| - espressocheese (QUIT: Quit: Client closed) (~espressoc@2601:8c1:c180:4630:d4b6:7e57:ea7:be8c) | 00:00 | |
| + robin (~robin@user/terpri) | 00:01 | |
| matt2 | minute: to be sure, i did just now connect to another external keyboard with the same result | 00:12 |
|---|---|---|
| minute | matt2: ok, then that's past my experience, maybe you can try what josch suggested | 00:18 |
| minute | i need some time off from work now :D (midnight on saturday) | 00:18 |
| josch | good night! | 00:19 |
| + espressocheese (~espressoc@2601:8c1:c180:4630:d4b6:7e57:ea7:be8c) | 00:19 | |
| josch | matt2: sorry, i also know of no way to recover a lost luks passphrase | 00:19 |
| josch | matt2: maybe take a full disk backup and then install a new system? that'd give you some time to remember | 00:19 |
| josch | last time i had a similar problem i knew that my passphrase was some combination of X and just wrote a small script to enumerate all possible permutations | 00:20 |
| matt2 | josch: that's basically what i'm hoping to get to. assuming that's the route i want to take, no issue eg flashing a micro sd and booting from that or is there a better way to get to where i can write and execute such a script? maybe doable at intramfs and i'm not thinking through it. would need to be in a similar spot to take a full disk backup i believe | 00:33 |
| josch | matt2: i think booting from sd-card is easiest because then you have a full Debian linux and are not constrained by the limits of the tiny initramfs | 00:35 |
| + chomwitt (~chomwitt@2a02:85f:9a74:ce00:1ac0:4dff:fedb:a3f1) | 00:35 | |
| espressocheese | hi, I dont think my speaker is working | 00:39 |
| espressocheese | can someone pls help? | 00:41 |
| josch | espressocheese: hrm... i only remember recent problems with rk3588 classic reform: https://community.mnt.re/t/rk3588-upgrade-no-audio-from-built-in-speakers/3640/6 | 00:43 |
| josch | espressocheese: can you pastebin your output of systemctl status reform-hw-setup.service somewhere? | 00:43 |
| espressocheese | hold on | 00:43 |
| + espressocheese55 (~espressoc@2601:8c1:c180:4630::ef4b) | 00:47 | |
| matt2 | josch: thanks much. will go that route and hopefully clear enough cobwebs to remember the password | 00:50 |
| josch | matt2: much success! i hope you get your data back! | 00:50 |
| espressocheese55 | × reform-hw-setup.service - MNT Reform Hardware Defaults Setup | 00:52 |
| espressocheese55 | Loaded: loaded (/usr/lib/systemd/system/reform-hw-setup.service; enabled; pres> | 00:52 |
| espressocheese55 | Active: failed (Result: exit-code) since Sat 2025-08-23 16:45:07 MDT; 6min ago | 00:52 |
| espressocheese55 | Invocation: bfc2c1c2dfaf4089ae3765d5860a3225 | 00:52 |
| espressocheese55 | Docs: man:reform-hw-setup(1) | 00:52 |
| espressocheese55 | Process: 652 ExecStart=/usr/bin/reform-hw-setup (code=exited, status=1/FAILURE) | 00:52 |
| espressocheese55 | Main PID: 652 (code=exited, status=1/FAILURE) | 00:52 |
| espressocheese55 | Mem peak: 1.8M | 00:52 |
| espressocheese55 | CPU: 80ms | 00:52 |
| espressocheese55 | Warning: some journal files were not opened due to insufficient permissions. | 00:52 |
| espressocheese55 | is that ok? or should I run it with admin? | 00:52 |
| josch | espressocheese: please send more than three lines of text using a pastebin service instead of pasting them verbatim into irc | 00:53 |
| josch | espressocheese: you could run this: | 00:53 |
| espressocheese55 | oh, sorry | 00:53 |
| espressocheese55 | run what josch | 00:55 |
| josch | one sec please | 00:55 |
