| - bkeys (QUIT: Ping timeout: 272 seconds) (~Thunderbi@98.19.131.193) | 00:06 | |
| elb | hmmm I have a diff for the handbook but no way to send it in | 00:09 |
|---|---|---|
| minute | elb: as a last resort you could submit it under "other problem" on support.mnt.re | 00:11 |
| minute | elb: are you powering the pocket with usb-c when doing these tests or from battery? | 00:11 |
| elb | battery | 00:15 |
| elb | I actually think I might have had a combination of two problems: bad cable or socket connection, followed by overheating and rebooting (?) with the case off | 00:16 |
| elb | I reassembled everything and I've transfered about 160 GB with no issues; the max I got previously was under 100 GB | 00:16 |
| elb | nope, spoke too soon, just hung at about 181 GB | 00:17 |
| minute | meh. i asked about the power because i suspected some brownout issue... same problem when connected to power supply? | 01:36 |
| + chrcav (~chrcav@user/chrcav) | 01:36 | |
| - mjw (QUIT: Ping timeout: 258 seconds) (~mjw@gnu.wildebeest.org) | 02:15 | |
| minute | installed 6.16.1 on my reform and somehow it still has 6.15.4 in a /boot/extlinux/extlinux.conf | 03:00 |
| minute | who generates this extlinux.conf? perhaps u-boot-menu? | 03:02 |
| minute | it mentions "u-boot-update" at the top | 03:03 |
| - chrcav (QUIT: Quit: leaving) (~chrcav@user/chrcav) | 03:04 | |
| minute | ok, installed u-boot-menu and ran sudo u-boot-update, now the new kernel is in that extlinux.conf | 03:08 |
| + emdevt (~emdevt@2001:7e8:fa17:7901:889:cfa1:9c37:6514) | 03:08 | |
| minute | ouch, now my reform is borked | 03:09 |
| minute | ah. it now thinks it's the hdmi version of rcore | 03:16 |
| minute | > dtb -> dtbs/6.16.1-mnt-reform-arm64/rockchip/rk3588-mnt-reform2.dtb | 03:17 |
| minute | why did that happen... u-boot says > Model: MNT Reform 2 with RCORE-DSI RK3588 Module | 03:17 |
| minute | hand correcting the dtb links in /boot and rebooting, lets see. very glad that i have 1. ssh server on my reform and 2. zeroconf or something that makes hostname lookups work locally | 03:19 |
| minute | it would be really great to figure out how we can get a serial console by default again... | 03:19 |
| minute | ok, 6.16.1 is up and brightness works much better now! | 03:21 |
| minute | > mntmn@reform:~$ sudo update-initramfs -u | 03:23 |
| minute | > update-initramfs: Generating /boot/initrd.img-6.16.1-mnt-reform-arm64 | 03:23 |
| minute | > Using DTB: rockchip/rk3588-mnt-reform2.dtb | 03:24 |
| minute | wtf | 03:24 |
| minute | so /usr/share/flash-kernel/db/all.db has "MNT Reform 2 with RCORE-DSI RK3588 Module". it also strangely has 2x "MNT Reform 2 with RCORE RK3588 Module" | 03:29 |
| - paperManu (QUIT: Ping timeout: 258 seconds) (~paperManu@72.10.129.226) | 03:29 | |
| minute | ah, the issue is i still have an old /etc/flash-kernel/machine | 03:30 |
| minute | josch: i think it might be useful for reform-check to complain about /etc/flash-kernel/... (not sure how many people except me end up in this situation like me though) | 03:31 |
| elb | minute: the very first time was plugged in; I unplugged it in case it was a noise or similar issue; I'm trying again plugged in now that I've reseated everything | 04:24 |
| elb | I believe this corrects the documentation for serial connection: http://paste.debian.net/1392957/ | 04:35 |
| elb | (and I did indeed replicate the hang plugged in) | 04:36 |
| elb | let me try an I/O stress test on the eMMC to make sure it isn't that I just didn't notice the problem because I wasn't doing heavy I/O (although I doubt it as I've done a number of large data transfers and, e.g., built the TeX kpathsea directories stuff) | 04:37 |
| - chomwitt (QUIT: Ping timeout: 248 seconds) (~chomwitt@2a02:85f:9a74:ce00:1ac0:4dff:fedb:a3f1) | 04:53 | |
| - skipwich (QUIT: Remote host closed the connection) (~skipwich@user/skipwich) | 05:02 | |
| + skipwich (~skipwich@user/skipwich) | 05:02 | |
| elb | 300 GB of storage-rate writes to the eMMC without any issues, it does appear to be PCIe and/or NVMe-related (although of course the eMMC isn't driving the block layer quite as hard) | 05:16 |
