thunderbi@user/oliverd) | 10:04 | |
| cararemixed | josch: I ran your fwupdmgr thing but it seems it downgraded me (I should have checked the version but refresh --force + update downgraded me). | 10:06 |
| cararemixed | On 2025 fw the systemd boot procedure will get stuck. I'm pulling down a system image to boot from an SD card again and try again. | 10:07 |
| + oliverD (~Thunderbi@user/oliverd) | 10:08 | |
| cararemixed | So a failsafe might be worth adding to allow it to boot past the failure. Sure it won't have a working system controller interface but getting a shell would at least let this be repaired on device. | 10:09 |
| josch | cararemixed: can you give me more details? To which version did it downgrade? How will the boot procedure get stuck? | 10:11 |
| cararemixed | 20251118 on both. | 10:11 |
| cararemixed | And for the boot loop... one sec I'll see if I can produce it and get an image. | 10:12 |
| cararemixed | https://usercontent.irccloud-cdn.com/file/Luo6HG9f/IMG_7794.JPG | 10:15 |
| - oliverD (QUIT: Quit: oliverD) (~Thunderbi@user/oliverd) | 10:18 | |
| cararemixed | Booting up on one of these: https://source.mnt.re/reform/reform-system-image/-/jobs/23576/artifacts/browse (some of the old image links seem dead btw) | 10:24 |
| josch | cararemixed: did "fwupdmgr update" ask you to which version to upgrade? I think you have to manually accept, right? | 10:26 |
| josch | cararemixed: i backed up old images to my server. See README.md for the links | 10:26 |
| cararemixed | I didn't check as I was multitasking and just said yes. | 10:27 |
| cararemixed | re old images: thanks. I rarely need them but I don't know where my old rescue card went. | 10:27 |
| josch | cararemixed: can you confirm that you saw progress bars when you ran "fwupd refresh"? I just ran that and got a timeout error. | 10:27 |
| cararemixed | Yes. When I did last night I saw bars. | 10:28 |
| cararemixed | And now I saw it again. | 10:28 |
| josch | minute: another regression: my audio is gone on RK3588 Pocket reform with latest MNT Debian unstable. Here are some parts from dmesg matching tlv320aic31 and some context: https://paste.debian.net/hidden/32bc8d71 | 10:28 |
| cararemixed | And it's the new release on this boot. My old system seemed to have some broken state I guess. I'll rerun the refresh on it once I can get in. | 10:29 |
| josch | cararemixed: ah i also saw that problem a week ago when booting the new kernel with old firmware. | 10:29 |
| josch | minute can likely help you -- i need to go to take care of the children, sorry | 10:30 |
| cararemixed | Id had been running a more recent firmware but not the major rework branch before all this just fine on the latest kernel | 10:30 |
| cararemixed | Only puzzle is why the system didn't refresh (this was 4am Berlin). | 10:32 |
| cararemixed | The other boot failure is more of a bug report. It'd be nice to let it fail into a slightly broken system than loop (it never times out but just adds more time to its wait) | 10:32 |
| cararemixed | back in | 10:33 |
| cararemixed | https://usercontent.irccloud-cdn.com/file/LG9apUPj/1786091687.JPG | 10:34 |
| cararemixed | Ok. Worked this time. Maybe some kind of cache needed busting somewhere | 10:35 |
| cararemixed | Beautiful work. It's so nice to not have to hack the module reload to get a working backlight again (which also broke gnome shell and this required brightnessctl s to be used every time)... quality of life restored. | 10:42 |
| - Guest29952 (QUIT: Quit: The Lounge - https://thelounge.chat) (~ndufresne@transit.collaboradmins.com) | 10:47 | |
| + Guest29952 (~ndufresne@transit.collaboradmins.com) | 10:50 | |
| - potash6 (QUIT: Quit: Ping timeout (120 seconds)) (~potash@user/foghorn) | 10:55 | |
| cararemixed | Ok. New oddity, charging via battery pack on the old mobo seems to reject my smart pd cable but work with a dumb cable now (reverse from before). Since its a dumb cable it also seems to charge my battery not my laptop. I'll do more testing. | 11:07 |
| + andreas-e (~Andreas@2a02-8434-b6a3-e901-facc-8e87-8e54-890d.rev.sfr.net) | 11:09 | |
| - andreas-e (QUIT: Client Quit) (~Andreas@2a02-8434-b6a3-e901-facc-8e87-8e54-890d.rev.sfr.net) | 11:10 | |
| minute | josch: thanks for test results! the paste about the audio issue is 404 though | 11:17 |
| minute | cararemixed: ah it is likely that the battery stuff is due to all the pd rework. i'll check again with my powerbank today | 11:18 |
| josch | minute: new try: https://mister-muffin.de/p/7cpT.txt | 11:19 |
| minute | cararemixed: maybe the following workaround could help: if you go to "console" in the oled menu, and then type (pdsink 1) and then enter, then (pdreset) | 11:19 |
| minute | cararemixed: this should force PD code to only be a sink and not source | 11:19 |
| minute | josch: thank you! weird @ unsupported freq. could be a side effect of a patch i did for qcom audio. shame that i didn't catch this in testing | 11:20 |
| minute | hmm i don't have this tlv error on my rk3588 pocket | 11:23 |
| minute | but my kernel is probably not the final one i released > 7.1.6-1+reform20260806T172143Z | 11:24 |
| minute | updating now | 11:24 |
| cararemixed | minute: I'll be out for a bit but I'll try it when I have a place to pull the pocket out. Thanks | 11:28 |
| minute | josch: hmm i can't reproduce the audio issue on rk3588 pocket reform | 11:29 |
| minute | my kernel is now > 7.1.6-1+reform20260807T053531Z | 11:30 |
| josch | minute: i rebooted and audio is back | 11:31 |
| minute | josch: ohh ok hm | 11:32 |
| josch | kernel 7.1.7 is currently building on MNT CI | 11:32 |
| minute | constantly new versions! | 11:33 |
| minute | funny (well...) bug @ that i broke holding hyper+enter to turn the machine on in the keyboard fw uff | 11:34 |
| minute | will fix today, at least the fw update pipeline is fixed now so that's a huge improvement to getting fixes to people | 11:34 |
| josch | thank you! | 11:35 |
| minute | cararemixed: sure! btw another workaround should be to connect the battery pack while pocket is off, because then it can be in sink mode only | 11:38 |
| josch | minute: i talked with sanders who agreed to delete our posts from the thread | 11:49 |
| cararemixed | I'll create a table of states/cables/charge behaviors (I have multiple ports on this brick with diff behaviors) | 11:49 |
| minute | josch: thank you! some threads have a habit of getting very confusing after a while | 12:02 |
