| minute | unfortunately my pocket doesn't have any more wifi reception here, so can't try atm | 00:05 |
|---|---|---|
| minute | (the hotel recently messed something up with the router, so only my next gets 1-2 bars of 2.4ghz reception) | 00:05 |
| cararemixed | running my migrate at the moment josch. I'll try it momentarily | 00:05 |
| minute | thx cararemixed! | 00:06 |
| + ohnoitsnoah (~ohnoitsno@static-23-234-92-117.cust.tzulo.com) | 00:06 | |
| - ohnoitsnoah_ (QUIT: Ping timeout: 264 seconds) (~ohnoitsno@72.181.111.171) | 00:06 | |
| cararemixed | josch: Specific things I should be reproducing beyond the upgrade? | 00:07 |
| josch | cararemixed: you could create a file /etc/apt/preferences.d/mozjs and fill it with: | 00:08 |
| josch | Package: libmozjs-140-0 | 00:09 |
| josch | Pin: version 140.6.0-1 | 00:09 |
| josch | Pin-Priority: 1001 | 00:09 |
| josch | Use apt-cache policy to verify that the pinning worked | 00:09 |
| josch | once you have this in place you can do "apt full-upgrade" without upgrading libmozjs-140-0 | 00:09 |
| josch | and if the theory is correct, your gdm will not break | 00:10 |
| cararemixed | Ok. Will do | 00:10 |
| cararemixed | hmmm. | 00:10 |
| cararemixed | Random glitch I get on power on sometimes. https://usercontent.irccloud-cdn.com/file/NiXJYLk0/IMG_7212.JPG | 00:11 |
| cararemixed | I power down and back up and it fixes. | 00:11 |
| josch | minute: to find the offending package i bisected snapshot.d.o timestamps and generated system images for each with this MR https://source.mnt.re/reform/reform-system-image/-/merge_requests/127 because i cannot be bothered to click through the setup wizard on each bisection step. You wondered if this is necessary. The current bug would make a feature like this useful (but i'm not finished yet) | 00:11 |
| josch | cararemixed: i have this too sometimes | 00:12 |
| cararemixed | josch: Just double checked the pin. Doing the full upgrade npw | 00:15 |
| minute | cararemixed: yeah, that glitch is known... some intermittent timing issue in the panel driver | 00:16 |
| minute | josch: ack | 00:17 |
| cararemixed | I remember I got it when I first received the pocket and thought I had a broken machine. Fortunately a quick powercycle got it on the right track. | 00:17 |
| minute | i think we can tackle this issue soon, kcxt was also looking into rewriting the driver | 00:20 |
| cararemixed | Ok now the only slow part other than slow wifi... | 00:21 |
| cararemixed | https://usercontent.irccloud-cdn.com/file/ThTfuwKr/IMG_7213.JPG | 00:22 |
| minute | cararemixed: we had this issue too at the office. turns out it was ipv6 related | 00:22 |
| minute | cararemixed: so we turn off ipv6 when installing system images... | 00:22 |
| cararemixed | Anyone know why my machine always hangs here. For indeterminate time. Sometimes over 30 mins | 00:22 |
| minute | cararemixed: yes, must be the same issue | 00:23 |
| cararemixed | Anyone know whyHmmm | 00:23 |
| cararemixed | My wifi is glitching now too. Hah | 00:23 |
| cararemixed | Switched to using my phone to chat. | 00:23 |
| minute | cararemixed: maybe you also have telekom, or fritzbox router? | 00:23 |
| cararemixed | It’s a cheap Vodafone box. | 00:24 |
| cararemixed | Always giving me issues. | 00:24 |
| cararemixed | I’m going to tether to my phone I think. It’ll pass quicker | 00:24 |
| minute | sudo sysctl -w net.ipv6.conf.all.disable_ipv6=1 | 00:25 |
| minute | this is what made that work for us in the mnt hq :D | 00:26 |
| minute | (and otherwise i usually remove apt-listbugs because it can be otherwise slow, but that's not recommended) | 00:26 |
| - ohnoitsnoah (QUIT: Ping timeout: 240 seconds) (~ohnoitsno@static-23-234-92-117.cust.tzulo.com) | 00:26 | |
| cararemixed | Right. | 00:26 |
| + ohnoitsnoah_ (~ohnoitsno@2603:8081:3c00:59f3::10a9) | 00:27 | |
| cararemixed | I switched to my phone anyway as I seem to be having service issues outside of that right now. | 00:27 |
| minute | alright, and there it works? | 00:27 |
| cararemixed | Yeah. Tethering on my phone is always faster too. | 00:27 |
| cararemixed | Just a telekom eSim but it works when I hit issues. | 00:28 |
| minute | nice | 00:28 |