| - antti (QUIT: Ping timeout: 244 seconds) (~antti@user/antti) | 05:45 | |
| + antti (~antti@user/antti) | 05:45 | |
| josch | minute: whoops, the check exists but only on a311d and ls1028a | 07:39 |
| josch | minute: i'll turn the check around and make it warn only if *not* on imx8mq, then we automatically cover all future SoMs :) | 07:39 |
| josch | ah no, we cannot turn it around | 07:41 |
| josch | minute: but you should've gotten this: | 07:41 |
| josch | E: unexpected content in /etc/flash-kernel/machine | 07:41 |
| josch | minute: do you have your reform-check output from before around by any chance? | 07:42 |
| josch | minute: you also should've gotten: | 07:47 |
| josch | E: your currently loaded dtb is not the one referenced by flash-kernel | 07:47 |
| josch | E: contents of /proc/device-tree/model: xxx | 07:47 |
| josch | E: contents of /etc/flash-kernel/machine: yyy | 07:47 |
| + reform27036 (~rene@p200300d0ff2803007fc2100dbdd09829.dip0.t-ipconnect.de) | 07:53 | |
| * reform27036 -> rwa_ | 07:53 | |
| - rwa_ (QUIT: Remote host closed the connection) (~rene@p200300d0ff2803007fc2100dbdd09829.dip0.t-ipconnect.de) | 07:56 | |
| + libercv (~libercv@user/libercv) | 09:22 | |
| + gustav28 (~gustav@c-92-32-80-134.bbcust.telenor.se) | 09:29 | |
| - libercv (QUIT: Quit: Konversation terminated!) (~libercv@user/libercv) | 09:33 | |
| - emdevt (QUIT: Ping timeout: 252 seconds) (~emdevt@2001:7e8:fa17:7901:889:cfa1:9c37:6514) | 09:36 | |
| - shtrophic (QUIT: Ping timeout: 276 seconds) (~m-hrdsqi@user/shtrophic) | 10:31 | |
| + shtrophic (~m-hrdsqi@user/shtrophic) | 10:51 | |
| + reform10021 (~rene@p200300d0ff2803007fc2100dbdd09829.dip0.t-ipconnect.de) | 11:07 | |
| * reform10021 -> rwa_ | 11:07 | |
| rwa_ | josch: speaking about reform-check - is it intentional to put out "warnings" about missing software packages (like gedit or micro) | 11:08 |
| rwa_ | I've deliberately removed them - no need to tell me every time i run reform-check ;) | 11:09 |
| rwa_ | i'd make it an "information" or remove it entirely and only keep warnings when critical packages like reform-tools are missing | 11:09 |
| - rwa_ (QUIT: Client Quit) (~rene@p200300d0ff2803007fc2100dbdd09829.dip0.t-ipconnect.de) | 11:10 | |
| - voltagex (QUIT: Read error: Connection reset by peer) (bb9d16d177@user/voltagex) | 11:20 | |
| - larme (QUIT: Read error: Connection reset by peer) (22544863b6@user/larme) | 11:20 | |
| - noam (QUIT: Read error: Connection reset by peer) (81879d1ffa@2a03:6000:1812:100::dfc) | 11:20 | |
| + voltagex__ (bb9d16d177@user/voltagex) | 11:21 | |
| - rwv (QUIT: Read error: Connection reset by peer) (1fa29d61cb@2a03:6000:1812:100::6b) | 11:21 | |
| + larme (22544863b6@user/larme) | 11:21 | |
| + noam_ (81879d1ffa@2a03:6000:1812:100::dfc) | 11:21 | |
| + rwv (1fa29d61cb@2a03:6000:1812:100::6b) | 11:21 | |
| + chomwitt (~chomwitt@2a02:85f:9a74:ce00:1ac0:4dff:fedb:a3f1) | 11:32 | |
| - n_to (QUIT: Quit: quitidiquit) (~n_to@2a03:4000:6:3662:24b1:57ff:fec6:76c1) | 12:16 | |
| + n_to (~n_to@2a03:4000:6:3662:24b1:57ff:fec6:76c1) | 12:17 | |
| - skipwich (QUIT: Ping timeout: 258 seconds) (~skipwich@user/skipwich) | 12:25 | |
| + mjw (~mjw@gnu.wildebeest.org) | 12:30 | |
| + skipwich (~skipwich@user/skipwich) | 12:30 | |
| + emdevt (~emdevt@2001:7e8:fa17:7901:889:cfa1:9c37:6514) | 12:36 | |
| minute | elb: thanks for the thorough checks! maybe an interaction between rk3588 pci host and the ssd controller then. there are some pcie related tweaks on kernel cmdline one could try | 12:48 |
| minute | elb: i haven't seen an ssd by sk hynix yet. we mostly ship sandisk/wd blue/green, or in the initial crowd supply batch, taifast with realtek controller | 12:50 |
| minute | josch: i don't have the output anymore but it didn't contain the word "flash", i did something like reform-check 2>&1 | less and searched for that | 12:52 |
| minute | bremner: gordon1: i wonder how i could support you better with cracking the hibernate nut. when i see phy link never came up, i wonder what's the difference between the initial link. for example, maybe the pci device's reset has to be asserted once before that and isn't? | 13:26 |
