| zeha | minute: ah i think i saw the irq <> usb printf issue some time ago. generally the biq irq handlers pico-sdk seems to encourage might not be a good idea? | 00:04 |
|---|---|---|
| + reform30345 (~mmmm@ip2504e528.dynamic.kabel-deutschland.de) | 00:11 | |
| reform30345 | Linux mmmm 6.15.2-mnt-reform-arm64 #1 SMP Debian 6.15.2-1~exp1+reform20250616T055357Z (2025-06-16) aarch64 GNU/Linux | 00:11 |
| reform30345 | this is from rk3588 classic reform | 00:11 |
| - chomwitt (QUIT: Quit: WeeChat 3.8) (~chomwitt@2a02:85f:9a1d:4100:1ac0:4dff:fedb:a3f1) | 00:12 | |
| reform30345 | but imx8mq also seems to work with audio, display wifi etc on 6.15 | 00:12 |
| - reform30345 (QUIT: Client Quit) (~mmmm@ip2504e528.dynamic.kabel-deutschland.de) | 00:12 | |
| + chomwitt (~chomwitt@2a02:85f:9a1d:4100:1ac0:4dff:fedb:a3f1) | 00:15 | |
| - Ar|stote|is (QUIT: Ping timeout: 248 seconds) (~linx@149.210.5.33) | 00:16 | |
| + Ar|stote|is (~linx@149.210.5.33) | 00:20 | |
| - AnimaInvicta (PART: !!unknown attribute: msg!!) (~AnimaInvi@88-120-179-216.subs.proxad.net) | 00:31 | |
| kfx | anyone know the kernel cmdline to get output to the serial port on rk3588? my output stops at "Starting kernel..." | 00:48 |
| josch | kfx: we store the vanilla kernel cmdline for each SoM in machine readable *.conf files. Here is the one for rk3588 classic reform: https://source.mnt.re/reform/reform-tools/-/blob/main/machines/MNT%20Reform%202%20with%20RCORE%20RK3588%20Module.conf?ref_type=heads#L23 | 00:54 |
| sigrid | kfx: earlycon=uart8250,mmio32,0xfeb50000 keep_bootcon | 00:54 |
| josch | you are looking for console=tty1 | 00:55 |
| josch | oh serial :D | 00:55 |
| - mjw (QUIT: Ping timeout: 244 seconds) (~mjw@gnu.wildebeest.org) | 00:59 | |
| kfx | thanks | 00:59 |
| + mjw (~mjw@gnu.wildebeest.org) | 01:05 | |
| - mjw (QUIT: Ping timeout: 244 seconds) (~mjw@gnu.wildebeest.org) | 01:09 | |
| minute | i actually don't know where turquoise output comes from, i also recently have some on my intel machine | 01:15 |
| - wiedi (QUIT: Read error: Connection reset by peer) (~wiedi@ip5f581498.dynamic.kabel-deutschland.de) | 01:16 | |
| - chomwitt (QUIT: Quit: WeeChat 3.8) (~chomwitt@2a02:85f:9a1d:4100:1ac0:4dff:fedb:a3f1) | 01:17 | |
| josch | hm okay | 01:19 |
| josch | also, it seems that the top of my rk3588 classic reform screen is getting squeezed quite a bit, no? https://mister-muffin.de/p/C8dc.jpg | 01:19 |
| josch | and sometimes (but very seldom and a reboot fixes it) my pocket reform display comes up like this: https://mister-muffin.de/p/Aeoc.jpg | 01:20 |
| + chomwitt (~chomwitt@2a02:85f:9a1d:4100:1ac0:4dff:fedb:a3f1) | 01:20 | |
| - chomwitt (QUIT: Quit: WeeChat 3.8) (~chomwitt@2a02:85f:9a1d:4100:1ac0:4dff:fedb:a3f1) | 01:25 | |
| minute | josch: oh, that squeeze is no good. i recommend to loosen those screws a bit | 01:31 |
| bremner | huh. never seen the second one | 01:31 |
| bremner | maybe a newer kernel fixes it (6.14.6 here) | 01:33 |
| minute | josch: the pocket glitch in the second one doesn't look very familiar | 01:33 |
| josch | minute: thanks a lot for your feedback! | 01:35 |
| minute | firefox is so good now on rk3588 with that mesa patch... i still can't believe it | 01:37 |
| minute | especially it runs in hidpi mode, which chromium couldn't do without crashing regularly | 01:38 |
| minute | (because of libsync bug) | 01:38 |
| josch | heh, coming from a311d, i found firefox on rk3588 blazing fast already :D | 01:38 |
| josch | shall we build patched mesa again? | 01:39 |
| minute | maybe we shall | 01:39 |
| minute | josch: it'll take a while until a proper fix is made and then trickled down to debian, so, i'd say lets do it | 01:43 |
| josch | minute: would you like me to add mojyacks first or second patch? | 01:44 |