| espressocheese55 | sounds good | 00:55 |
| espressocheese | Im signed in on two places, so thats why I have two accounts | 00:58 |
| josch | espressocheese: sudo journalctl --boot --no-pager --unit reform-hw-setup.service | curl -F 'arg=<-' https://mister-muffin.de/paste | 00:58 |
| josch | espressocheese: that will redirect the output of journalctl for the reform-hw-setup.service unit of the current boot into my pastebin | 00:59 |
| josch | espressocheese: the command will output the url that you can paste here | 00:59 |
| espressocheese55 | done | 01:02 |
| espressocheese55 | did you get it? | 01:03 |
| josch | espressocheese: you have to send me the link the command printed :) | 01:10 |
| espressocheese55 | https://mister-muffin.de/p/69Pz.txt | 01:11 |
| josch | aha indeed: | 01:11 |
| josch | Aug 23 16:45:07 gal-pocket-mnt reform-hw-setup[850]: amixer: Unable to find simple control 'Speaker Driver',0 | 01:11 |
| josch | this is a problem that is on my todo list | 01:12 |
| josch | it is the problem of the forum thread here: https://community.mnt.re/t/rk3588-upgrade-no-audio-from-built-in-speakers/3640/6 | 01:12 |
| josch | espressocheese: which kernel are you running? | 01:12 |
| espressocheese55 | idk | 01:12 |
| josch | espressocheese: uname -k | 01:12 |
| espressocheese55 | I get invalid option | 01:13 |
| josch | whoops sorry | 01:13 |
| josch | uname -r | 01:13 |
| espressocheese55 | 6.15.4-mnt-reform-arm64 | 01:14 |
| espressocheese55 | I just got the device today; | 01:14 |
| josch | thank you | 01:15 |
| espressocheese55 | sure | 01:15 |
| josch | espressocheese: please watch that forum thread | 01:15 |
| espressocheese55 | which one? | 01:15 |
| josch | it is the problem of the forum thread here: https://community.mnt.re/t/rk3588-upgrade-no-audio-from-built-in-speakers/3640/6 | 01:16 |
| espressocheese55 | im on the rk chip | 01:16 |
| josch | good, i have a pocket with rk3588 here as well so i should be able to reproduce and fix the issue | 01:16 |
| espressocheese55 | nice | 01:17 |
| espressocheese55 | kb is slightly cramped' | 01:17 |
| espressocheese55 | is there a good desktop irc client for the pocket reform? | 01:19 |
| - aphistic (QUIT: Server closed connection) (sid347194@id-347194.ilkley.irccloud.com) | 01:19 | |
| + aphistic (sid347194@id-347194.ilkley.irccloud.com) | 01:19 | |
| minute | josch: btw note for later: today, with the new backlight pwm freq, one can hear that freq on usb audio very distinctly, except if near 100% or near 0% | 01:20 |
| espressocheese55 | and is it just for me, or does the irc connection close for everyone when they leave? | 01:20 |
| josch | minute: yes, others noted that as well :/ | 01:20 |
| minute | josch: ok, on monday i need to find a new set of frequencies | 01:21 |
| josch | whoops, it's late already... | 01:22 |
| espressocheese55 | lol | 01:22 |
| espressocheese55 | I guess youre in europe so that makes sense | 01:22 |
| espressocheese55 | did anyone try running oss-20b on rk3588? | 01:30 |
| - kfx (PART: .) (~kfx@wopr.sciops.net) | 01:34 | |
| espressocheese55 | I assume oss-20 would run reasonably well | 01:37 |
| gordon1 | there is a llama.cpp issue https://github.com/ggml-org/llama.cpp/issues/722 that goes over the rk3588 npu performance, and tl;dr; it is not amazing | 01:40 |
| - espressocheese55 (QUIT: Quit: Client closed) (~espressoc@2601:8c1:c180:4630::ef4b) | 01:41 | |
| + espressocheese14 (~espressoc@2601:8c1:c180:4630::ef4b) | 01:41 | |
| espressocheese14 | hi | 01:41 |