| josch | minute: oh unfortunate... i'd have liked to fix that bug :( | 13:59 |
| minute | josch: i guess i can restore that situation by hand | 13:59 |
| josch | if you like :) | 14:00 |
| minute | josch: btw i remember there was some control character at the end of the machine file, oddly | 14:00 |
| josch | i can try myself as well but only later today | 14:00 |
| - chomwitt (QUIT: Ping timeout: 248 seconds) (~chomwitt@2a02:85f:9a74:ce00:1ac0:4dff:fedb:a3f1) | 14:03 | |
| gordon1 | minute: i think the only thing is missing is organization on my part now, i can use my work laptop to do the console thing i think | 14:07 |
| gordon1 | but yeah, after conquering the slicer and box64, i think hibernate is the only outstanding issue on my reform todo list | 14:20 |
| bremner | I have learned about the mechanics of dealing with kernel patches, u-boot, etc..., but I still don't know much about kernel debugging, so I have just been trying patches that others have suggested. | 14:22 |
| josch | bremner: one of us! one of us! :D | 14:23 |
| bremner | heh. Well, at least I haven't made my pocket reform unbootable yet. | 14:23 |
| josch | buuh ;) | 14:23 |
| bremner | I know, I need to level up :) | 14:24 |
| bremner | I have made grub based systems unbootable a few times, so clearly u-boot is safer than grub :eyeroll: | 14:24 |
| josch | o0 | 14:25 |
| josch | i haven't used grup for too long it seems | 14:25 |
| bremner | iirc, my real problems have been related to encrypted filesystems | 14:25 |
| + paperManu (~paperManu@72.10.129.226) | 14:27 | |
| gordon1 | torxification of mnt reform is complete | 14:36 |
| gordon1 | unfortunately i lost battery cal in the process | 14:36 |
| minute | :0 | 14:43 |
| minute | gordon1: on reform, battery cal is normally just a matter of charging up full | 14:43 |
| minute | (it detects 100%) | 14:43 |
| + chomwitt (~chomwitt@2a02:85f:9a74:ce00:1ac0:4dff:fedb:a3f1) | 14:43 | |
| gordon1 | sure, i meant i had to disconnect the battery while changing screws | 14:44 |
| gordon1 | so now it is ??? and i need to charge it | 14:44 |
| minute | gordon1: yeah. | 14:44 |
| gordon1 | it is in the process of fixing itself | 14:44 |
| gordon1 | minute: those screws with a nylon strap, is it something you can get off the shelf or was it applied during the assembly? | 14:45 |
| gordon1 | i mean the blue stuff used as threadlocker | 14:46 |
| minute | gordon1: we were able to order them like that from the screw manufacturer. they call it "nylon dispensing" | 14:47 |
| minute | (fwiw our current screw supplier is "snippin") | 14:47 |
| gordon1 | cool, i need to search for those but with torx | 14:47 |
| gordon1 | used loctite instead for now | 14:47 |
| minute | yeah we used loctite for a while but it was tedious applying it in production. fine for one offs of course | 14:47 |
| gordon1 | oh i bet it is not fun | 14:48 |
| gordon1 | doing it for 1000s times instead of 10 | 14:48 |
| gordon1 | good that there is such an option with nylon ring now | 14:48 |
| bremner | oh lol. Such a software guy, I read torxification as something to do with Tor | 14:53 |
| - chartreuse (QUIT: Ping timeout: 260 seconds) (~chartreus@S0106908d78501d1d.cg.shawcable.net) | 15:21 | |
| josch | gordon1: i was pondering about torxifying my reform as well but never found a good source for M2 screws with torx or a torx M2 screwdriver that does not suck... But i already have several screw heads at the verge of giving up and i'd like to replace them with a better shape | 15:30 |
| gordon1 | josch: i wanted to go with accuscrew, looked it up, coughed on price and ordered them all from ebay (and had some other lying around already) | 15:31 |
| josch | violet, ex-parrot: i remember that you are running debian stable on your reform? I just upgraded to trixie and while i was running kernel 6.12 before as well, the new mesa version (25.0) gives me a flicker of the full screen every few seconds... very annoying... Do you have that too? | 15:32 |
| gordon1 | i also ordered wrong screws for the heatsink, so it is technically not 100% torxified | 15:33 |