| minute | josch: second | 01:45 |
| minute | josch: btw my 4G reception is not very good yet, so i can't really recommend an antenna yet | 01:46 |
| minute | i did a little test at home and got like 25mbit down from congstar | 01:46 |
| minute | good enough for email i guess :D | 01:46 |
| josch | ah there is a problem | 01:47 |
| josch | we are currently pulling from experimental | 01:47 |
| josch | and mesa happens to have an upload in experimental | 01:47 |
| josch | but that one fails to build from source | 01:47 |
| josch | so we'd have to pull kernel from experimental but normal packages from unstable | 01:48 |
| josch | except of course reform-tools which needs to come from experimental... | 01:48 |
| josch | minute: i already tried the mesa thing with the first patch 2 days ago here: https://source.mnt.re/reform/reform-debian-packages/-/merge_requests/117 | 01:48 |
| minute | josch: ok, maybe we wait until mesa is ok then in experimental? | 01:49 |
| josch | no, i'll just hardcode the specific suites tomorrow for the time being | 01:50 |
| josch | it is likely that nobody cares about experimental | 01:50 |
| josch | we need trixie to be released and then things become easier again :) | 01:50 |
| josch | hrm... then i should put my congstar card in and try that -- i get much less from o2 which might be network provider related and not antenna related... | 01:52 |
| josch | maybe it also helps to attach the auxilary lte antenna | 01:52 |
| minute | hmm, so far i haven't got an aux one | 01:56 |
| minute | but in my home i think i have exceptionally good reception | 01:57 |
| josch | the challenge for me is, that i usually rely on lte while in the train and that is understandably a challenge :) | 01:58 |
| minute | i see... i'm on a long train ride on wednesday so also looking forward to how that'll work out | 02:00 |
| josch | good luck! :) | 02:01 |
| josch | not having wifi right now is not so bad because regio does not have wifi anyways :D | 02:01 |
| minute | ah well :D | 02:03 |
| josch | oh and do you have a thermal pad installed on your quectel modem cover? | 02:04 |
| josch | (the red pad) | 02:04 |
| minute | ah no that is a leftover from olden tymes when i couldn't properly mount this modem | 02:05 |
| minute | it's just a piece of foam tape | 02:06 |
| josch | ah kk | 02:06 |
| josch | okay, mesa is building on MNT CI :) | 02:20 |
| josch | MR is here: https://source.mnt.re/reform/reform-debian-packages/-/merge_requests/117 | 02:23 |
| josch | will check tomorrow whether it succeeded | 02:23 |
| spew | ah yes the olden tymes when the only thing we mounted were horses | 02:24 |
| - paperManu (QUIT: Ping timeout: 265 seconds) (~paperManu@72.10.128.164) | 03:47 | |
| - kop316 (QUIT: Remote host closed the connection) (m-6f6zq6@static.138.159.90.157.clients.your-server.de) | 04:01 | |
| + kop316 (m-6f6zq6@static.138.159.90.157.clients.your-server.de) | 04:04 | |
| - kop316 (QUIT: Remote host closed the connection) (m-6f6zq6@static.138.159.90.157.clients.your-server.de) | 04:04 | |
| + kop316 (m-6f6zq6@static.138.159.90.157.clients.your-server.de) | 04:04 | |
| - colinsane (QUIT: Quit: bye) (~colinunin@97-113-145-89.tukw.qwest.net) | 05:55 | |
| - robin (QUIT: Remote host closed the connection) (~robin@user/terpri) | 05:58 | |
| + colinsane (~colinunin@97-113-145-89.tukw.qwest.net) | 06:28 | |
| - colinsane (QUIT: Ping timeout: 272 seconds) (~colinunin@97-113-145-89.tukw.qwest.net) | 06:36 | |
| kfx | got an alpine kernel package that mostly boots... but it only detects the emmc and not the sd card reader. guess it's time to find the missing driver | 07:31 |
| + wiedi (~wiedi@ip5f581498.dynamic.kabel-deutschland.de) | 07:36 | |
| josch | kfx: same dtb? | 07:43 |
| * voltagex_ -> voltagex | 07:45 | |