| minute | cararemixed: cool thanks | 12:03 |
| josch | absolutely -- i wonder if there is a moderation tool which allows me to break out messages and move them to a new thread | 12:03 |
| cararemixed | Did the latest of like 60ish miutes ago kernel/module build include changes? I had shut my pocket off and did a bunch of testing and no longer see the weird PD charge inversion. Count me confused. | 12:30 |
| cararemixed | I'll keep an eye out for the issue if it's spurious but afaict, all modes/cables work as they should right now. | 12:32 |
| cararemixed | FWIW I tend to keep my kernel very fresh when I've got my other machines in reach so I can hit issues early and report them. But at the moment, this machine feels the most solid it's been since I bought it. Big win on this update. Thanks for everyone who worked on it but especially minute for the sleuthing USB-C madness. The upcoming hw-upgrades are going to make this device even more amazing. | 12:41 |
| + reform18198 (~franzo@2a01:5241:cfa:9500:9322:b0a9:440a:b453) | 13:14 | |
| reform18198 | hi, I have a problem | 13:14 |
| reform18198 | I did sudo apt upgrade without knowing that a new kernel/firmware rolled out... long story short.. | 13:14 |
| reform18198 | now my battery indicator says 0% and I can't upgrade the firmware | 13:15 |
| reform18198 | how can I boot with an older kernel version without grub? | 13:15 |
| reform18198 | what happens if I try sudo apt remove linux-image-version-blabla? | 13:19 |
| cararemixed | I usually use a spare sd card around to recover from weird states | 13:39 |
| cararemixed | I had a similar issue on my pocket with old fw and the new kernel. I used a slightly older sd card imagine to boot an older kernel and flash fw | 13:40 |
| cararemixed | less chance of messing the primary install up | 13:41 |
| reform18198 | I uninstalled latest kernel so now it should boot with the old one,correct? | 13:44 |
| reform18198 | I am rebooting, wish me good luck | 13:52 |
| - reform18198 (QUIT: Remote host closed the connection) (~franzo@2a01:5241:cfa:9500:9322:b0a9:440a:b453) | 13:53 | |
| + reform26880 (~franzo@2a01:5241:cfa:9500:9322:b0a9:440a:b453) | 14:01 | |
| reform26880 | all good, installing the new kernel | 14:02 |
| - reform26880 (QUIT: Remote host closed the connection) (~franzo@2a01:5241:cfa:9500:9322:b0a9:440a:b453) | 14:02 | |
| + paperManu (~paperManu@204.48.0.160) | 14:08 | |
| cararemixed | New issue spotted. Wall charger now, I get this oscillation between states | 14:17 |
| cararemixed | https://usercontent.irccloud-cdn.com/file/Hna23SSt/IMG_7797.MP4 | 14:17 |
| josch | minute: i had a similar problem as reform18198 above. If the system thinks that it's at 0% battery, it will stop you from upgrading the firmware. To work around that I did "rmmod reform2_lpc" and then the system no longer thinks that there is a battery. Maybe add that workaround (or maybe you know a better one?) to your forum post? | 14:18 |
| cararemixed | Ok. Tested multiple cables. The wall charger is failing no matter which kind of cable so it might be an electrical fault in it or similar. | 14:22 |
| cararemixed | It's an apple 140W charger so you'd imagine it'd handle this load ok | 14:23 |
| cararemixed | Same issue with an apple 30W adapter. So weird | 14:39 |
| cararemixed | My other adapters work fine. | 14:42 |
| cararemixed | I guess my old (not sure old though, they're both GaN) apple chargers are just not really compliant with whatever. They charge other things and used to charge this pocket reform. I really only used it because I had a longer extension cable to move it further from the wall. | 14:45 |
| cararemixed | Ironically I think my macbook that I sometimes dig out has the same issue so even apple hardware seems to find problems with it. | 14:45 |
| cararemixed | If others have more recent power adapters from apple to try, I'd be curious if you can reproduce. | 14:46 |
| minute | josch: ah weird @ not wanting to update the firmware | 15:01 |
| minute | josch: and yes, discourse has "move things to thread" features afaik | 15:01 |
| minute | cararemixed: the kernel is not involved in PD charging things, only the sysctl fw, which wasn't updated after last nights update. but PD is non deterministic when the pocket is on because it is in "toggle" mode (DRP) so it's kind of random if it connects as source or sink initially | 15:02 |
| cararemixed | I see. Thanks. So unlucky trials and then the reverse after | 15:03 |
| josch | minute: are you somehow able to upgrade the firmware at low battery? That would we useful to know how | 15:13 |
| josch | minute: ah i found out how to do it. It's not in the settings menu for a reply but you first have to click on the wrench icon on top of the whole thread, then click "select posts", then select the posts you want to move and *then* you get the option to move these to another thread. I'll make use of this from now on! | 15:22 |
| minute | josch: weird issue on a test system here (rk3588): mnt-sc doesn't get included in the initramfs | 15:30 |
| minute | josch: about the upgrade, i would assume there's an override option that can be passed to fwupdmgr update | 15:31 |
| minute | but also it wasn't auto loaded by the kernel later, i had to modprobe it | 15:36 |
| minute | josch: hmm i didn't pay enough attention to your "reload reform2-lpc" thing you added to reform-hw-setup btw, why is it being reloaded for all pocket reform models? | 15:41 |
| - chrcav (QUIT: Ping timeout: 248 seconds) (~chrcav@user/chrcav) | 15:43 | |
| minute | hmmmm i have a feeling the spi stuff broke on 7.1.7 because i still have 7.1.6 on my pocket. lets upgrade | 15:44 |
| sigrid | that'd be pretty strange since there's only a single commit between those versions in mainline kernel, no? | 15:45 |
| minute | yeah but i mean maybe in our CI there's a difference | 15:45 |
| sigrid | oh ok | 15:45 |
| minute | hmm but yeah, josch's new builds only add a qcom change, unrelated | 15:47 |
| amospalla | installing new kernel on Debian Trixie | 15:47 |
| minute | ok installed the same kernel package on my pocket reform, and i don't have this issue, mnt-sc is loaded 3.8 seconds into the boot by initramfs | 15:49 |
| minute | cararemixed: your new power supply problems are almost certainly due to my restructured firmware, and i need to fix new bugs to accomodate those chargers again | 15:56 |
| minute | i think we have some apple chargers here | 15:56 |