| josch | i also have a reprouducible segfault of sway but i seem to be unable to run sway with gdb as the screen simply freezes with the last message from gdb on the screen "detaching after vfork from child process" -- does somebody have any experience with getting a traceback from sway when it segfaults? | 15:34 |
| minute | josch: i remember i did that once. did you try sshing in and attaching to running sway from gdb? | 15:41 |
| minute | (from another machine) | 15:41 |
| josch | good idea, i'll try that next | 15:41 |
| elb | minute: OK, good to know ... I'll see if I can get my hands on a WD, I might even have one available around here. I got this one because it appeared to have a very low expected power draw. | 15:49 |
| + bkeys (~Thunderbi@134.22.115.162) | 16:00 | |
| - bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@134.22.115.162) | 16:12 | |
| + bkeys (~Thunderbi@134.22.115.162) | 16:13 | |
| - bkeys (QUIT: Ping timeout: 258 seconds) (~Thunderbi@134.22.115.162) | 16:17 | |
| minute | elb: ah, bummer | 16:22 |
| elb | well shucks, I just checked to see if there's a new nvme firmware and I have the latest | 16:44 |
| + bkeys (~Thunderbi@h193.131.19.98.dynamic.ip.windstream.net) | 16:46 | |
| - emdevt (QUIT: Ping timeout: 248 seconds) (~emdevt@2001:7e8:fa17:7901:889:cfa1:9c37:6514) | 16:54 | |
| + reform21783 (~rene@p200300d0ff2803007fc2100dbdd09829.dip0.t-ipconnect.de) | 17:07 | |
| * reform21783 -> rwa_ | 17:07 | |
| - rwa_ (QUIT: Remote host closed the connection) (~rene@p200300d0ff2803007fc2100dbdd09829.dip0.t-ipconnect.de) | 17:17 | |
| josch | are there docs for what the orange light means for the classic reform motherboard 2.5 eithernet RJ45 socket? | 17:50 |
| josch | oh dear, the cable might be bad... | 17:52 |
| ch | i vaguely remember yellow leds on nics to mean '100M' or such :) | 17:56 |
| josch | it maybe worse | 18:04 |
| josch | same cable and the pocket reform gives me "link up" "link down" messages on dmesg | 18:04 |
| josch | nothing with a311d classic reform mb25 | 18:04 |
| josch | (i just updated to trixie and wanted to test all the things) | 18:04 |
| josch | (display still flickers) | 18:04 |
| ch | sounds enough like broken cable | 18:05 |
| josch | then why does it work with the pocket? | 18:05 |
| ch | i thought it says link up _and_ down? | 18:05 |
| josch | on the pocket, yes | 18:06 |
| josch | no dmesg activity on a311d classic reform | 18:07 |
| josch | everything works on the pocket with the same cable -- i get dhcp and everything | 18:07 |
| + d_m (~d_m@user/d-m/x-5109880) | 18:20 | |
| josch | ah okay | 18:26 |
| josch | i manually performed the workaround for ethernet PHY reset problem from reform-hw-setup | 18:26 |
| josch | (unloading and re-loading modules) | 18:26 |
| josch | and now i get "Link is Up - 1Gbps/Full - flow control rx/tx" | 18:26 |
| josch | hrm... | 18:26 |
| josch | aha okay, i refreshed my memory on the a311d display ficker issue | 18:38 |
| josch | it's this one: https://community.mnt.re/t/display-flickering-on-pocket-reform-related-to-mesa-25-0-0/3133 | 18:38 |
| josch | which is this debian bug: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1099032 | 18:38 |
| josch | but mesa says it's not their bug: https://gitlab.freedesktop.org/mesa/mesa/-/issues/12718 | 18:38 |
| josch | and disabling afbc 32x8 support in display planes is a workaround for which i created this MR: https://source.mnt.re/reform/reform-debian-packages/-/merge_requests/93 | 18:39 |
| josch | unfortunately my list does not have any A311D users for any IRC participant idling right now... | 18:39 |
| - chomwitt (QUIT: Ping timeout: 248 seconds) (~chomwitt@2a02:85f:9a74:ce00:1ac0:4dff:fedb:a3f1) | 19:19 | |
| + libercv (~libercv@user/libercv) | 19:33 | |
| - libercv (QUIT: Quit: Konversation terminated!) (~libercv@user/libercv) | 20:19 | |
| vkoskiv | I use A311D | 20:34 |
| - gustav28 (QUIT: Quit: Quit) (~gustav@c-92-32-80-134.bbcust.telenor.se) | 20:42 | |
| josch | vkoskiv: thank you! what mesa version and kernel are you on? | 20:46 |