| kfx | josch: yes | 07:52 |
| kfx | it's the 6.14 patches and dtb from your repo | 07:52 |
| josch | hrm... | 07:53 |
| josch | (it's also MNT's repo -- i am just a random contributor) :D | 07:54 |
| kfx | I suspect something in the kernel config is =m that should have been =y and didn't make it into the initrd | 07:58 |
| kfx | will be an easy fix | 07:58 |
| josch | minute: shiguru from the forum now also successfully tested 6.15 with imx8mplus | 09:25 |
| josch | minute: unfortunately there is a regression on a311d: the backlight only comes on once gdm starts | 09:25 |
| josch | before that, the screen remains dark | 09:25 |
| josch | any volunteers to bisect between 6.14 and 6.15 to find the breaking commit? | 09:25 |
| + mjw (~mjw@gnu.wildebeest.org) | 10:15 | |
| + andreas-e (~Andreas@2001:861:c4:f2f0::c64) | 10:23 | |
| minute | josch: backlight on pocket or classic? | 10:38 |
| josch | minute: a311d classic reform | 10:42 |
| josch | minute: the other platforms i have work fine (rk3588 classic, rk3588 pocket, imx8mq classic) | 10:43 |
| minute | josch: neat | 10:44 |
| grimmware | minute, josch: how would we feel about a pinned thread on the forum for announcing regressions, breaking changes or other things that people are liable to ask about? I ask because I also spotted the new loglevel change mentioned in https://community.mnt.re/t/login-menu-appears-is-then-covered-by-errors/3492/3 and nearly asked about it in here which would have led to y'all repeating yourselves. I wasn't aware that this might | 10:44 |
| grimmware | mess with tuigreet which I'm still using. I think if the thread were locked down to posts from minute, josch and maybe some other people who you can trust to only post relevant things in there so it doesn't get lost in discussion or tangents then it would serve as a sensible thread for people to watch. | 10:44 |
| josch | yes, this kernel upgrade seems to be a lot smoother than the last :) | 10:44 |
| josch | grimmware: i'm fine with pinning threads | 10:45 |
| josch | grimmware: instead of having one thread to check, maybe make a new one each time some announcement has to be made and pin it? | 10:45 |
| grimmware | also: do y'all need a test of 6.15 on a v1 display rk3588 pocket? | 10:45 |
| minute | josch: about the backlight, maybe a driver is missing in initramfs | 10:46 |
| josch | grimmware: i have only tested 6.15 with rk3588 v2 display pocket, so yes please :) | 10:46 |
| josch | minute: but then i wonder why that driver was included in 6.14 and not in 6.15? | 10:46 |
| minute | josch: something related to display, audio, pwm... could be anything | 10:46 |
| minute | anything that's in the dependency chain | 10:47 |
| minute | josch: one would need to check if they restructured anything on meson | 10:47 |
| josch | minute: sounds like me already having typed "make bindeb-pkg -j8" in my linux git clone will be the way to find out what's going on :) | 10:47 |
| josch | i have *two* rk3588 at home now, should be much more fun to bisect | 10:48 |
| minute | josch: haha ok... if i was debugging this i'd use break=modules or break=bottom or sth like that and do lsmod diff on the two kernels | 10:52 |
| minute | and compare details of the backlight and pwm nodes | 10:52 |
| josch | oh also a good idea! | 10:52 |
| josch | thank you for the tips -- seems like this is my project for today XD | 10:52 |
| minute | cool :3 | 10:57 |
| minute | my project will be to repair the next which i broke last night | 10:58 |
| minute | managed to damage a gpio of the rp2350a | 10:58 |
| + NEVERVERY4 (~NEVERVERY@38.137.80.140) | 10:59 | |
| josch | i managed to ripp of a mhf4 socket from its pad because the antenna cable was sitting so tight this morning... :( | 10:59 |