| cararemixed | I can always drop a spare off if you want one to test | 15:57 |
| minute | cararemixed: cool thanks, i'll get back to you on that | 15:58 |
| minute | josch: fwupdmgr --no-safety-check update | 15:58 |
| minute | josch: (fwupdmgr --help has all the options) | 15:58 |
| josch | minute: I don't think I committed the reform2-lpc reload. I don't see it in git. | 16:01 |
| amospalla | worked on my pocket, Debian/Trixie, panel v1, rk3588 | 16:01 |
| josch | minute: the qcom change is just a revert and then a revert of the revert so no-op in the end. | 16:02 |
| josch | minute: thank you for --no-safety-check -- I absolutely missed that in the more than 100 lines of --help output :D | 16:03 |
| minute | josch: damn, you're right @ reform2-lpc. i wonder where that machine got that from, let's see, might be a local patch then | 16:03 |
| josch | amospalla: you upgraded to 7.1.3 from trixie-backports? | 16:03 |
| amospalla | josch: yes | 16:04 |
| josch | amospalla: then you beat me to it. That was on my todo list. It works for you then? | 16:04 |
| amospalla | I've checking debian packages all morning. | 16:04 |
| amospalla | Yes, but with one caveat, display takes 20 seconds to show, since kernel boots. But it already had some problem since I moved to barebox, so I don't know if it is something related to barebox or not. I'm posting this on the forums too. | 16:05 |
| minute | ah that machine still has reform-tools 1.8.7 | 16:05 |
| minute | i totally forgot that reform-tools has to be updated separately last night, and didn't mention it in my post :0 | 16:06 |
| minute | but it also shouldn't be critical for this, hm. | 16:06 |
| amospalla | The good thing is that, even it takes time to show, now it does not need the dirty hack I was using before and ends working sooner or later. | 16:08 |
| josch | minute: reform-tools 1.88 is necssary to include mnt-sc in the initramfs | 16:08 |
| minute | josch: aaa | 16:08 |
| minute | right, so then i must add that to the instructions now | 16:08 |
| josch | thank you! | 16:08 |
| amospalla | Do you want me to notice this thing about the display on the forum post? | 16:08 |
| minute | amospalla: that's exactly the problem i just discovered. you need to do sudo apt install reform-tools | 16:12 |
| minute | amospalla: then the display will come up after a few seconds and not after 20 sseconds | 16:12 |
| minute | amospalla: also, without that, mnt-sc driver will not be loaded at all so no brightness control | 16:13 |
| amospalla | minute: I need 1.88 right? I'll paciently wait, thank you :) | 16:14 |
| josch | amospalla: why wait? | 16:14 |
| amospalla | oh, it is already updated on repos? let me see | 16:14 |
| minute | amospalla: yeah it was rolled out before the other updates. i just had forgotten to include it in the instructions last night | 16:14 |
| josch | amospalla: oh it is not whoops | 16:15 |
| josch | fixing... | 16:15 |
| minute | josch: it is | 16:15 |
| minute | josch: at least i was able to install it on the test machine | 16:15 |
| amospalla | Mine is on 1.87-1~bpo13+1+reform20251209T125945Z~bpo13+1 | 16:15 |
| minute | ah sorry trixie vs unstable i guess? | 16:15 |
| minute | vs mnt repo i mean | 16:15 |
| + andreas-e (~Andreas@2a02-8434-b6a3-e901-facc-8e87-8e54-890d.rev.sfr.net) | 16:15 | |
| amospalla | yes, Trixie, waiting some seconds won't hurt me. | 16:16 |
| josch | amospalla: thank you for finding this issue | 16:17 |
| amospalla | oh no, please, thank you. | 16:17 |
| josch | amospalla: uploaded -- will probably be in the repo in about 6 hours | 16:19 |
| amospalla | I love how OLED shows the minimum battery consumption while powered off instead of zero. | 16:19 |
| amospalla | thank you :) | 16:19 |
| minute | at some point would be nice to have something like reform-update-system that does all the required/common updates at once :D or a gtk4 version of reform-check... | 16:23 |
| josch | minute: we had similar situations in the past regarding the u-boot version which we also cannot update via a package upgrade | 16:29 |
| - _justin_kelly4 (QUIT: Ping timeout: 250 seconds) (~justinkel@user/justin-kelly/x-6011154) | 16:40 | |
| minute | josch: right, would be great to move everyone on rk3588 pocket to barebox | 16:41 |
| minute | josch: so that script could do all those things :D | 16:41 |
| minute | (why barebox? because it makes it much easier to recover from installation problems) | 16:41 |
| + _justin_kelly4 (~justinkel@user/justin-kelly/x-6011154) | 16:50 | |
| + jnerula (~jnerula@li1009-93.members.linode.com) | 16:51 | |
| minute | fix pushed for long-press to powerup. i also want to fix one other little annoyance | 17:00 |
| minute | (that's been around longer): when you press hyper-enter to open the menu, the menu staying open depends on whether you lift hyper or enter first | 17:01 |
| minute | ahh yes, fixed that now, so much better | 17:04 |
| minute | gonna do a quick look at apple chargers before i tag that | 17:07 |
| andreas-e | Hello! I am doing something difficult to admit, that is making my MNT Reform a bit more mainstream ;-) | 17:10 |
| andreas-e | So instead of using sway, I wanted to try out wayfire. But when I launch it, I just get a black screen with an arrow cursor (that reacts to the trackball). | 17:11 |
| andreas-e | After killing it, I see an error message from libseat: "Could not connect to socket /run/seatd.sock: No such file or directory" | 17:12 |
| andreas-e | Do you have an idea what I could do? | 17:12 |
| andreas-e | Or alternatively, could I just install xfce and run startxfce4? | 17:12 |
| andreas-e | I do not care so much about the exact system; but indeed xfce4 is what I use elsewhere. | 17:13 |
| minute | andreas-e: why not use gnome? | 17:14 |
| minute | andreas-e: we ship that as the "mainstream default" | 17:14 |
| minute | (i personally use it) | 17:14 |
| minute | andreas-e: about xfce, it depends. what cpu do you have? | 17:15 |
| andreas-e | The oldest one. I would definitely prefer something simple... And I am not used to gnome. Actually in the past I have migrated from KDE to xfce4 so be more compatible with weaker systems. | 17:16 |
| minute | oh wow i have just found 100% evidence for keyboard->mainboard signal issues happening when opening/closing the clamshell/rotating the hinges | 17:17 |
| minute | because sysctl complains about syntax errors from the keyboard while i'm rotating it :D | 17:17 |