| josch | oh indeed you do, i'm not smart enough to grep by own list which had "a311d" listed as "MNT Reform 2 with BPI-CM4 Module" | 20:47 |
| irimi1 | hey, just wanted to ask whether there's a definite solution for the problem with the USB connection to the keyboard yet, or whether trying to upgrade the firmware is still my best shot ^^ | 20:50 |
| minute | irimi1: what's your current keyboard firmware version? | 21:09 |
| irimi1 | how do i check that again? | 21:11 |
| irimi1 | i don't think i updated it at all though, except some debian package took care of that | 21:11 |
| josch | irimi1: debian packages do not update your firmware | 21:35 |
| irimi1 | minute, josch: then it's probably still what it came with, but i don't know how to check | 21:39 |
| josch | irimi1: standalone or built-in? pocket? classic reform? | 21:41 |
| irimi1 | 21:42 | |
| josch | irimi1: ch is working on a script which will be part of the next reform-tools release which will give you that information easily as part of reform-check | 21:48 |
| josch | i'll look up the command to obtain it later unless somebody else beats me to it :) | 21:48 |
| ch | you can grab the script from https://source.mnt.re/reform/reform-tools/-/merge_requests/135 | 21:48 |
| ch | josch: tbh i think it could be merged as is and i can iterate on the json stuff later | 21:49 |
| ch | or next week | 21:49 |
| ch | well not next week | 21:49 |
| irimi1 | ch, josh: that tool is giving me an empty list :) | 21:52 |
| ch | ah because your firmware is too old for this | 21:52 |
| ch | then the only real way is by asking the oled | 21:52 |
| irimi1 | i'm guessing what i find at "system status"? which line do you need? | 21:53 |
| irimi1 | ah, probably the one including 'HID' | 21:54 |
| irimi1 | PREFHID20240416 | 21:55 |
| josch | irimi1: if the answer is "too old" then updating is done manually and not via fwupd: https://source.mnt.re/reform/pocket-reform/-/releases | 22:02 |
| + matt2 (~matt@97-122-64-98.hlrn.qwest.net) | 22:10 | |
| irimi1 | josch: it doesn't find the keyboard device. i think it wasn't listed when i consulted ch's help a few weeks back, either | 22:11 |
| minute | josch: fwupd doesn't support the keyboard fw yet | 22:23 |
| minute | irimi1: mnt.re/update-pocket-keyboard | 22:24 |
| minute | (as in `curl -sfL mnt.re/update-pocket-keyboard | sudo bash -`) | 22:24 |
| minute | (if you dare) | 22:24 |
| minute | irimi1: it's best if you have an external keyboard with usb-c to be able to recover from any issues! | 22:25 |
| minute | irimi1: (and don't turn the pocket off if there are any issues, otherwise it's hard to turn it back on) | 22:25 |
| irimi1 | that probably wouldn't find the device either though, right? | 22:27 |
| irimi1 | yea: Error: MNT Pocket Reform Input device not found or more than one found. | 22:27 |
| minute | irimi1: well i recently saw that that april firmware isn't detected yeah. but you can put the keyboard into flashing mode in the oled menu + press X before running this script. BUT you need an external keyboard before you do that | 22:28 |
| minute | (or ssh into the machine, that's also possible if you set that up before) | 22:29 |
| irimi1 | yeah i am ssh'd in right now. the external keyboard thing is a bit annoying due to the different keymap | 22:29 |
| gordon1 | nope, ffmpeg-8 cannot do vulkan hwdec on panvk http://0x0.st/Kina.txt | 22:29 |
| minute | irimi1: ok, ssh should be fine then. just don't power off if you encounter issues, instead, write me here | 22:30 |
| gordon1 | same with h264 | 22:30 |
| irimi1 | minute: still the same error | 22:30 |
| minute | gordon1: ah. those codecs are not supported by the compute shader vulkan stuff. only prores and 2 others | 22:30 |
| irimi1 | OLED says "Entered firmware update mode." | 22:30 |
| minute | irimi1: ok, can i see lsusb output? | 22:31 |
| gordon1 | hmm | 22:31 |
| irimi1 | minute: https://paste.systemli.org/?1c8fc8f0767cf835#65VkRTm72vasYfNiXQSMyHC9dqonrUHaS251x4mRiNdq | 22:31 |
| gordon1 | right, not very helpful then | 22:31 |
| minute | irimi1: thanks. ok uhm, what is that raspberry pi pico there | 22:32 |