| cararemixed | generating initrd. Will reboot shortly. | 00:28 |
| cararemixed | Works | 00:30 |
| cararemixed | Thanks josch | 00:30 |
| minute | awesome, thanks for confirming cararemixed | 00:31 |
| josch | cararemixed: thank you! | 00:31 |
| - chomwitt (QUIT: Ping timeout: 256 seconds) (~chomwitt@2a02:85f:9a0a:f200:1ac0:4dff:fedb:a3f1) | 00:33 | |
| cararemixed | Still seeing emacs crash. I haven’t had a chance to hunt this one down. https://usercontent.irccloud-cdn.com/file/lRFH46pl/1768779407.JPG | 00:37 |
| cararemixed | So I guess it’s not my local changes accumulated over time. | 00:37 |
| cararemixed | Specifically crashes in some X code paths. | 00:38 |
| cararemixed | I spend more time with neovim and helix but this one puzzles me for the few times I want to run emacs | 00:40 |
| cararemixed | I need to get some sleep though. seems like more compiler flag issues here. Probably another library to pin. | 00:50 |
| cararemixed | The jit makes it a bit harder to understand so I need to figure out how to disable it. ltrace shows a few useful leads at least but my brain isn’t awake enough to do this right now. | 00:53 |
| cararemixed | I think it’s tree sitter. Not sure yet. | 00:56 |
| - siviq (QUIT: Quit: Client closed) (~siviq@user/siviq) | 00:56 | |
| - pomel0 (QUIT: Ping timeout: 260 seconds) (~pomel0@user/pomel0) | 01:00 | |
| josch | aha, we were too slow: | 01:01 |
| josch | https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1125777 | 01:01 |
| josch | so rebuilding gjs will fix this | 01:03 |
| + gianmarcogg03_ (~quassel@user/gianmarcogg03) | 01:06 | |
| cararemixed | Hmm. Actually emacs -nw works so now I don’t know. Might actually be some of the x interactions after all. | 01:07 |
| - gianmarcogg03 (QUIT: *.net *.split) (~quassel@user/gianmarcogg03) | 01:09 | |
| - cow321 (QUIT: *.net *.split) (~deflated8@user/meow/deflated8837) | 01:09 | |
| - angelwood (QUIT: *.net *.split) (~angelwood@user/angelwood) | 01:09 | |
| - rick_ (QUIT: *.net *.split) (~m-x42ph2@m.n0id.space) | 01:09 | |
| + cow321 (~deflated8@user/meow/deflated8837) | 01:14 | |
| - spew (QUIT: Quit: nyaa~) (~spew@user/spew) | 01:16 | |
| + enwu1 (~enwu@user/enwu) | 01:16 | |
| + angelwood (~angelwood@user/angelwood) | 01:17 | |
| + ex--parrot (~fincham@user/ex-parrot) | 01:18 | |
| + buckket1 (~buckket@vps.buckket.org) | 01:20 | |
| + timide_ (~timide@user/timide) | 01:21 | |
| - manis (QUIT: Ping timeout: 246 seconds) (01a66df340@185.72.67.185) | 01:21 | |
| - ex-parrot (QUIT: Ping timeout: 246 seconds) (~fincham@user/ex-parrot) | 01:21 | |
| - enwu (QUIT: Ping timeout: 246 seconds) (~enwu@user/enwu) | 01:21 | |
| - buckket (QUIT: Ping timeout: 246 seconds) (~buckket@vps.buckket.org) | 01:21 | |
| - timide (QUIT: Ping timeout: 246 seconds) (~timide@user/timide) | 01:21 | |
| * enwu1 -> enwu | 01:21 | |
| + pomel0 (~pomel0@user/pomel0) | 01:27 | |
| josch | the problem was also seen on a apple macbook pro with the apple m1 chip | 01:35 |
| josch | the problem should be taken care of soon -- the affected DD said they will perform the fixing upload of gjs | 01:36 |
| - bkeys (QUIT: Ping timeout: 240 seconds) (~Thunderbi@134.22.115.162) | 01:59 | |
| - pomel0 (QUIT: Ping timeout: 256 seconds) (~pomel0@user/pomel0) | 01:59 | |
| + pomel0 (~pomel0@user/pomel0) | 01:59 | |
| + bkeys (~Thunderbi@98.sub-174-229-19.myvzw.com) | 02:03 | |
| + bkeys1 (~Thunderbi@h69.128.19.98.dynamic.ip.windstream.net) | 02:28 | |
| - bkeys (QUIT: Read error: Connection reset by peer) (~Thunderbi@98.sub-174-229-19.myvzw.com) | 02:30 | |
| * bkeys1 -> bkeys | 02:30 | |
| - vagrantc (QUIT: Ping timeout: 244 seconds) (~vagrant@2600:3c01:e000:21:7:77:0:50) | 02:31 | |
| - paperManu (QUIT: Ping timeout: 240 seconds) (~paperManu@146.71.9.156) | 02:34 | |
| - mjw (QUIT: Ping timeout: 244 seconds) (~mjw@gnu.wildebeest.org) | 02:55 | |
| - AnimaInvicta (PART: !!unknown attribute: msg!!) (~AnimaInvi@88-120-179-216.subs.proxad.net) | 02:55 | |
| - bkeys (QUIT: Read error: Connection reset by peer) (~Thunderbi@h69.128.19.98.dynamic.ip.windstream.net) | 03:08 | |
| + bkeys (~Thunderbi@98.19.128.69) | 03:08 | |
| - bkeys (QUIT: Read error: Connection reset by peer) (~Thunderbi@98.19.128.69) | 03:19 | |