| qbit | whoa | 17:18 |
| minute | so something causes data bits on the uart when moving that | 17:18 |
| qbit | emf from cables being.. pinched? | 17:18 |
| qbit | emf from .. metal parts moving past each other? (i have no idea what I am talking about :D) | 17:19 |
| minute | no idea exactly how that works yet | 17:19 |
| minute | andreas-e: so you have imx8mq? | 17:19 |
| minute | andreas-e: i definitely wouldn't recommend wayfire, we no longer maintain that | 17:19 |
| minute | andreas-e: not 100% sure but maybe the following works nowadays: startxfce4 --wayland | 17:20 |
| minute | andreas-e: apparently xfce4 4.20 introduced wayland support | 17:21 |
| minute | andreas-e: so try to install it! | 17:21 |
| andreas-e | Yes, I think I have imx8mq (there was a second, plus variant a bit later, no? I definitely have the original one from the crowdfunding). | 17:21 |
| andreas-e | So I will give xfce4 a try. | 17:21 |
| - amospalla (QUIT: Ping timeout: 264 seconds) (~jordi@user/amospalla) | 17:23 | |
| + amospalla (~jordi@user/amospalla) | 17:24 | |
| minute | amospalla: cat /proc/device-tree/model | 17:25 |
| minute | andreas-e: sorry i meant you, cat /proc/device-tree/model | 17:25 |
| minute | andreas-e: that will tell you exactly the model you have | 17:25 |
| + chrcav (~chrcav@user/chrcav) | 17:28 | |
| minute | cararemixed: what i can see so far is that charging with an active cable doesn't work when the active cable is plugged into pocket reform _first_ and then plugged into the supply. it only works the other way around | 17:28 |
| minute | i could probably time out sending source caps and force sink mode... | 17:29 |
| minute | ah yeah depending on the toggle lottery, whenever it tries SRC mode on a power supply, things fail | 17:30 |
| - amospalla (QUIT: Ping timeout: 244 seconds) (~jordi@user/amospalla) | 17:31 | |
| cararemixed | Oh. I'll go try. That's probably what happened | 17:32 |
| cararemixed | Confirmed that's what I see too (wasn't doubting you). Good to know for now. | 17:33 |
| + amospalla (~jordi@user/amospalla) | 17:35 | |
| minute | (not apple specific btw, happens on all my supplies. working on a better way to handle this) | 17:36 |
| andreas-e | Speaking of charging; I had the impression that when I run "halt", then apparently the machine does not turn off completely, but still uses some battery. Also the orange LED next to the network cable is still turned on. To completely turn off, I need to press circle+0. Is that true? | 17:36 |
| minute | andreas-e: do you have reform2-lpc installed? is your software up to date? | 17:40 |
| andreas-e | I am on up-to-date trixie, and have reform2-lpc-dkms installed. | 17:42 |
| andreas-e | But I am not certain about my apt/sources.list. The file contains the repositories from Debian itself. | 17:43 |
| andreas-e | In the sources.list.d subdirectory, I have a file with some metadata speaking about https://reform.debian.net/debian, trixie and main, followed by a PGP key block. | 17:44 |
| + wickedshell (~wickedshe@2601:8c0:c7c:3572::1b90) | 17:45 | |
| wickedshell | Long time no chat. (I have been quite lazy at fixing my IRC bouncer for some reason). | 17:46 |
| wickedshell | First: congrats on shipping the new system and keyboard firmware! | 17:46 |
| minute | wickedshell: wb | 17:46 |
| minute | andreas-e: oh trixie, that's unofficial... | 17:46 |
| minute | andreas-e: like, it's josch's release. but have you done an apt full-upgrade recently? | 17:47 |
| wickedshell | Second: May I have some help? I appear to be stuck at a black screen, I can tell it's doing something as power levels fluxuate. (A guess of the process on when to enter passwords doesn't actually get me to a networked | 17:47 |
| minute | andreas-e: also i'm not 100% sure about halt. does "poweroff" work? | 17:47 |
| minute | wickedshell: ok, system specs please | 17:48 |
| wickedshell | yeah, sorry I hit enter way early by accident, still putting it together | 17:48 |
| andreas-e | Okay! Indeed I followd josch's recommendation. I am old, I did a "apt-get dist-upgrade", which appears to be essentially the same as "apt full-upgrade". | 17:48 |
| wickedshell | machine. As a quick overview: pocket reform (not v2 mb), rk processor, barebox installed on uSD card, system installed to nVME drive with FDE | 17:48 |
| andreas-e | poweroff behaves the same as halt. | 17:49 |
| minute | andreas-e: does this show something? sudo dmesg | grep reform2-lpc | 17:49 |
| minute | andreas-e: do you have battery percentage in sway? | 17:50 |
| josch | should probably be grep reform2_lpc | 17:50 |
| wickedshell | (Sway being by wm once booted, although the login does appear to be managed by gnome at the moment) | 17:50 |
| minute | wickedshell: ok, any idea which display panel version? and, you got display in barebox? | 17:50 |
| andreas-e | dmesg shows nothing interesting with "grep reform2" | 17:51 |
| wickedshell | Not sure on panel version, my machine was a CrowdSupply one, but ordered well after the crowdfunding phase. And no display in barebox after the update. | 17:51 |
| minute | andreas-e: sorry i'm multitasking too much. i meant: lsmod | grep lpc | 17:51 |
| minute | wickedshell: ok, that's concerning @ no display in barebox | 17:52 |
| wickedshell | Anything convient for checking on that? I can take it apart if needed. (Also no rush on getting to me) | 17:52 |
| minute | wickedshell: even after power cycle, yes? | 17:52 |
| josch | wickedshell: reform-check outputs your display version | 17:52 |
| minute | josch: they don't have display i guess... | 17:52 |
| andreas-e | Sway shows no battery, and the lsmod returns nothing. Sorry, this "halt" not fully working is not a big deal, so unless you think it is a bug "on your end", we do not have to debug it. | 17:52 |
| minute | wickedshell: do you have an HDMI monitor you could connect? | 17:52 |
| minute | andreas-e: ok but you're missing major features of the system | 17:53 |
| minute | andreas-e: you're missing the reform2-lpc driver somehow | 17:53 |
| wickedshell | minute: Sorry, no I lied and got confused by accident: barebox *DOES* have a display. So sorry. It goes out after, and entering password at appropriate time in boot process doesn't seem to go anywhere | 17:53 |
| minute | andreas-e: try sudo modprobe reform2-lpc, any error? | 17:53 |
| wickedshell | And yes on hdmi, give me a minute to find the appropriate adapters | 17:53 |