| gordon1 | oh well | 22:32 |
| irimi1 | minute: good question. i've only connected the power supply and ethernet | 22:32 |
| minute | irimi1: omg your keyboard fw version is 2024 (!) | 22:33 |
| minute | not 2025 | 22:33 |
| minute | irimi1: did you get this device recently? | 22:33 |
| minute | irimi1: maybe from CS stock? | 22:33 |
| irimi1 | nope, from the campaign. one of the unfortunate ones who had to wait due to the display issues, i think | 22:33 |
| minute | irimi1: ok everything is ancient on there. the system controller is probably the "pi pico", before it had its own usb id | 22:34 |
| minute | irimi1: but wait, you got a unit recently that was from the CS campaign?! | 22:34 |
| irimi1 | no, not recently | 22:34 |
| minute | ok | 22:34 |
| minute | irimi1: processor is imx8mp? | 22:35 |
| - ZetaR (QUIT: Server closed connection) (~user@c-98-208-141-90.hsd1.fl.comcast.net) | 22:35 | |
| irimi1 | yes | 22:35 |
| josch | minute: good to see you back in irc during the past days :) | 22:35 |
| + ZetaR (~user@c-98-208-141-90.hsd1.fl.comcast.net) | 22:35 | |
| irimi1 | i wanna say i got it early 2025, but the shipping notification says it was shipped 2024-10 | 22:36 |
| minute | irimi1: ok. my thoughts are like this: first, we should update u-boot to make sure you have proper USB hub reset. then reboot, but only if you've set up ssh so you can log in again via ssh as the keyboard is currently not responsive (flashing mode) | 22:36 |
| minute | irimi1: what kernel are you on? | 22:37 |
| irimi1 | 6.15.6-mnt-reform-arm64 | 22:37 |
| irimi1 | (just ran an update) | 22:37 |
| minute | ok, that's recent | 22:37 |
| minute | i assume you also have recent reform-tools, right? | 22:37 |
| minute | then you can run: `sudo reform-flash-uboot emmc` | 22:37 |
| irimi1 | if that's a debian package, then yes. if that's something i would have gotten another way, then no | 22:38 |
| minute | yes, debian package | 22:38 |
| irimi1 | the only way to get the keyboard out of update mode is to power it off, right? | 22:39 |
| minute | irimi1: technically there is another way but that's more complicated. the easiest is to power cycle using the standby switch | 22:39 |
| minute | irimi1: this only works if you haven't flashed a broken firmware or sth though (you haven't) | 22:40 |
| irimi1 | alright, i'm gonna do that with the reboot then. | 22:40 |
| minute | my reasoning about the reboot is that if usb hub resets, there's a chance it will see the keyboard, still in flashing mode | 22:40 |
| minute | and then you can try the script again | 22:40 |
| irimi1 | oh so i should leave it in flashing mode? | 22:40 |
| minute | yeah sure. that's why i said, make sure you can ssh in automatically, or have external keyboard | 22:41 |
| minute | i mean, it doesn't matter that much | 22:41 |
| minute | whatever is more convenient | 22:41 |
| minute | but you updated uboot? | 22:41 |
| irimi1 | not yet, i'm about to :) | 22:41 |
| irimi1 | just to make sure: 'emmc' is right when i migrated to the SSH? | 22:42 |
| irimi1 | *SSD | 22:42 |
| irimi1 | minute: looking a bit better now: https://paste.systemli.org/?8adfb9a077c85dd8#FAFsuacFEkgvmoB7xdDsACFNjeZ9PDYu58h5qgTCSmMJ | 22:47 |
| irimi1 | guessing the 'Raspberry Pi RP2 Boot' is the keyboard controller | 22:47 |
| minute | irimi1: yes | 22:48 |
| irimi1 | now which way do i update the ancient firmware now? | 22:48 |
| minute | irimi1: i think my script should just work now | 22:48 |
| irimi1 | the update-pocket-keyboard one? | 22:48 |
| minute | yeah it should recognize the 2e8a:0003 | 22:49 |
| irimi1 | i also have a update-keyboard-firmware.sh | 22:49 |
| irimi1 | ok, running the former then | 22:49 |
| minute | irimi1: how does it look? | 22:50 |
| irimi1 | the device disappeared just in time ^^ | 22:50 |
| minute | wat | 22:50 |
| irimi1 | not in the lsusb output anymore | 22:50 |
| minute | ok that's not good | 22:50 |
| minute | irimi1: can you replug the cables of the keyboard, on the keyboard side and the motherboard side? maybe something is loose | 22:51 |
| minute | (slightly) | 22:51 |