| espressocheese14 | ill try to run gpt-oss | 01:44 |
| espressocheese14 | bc its moe, so it should run reasonably well, I asume?' | 01:44 |
| espressocheese14 | also, how do I know if the ssd is mounted? | 01:45 |
| gordon1 | what are you planning to run it with? | 01:45 |
| espressocheese14 | run what gpt-oss? the rk3588 | 01:45 |
| gordon1 | i mean what software? or is it a full software package, not just the model? | 01:46 |
| espressocheese14 | ollama | 01:46 |
| gordon1 | > how do I know if the ssd is mounted? | 01:46 |
| gordon1 | mount | grep nvme0 | 01:46 |
| gordon1 | you did miss my message, didn't you? | 01:47 |
| gordon1 | <gordon1> there is a llama.cpp issue https://github.com/ggml-org/llama.cpp/issues/722 that goes over the rk3588 npu performance, and tl;dr; it is not amazing | 01:47 |
| gordon1 | i think llama.cpp does not support rk3588 NPU | 01:47 |
| espressocheese14 | I saw that messaage | 01:47 |
| espressocheese14 | thanks | 01:47 |
| espressocheese14 | and the ssd is not mounted | 01:47 |
| gordon1 | it is closed as not planned, and ithink cpu-only run will be even slower | 01:47 |
| espressocheese14 | How do I move my home directory over? | 01:48 |
| gordon1 | you're running stock debian, aren't you? | 01:48 |
| espressocheese14 | im running the default os | 01:49 |
| espressocheese14 | whatever came with the device | 01:49 |
| gordon1 | yeah, i not the best person to asnwer that, probably a good idea to wait for someone who knows about it, iirc there is a script to migrate to nvme | 01:50 |
| minute | sudo reform-setup-encrypted-disk | 01:50 |
| gordon1 | espressocheese14: actually you probably right, moe can run much better when not offloaded to gpu with dedicated ram, so it is worth trying | 01:52 |
| espressocheese14 | ill let you know | 01:52 |
| espressocheese14 | oss-20b only has 3.6b active and it fits on rk3588 | 01:53 |
| minute | apropos, (i am extremely skeptical of llm's usefulness for anything), and i didn't want to use one but i googled for "prime number between 50000 and 60000", and that ai overview thing said, 50009 | 01:54 |
| gordon1 | well, i'm fitting 40b into 24gb of ram w/o moe, so there is definitely enough ram | 01:54 |
| minute | then i googled, "is 50009 prime?" | 01:54 |
| minute | and that AI overview literally writes: No, 50009 is not a prime number because it is divisible by 3 and 16669.1, making it a composite number. | 01:54 |
| minute | LOL | 01:54 |
| gordon1 | minute: yeah, they're not amazing at math | 01:54 |
| espressocheese14 | but they should be good at other stuff | 01:55 |
| minute | funnily, the numbers in the first question are images | 01:55 |
| gordon1 | they're not amazing at anything pretty much, but they're ok at offloading some tasks | 01:55 |
| minute | so they do some trick to work around something there | 01:55 |
| espressocheese14 | could probablyt finetune for matrh | 01:55 |
| minute | not gonna happen :D anyway, it's a pet peeve of mine, i hope you'll enjoy your pocket reform in any case | 01:56 |
| espressocheese14 | lol | 01:56 |
| gordon1 | there are some models that claim to be good at math, but they really aren't | 01:56 |
| espressocheese14 | no finetuning on pocket reform | 01:56 |
| minute | btw 50009 is also not prime. it is 43 x 1163 | 01:57 |
| minute | don't believe what the llms say... | 01:58 |
| espressocheese14 | pretty bad token per second but still kinda usable | 01:58 |