| minute | wickedshell: ok great. because if you get an image on hdmi, it will be easier to debug this | 17:53 |
| andreas-e | modprobe: FATAL: Module reform2-lpc not found in directory /lib/modules/6.12.94-mnt-reform-arm64 | 17:54 |
| wickedshell | I got confused because I had tried removing the uSD card before to use the bootloader from eMMC (If I understood the fallbacks correctly and that had no display) | 17:54 |
| minute | wickedshell: yep that makes sense. | 17:54 |
| josch | andreas-e: dpkg -l reform2-lpc-dkms | 17:54 |
| andreas-e | 1.71-2+reform20251209T125945Z+1 | 17:55 |
| andreas-e | Maybe this is an outdated version from before the trixie update? | 17:55 |
| andreas-e | My reform-tools are of the same version. | 17:56 |
| minute | andreas-e: maybe missing kernel headers so it doesn't get built | 17:58 |
| minute | andreas-e: sudo apt install linux-headers-mnt-reform-arm64 | 17:59 |
| josch | andr if that is the case, reform-check will tell you about it | 17:59 |
| josch | andreas-e: also, for any bug you find in the stuff from reform.d.n, first try to reproduce the bug with a fresh MNT system image. If the bug is not there, then it's not a problem for MNT to spend much time on. | 18:01 |
| josch | And in that case: bug report in the Debian bug tracker or directly to me. | 18:01 |
| wickedshell | minute okay, my actual monitor did not work with barebox (4k LG thing so not fully surprised), PiKVM got me a screen through the console boot process, the moment it should have booted to gnome login screen disappeared, gonna try tweaking the settings to see if that gets it back | 18:01 |
| andreas-e | Strange, I did have kernel headers, but not for the latest kernel. Okay josch, I see! | 18:01 |
| andreas-e | minute: Thanks a lot, now I have the module! | 18:02 |
| wickedshell | typing through the gnome login screen gets me into a sway session though | 18:02 |
| wickedshell | And sway is showing a correct understanding of the battery | 18:05 |
| josch | andreas-e: ah nice, glad that your problem is solved! | 18:05 |
| wickedshell | reform panel version 1 | 18:05 |
| wickedshell | Do you want the full output of reform-check? | 18:06 |
| josch | can't hurt :) | 18:06 |
| wickedshell | Right? I asked then realized that :P | 18:06 |
| andreas-e | josch: Well, it is not really solved, since "halt" still does not turn the machine off completely ;) But this is not a problem, I can always do a circle+0. | 18:08 |
| minute | andreas-e: it'll probably work next time, not sure about "halt", i never use that, but try "systemctl poweroff" or "poweroff" next time | 18:09 |
| andreas-e | minute: And many thanks concerning the window manager! "startxfce4 --wayland" works, and I would never have guessed the additional "--wayland" parameter. This printed a message that it needed labwc, and after installing it, I get my xfce environment. | 18:10 |
| minute | andreas-e: very nice & good to know | 18:10 |
| wickedshell | minute josch https://pastebin.com/7yejL5tq | 18:10 |
| andreas-e | Indeed, "poweroff" works! Thanks! | 18:11 |
| minute | wickedshell: ok can you also pastebin your full dmesg? | 18:11 |
| wickedshell | /boot/boot.scr has some wild looking characters at the start | 18:11 |
| minute | wickedshell: also, to clarify, after logging into sway, did the internal display come on or it's still blank and you're only seeing hdmi? | 18:11 |
| wickedshell | standby | 18:11 |
| minute | wickedshell: that's normal @ boot.scr | 18:11 |
| minute | but it's not used by barebox anyway | 18:12 |
| wickedshell | dmesg: https://pastebin.com/GbpTGZ0p | 18:13 |
| minute | wickedshell: many thanks. that shows what happened | 18:15 |
| minute | wickedshell: > dw-mipi-dsi2 fde30000.dsi: command interface is busy | 18:15 |
| minute | wickedshell: > panel-mnt-pocket-reform fde30000.dsi.0: [display v1] failed to set pixel format: -110 | 18:15 |
| wickedshell | wlr-randr does see something on dsi-1 at 1200x1920 at 54.0219999 Hz. The dimesnions seem plausible, but that's an odd rate | 18:16 |
| minute | wickedshell: and so on, so the second time it was initialized, the display got into a bad state | 18:16 |
| minute | wickedshell: that's normal | 18:16 |
| minute | wickedshell: please try the following: | 18:16 |
| - mjw (QUIT: Ping timeout: 256 seconds) (~mjw@206.12.14.47) | 18:16 | |
| minute | wickedshell: swaymsg output DSI-1 disable; swaymsg output DSI-1 enable; swaymsg output DSI-1 mode 1200x1920 | 18:16 |
| wickedshell | Ran, still no output | 18:17 |
| wickedshell | Same round of command intefrace busy/pixel format stuff | 18:17 |
| minute | wickedshell: ok, would you be so kind and pastebin another dmesg? there should be new output | 18:18 |
| minute | wickedshell: ah ok | 18:18 |
| minute | wickedshell: so it doesn't recover | 18:18 |
| wickedshell | I can still get the full dmesg for you though if you want? I know how users interpretaing logs goes :P | 18:18 |
| minute | wickedshell: yes, would be nice | 18:18 |
| wickedshell | minute https://pastebin.com/6L9AnYLb | 18:19 |
| minute | wickedshell: thanks, yeah dsi is stuck somehow. probably the display being upset. btw you also tried this with uboot, right? | 18:20 |
| minute | i did the display v1 tests only with a machine that has uboot... | 18:20 |
| wickedshell | If I pull the uSD card I end up back in uBoot right? (I haven't tried since connecting the monitor) | 18:21 |
| wickedshell | No output after pulling the uSD card that hare barebox on it... | 18:22 |
| minute | wickedshell: also no hdmi output? | 18:22 |
| wickedshell | Yeah | 18:22 |
| minute | huh | 18:22 |
| wickedshell | And notably less power consumption as well when on | 18:22 |
| minute | mhm maybe that uboot is old somehow? | 18:22 |
| wickedshell | Was sitting at ~0.7A idle with barebox, down to ~0.5 with uBoot | 18:22 |
| minute | wickedshell: ah you don't have uboot at all | 18:22 |
| minute | wickedshell: otherwise you wouldn't be able to boot barebox from sd card | 18:23 |
| minute | wickedshell: try reinstalling it by doing what reform-check suggests: sudo reform-flash-bootloader emmc | 18:23 |
| wickedshell | Ah, fair :P | 18:23 |
| wickedshell | Wilco | 18:23 |
| minute | (you can always nuke it again) | 18:23 |
| wickedshell | To be fair I haven't actually installed openbsd yet which was the point of barebox for me. Unfrotunately embedded rust development on it doesn't really work on openbsd that I've been able to solve, at least without having to manually maintain the full toolchain :/ | 18:25 |