| + bkeys (~Thunderbi@98.19.128.69) | 03:19 | |
| elb | ok I just reproduced the cryptsetup: waiting for encrypted source device boot failure, let me see what happens when I get into initramfs | 03:21 |
| elb | no nvme devices | 03:21 |
| elb | (replied on the forum) | 03:31 |
| - nsc (QUIT: Ping timeout: 252 seconds) (~nicolas@i5C74DCE4.versanet.de) | 03:51 | |
| + nsc (~nicolas@16-97-142-46.pool.kielnet.net) | 03:53 | |
| bremner | cararemixed: if you have a reproducer starting with "emacs -Q", I can test it | 04:30 |
| bremner | oof, getting 21kb/s fetching collabora's kernel repo. | 04:36 |
| bremner | A long shot, but are there any mirrors for https://gitlab.collabora.com/hardware-enablement/rockchip-3588/linux ? | 05:06 |
| - voltaire28_ (QUIT: Ping timeout: 240 seconds) (~jlafon@28.162.2.93.rev.sfr.net) | 05:11 | |
| - Ar|stote|is (QUIT: Quit: https://quassel-irc.org - Chat comfortably. Anywhere.) (~linx@149.210.3.83) | 05:25 | |
| + Ar|stote|is (~linx@149.210.3.83) | 05:26 | |
| - Ar|stote|is (QUIT: Client Quit) (~linx@149.210.3.83) | 05:27 | |
| + Ar|stote|is (~linx@149.210.3.83) | 05:27 | |
| - Ar|stote|is (QUIT: Client Quit) (~linx@149.210.3.83) | 05:30 | |
| + Ar|stote|is (~linx@149.210.3.83) | 05:31 | |
| + manis (01a66df340@185.72.67.185) | 05:31 | |
| - Ar|stote|is (QUIT: Client Quit) (~linx@149.210.3.83) | 05:33 | |
| + Ar|stote|is (~linx@149.210.3.83) | 05:33 | |
| - schalken (QUIT: Quit: Leaving) (~schalken@117-118-178-69.gci.net) | 05:43 | |
| + schalken (~schalken@117-118-178-69.gci.net) | 05:43 | |
| + johl (~johl@wikidata/Jens-Ohlig) | 06:12 | |
| - johl (QUIT: Ping timeout: 246 seconds) (~johl@wikidata/Jens-Ohlig) | 06:21 | |
| bremner | nvm. Be the mirror you want to see in the world I gues | 06:50 |
| bremner | wish me luck, rebooting into 6.19-rc6, too lazy to reattach serial console... | 06:50 |
| bremner | cool it booted | 06:51 |
| + chomwitt (~chomwitt@2a02:85f:9a0a:f200:1ac0:4dff:fedb:a3f1) | 07:22 | |
| josch | [i-also-like-to-live-dangerously.jpg] | 07:44 |
| * gianmarcogg03_ -> gianmarcogg03 | 09:01 | |
| josch | minute: libgjs and libmozjs are fixed in unstable now. Using Debian testing would not've helped here btw. The bug also affected testing. | 10:03 |
| minute | josch: oh wow ok | 10:58 |
| minute | josch: we could and probably should anyway set up a CI system @ mnt that regularly does apt upgrades and reboots and checks if the desktop comes up | 10:59 |
| minute | i feel like there should be some framework for this? | 10:59 |
| josch | in debian this is done by openqa | 11:04 |
| josch | it takes screenshots (but i'm told it could also have a camera connected to test "the screen turns on") and then compares to the expected image or does OCR to find text or click buttons | 11:04 |
| josch | for bugs like the last, it would've already been cought by a simple qemu vm doing upgrade tests | 11:06 |
| Svp | seems theres been talks about removing gtk2 from debian, hm... | 11:34 |
| minute | josch: right | 11:36 |
| minute | Svp: oh? which apps are still using gtk2? (was kicad/wxwidgets ever upgraded to gtk3+?) | 11:37 |
| Svp | off the top of my head i can only think of hexchat | 11:37 |
| minute | josch: i can post PSA now that it's safe to upgrade / and that people who have access to their borked system should apt update and apt upgrade to fix gdm? (or just apt update and apt install libmozjs-140-0?) | 11:38 |
| josch | minute: i didn't verify this myself yet. (at work right now) I'll try to flash a system image and test this as soon as I can (maybe before 15:00) but if you beat me to it please go ahead :) | 11:41 |
| minute | josch: i'll hold my horses :D i have very bad wifi here since yesterday, so not taking risks until that's fixed | 11:45 |
| minute | (hotel told me they'll try to fix it today) | 11:45 |
| amk | Hey I've a small patch for reform-power-daemon, would it be possible to get my gitlab account approved? Thanks :) | 12:03 |
| + mjw (~mjw@gnu.wildebeest.org) | 12:23 | |
| minute | amk: sure, what's the username? | 12:35 |
| amk | i think i put down amk | 12:35 |
| + paperManu (~paperManu@146.71.9.156) | 12:48 | |