| - NEVERVERY4 (QUIT: Read error: Connection reset by peer) (~NEVERVERY@38.137.80.140) | 11:04 | |
| minute | josch: :// | 11:22 |
| minute | josch: btw regarding lte antennas, mouser and digikey have many models | 11:22 |
| grimmware | when testing new kernels are y'all installing every package in the archive.zip or just selected ones? | 11:22 |
| grimmware | I've historically been picking the ones I know need to be tested and then iteratively resolving the broken dependencies each time because it falls off the end of my shell history and obviously that's a massive ballache | 11:24 |
| josch | minute: indeed they do but they also have a shipping fee of 18 EUR if you don't order at least 50 EUR of stuff which makes me usually avoid them for ordering parts which (for example in case of these antennas) cost less than 2 EUR :/ | 11:28 |
| minute | grimmware: i only install the one with the biggest file size :D | 11:28 |
| josch | minute: this is also why i bought my laird antenna from shop.mntre.com instead of getting it from digikey/mouser back in the day | 11:28 |
| josch | grimmware: what is an archive.zip? | 11:29 |
| grimmware | sorry, artifacts.zip even | 11:33 |
| grimmware | my brain is mud today | 11:33 |
| josch | grimmware: i've never used that zip archive | 11:35 |
| josch | i usually find the pipeline which i triggered or which belongs to my MR and then open the reprepro job log of the pipeline i want, like this one: https://source.mnt.re/reform/reform-debian-packages/-/jobs/11212 | 11:35 |
| josch | at the eight-to-last line the job log says: | 11:36 |
| josch | deb [trusted=yes arch=arm64] https://source.mnt.re/reform/reform-debian-packages/-/jobs/11212/artifacts/raw/repo reform main | 11:36 |
| josch | i copypaste that into my /etc/apt/sources.list, then run "apt update" and then "apt install" the new kernel | 11:36 |
| gsora | trying the mesa patch as we speak! | 12:09 |
| + paperManu (~paperManu@72.10.128.164) | 12:11 | |
| gsora | firefox feels a lot smoother in web apps | 12:13 |
| gsora | almost chromium-like | 12:13 |
| minute | gsora: for me it's faster than chromium for some stuff, like mattermost | 12:18 |
| minute | zeha: GPN y/n? | 12:18 |
| gsora | i'm scrolling my bsky feed and it's so smooth | 12:19 |
| gsora | github is also very smooth now! | 12:21 |
| gsora | once vulkan 1.4 hits, nothing will stop panfrost! | 12:23 |
| - iank_ (QUIT: Quit: ZNC 1.8.2+deb2+deb11u1 - https://znc.in) (~iank@fsf/staff/iank) | 12:28 | |
| + iank (~iank@fsf/staff/iank) | 12:29 | |
| josch | i just got another 10:11 hours of pocket runtime with the bigger batteries, so this was not a fluke but seems to be reproducible. | 12:32 |
| josch | minute: this also means that i got 24 hours of runtime without any reboot after downgrading the firmware -- i will now upgrade the firmware, leave it plug in, move nothing and see what happens. If it reboots again, then it's the sysctl firmware and not my hardware, i'd say. | 12:33 |
| - mjw (QUIT: Ping timeout: 252 seconds) (~mjw@gnu.wildebeest.org) | 12:43 | |
| * Guest1629 -> mjw | 12:43 | |
| minute | josch: ok, sounds good! i hope it's just the watchdog timeout. but... the wd reset should also keep the system on, need to look into that | 12:53 |
| + gustav28 (~gustav@c-78-82-38-186.bbcust.telenor.se) | 13:02 | |
| minute | ok, next's rp2350a is swapped and thus fixed | 13:23 |
| gordon1 | minute: are you using hot air to swap QFNs with addition to miniware or just the hotplate on its own? | 13:26 |
| minute | gordon1: hot air yes | 13:26 |
| minute | gordon1: miniware just as preheater | 13:26 |
| gordon1 | ah, ok, good, i thought i missed some deep dark art of swapping chips w/o hot air | 13:27 |