| wickedshell | just had the oddest experience with uBoot where it picked up a vaguely functional arch install off the uSD card (apparently that's the last thing I used it for) | 18:28 |
| minute | lol | 18:28 |
| wickedshell | I entered my FDE password then got dropped to a an arch linux arm login prompt... | 18:29 |
| wickedshell | ANyways, still no display on a uboot system | 18:29 |
| minute | hm, uart/usb issue not reproducible anymore after re-plugging in the jsts in the keyboard | 18:29 |
| minute | wickedshell: also no HDMI? | 18:29 |
| wickedshell | HDMI is back with uBoot now | 18:29 |
| minute | wickedshell: aha, good | 18:30 |
| wickedshell | And I'm in again | 18:30 |
| minute | wickedshell: ok then your display v1 behaves slightly differently than other display v1, i've kind of feared that | 18:30 |
| wickedshell | There are a pair of couldn't parse response messages from system controller in the log as well | 18:30 |
| minute | i guess there are two variants, but don't yet know how to recognize them | 18:30 |
| minute | wickedshell: yeah those are normally harmless because there's tons of retries | 18:30 |
| minute | wickedshell: i'll need to dig around or boxes to see if i can find another display v1 that fails in the same way | 18:31 |
| minute | s/or/our | 18:31 |
| minute | wickedshell: now it would be best if you went back to barebox, and then interrupt the boot ("m") and then select an older kernel | 18:31 |
| wickedshell | Well happy to be remote hands/testing of anything you want, I have to admit that display stuff is definitly outside my wheelhouse | 18:31 |
| minute | wickedshell: this way we can see if the issue is mostly in the panel driver or also connected to SC FW | 18:32 |
| minute | wickedshell: like, possibly the display comes back up with older kernel from before last night | 18:32 |
| wickedshell | If I go back to the older kernel I also need to roll the firmware on the keyboard/controller back right? | 18:32 |
| minute | wickedshell: not immediately | 18:32 |
| minute | wickedshell: lets try the old kernel only first | 18:32 |
| wickedshell | wilco | 18:32 |
| + reform23140 (~lnsy@107-207-30-23.lightspeed.irvnca.sbcglobal.net) | 18:32 | |
| * reform23140 -> lnsy | 18:34 | |
| - lnsy (QUIT: Client Quit) (~lnsy@107-207-30-23.lightspeed.irvnca.sbcglobal.net) | 18:34 | |
| minute | cararemixed: i pusheda potential fix for your charger issues | 18:35 |
| minute | s/pusheda/pushed a/ | 18:36 |
| wickedshell | minute display on old one, although it now thinks battery is 6% and not charging (actually is 100% and plugged in) | 18:38 |
| cararemixed | minute: the 5 retries commit? I can try and compile and flash here or wait for CI. | 18:38 |
| wickedshell | Felt slower to boot, but I can't actually defend that | 18:39 |
| wickedshell | Will grab you dmesg though, as I do see some stuff that maybe means more to you | 18:40 |
| cararemixed | wickedshell: 6% was what I used to get on the old lpc module + firmware when the module wasn't initialized properly. Reloading would usually fix it. | 18:40 |
| wickedshell | minute https://pastebin.com/xhSitxFL with old kernel | 18:41 |
| cararemixed | ACTION is running build.sh locally. | 18:42 |
| cararemixed | Seems like there might be an issue with this patch. https://usercontent.irccloud-cdn.com/file/LWbs2WVz/1786121097.JPG | 18:45 |
| cararemixed | minute: ☝️ | 18:45 |
| cararemixed | https://usercontent.irccloud-cdn.com/file/o2zxAXRT/IMG_7800.JPG | 18:48 |
| minute | cararemixed: mhm, i mean during the update that is to be expected | 18:49 |
| minute | cararemixed: maybe needs a standby switch power cycle. or sometimes it hangs if updated at an unfortunate time and then needs batteries reconnected. didn't happen to me in a while though and i did a lot of updates | 18:51 |
| minute | wickedshell: that is normal and expected @ 6% | 18:51 |
| + mjw (~mjw@206.12.14.47) | 18:51 | |
| minute | wickedshell: but cool that the display works. so the problem will be fixable by tuning the panel driver. | 18:52 |
| cararemixed | I'll fiddle with the controller power. One sec... gotta fetch some tools. | 18:52 |
| minute | wickedshell: i just found a display here that exhibits the same issue. | 18:52 |
| cararemixed | minute: hmm. well cutting power left it dead until I plugged it in (with charged batteries). | 18:58 |
| cararemixed | I physically disconnected my batteries in this case as my side switch snapped at some point so it's stuck on. | 18:58 |
| cararemixed | Let me see if it crashes again with some plugging/unplugging. | 18:59 |
| cararemixed | Yeah. It crashes hard on me consistently when starting with a PD cable plugged into the MNT first battery second. | 19:00 |
| cararemixed | Ok. Not consistent but twice in a row | 19:02 |
| cararemixed | Once it showed a ridiculous current though. Over 7A which seems very wrong. | 19:03 |
| cararemixed | -7A that is | 19:03 |
| cararemixed | So potentially something is triggering a fast charge state and killing the circuit. | 19:03 |
| cararemixed | The module still functions fine. So it doesn't appear to have damaged anything if that rate is to believed. | 19:04 |
| cararemixed | I can still crash it by leaving it plugged in on the pocket side and unplugging it on the other after repeated attempts. | 19:05 |
| - RandyK (QUIT: Ping timeout: 252 seconds) (~RandyK@user/randyk) | 19:05 | |
| + RandyK (~RandyK@user/randyk) | 19:06 | |
| cararemixed | It only happens on my 100W rated negotiated port which both charges and discharges the battery. | 19:08 |
| cararemixed | The other port is fine after repeated attempts. | 19:09 |
| cararemixed | Hope this helps? I also have more than one of these batteries so ... if you need samples to reproduce LMK. Seems like I have all the broken hardware bits. | 19:10 |
| jfred | minute: looks like I have the same DSI issue with display v1 that wickedshell has | 19:12 |
| - aloo_shu (QUIT: Ping timeout: 265 seconds) (~aloo_shu@90.166.99.148) | 19:12 | |
| + aloo_shu_ (~aloo_shu@85.51.16.21) | 19:13 | |
| * aloo_shu_ -> aloo_shu | 19:14 | |
| minute | jfred: thanks for reporting, i just saw that i must have introduced some last minute bug, because the main testing disp v1 also doesn't work here, mea culpa. working now to correct this | 19:14 |
| jfred | no worries, thanks for working on it :) | 19:15 |
| minute | cararemixed: hmm 7A can't be true | 19:15 |