| irimi1 | i did that earlier (and a few weeks back), but i can always just do that again.. | 22:51 |
| irimi1 | that's the tail of my dmesg btw: https://paste.systemli.org/?50987f132b4b057c#E4PkKEpuhDjHRuz9YEvceba6ug7iFsfvdEM55o3GWR7R | 22:52 |
| minute | irimi1: thanks, it looks like how it would look like with a very bad connection | 22:52 |
| minute | irimi1: it's also not very safe to flash it if it's like that... could take many tries then | 22:52 |
| irimi1 | that's the problem i originally wanted to solve btw. | 22:52 |
| minute | irimi1: yes. the firmware, when running, also had some problems like this. but now it's not running | 22:53 |
| minute | irimi1: i.e. the bootloader should be stable. but it doesn't look like the connection is stable anyway | 22:53 |
| + espressocheese (~espressoc@2601:8c1:c180:4630:d4b6:7e57:ea7:be8c) | 22:53 | |
| irimi1 | is there a plan b for when reseating the cable doesn't work? | 22:54 |
| minute | irimi1: are you SSHed in via that wifi card that i see in lsusb? or ethernet? | 22:54 |
| minute | irimi1: (or imx8mp wifi) | 22:54 |
| irimi1 | i'm hoping via ethernet :D | 22:54 |
| irimi1 | nmtui shows no wireless connection | 22:55 |
| minute | irimi1: you could remove the wifi card to see if there's any weird interaction on the usb bus through the hub | 22:55 |
| espressocheese | Hi! I got the mnt reform pocket, and I have a couple of questions. First, I have an ssd installed, and I'm not sure how to make it boot. Do I need an sd card? Second, my laptop doesnt close all the way, is there something I can do about that? | 22:55 |
| minute | espressocheese: the last thing is very unusual, can you show pix or video of how it doesn't close? and you're positive there's nothing in the way? | 22:56 |
| minute | irimi1: which country are you in? asking because of the best course for getting you some replacement parts | 22:56 |
| espressocheese | Thanks for responding, the wifi antenna was too short to put on headphone board, so I put it on back cover. Also, I did not get a display adapter | 22:57 |
| espressocheese | I'll send an image just a sec | 22:57 |
| irimi1 | minute: Germany. i'm c3manu on chaos.social, i came to say hi at GPN :) | 22:58 |
| irimi1 | minute: i'm gonna try reseating the cable tomorrow, and disconnecting the wifi card after that. it's probably best to not do that with ADHD meds, plus i can run an off-site backup over night. | 22:59 |
| irimi1 | minute (also josh, ch): thank you so much for your help already :) | 22:59 |
| minute | irimi1: ah nice, then the plan b or c is easy, we fix your device or we send you replacement keyboard directly (depending on stock) | 22:59 |
| minute | espressocheese: display adapter? | 23:00 |
| irimi1 | minute: lovely :) | 23:00 |
| espressocheese | hold on, idk how to send an image with irc | 23:00 |
| espressocheese | "17. Remove the protective film from the Display (20) and place the Display Frame | 23:00 |
| espressocheese | Gasket (7)." | 23:00 |
| espressocheese | I never got the gasket | 23:00 |
| minute | espressocheese: oh the gasket... waa. are you sure? and you got this device in the last days? | 23:01 |
| espressocheese | yes | 23:01 |
| minute | espressocheese: the gasket is maybe hard to spot, it's transparent and 0.3mm thin | 23:01 |
| espressocheese | I just got the device | 23:01 |
| irimi1 | minute: i'm gonna check in tomorrow to let you know how it went | 23:01 |
| minute | i will ask how it is usually packed | 23:01 |
| espressocheese | I can check again but I havent seen it | 23:01 |
| espressocheese | Is there an email I can send the image to? bc I don't know how to send over irc | 23:02 |
| espressocheese | Or whats app | 23:02 |
| minute | espressocheese: yeah you can send to me, lukas@mntre.com | 23:02 |
| espressocheese | I did | 23:03 |
| espressocheese | I can force it closed, but thats how far it goes without forcing it | 23:03 |
| minute | espressocheese: ok don't force that | 23:05 |
| espressocheese | of course, lol | 23:05 |
| espressocheese | Did you see the image? | 23:05 |
| espressocheese | That's how far it closes naturally | 23:06 |