| gordon1 | had to mill aluminium jaws for my stick vise since 3d printed ones are not really compatible with hot air | 13:28 |
| minute | pretty cool https://github.com/peterharperuk/pico-examples/blob/powman/powman/powman_example.c | 13:44 |
| minute | gordon1: what are you using to mill aluminum? | 13:44 |
| minute | i feel i miss some nice enclosed little cnc thing but not sure if something exists that's not a lot of pain to maintain | 13:44 |
| gordon1 | minute: i used manual mill in our local hackspace | 13:47 |
| gordon1 | small cncs sound like a good idea, but in reality it needs a lot of rigidity to mill something reasonably well | 13:48 |
| minute | ohh i see | 13:48 |
| gordon1 | i can't suggest particular model, but if you gonna go with small one - find one with the most choky frame, the thicker the better | 13:48 |
| Zaba | I think stepcraft has some decent small cnc mills | 13:50 |
| gordon1 | that's said, at the hackspace we have medium sized one, roughly 1000x1000x1000 bulk dimensions, and it is perfectly capable of milling titanium | 13:51 |
| gordon1 | got it from the uni as a donation, gutted it and put smoothieboard in and replaced steppers and spindel | 13:52 |
| gordon1 | minute: if you aim for something that big, it's gonna be useful machine https://cardiffhackspace.co.uk/wp-content/uploads/2023/03/AED9E4DD-D580-4E23-99AF-1440083E4D88-768x1024.jpg (sorry, no banana for scale, but it is roughly a bit less than 1m across) | 13:55 |
| abortretryfail | That miniware heater is fancy. I use a pancake griddle. :) | 13:57 |
| gordon1 | it's awesome to reflow little usb thumbstick-sized boards | 14:02 |
| gordon1 | a bit too small to reflow reform mb | 14:03 |
| gordon1 | but i never actually had a thought to use it as a preheater before today, it makes total sense | 14:03 |
| minute | yeah, really good if you need to swap chips | 14:07 |
| minute | BoostisBetter: do you have the new backplate already? the one for better wifi reception i mean | 14:17 |
| minute | BoostisBetter: for pocket | 14:17 |
| - paperManu (QUIT: Ping timeout: 272 seconds) (~paperManu@72.10.128.164) | 14:44 | |
| + bkeys1 (~Thunderbi@h211.16.186.173.dynamic.ip.windstream.net) | 15:45 | |
| - bkeys (QUIT: Ping timeout: 252 seconds) (~Thunderbi@h211.16.186.173.dynamic.ip.windstream.net) | 15:45 | |
| * bkeys1 -> bkeys | 15:45 | |
| bkeys | https://copr.fedorainfracloud.org/coprs/bkeys/kernel-reform/build/9171148/ | 16:03 |
| bkeys | In theory now we got the Collabara kernel + Reform patches being built as an RPM | 16:03 |
| + reformenjoyer (~reformenj@bras-base-oshwon9537w-grc-37-142-198-156-208.dsl.bell.ca) | 16:19 | |
| reformenjoyer | Hello | 16:19 |
| reformenjoyer | I got my Reform, and was happy | 16:20 |
| reformenjoyer | I opened it up, I plugged in the batteries, I turned it on (success!) and I gave it a full apt update & upgrade | 16:20 |
| reformenjoyer | the battery meter said 70% | 16:20 |
| reformenjoyer | I rebooted a couple times just to be sure | 16:21 |
| reformenjoyer | then I shut it down, using 'shut down' from the power menu in the OS | 16:21 |
| reformenjoyer | the next day, I try to turn it on, and it is dead | 16:21 |
| reformenjoyer | I plug it in to a charger, and it shows batteries at 2.1, 2.3, and 0.0 volts | 16:22 |
| reformenjoyer | Was shutting it down from within linux not a safe operation? | 16:23 |
| reformenjoyer | I got less than an hour of uptime out of this machine in total so far and I have to replace the batteries? | 16:23 |
| reformenjoyer | Please advise | 16:23 |
| andreas-e | I am not sure it is enough to shutdown from Linux (with "poweroff" or "halt" as root). | 16:25 |
| andreas-e | To be sure, I always do a "circle key" 0 now. | 16:25 |
| andreas-e | Looking at the little oled display, it still seems to do something. | 16:25 |