| minute | cararemixed: like, i'm sure that's just garbled data | 19:15 |
| cararemixed | Yeah. I didn't get a screen shot in time. | 19:15 |
| cararemixed | Probably. Before it crashes it will freeze and hold some value for 5 seconds or so and then hit 0% | 19:16 |
| - Gooberpatrol66 (QUIT: Read error: Connection reset by peer) (~Gooberpat@user/gooberpatrol66) | 19:16 | |
| cararemixed | Power cycle requires unplugging batteries, plugging them back in and then ASLO plugging power in to get the controller to turn on for whatever reason. Otherwise it remains unresponsive. | 19:16 |
| minute | cararemixed: hmm thanks for the report in any case, i need to do more testing to repro this / will get back to you about sampling your power supplies. are you able to boot to flash back the older version? | 19:20 |
| minute | cararemixed: plugging the power in after replugging the batteries is normal. it unlocks the battery safety | 19:20 |
| minute | cararemixed: i mean, that it would only start working then | 19:20 |
| minute | cararemixed: i'm gonna need to fix the disp v1 issue now and then i can get back on that issue. | 19:21 |
| cararemixed | minute: I can use the current version to boot. I just had to hard power cycle the controller and avoid the cursed wiring sequence. | 19:21 |
| cararemixed | I'll revert if I run into the crash again but I'm curious if there are other issues so I'll keep it and roll forward with the repo if things work otherwise. | 19:22 |
| minute | cararemixed: ah ok, now i get it | 19:22 |
| minute | cararemixed: many thanks! | 19:22 |
| - bremner (QUIT: Remote host closed the connection) (~bremner@notmuch/developer/bremner) | 19:30 | |
| + bremner (~bremner@notmuch/developer/bremner) | 19:30 | |
| - mjw (QUIT: Ping timeout: 264 seconds) (~mjw@206.12.14.47) | 19:38 | |
| andreas-e | I have never used the HDMI output before, and just tried. Nothing happens, neither from the tty, nor from xfce4 (where in the settings the external monitor does not show up). On the very old Reform version, do I have to do anything special? | 19:41 |
| wickedshell | andreas-e not sure if it's relevant and I didn't really dive into it, but my 4K monitor wasn't very happy to talk to it, but not sure if that's some cabling thing, swapping to pikvm at a lower resolution was fine. But this is honestly the first time I've tried using it, so I wouldn't take that statement too far | 19:44 |
| minute | ah might have found the problem with the panel | 19:44 |
| minute | wickedshell: andreas-e: that problem is imx8mq specific | 19:44 |
| minute | andreas-e: there are 2 different dtb modes. dual and single screen | 19:45 |
| andreas-e | Single screen would be enough; that means either internal or external | 19:45 |
| minute | andreas-e: check the tool reform-display-config: https://source.mnt.re/reform/reform-tools/-/blob/main/bin/reform-display-config?ref_type=heads | 19:45 |
| minute | andreas-e: that's not how it works though... | 19:46 |
| minute | andreas-e: if you need hdmi, do: sudo reform-display-config dual | 19:46 |
| minute | andreas-e: and then reboot. hopefully it works | 19:46 |
| wickedshell | Interesting note on wifi btw vs the imx8m+ vs rk: I left the external wifi card mnt resells that I had in the imx setup, and it performs sooo, sooo, soooo much better on the rk which I hadn't expected | 19:46 |
| minute | ah it might also be that imx8mq hdmi is currently broken, but not sure, i think josch was trying to fix it for a while | 19:46 |
| wickedshell | (same antenna placement) | 19:48 |
| andreas-e | minute: Thanks, I will give it a try, unless josch tells me otherwise. My fear would only be that after doing some hard changes, no display will actually work... And I am confused why this is necessary. Why would one choose single then? | 19:49 |
| - aloo_shu (QUIT: Remote host closed the connection) (~aloo_shu@85.51.16.21) | 19:49 | |
| + aloo_shu (~aloo_shu@85.51.16.21) | 19:51 | |
| minute | andreas-e: i spent a lot of time doing that workaround back then. it has to do with the bus bandwidth on that chip | 19:52 |
| minute | andreas-e: and the available display controllers | 19:52 |
| minute | andreas-e: on single mode, the better display controller (DCSS) can serve the internal display. on HDMI, it's needed for HDMI. so it's a tradeoff | 19:52 |
| minute | andreas-e: on dual mode, i mean. | 19:52 |
| minute | andreas-e: so in single mode, the internal display has higher refresh and the gpu should be a bit faster because of framebuffer compression | 19:53 |
| andreas-e | Okay, I see. | 19:53 |
| andreas-e | Interesting! | 19:53 |
| + mjw (~mjw@206.12.14.47) | 19:55 | |
| minute | i mean that chip is very old and they had a bunch more constraints... | 19:55 |
| minute | > The i.MX 8M series were announced on January 4 at CES 2017 | 19:55 |
| minute | still never seen an imx8QM though. that was the one i wanted to use originally... but not sure if it's GPU is even supported _now_ | 19:57 |
| minute | also supposedly very hot | 19:57 |
| - brennen (QUIT: Quit: WeeChat 2.8) (~brennen@user/brennen) | 19:58 | |
| + brennen (~brennen@user/brennen) | 19:58 | |
| andreas-e | HDMI works! Thanks a lot! | 20:04 |
| minute | andreas-e: great! | 20:09 |
| josch | minute: i didn't test imx8mq hdmi of kernel 7.0 or later but since andreas-e has the debian stable kernel, that "should" work. | 20:09 |
| josch | ah okay it works :) | 20:09 |
| andreas-e | The kernel is 6.12.94. | 20:09 |
| josch | andreas-e: if it works then that's all that matters | 20:11 |
| josch | i haven't spent time with imx8mq for a few months now, so no idea what does and doesn't work with modern kernels :( | 20:11 |
| andreas-e | Definitely! And I understand that one would rather spend time with a newer processor module. That said, for a given task you would spend more time with the old one ;-) | 20:13 |
| josch | andreas-e: imx8mq is a never-ending saga. The patch was at version 22 (!!) and then its author Sandor Yu left NXP. Work was now resumed by Laurentiu Palcu (also from NXP) but the last message on dri-devel from May 19 with v23 of the patch stack only has replies from a claude review bot as far as i can see... | 20:19 |
| andreas-e | Oh, so the bets are up whether this will arrive before the last person stops using the module! | 20:22 |
| + Gooberpatrol66 (~Gooberpat@user/gooberpatrol66) | 20:23 | |
| andreas-e | Would this end up in the kernel or in a driver? | 20:30 |