| minute | amk: ok, check if i approved the right account :D | 12:49 |
| amk | minute: yep im in, thanks! | 12:50 |
| minute | great | 12:54 |
| - chomwitt (QUIT: Ping timeout: 250 seconds) (~chomwitt@2a02:85f:9a0a:f200:1ac0:4dff:fedb:a3f1) | 12:55 | |
| josch | sorry, will take longer i don't have an sd-card with me | 13:23 |
| - lidstah (QUIT: Ping timeout: 252 seconds) (~lidstah@gateway/tor-sasl/lidstah) | 13:30 | |
| + lidstah (~lidstah@gateway/tor-sasl/lidstah) | 13:32 | |
| - mjw (QUIT: Ping timeout: 246 seconds) (~mjw@gnu.wildebeest.org) | 13:36 | |
| * Guest713 -> mjw | 13:58 | |
| + siviq (~siviq@user/siviq) | 13:59 | |
| - paperManu (QUIT: Ping timeout: 246 seconds) (~paperManu@146.71.9.156) | 14:04 | |
| cararemixed | bremner: this is w/o config so same as -Q, it will crash in GUI mode before rendering a window. | 14:14 |
| cararemixed | My spouse is in town so I am spending a little less time at the keyboard. Also recovering (TMI: Unq n HGV juvpu ernyyl ehvarq zl jrrxraq) | 14:18 |
| cararemixed | I also need to focus on writing the charter for a little collective I'm starting (glitchwitches.org). I keep putting it off and tinkering with things. | 14:19 |
| josch | cararemixed: ouch -- speedy recovery to you! | 14:21 |
| + paperManu (~paperManu@modemcable141.205-200-24.mc.videotron.ca) | 14:30 | |
| - pomel0 (QUIT: Remote host closed the connection) (~pomel0@user/pomel0) | 14:50 | |
| - bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@98.19.128.69) | 14:51 | |
| + bkeys (~Thunderbi@98.19.128.69) | 14:52 | |
| - siviq (QUIT: Quit: Client closed) (~siviq@user/siviq) | 15:05 | |
| josch | minute: could you please enable gitlab CI for https://source.mnt.re/amk/reform-tools/ thank you! | 15:18 |
| - gianmarcogg03 (QUIT: Ping timeout: 260 seconds) (~quassel@user/gianmarcogg03) | 15:30 | |
| + gianmarcogg03 (~quassel@user/gianmarcogg03) | 15:30 | |
| - RandyK (QUIT: Ping timeout: 252 seconds) (~RandyK@user/randyk) | 15:31 | |
| + RandyK (~RandyK@user/randyk) | 15:31 | |
| minute | josch: ok done! | 15:39 |
| josch | thank you! | 15:40 |
| josch | i'm now making sure that the libmozjs/gdm issue is resolved | 15:40 |
| minute | nice, many thanks! | 15:40 |
| + spew (~spew@user/spew) | 15:43 | |
| - bkeys (QUIT: Ping timeout: 264 seconds) (~Thunderbi@98.19.128.69) | 16:06 | |
| + chomwitt (~chomwitt@2a02:85f:9a0a:f200:1ac0:4dff:fedb:a3f1) | 16:09 | |
| + bkeys (~Thunderbi@172.56.65.184) | 16:12 | |
| - paperManu (QUIT: Ping timeout: 246 seconds) (~paperManu@modemcable141.205-200-24.mc.videotron.ca) | 16:13 | |
| + paperManu (~paperManu@modemcable141.205-200-24.mc.videotron.ca) | 16:15 | |
| + siviq (~siviq@user/siviq) | 16:24 | |
| josch | okay, i have good news and bad news | 16:27 |
| josch | good news: the problem is fixed in most recent debian unstable | 16:27 |
| josch | bad news: the recovery could be a bit tricky if you don't have a uart adapter or ssh access | 16:27 |
| josch | (because due to lack of graphics support in u-boot we cannot tell users to just boot into single-user mode) | 16:27 |
| josch | the problem is that for some reason, after switching to another tty, that tty becomes unresponsive after a few seconds to half a minute | 16:28 |
| josch | and at that point i cannot even switch to another tty anymore | 16:28 |
| josch | this is what might've bitten johl a few days ago -- doesn't matter that you pressed the right keys, the text terminal is just stuck | 16:28 |
| josch | this is also shown by the fact that when shutting down the system, the dmesg output which appears on the currently active tty starts printing from the top left first line instead of from the line where the tty was when it froze as it usually does | 16:30 |
| josch | and without a serial adapter or ssh access, users would have to boot from sd-card, mount their system, chroot into it and then upgrade... | 16:30 |
| ch | time for a special sdcard ;) | 16:32 |
| + mark_ (~mjw@gnu.wildebeest.org) | 16:32 | |
| josch | ch: what is that special sd-card supposed to do? | 16:33 |
| ch | start, mount, chroot, get network, upgrade? :) | 16:34 |
| josch | and you just mount things willy-nilly? | 16:34 |