| andreas-e | I had the first version, which discharged even when turned off; you can revive the batteries with special chargers. | 16:27 |
| reformenjoyer | handbook 4.16 says shutdown is safe / recommended | 16:29 |
| reformenjoyer | No warnings about this battery stuff | 16:29 |
| andreas-e | I do not know. Other people in this channel will be more competent to answer the question. | 16:30 |
| reformenjoyer | thanks | 16:31 |
| bkeys | reformenjoyer: Is it a Reform Next or a Classic Reform? | 16:39 |
| kop316 | so i flashed the newest keyboard firmware, but the batteries still don't charge. plugged in, its gone from 48-40 over a few days | 16:41 |
| reformenjoyer | Reform Classic. I wasnt aware Nexts were even shipping | 16:41 |
| kop316 | it seems that when i plug it in, one of the batteries spikes to 4.2v, then goes back to 3.7v | 16:42 |
| abortretryfail | First versions of the Reform didn't have protected battery boards. | 16:42 |
| + robin (~robin@user/terpri) | 16:42 | |
| reformenjoyer | bkeys abortretryfail : It arrived in the mail with a third edition handbook - pretty sure it should be whatever the latest hardware revision is. It came with a third edition handbook. How can I even tell? | 16:44 |
| bkeys | abortretryfail: Are those still shipping? | 16:44 |
| abortretryfail | idk, i bought mine years ago. :) | 16:44 |
| bkeys | Same here mine had unprotected boards I have since bought protected boards | 16:45 |
| abortretryfail | reformenjoyer: you can look at the battery boards themselves. If they have components on them, they're the protected ones. | 16:45 |
| bkeys | But it feels like it's been forever since that has been fixed | 16:45 |
| abortretryfail | The old ones were just clips with a connector on a PCB. | 16:45 |
| reformenjoyer | pretty sure these have components and are protected boards | 16:52 |
| reformenjoyer | This is my most expensive ever paperweight now | 16:53 |
| andreas-e | As I said, whatever happened, it should be possible to charge the batteries with a specialised charger. | 16:58 |
| - reformenjoyer (QUIT: Quit: Client closed) (~reformenj@bras-base-oshwon9537w-grc-37-142-198-156-208.dsl.bell.ca) | 16:59 | |
| abortretryfail | "specialized" is a bit of an over-statement. Any charger that fits 18650 cells and has a LiFePO4 mode aught to work. | 16:59 |
| andreas-e | After taking them out and using a lab type power supply. | 16:59 |
| abortretryfail | I used a Tenergy T4s for mine several times. It works fine. | 17:00 |
| andreas-e | I had the problem that as soon as the voltage was below 1.25V, the 18650/AA/AAA charger mistook them for an AA/AAA type battery and stopped charging them at 1.25V :) | 17:00 |
| andreas-e | So I needed the lab power supply to give them a little boost above that, and then the charger itself worked. | 17:01 |
| abortretryfail | lol that's easy, just parallel a dead one with a good cell via a resistor for a bit. | 17:01 |
| andreas-e | Well, that is a bit of specialised handling, is it not? | 17:02 |
| abortretryfail | Tbh with this charger I just had to pop the battery out and put it back in a few times and that got the voltage up. | 17:02 |
| andreas-e | That may also work. It did for some of my cells, but not all of them. | 17:04 |
| abortretryfail | With the protected boards hopefully nobody will ever have to deal with that again unless they leave the thing sitting for a very long time | 17:05 |
| + reformenjoyer (~reformenj@bras-base-oshwon9537w-grc-37-142-198-156-208.dsl.bell.ca) | 17:12 | |
| - andreas-e (QUIT: Read error: Connection reset by peer) (~Andreas@2001:861:c4:f2f0::c64) | 17:21 | |
| + vkoskiv (~vkoskiv@109-204-186-7.cust.valoonet.fi) | 17:37 | |
| - reformenjoyer (QUIT: Ping timeout: 272 seconds) (~reformenj@bras-base-oshwon9537w-grc-37-142-198-156-208.dsl.bell.ca) | 17:51 | |