| gordon1 | minute: so i think successful use i saw recently is to feed the yt-dlp transcript of a youtube video into the LLM and as it to summarize it for you, works reasonably well if you don't really care about watching it | 01:59 |
| minute | gordon1: ok! but what iffff... there was something important that it missed :D | 01:59 |
| gordon1 | yeah, it's for unimportant stuff | 01:59 |
| minute | gordon1: or got it wrong and you have that in your homework assignment and then it'll be an embarrasment | 02:00 |
| gordon1 | for when you can't be bothered to watch it yourself | 02:00 |
| gordon1 | yeah, that's definitenly not the use-case for llms | 02:00 |
| minute | they're also really bad at programming | 02:01 |
| gordon1 | that is true | 02:01 |
| minute | recently i found this trackball x/y bug in gnome | 02:01 |
| minute | and it was a nice algorithmic error in a function | 02:01 |
| - espressocheese14 (QUIT: Quit: Client closed) (~espressoc@2601:8c1:c180:4630::ef4b) | 02:01 | |
| minute | so i thought, great testcase for those billions of dollars! | 02:01 |
| minute | like, can it spot/explain the error | 02:02 |
| minute | claude went into some frenzy of not being able to reproduce its theory, it was quite hilarious | 02:02 |
| gordon1 | i did successfully use it few times with devstral but the effort spent into making it do what i asked to wasn't worth it | 02:03 |
| espressocheese | help, I bricked my pocket reform | 02:03 |
| minute | happens! | 02:03 |
| minute | actually, bricking is probably rare | 02:03 |
| espressocheese | I tried the ssd thing and now it will not boot | 02:04 |
| espressocheese | im stuck in initramfs | 02:04 |
| minute | nothing on screen at all? ah, initramfs | 02:04 |
| + xktr (~xktr@user/xktr) | 02:05 | |
| gordon1 | minute: i had a funny moment when we tested devstral and we asked it to write some simple program in haskell, it tried three times then gave up and wrote it in python | 02:05 |
| minute | gordon1: yeah they also have a really hard time with C, basically anything except python | 02:05 |
| + echeese123 (~echeese12@2606:40:8fc7:a3e::60:71b0) | 02:05 | |
| espressocheese | so how do I fix it? | 02:06 |
| minute | espressocheese: any error messages? | 02:06 |
| echeese123 | Yes, the password was wrong | 02:06 |
| espressocheese | I dont think I put a password, but it kept insisting for one | 02:06 |
| minute | well, sure you did | 02:06 |
| minute | reform-setup-encrypted-disk asks for one | 02:06 |
| espressocheese | pretty sure I just put empty | 02:07 |
| espressocheese | I just hit enter | 02:07 |
| minute | josch: is that possible? | 02:07 |
| espressocheese | apparently | 02:07 |
| espressocheese | lol | 02:07 |
| gordon1 | oh | 02:07 |
| minute | espressocheese: maybe you typed your usual password? | 02:07 |
| echeese123 | Nope | 02:08 |
| echeese123 | Nope | 02:08 |
| minute | no idea then except flashing a microSD card with a fresh system and trying again | 02:09 |
| echeese123 | Ok, I'll try that | 02:09 |
| - echeese123 (QUIT: Client Quit) (~echeese12@2606:40:8fc7:a3e::60:71b0) | 02:10 | |
| gordon1 | apparently you can http://0x0.st/KiRK.txt | 02:10 |
| gordon1 | til | 02:10 |
| minute | gordon1: uh oh | 02:10 |
| + echeese123 (~echeese12@2606:40:8fc7:a3e::60:71b0) | 02:10 | |
| minute | gordon1: and then the drive isn't decryptable or what... | 02:10 |
| echeese123 | So nothing I can do except reinstall? | 02:10 |
| minute | echeese123: if you encrypted the drive and don't know the passphrase, and enter isn't accepted, then yes. you're the first person to do this i think | 02:11 |
| minute | uncharted territory | 02:11 |