| josch | i've been pondering a small rescue utility which does some of the mount/chroot magic but am faced with the perfect being the enemy of the good | 16:36 |
| ch | well the partitions should have labels, right? | 16:36 |
| ch | if there's something non-standard, bail out | 16:36 |
| josch | the partition have labels and some are hidden behind luks | 16:37 |
| minute | josch: when you switch to the other tty while it still works, can you try `sudo stop gdm.service`? (or `sudo stop gdm3.service`) and then probably also `sudo disable gdm.service` | 16:41 |
| minute | josch: because what i think happens is that systemd tries to restart gdm after a certain time | 16:42 |
| minute | josch: and yanks you out of your vt | 16:42 |
| josch | minute: okay but i cannot try this now because the system on this sd-card is fixed now | 16:42 |
| minute | ah ok | 16:42 |
| josch | and i have to leave in an hour which is not enough time to get that far | 16:42 |
| josch | maybe tonight | 16:43 |
| minute | i think instead of chrooting etc, a rescue SD card could modify the boot process so that gdm is disabled | 16:43 |
| minute | or goes into rescue mode | 16:43 |
| minute | maybe systemd can be convinced not to go to graphical target | 16:44 |
| minute | via kernel params | 16:44 |
| josch | you mean a rescue sd-card should edit boot.scr or extlinux.conf on the /boot partition automatically and when the user reboots, linux boots with different options? | 16:44 |
| minute | josch: no it would just append more kernel params | 16:44 |
| minute | bootargs | 16:45 |
| minute | (not sure if that still works with extlinux) | 16:45 |
| josch | how would it do that? | 16:45 |
| minute | setenv bootargs | 16:45 |
| josch | but that's not permanent | 16:45 |
| minute | why does it need to be permanent? | 16:46 |
| minute | it would just allow someone to log in normally in text console mode, where they can fix their system | 16:46 |
| minute | instead of launching fragile gdm | 16:46 |
| josch | so u-boot loads a boot.scr or extlinux.conf from sd-card with the bootargs we want but how does this boot.scr/extlinux.conf then load the right kernel from elsewhere? | 16:47 |
| minute | elsewhere? | 16:47 |
| josch | yes, the system we actually want to boot is on emmc/nvme | 16:47 |
| minute | right, normally on emmc i guess | 16:47 |
| minute | i would say it would just load the extlinux.conf from the emmc /boot | 16:48 |
| minute | and mess with the params | 16:48 |
| josch | and how do we convince the system to load the modified u-boot from sd-card instead using the u-boot from emmc? | 16:50 |
| minute | josch: right, i forgot about that, sorry. butttt the emmc uboot prefers extlinux.conf from sd card | 16:50 |
| josch | yes, but extlinux.conf from sd-card knows nothing about the kernel versions it could be loading from /boot on emmc | 16:51 |
| minute | josch: but that doesn't know the... yeah. | 16:51 |
| minute | le mess | 16:51 |
| josch | mostly boils down to not having graphics at bootloader stage (be it u-boot, barebox whatever) | 16:51 |
| minute | small custom linux disk with recovery tool in initramfs then? | 16:51 |
| minute | could be stripped down version of system-image and a script | 16:52 |
| josch | what would the advantage be of another bootable image instead of adding a rescue script to reform-tools and shipping that in the default image? | 16:53 |
| minute | but yeah, like ch says, that helps with discovering and chrooting into existing system | 16:53 |
| minute | josch: how to launch that script? | 16:53 |
| minute | ah, manually | 16:53 |
| josch | we assume the system from sd-card boots normally | 16:53 |
| josch | then the user can connect to wifi and do any other preparation | 16:53 |
| josch | and then run reform-rescue-shell | 16:53 |
| josch | (tentative name of what i have here right now) | 16:54 |
| minute | and later a button in reform-setup-wizard? | 16:54 |
| minute | (that would launch this script) | 16:54 |
| josch | yes, like it would launch reform-migrate or reform-setup-encrypted-disk | 16:55 |
| josch | but at the point where reform-setup-wizard becomes a "normal program" you could also turn things around: | 16:55 |