| bkeys | abortretryfail: I did waiting for my rk3588 upgrade and I gotta get new batteries now | 17:52 |
| bkeys | Wait so we can resurrect LifePO4 batteries? | 17:53 |
| - vkoskiv (QUIT: Remote host closed the connection) (~vkoskiv@109-204-186-7.cust.valoonet.fi) | 18:05 | |
| + colinsane (~colinunin@97-113-74-37.tukw.qwest.net) | 18:11 | |
| + vkoskiv (~vkoskiv@109-204-186-7.cust.valoonet.fi) | 18:12 | |
| abortretryfail | Yeah. It's not guaranteed, but you can try. | 18:18 |
| - colinsane (QUIT: Ping timeout: 276 seconds) (~colinunin@97-113-74-37.tukw.qwest.net) | 18:19 | |
| josch | unfortunately the people involved in the discussion 2 hours ago are already offline | 18:20 |
| josch | but: reform next are not shipping yet and are probably still months off | 18:20 |
| josch | the rk3588 classic reform batch which recently shipped might have an unfortunate bug in the lpc/keyboard firmware which prevents the reform from booting up | 18:21 |
| josch | it is unlikely that any of the batteries are dead because new classic reform ship with protected battery boards for a long while now | 18:21 |
| josch | in cases like these, good advice is to write support@mntre.com | 18:21 |
| minute | when reformenjoyer comes back, tell them the cells look like 0.0v when the protection is active | 18:40 |
| minute | you just need to let it sit for half an hour or so with the power supply connected | 18:40 |
| bremner | that's true for the (recent) pockets as well, right? | 18:49 |
| abortretryfail | Pocket won't charge a 0.0V discharge-protected cell. I was helping a friend troubleshoot hers and one cell had to be charged via an external charger to wake it back up. | 18:58 |
| abortretryfail | Well, I can't remember if that was before or after I updated the sysctl firmware. It maybe could. :) | 19:01 |
| bremner | hmm. Because I think I discharged mine all the way to zero, and it woke up fine when charged into USB PD | 19:06 |
| + AnimaInvicta (~AnimaInvi@88-120-179-216.subs.proxad.net) | 19:07 | |
| abortretryfail | Did they measure zero on a voltmeter, or was it just refusing to light up the OLED? | 19:08 |
| bremner | the latter | 19:08 |
| abortretryfail | What I'm talking about was one of the LiPo cells reading 0 on a voltmeter in addition to the OLED refusing to light up. | 19:09 |
| bremner | I thought the point of discharge protection was to prevent reaching 0v? | 19:09 |
| abortretryfail | The cells themselves have a little PCB on them with protection iirc. | 19:09 |
| abortretryfail | Yeah, it should prevent the cell itself from getting that low by disconnecting the load. | 19:09 |
| kfx | in the reform, the discharge protection is in the battery carrier board | 19:11 |
| kfx | you can use protected cells with it, but most cells are unprotected | 19:11 |
| abortretryfail | Right. In both cases, from the motherboard's perspective a discharge-protected cell reads 0.0V because it was disconnected. | 19:13 |
| - elektron (QUIT: Ping timeout: 252 seconds) (~elektron@apoc.halo.nu) | 19:14 | |
| + elektron (~elektron@apoc.halo.nu) | 19:14 | |
| kfx | abortretryfail: crucially, a voltage tester will read 0.0 on a protected cell that has discharged to cutoff, but will still read nonzero on a nonprotected cell | 19:29 |
| kfx | just brought it up as another diagnostic input | 19:29 |
| bkeys | https://copr.fedorainfracloud.org/coprs/bkeys/kernel-reform/build/9171148/ | 19:43 |
| bkeys | I hope this is the holy kernel that will finally enable rk3588 reform with Fedora | 19:43 |
| bkeys | The build should be finished within the half hour | 19:44 |
| bkeys | Gah it isn't, the stupid srpm build verifies the image -_- | 19:45 |