| minute | i believe the imx8mq has a 15 year guaranteed production time | 20:30 |
| + AnimaInvicta (~AnimaInvi@88-169-191-86.subs.proxad.net) | 20:32 | |
| - _justin_kelly4 (QUIT: Ping timeout: 252 seconds) (~justinkel@user/justin-kelly/x-6011154) | 20:51 | |
| + lidstah2 (~lidstah@gateway/tor-sasl/lidstah) | 20:54 | |
| - lidstah (QUIT: Ping timeout: 252 seconds) (~lidstah@gateway/tor-sasl/lidstah) | 20:56 | |
| - lidstah2 (QUIT: Remote host closed the connection) (~lidstah@gateway/tor-sasl/lidstah) | 21:04 | |
| + lidstah (~lidstah@gateway/tor-sasl/lidstah) | 21:06 | |
| - lidstah (QUIT: Remote host closed the connection) (~lidstah@gateway/tor-sasl/lidstah) | 21:06 | |
| + lidstah (~lidstah@gateway/tor-sasl/lidstah) | 21:07 | |
| - lidstah (QUIT: Remote host closed the connection) (~lidstah@gateway/tor-sasl/lidstah) | 21:07 | |
| + lidstah (~lidstah@gateway/tor-sasl/lidstah) | 21:08 | |
| + reform26232 (~esi@stud-0ed94.gardur.hi.is) | 21:12 | |
| - reform26232 (QUIT: Client Quit) (~esi@stud-0ed94.gardur.hi.is) | 21:12 | |
| - wickedshell (QUIT: Quit: Client closed) (~wickedshe@2601:8c0:c7c:3572::1b90) | 21:18 | |
| + Esi (sid14426@helmsley.irccloud.com) | 21:28 | |
| Esi | I am now a proud owner of a MNT Pocket. thank you those responsible for an amazing device <3 | 21:29 |
| andreas-e | Nice! | 21:31 |
| - rodion_goritskov (QUIT: Ping timeout: 264 seconds) (~rodion_go@mail.goritskov.com) | 21:32 | |
| minute | Esi: awesome! | 21:32 |
| Esi | some possible "fun" problems but i am traversing the forum and so on to see if there's anything there before I create a new post or ask here :3 | 21:33 |
| minute | ok i made some progress on the dispv1 issue | 21:38 |
| minute | the main thing that's disturbing it is the reset :D | 21:38 |
| - mjw (QUIT: Ping timeout: 264 seconds) (~mjw@206.12.14.47) | 21:39 | |
| + elb (~elb@user/elb) | 21:43 | |
| josch | Esi: if this channel is silent, there is little reason not to ask :) | 21:50 |
| minute | fix https://source.mnt.re/reform/reform-debian-packages/-/commit/d727cac29590ce7d4b872dde106662c706f764c9 | 22:03 |
| Esi | hmmm, should the gnome-software be completely empty with no repos? 🤔 | 22:04 |
| minute | Esi: yeah gnome-software is a little strange when you don't have flathub set up. otherwise it will only fill up later with debian packages, i think after your first "sudo apt update" in a terminal... | 22:11 |
| minute | kind of defeats the purpose of a GUI... | 22:11 |
| josch | maybe we haven't set up gnome right. I think the packagekit component is responsible to check for updates and maybe it should but doesn't do that? | 22:13 |
| minute | on my own systems i always mask the packagekit services because they're so disruptive... | 22:16 |
| minute | we could also look into shipping bazaar instead, and replace it on the dock... just confusing when it shows up in applications search | 22:17 |
| minute | not sure if gnome-software can just be uninstalled though | 22:18 |
| minute | nope, gnome-core depends on it | 22:19 |
| minute | but maybe packagekit can be customized in a way that it won't do stuff automatically in the background all the time and peg the cpu at 100% | 22:20 |
| Esi | so i guess gnome-software should be only used for flatpaks and such on the pocket? or at least that's the intention? | 22:27 |
| - S0rin (QUIT: Quit: WeeChat 4.6.3) (~S0rin@user/s0rin) | 22:30 | |
| minute | Esi: it's bundled with Gnome. and it can work for debian packages. it just needs to be nudged a bit to fill its catalog | 22:33 |
| minute | Esi: i think if you do a `sudo apt update` and then restart gnome-software a bit later, it should be there... | 22:34 |
| - lidstah (QUIT: Remote host closed the connection) (~lidstah@gateway/tor-sasl/lidstah) | 22:45 | |
| + lidstah (~lidstah@gateway/tor-sasl/lidstah) | 22:46 | |
| + S0rin (~S0rin@user/s0rin) | 22:54 | |
| + mjw (~mjw@206.12.14.68) | 23:01 | |
| rick_ | just get rid of gnome, problem solved (is it visible that i don't really like gnome) XDD | 23:03 |
| - lidstah (QUIT: Remote host closed the connection) (~lidstah@gateway/tor-sasl/lidstah) | 23:03 | |
| + lidstah (~lidstah@gateway/tor-sasl/lidstah) | 23:04 | |
| rick_ | one a more serious side.. wouldn't it be better to ship a more lightwight desktop env in general? like xfce? i wish xfce would finally have full wayland support | 23:06 |
| rick_ | i do love my sway, but i understand thats its to 'strange' for some users | 23:06 |
| + lidstah2 (~lidstah@gateway/tor-sasl/lidstah) | 23:06 | |
| - lidstah (QUIT: Remote host closed the connection) (~lidstah@gateway/tor-sasl/lidstah) | 23:07 | |
| josch | rick_: before gnome there was wayland as another option to sway but it had rougher edges than gnome and received far less support and had more bugs even though it is more simple. | 23:07 |
| Esi | minute: apt update done and restarted gnome-software. no luck. also tried restarting the pocket | 23:08 |
| rick_ | josch ahh sad.. btw you must mean sth different than wayland, am i right? wayland is the 'protocol', just like x11 | 23:10 |
| josch | ups | 23:10 |
| josch | rick_: i meant wayfire | 23:10 |
| josch | rick_: we had our own reform wayfire theme but that's discontinued now | 23:11 |
| rick_ | josch oh? never heard of that.. need to check that out 0.0 | 23:11 |
| josch | rick_: https://community.mnt.re/t/intention-to-remove-reform-firedecor-for-wayfire-from-debian/4009 | 23:13 |
| rick_ | josch oh thanks! | 23:13 |
| Esi | https://www.irccloud.com/pastebin/ePSuh1SX/gnomesoftware | 23:14 |
| - lidstah2 (QUIT: Remote host closed the connection) (~lidstah@gateway/tor-sasl/lidstah) | 23:16 | |
| + lidstah (~lidstah@gateway/tor-sasl/lidstah) | 23:19 | |
| Esi | installing gnome-software-plugin-deb got me halfway there. it lists installed software now at least | 23:29 |
| Esi | sudo apt upgrade and a restart and it's working fine now | 23:30 |
| josch | minute: gnome-software-plugin-deb is a Recommends of gnome-software and we don't install Recommends in mkimage.sh | 23:31 |
| josch | amospalla: reform-tools 1.88 is in the reform.debian.net repo for trixie-backports now | 23:33 |
| andreas-e | rick_: This picks up some discussion we had earlier today. I tried to launch wayfire (it is suggested together with sway when logging in), but only got a black screen with a cursor. Then I ended up installing xfce, which needs to be started with "startxfce4 --wayland". | 23:56 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