| josch | create syste images which auto-login the user into a desktop | 16:55 |
| josch | from there they can either do rescue stuff or they click a button on the desktop which launches the setup-wizard | 16:56 |
| minute | yep | 16:56 |
| minute | probably the way to go | 16:56 |
| josch | this is how calamares in debian-live does it | 16:56 |
| minute | yeah i know | 16:56 |
| minute | anyway, the script is the most useful first step | 16:57 |
| - mark_ (QUIT: Ping timeout: 260 seconds) (~mjw@gnu.wildebeest.org) | 17:10 | |
| - bkeys (QUIT: Ping timeout: 246 seconds) (~Thunderbi@172.56.65.184) | 17:25 | |
| josch | the problem is all the spagetti code :/ | 17:38 |
| josch | this is what i wrote down without having run it yet: https://paste.debian.net/hidden/04a25b03 | 17:39 |
| josch | and it's already over 160 loc... | 17:39 |
| minute | josch: maybe soon it's time for some library code :D | 17:46 |
| minute | for looking up machine config etc | 17:46 |
| Svp | for some unclear reason my mbv3's rtc is always drifting by about a minute. hm... | 17:51 |
| minute | weird | 17:57 |
| minute | Svp: weak coin cell maybe? | 17:57 |
| + johl (~johl@wikidata/Jens-Ohlig) | 17:57 | |
| Svp | that'd be weird unless it came with one that was already a bit low. actually did it come with one, i cant remember... | 17:58 |
| johl | Tried out another external keyboard, no luck. What got me to the text console, was hyper+super (mnt logo key)+3 | 18:00 |
| johl | I don't know if that's only applicable for my firmware. | 18:00 |
| johl | Anyway, updated all the Debian things. Shut down the computer, turned it on. Now it's completely bricked, won't boot. Some kernel problem perhaps? | 18:02 |
| johl | It's getting frustrating :( | 18:03 |
| - aloo_shu (QUIT: Ping timeout: 265 seconds) (~aloo_shu@85.51.18.67) | 18:04 | |
| johl | Any ideas on how to fix this? | 18:04 |
| minute | johl: wow interesting @ that combo | 18:04 |
| minute | johl: sorry for the frustrations ;/ did you try to boot it up more than once? | 18:05 |
| johl | Yes | 18:05 |
| minute | johl: just black screen? | 18:05 |
| minute | johl: i.e. no text output at all? | 18:05 |
| johl | Yes. Keyboard lights up, but the screen stays black | 18:06 |
| minute | johl: ok, that's quite unusual. do you have a microSD card? then you can boot from a fresh system image and then we can look into recovery | 18:06 |
| minute | johl: and josch is working on a recovery tool that will make all of this a bit easier soon | 18:07 |
| johl | I thought that doing a full update with aptitude would fix things as I saw that the bug josch found was already fixed and rolled out this morning around 11. But it also installed a new kernel, could that be it? | 18:09 |
| johl | I'll get a microSD card. | 18:09 |
| johl | brb, shopping :) | 18:12 |
| - johl (QUIT: Quit: leaving) (~johl@wikidata/Jens-Ohlig) | 18:12 | |
| minute | anyone else tested kernel 6.18.5 on pocket reform? | 18:14 |
| gordon1 | only on classic | 18:16 |
| - siviq (QUIT: Quit: Client closed) (~siviq@user/siviq) | 18:16 | |
| + siviq (~siviq@user/siviq) | 18:16 | |
| ch | Linux zeppelin 6.18.5-mnt-reform-arm64 #1 SMP PREEMPT Debian 6.18.5-1+reform20260117T144105Z (2026-01-1 aarch64 GNU/Linux | 18:22 |
| ch | it seems a bit slow to boot | 18:22 |
| ch | packagekit/gnome-software decides to run directly after boot? ok | 18:23 |
| ch | iwlwifi crashed now | 18:24 |
| ch | pcieport 0003:30:00.0: bridge configuration invalid ([bus 00-00]), reconfiguring | 18:26 |
| ch | 1sec later iwlwifi crashed | 18:26 |
| ch | sudo takes forever, maybe because it cannot network | 18:26 |
| - siviq (QUIT: Ping timeout: 272 seconds) (~siviq@user/siviq) | 18:31 | |
| jfred | I just discovered the "reset keyboard" option in the OLED menu (not sure if that's new) which just saved me from a hard reboot when my keyboard disconnected from my pocket reform ^^; | 18:33 |
| jfred | could have been there the whole time and I've just never noticed it haha | 18:34 |
| jfred | dmesg logs from that in case it's helpful for anything: https://paste.sr.ht/~jfred/0e543b31c55de98caa0fb427e2a65286b785fc9a | 18:36 |
| minute | ch: uff | 18:38 |
| minute | jfred: yeah it's in there since a while :D | 18:39 |