| kfx | bkeys: can you post your kernel .config somewhere? when I run make olddefconfig it strips out CONFIG_MUX_MMIO for me | 19:49 |
| bkeys | kfx: Right now I'm just using the Fedora default config | 19:49 |
| kfx | ok will look. thanks | 19:49 |
| bkeys | I fixed the issue of it putting the upstream kernel and it should be building the collabara kernel now on this new build | 19:51 |
| bkeys | I have wanted to do my game development hobby on the Reform using Fedora since before I even bought mine all those years ago, I will finally achieve this goal after having this thing for all these years I am determined | 19:52 |
| kfx | I've been running a completely different operating system on mine since I got it. this is the first time I'm having to build device-specific kernels and boy has the kernel got even more byzantine than I remembered | 19:53 |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:50) | 19:59 | |
| bkeys | What OS are you running on it? | 19:59 |
| + mark_ (~mjw@gnu.wildebeest.org) | 20:00 | |
| kfx | I was running 9front on imx8mq; currently trying to get a working alpine linux on rk3588 | 20:08 |
| bkeys | Yeah at some point I gotta run a diff of josch's kernel config and the Fedora config if any issues arise out of the config | 20:14 |
| bkeys | Right now my goal is to get the Collabara kernel + Reform patches built and packaged as RPM and I'll see what issues are left after that. Knowing my experience it will fix nothing but the fact the system image "just works" with edk2 means I know Fedora can do the same | 20:14 |
| bkeys | https://copr.fedorainfracloud.org/coprs/bkeys/kernel-reform/build/9172180/ | 20:21 |
| bkeys | Alright hopefully this is the build I have been dreaming of | 20:21 |
| minute | bkeys: i hadn't been around for the edk2/efi optimized reform image test, so it can boot now without special configuration in edk2? | 21:41 |
| * mjw -> Guest2002 | 21:47 | |
| - Guest2002 (QUIT: Killed (silver.libera.chat (Nickname regained by services))) (~mjw@2001:1c06:2486:a800:7602:5eff:dc71:a72c) | 21:47 | |
| * mark_ -> mjw | 21:47 | |
| + Guest2002 (~mjw@2001:1c06:2486:a800:7602:5eff:dc71:a72c) | 21:47 | |
| - gustav28 (QUIT: Quit: Quit) (~gustav@c-78-82-38-186.bbcust.telenor.se) | 22:15 | |
| - vkoskiv (QUIT: Ping timeout: 252 seconds) (~vkoskiv@109-204-186-7.cust.valoonet.fi) | 22:16 | |
| - hairu (QUIT: Remote host closed the connection) (m-uotkmd@user/hairu) | 22:20 | |
| + hairu (m-uotkmd@user/hairu) | 22:21 | |
| - hairu (QUIT: Remote host closed the connection) (m-uotkmd@user/hairu) | 22:34 | |
| + hairu (m-uotkmd@user/hairu) | 22:36 | |
| bkeys | minute: Yeah I can boot josch's image with no EFI shell or anything it's complete plug and play | 22:53 |
| minute | bkeys: ah that's really awesome | 22:54 |
| bkeys | Yeah it is, I'm working on a Fedora kernel that will hopefully do the same thing | 22:54 |
| bkeys | AFAIK all the hardware on the system image when I booted it worked, GPU accel and everything. | 22:55 |
| minute | nice. | 22:56 |
| + paperManu (~paperManu@72.10.128.164) | 23:25 | |
| josch | I want to give credit here to Helmut Grohne who has figured out how to do this with Qemu and from whom I largly copied things: https://salsa.debian.org/helmutg/debvm/-/merge_requests/49/diffs | 23:25 |
| - GNUmoon (QUIT: Remote host closed the connection) (~GNUmoon@gateway/tor-sasl/gnumoon) | 23:30 | |
| - RandyK (QUIT: Remote host closed the connection) (~RandyK@user/randyk) | 23:30 | |
| + RandyK (~RandyK@user/randyk) | 23:31 | |
| + GNUmoon (~GNUmoon@gateway/tor-sasl/gnumoon) | 23:31 | |
| josch | i just had my hands on the pocket keyboard and suddenly the backlight turned from purple to blue -- which buttons did i accidentally press? | 23:48 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