| gordon1 | ... and you can open in with cryptsetup luksOpen by just hitting enter | 02:11 |
| minute | echeese123: have you tried just pressing enter for the passphrase on boot? | 02:11 |
| gordon1 | so i guess it might be issue in initramfs that does not allow for empty password? | 02:11 |
| minute | ah could be | 02:11 |
| minute | i don't remember the commands to manually open the disk in initramfs | 02:12 |
| gordon1 | echeese123: anyway, it is possible to add a password to existing volume from initramfs, if you want to do that | 02:12 |
| gordon1 | if there is a rescue shell | 02:13 |
| gordon1 | is there? | 02:13 |
| echeese123 | Idk | 02:14 |
| gordon1 | sorry, i'm not familiar with debian initrd, what would happen if you enter password incorrectly three times? will it drop you to a shell? | 02:14 |
| echeese123 | Of some sort | 02:15 |
| echeese123 | Can we continue in an hour or so? | 02:16 |
| gordon1 | well, i definitely will be sleeping by then | 02:16 |
| - echeese123 (QUIT: Quit: Client closed) (~echeese12@2606:40:8fc7:a3e::60:71b0) | 02:17 | |
| + echeese123 (~echeese12@2606:40:8fc7:a3e::60:71b0) | 02:17 | |
| echeese123 | Ah, I see... Hopefully someone will help | 02:17 |
| echeese123 | Otherwise I'll try it tomorrow | 02:17 |
| - echeese123 (QUIT: Client Quit) (~echeese12@2606:40:8fc7:a3e::60:71b0) | 02:18 | |
| + echeese123 (~echeese12@2606:40:8fc7:a3e::60:71b0) | 02:19 | |
| - echeese123 (QUIT: Client Quit) (~echeese12@2606:40:8fc7:a3e::60:71b0) | 02:20 | |
| - spew (QUIT: Server closed connection) (e9e8135d87@user/spew) | 02:28 | |
| + spew (e9e8135d87@user/spew) | 02:28 | |
| - espressocheese (QUIT: Ping timeout: 250 seconds) (~espressoc@2601:8c1:c180:4630:d4b6:7e57:ea7:be8c) | 02:54 | |
| - paperManu (QUIT: Ping timeout: 258 seconds) (~paperManu@72.10.129.226) | 03:11 | |
| + espressocheese (~espressoc@2601:8c1:c180:4630:d4b6:7e57:ea7:be8c) | 03:23 | |
| espressocheese | hi, is anyone up who can help me with my mnt pocket reform> | 03:23 |
| espressocheese | im stuck at the initramfs screen | 03:25 |
| + emdevt (~emdevt@2001:7e8:fa17:7901:889:cfa1:9c37:6514) | 03:26 | |
| espressocheese | hi | 03:26 |
| espressocheese | could you help me pls? | 03:26 |
| espressocheese | im stuck on initramfs screen | 03:27 |
| espressocheese | im in, but I want to fix it permenantely | 03:33 |
| + espressocheese21 (~espressoc@2601:8c1:c180:4630::ef4b) | 03:43 | |
| - matt2 (QUIT: Quit: WeeChat 4.6.1) (~matt@97-122-64-98.hlrn.qwest.net) | 04:00 | |
| - op_4 (QUIT: Remote host closed the connection) (~tslil@user/op-4/x-9116473) | 04:05 | |
| - mjw (QUIT: Ping timeout: 256 seconds) (~mjw@gnu.wildebeest.org) | 04:05 | |
| + op_4 (~tslil@user/op-4/x-9116473) | 04:05 | |
| - espressocheese21 (QUIT: Quit: Client closed) (~espressoc@2601:8c1:c180:4630::ef4b) | 04:30 | |
| + espressocheese67 (~espressoc@2601:8c1:c180:4630::ef4b) | 04:34 | |
| - zoglesby (QUIT: Server closed connection) (f0f8ca1525@user/zoglesby) | 04:58 | |
| + zoglesby (f0f8ca1525@user/zoglesby) | 04:58 | |
| - skipwich (QUIT: Remote host closed the connection) (~skipwich@user/skipwich) | 05:01 | |
| + skipwich (~skipwich@user/skipwich) | 05:01 | |
| ch | usually an update-initramfs -u should fix such issues | 05:06 |
| ch | might depend on the actual issue | 05:06 |
| - espressocheese67 (QUIT: Quit: Client closed) (~espressoc@2601:8c1:c180:4630::ef4b) | 05:07 | |
| + espressocheese87 (~espressoc@2601:8c1:c180:4630:d4b6:7e57:ea7:be8c) | 05:14 | |