| jfred | what this tells me is I need to explore menus more often haha | 18:39 |
| ch | seems ok after power cycle | 18:41 |
| ch | random internet posts also suggest that on x86 hardware, so maybe thats just an iwlwifi fw/hw problem | 18:41 |
| jfred | also interestingly the gtk4 app I'm working on seems to hang the GPU briefly on startup, noticed that in the dmesg logs | 18:42 |
| minute | ch: yeah i had very rare iwlwifi issues also | 19:02 |
| rwa_ | josch: re https://source.mnt.re/reform/reform-debian-packages/-/merge_requests/161 - i added the job repo and installed kernel 6.18.5. unfortunately the display turns of after a few seconds instead of showing the systemd logs | 19:07 |
| rwa_ | maybe i missed something during install, flash-kernel and reform-tools are already updated, will have a closer look later | 19:07 |
| minute | uff | 19:10 |
| + siviq (~siviq@user/siviq) | 19:15 | |
| gordon1 | rwa_: did you try to adjust backlight after it has turned itself off? | 19:17 |
| rwa_ | yep, doesn't do anything | 19:17 |
| rwa_ | booting into 6.17.13 still works fine, but with screen flickering included :) | 19:18 |
| gordon1 | weird | 19:19 |
| rwa_ | as i said, not sure if i missed some package update that should be there, but i think everything needed is there in /boot for kernel 6.18.5 | 19:21 |
| gordon1 | modules in /lib/modules for 6.18.5? | 19:22 |
| rwa_ | let me check | 19:22 |
| + mark_ (~mjw@gnu.wildebeest.org) | 19:24 | |
| rwa_ | hm, reform2-lpc-dkms is missing for 6.18.5 - maybe i forgot to install kernel headers | 19:25 |
| gordon1 | i don't think i've seen reform2-lpc-dkms module, are we talking about classic reform? | 19:27 |
| rwa_ | yeah | 19:27 |
| gordon1 | oh that's the debian package name, not the module name, isn't it? | 19:29 |
| rwa_ | reform2_lpc is the module name | 19:30 |
| minute | on pocket, reform2-lpc-dkms is needed for display backlight control for display v2 | 19:30 |
| minute | but the display will otherwise be just in a "max brightness" state | 19:30 |
| gordon1 | you don't have to have reform2_lpc to get the display working, i actually routinely reboot the new kernel w/o reform2_lpc and compile it afterwards just because i can't be bothered to read man depmod | 19:30 |
| minute | hmm except maybe if the brightness was set to 0 with a working driver before reboot... | 19:30 |
| minute | (talking about pocket only) | 19:31 |
| gordon1 | yeah i don't think it can affect backlight on classic in any way, can it? | 19:33 |
| minute | nope | 19:33 |
| cararemixed | Catching up... I'm running 6.18.5-1+reform20260117T144105Z just fine | 19:35 |
| cararemixed | rk3588 pocket | 19:35 |
| cararemixed | old IO board but I'm not sure that would matter here | 19:36 |
| rwa_ | cararemixed: i don't think it's a problem with 6.18.5 in general, but either with the patched version from the MR linked above or me being too stupid to install a kernel from a jobrepo (although i did that successfully a while ago) ;) | 19:37 |
| minute | cararemixed: thanks! i wonder what broke for johl... | 19:38 |
| minute | (there's 2 separate topics going on here atm) | 19:39 |
| rwa_ | ah i see, reading through the chat with jo_hl | 19:39 |
| cararemixed | Not sure. I am watching the fresh install I made last night carefully for things. I still can crash emacs in gtk mode which is the only regression on my personal list left. | 19:39 |
| rwa_ | that gdm breakage did some colateral damage to a few people it seems :( | 19:40 |
| cararemixed | Last time I got that I did the SD card boot + mount/chroot sequence to fix it. | 19:41 |
| cararemixed | I do think it'd be nice to include a script to automate the steps beyond the more rescue focused flow. It's handy to drop into alternate installs/userspaces. | 19:42 |
| cararemixed | UID mappings could get messy but if people are using their installed user then it is at least consistent. | 19:43 |
| rwa_ | tbh this is one of those why i'd like to stay away from Debian unstable...i'd happily test anything related to Reform (firmware, tools, lpc module and stuff), but i don't want to mess around with this sort of software bugs on a system thats meant to be a (almost) daily driver :/ | 19:46 |