| minute | espressocheese: yes. def. not normal. can you send more angles, how it looks like near the rear? | 23:06 |
| minute | i remember some case where someone mounted some components upside down, like the hinges or something | 23:06 |
| minute | espressocheese: and it's not the magnets repelling, yeah? (probably not) | 23:07 |
| minute | i.e. magnets not attracting but repelling | 23:07 |
| espressocheese | I sent you more angles, I think thats how it is even without magnets | 23:08 |
| espressocheese | It looks like it may be a bad hinge | 23:08 |
| minute | espressocheese: my colleague says about the gasket, it's packed with > all closing panels (top, bottom, display panel and keyboard panel) | 23:09 |
| espressocheese | I did not see it | 23:09 |
| espressocheese | I found it | 23:09 |
| minute | great | 23:10 |
| espressocheese | just fyi: I took off the top magnets and its the same thing | 23:12 |
| espressocheese | so the closing issue is not the magnets | 23:15 |
| minute | espressocheese: thanks for more pics. can you open it 180 degrees and take off the bezels and show a photo incl how the hinges are mounted? maybe something will be more obvious | 23:15 |
| espressocheese | I sent a pic of the left hinge, did I do them backwards or something? | 23:17 |
| minute | espressocheese: looks ok on that side i think. and on keyboard side? | 23:19 |
| espressocheese | its the small screw part, looks fine to me | 23:20 |
| minute | espressocheese: anything that could block it? | 23:20 |
| espressocheese | no | 23:20 |
| minute | espressocheese: hmm then i'm kind of stumped, at least tonight. i can show it to my colleague on monday. if i were you i would remove the display and keyboard, and maybe the cables in the hinges, and see if it closes properly or not, and add parts one by one to see what could be the culprit | 23:21 |
| espressocheese | im trying to swap the hinges | 23:22 |
| espressocheese | THAT WORKED | 23:25 |
| - espressocheese (QUIT: Quit: Client closed) (~espressoc@2601:8c1:c180:4630:d4b6:7e57:ea7:be8c) | 23:32 | |
| matt2 | i have a pocket reform with encrypted nmve setup using the bundled script. been great. i put down the machine for a while waiting for a new charger board. in the meantime looks like i either forgot the password to decrypt it or the luks header is corrupted or ??. in any case basic chain of errors on my part and i can't access it. to boot into eg a shell i can access that drive from to run a few | 23:35 |
| matt2 | things to attempt to recover what can i do? as simple as reflashing the system image to the microsd and booting again or something else? | 23:35 |
| - xktr (QUIT: Quit: leaving) (~xktr@user/xktr) | 23:36 | |
| - robin (QUIT: Remote host closed the connection) (~robin@user/terpri) | 23:47 | |
| minute | matt2: tried connecting an external keyboard to type the password? to rule out kbd issues | 23:49 |
| + espressocheese (~espressoc@2601:8c1:c180:4630:d4b6:7e57:ea7:be8c) | 23:49 | |
| josch | (or boot from an ssd and try to decrypt using cryptsetup luksOpen) | 23:50 |
| josch | s/ssd/sd-card/ | 23:50 |
| espressocheese | I think I got kicked out, can someone pls help me with the software? :) | 23:50 |
| josch | espressocheese: your last message was "THAT WORKED" what was your software problem? | 23:52 |
| espressocheese | the reason I said that worked was bc I swapped the hinges and now it closes | 23:52 |
| josch | yes, what is your software problem? | 23:53 |
| espressocheese | it doesnt boot | 23:53 |
| espressocheese | do I need an sd card to boot? | 23:53 |
| josch | the device should come with a bootable image flashed to emmc | 23:54 |
| josch | but you could flash an image to an sd-card and try to boot that, yes | 23:54 |
| espressocheese | I got the self build and it doesnt boot | 23:54 |
| matt2 | minute: after N failed attempts i got dumped into an intramfs promt and was able to type the password there to see that the password i am typing is what i expect so i don't believe it's a layout issue. | 23:54 |
| espressocheese | Just had to flip the switch, oops | 23:57 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!