| cararemixed | Yeah. Honestly it's partly why I prefer other rolling release distros. They are less stable than debian stable but more stable than debian unstable for whatever reason. | 19:56 |
| cararemixed | The biggest blocker I usually have is access to newer libraries and compilers for various projects. Being on stable usually means I end up having to build entire toolchains by hand or switching to nix flakes... which is actually fine but also its own issue with some apps interacting with things like mesa. | 19:58 |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:50) | 20:02 | |
| jfred | I'm curious if there's any hope for graphics in u-boot on some of the newer modules. That could help with troubleshooting/recovery without needing another machine | 20:11 |
| cararemixed | I wonder if keyboard interaction with that display could offer boot feedback. Its work but maybe less work than drivers for various display modules. | 20:13 |
| vagrantc | well, figured out a few things ... arm-trusted-firmware/trusted-firmware-a 2.14 does not boot the mnt/reform2-rk3588 ... and u-boot 2026.01 seems to only try to boot from eMMC | 20:33 |
| minute | cararemixed: yeah keyboard boot menu might be a viable thing. usb already exists, we only need to out how to do custom hid raw requests in uboot | 20:42 |
| gordon1 | that would be so good | 20:43 |
| minute | ok i'll put it in my todo, but feel free to beat me to the goal | 20:43 |
| - paperManu (QUIT: Ping timeout: 264 seconds) (~paperManu@modemcable141.205-200-24.mc.videotron.ca) | 20:57 | |
| + voltaire28_ (~jlafon@28.162.2.93.rev.sfr.net) | 21:02 | |
| gordon1 | it's amazing how u-boot code has little to no mention of HID | 21:06 |
| gordon1 | i wonder if there's just hardcoded thing for handling keyboard descriptors or something | 21:08 |
| gordon1 | yep | 21:12 |
| gordon1 | HID_BOOT i assume | 21:13 |
| gordon1 | https://github.com/u-boot/u-boot/blob/master/common/usb_kbd.c#L708-L713 | 21:14 |
| + aloo_shu (~aloo_shu@85.51.18.67) | 21:21 | |
| + reform21134 (~ryan@user/wirelessrpm) | 21:24 | |
| - mjw (QUIT: Killed (tungsten.libera.chat (Nickname regained by services))) (~mjw@2001:1c06:2486:a800:a09a:fc1c:5a8:e74d) | 21:28 | |
| * mark_ -> mjw | 21:28 | |
| + Guest2068 (~mjw@2001:1c06:2486:a800:a09a:fc1c:5a8:e74d) | 21:28 | |
| + paperManu (~paperManu@146.71.9.156) | 21:32 | |
| + paperManu_ (~paperManu@146.71.9.156) | 21:36 | |
| + rick_ (~m-x42ph2@m.n0id.space) | 21:40 | |
| - voltaire28_ (QUIT: Quit: WeeChat 4.7.2) (~jlafon@28.162.2.93.rev.sfr.net) | 21:41 | |
| + reform25277 (~matt@user/reform11910) | 21:42 | |
| - reform25277 (QUIT: Remote host closed the connection) (~matt@user/reform11910) | 21:45 | |
| minute | gordon1: oh it only supports boot hid protocol? | 21:50 |
| + mjv (~matt@user/reform11910) | 21:50 | |
| gordon1 | yep, seems like it | 21:50 |
| minute | this could be used https://github.com/u-boot/u-boot/blob/6b2d05748cf3cd6ba417a96c00602b0122e10af6/common/usb_kbd.c#L170C13-L170C27 | 21:55 |
| minute | we can just send usb control msg with the pixels | 21:56 |
| minute | so 1. make custom video driver 2. set up 128x32 (?) video framebuffer in there 3. discover the usb keyboard somehow 4. convert the pixels to 1-bit color on the fly 5. blast to usb_control_msg | 21:59 |
| gordon1 | hmm | 22:00 |
| + voltaire28_ (~jlafon@28.162.2.93.rev.sfr.net) | 22:03 | |
| minute | could be used as the base i think https://github.com/u-boot/u-boot/blob/master/drivers/video/simplefb.c | 22:03 |
| josch | everybody feel free to pass this on to users with the gdm issue who struggle with chrooting into their system from sd-card but who feel that they are motivated to report bugs they encounter: https://community.mnt.re/t/testers-wanted-for-new-tool-reform-rescue-shell-maybe-interesting-for-users-with-the-gdm-issue/4153/1 | 22:31 |
| - mjv (QUIT: Remote host closed the connection) (~matt@user/reform11910) | 23:23 | |
| - ohnoitsnoah_ (QUIT: Quit: Konversation terminated!) (~ohnoitsno@2603:8081:3c00:59f3::10a9) | 23:31 | |
| - vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:50) | 23:43 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