| espressocheese87 | hi, I got it working. But I'd like to use the GPU/NPU for AI models. I dont think the default system supports that, can someone pls help with that? thanks! | 05:15 |
| - espressocheese (QUIT: Ping timeout: 250 seconds) (~espressoc@2601:8c1:c180:4630:d4b6:7e57:ea7:be8c) | 05:16 | |
| - chomwitt (QUIT: Ping timeout: 248 seconds) (~chomwitt@2a02:85f:9a74:ce00:1ac0:4dff:fedb:a3f1) | 05:24 | |
| - RandyK (QUIT: Remote host closed the connection) (~RandyK@user/randyk) | 08:10 | |
| + RandyK (~RandyK@user/randyk) | 08:10 | |
| josch | espressocheese87: the last time (and i think even only time?) somebody talked about the npu of the rk3588 was here: https://mntre.com/reform-irc-logs/2025-08-14.log.html#t19:35:01 | 08:25 |
| - aperezdc (QUIT: Server closed connection) (~aperezdc@46.23.89.43) | 08:29 | |
| + aperezdc (~aperezdc@2a03:6000:6e61:633::43) | 08:29 | |
| - f_ (QUIT: Server closed connection) (16abab341f@postmarketOS/funderscore) | 09:41 | |
| + f_ (16abab341f@postmarketOS/funderscore) | 09:41 | |
| + gustav28 (~gustav@c-92-32-80-134.bbcust.telenor.se) | 10:02 | |
| + reform4343 (~rene@p200300d0ff1c1f007fc2100dbdd09829.dip0.t-ipconnect.de) | 10:50 | |
| * reform4343 -> rwa_ | 10:50 | |
| - espressocheese87 (QUIT: Ping timeout: 250 seconds) (~espressoc@2601:8c1:c180:4630:d4b6:7e57:ea7:be8c) | 11:28 | |
| - rwa_ (QUIT: Quit: Leaving) (~rene@p200300d0ff1c1f007fc2100dbdd09829.dip0.t-ipconnect.de) | 11:33 | |
| - vkoskiv (QUIT: Server closed connection) (~vkoskiv@109-204-186-7.cust.valoonet.fi) | 11:49 | |
| + vkoskiv (~vkoskiv@109-204-186-7.cust.valoonet.fi) | 11:50 | |
| minute | pretty cool, via casey https://sigma-star.at/blog/2023/07/embedded-go-prog/ | 12:19 |
| + chomwitt (~chomwitt@2a02:85f:9a74:ce00:1ac0:4dff:fedb:a3f1) | 12:26 | |
| f_ | so I just merged matzipan's port of the a311d reform2 in pmOS, if anyone wants to give it a try ^^ https://wiki.postmarketos.org/wiki/MNT_Research_Reform_2_with_BPI-CM4_(mnt-reform2-bpi-cm4) | 12:41 |
| - Zaba (QUIT: Server closed connection) (80b9b4b35e@2a03:6000:1812:100::116) | 13:13 | |
| + Zaba (80b9b4b35e@2a03:6000:1812:100::116) | 13:13 | |
| - sbp (QUIT: Server closed connection) (~sbp@apache/doge/sbp) | 13:21 | |
| + sbp (~sbp@2001:19f0:6800:1102:5400:ff:fe11:39a1) | 13:21 | |
| - ndufresne (QUIT: Server closed connection) (~ndufresne@apple.collaboradmins.com) | 13:36 | |
| + ndufresne (~ndufresne@apple.collaboradmins.com) | 13:37 | |
| + mjw (~mjw@gnu.wildebeest.org) | 13:43 | |
| + paperManu (~paperManu@72.10.129.226) | 13:50 | |
| + mark_ (~mjw@gnu.wildebeest.org) | 13:55 | |
| - mjw (QUIT: Ping timeout: 256 seconds) (~mjw@gnu.wildebeest.org) | 13:57 | |
| * mark_ -> mjw | 14:42 | |
| - burley (QUIT: Server closed connection) (fc32d5b940@user/burley) | 14:50 | |
| + burley (fc32d5b940@user/burley) | 14:50 | |
| - sad_plan (QUIT: Server closed connection) (sadplan@tilde.club) | 15:26 | |
| + sadplan (sadplan@tilde.club) | 15:26 | |
| - sbp (QUIT: Changing host) (~sbp@2001:19f0:6800:1102:5400:ff:fe11:39a1) | 15:41 | |
| + sbp (~sbp@apache/doge/sbp) | 15:41 | |
| irimi1 | minute: okay, i've disconnected the wifi card antenna (only had one connected; from the additional module), tried without the lid (as a way to test without even the disconnected antenna cable), pressed the plugs in pretty firmly, moved the cable around a bit to try different positions, and none of that worked unfortunately :/ | 16:26 |
| irimi1 | minute: is there anything else i could try before i put the lids back on properly? | 16:27 |
| jfred | Whoa, that's weird... I just umplugged my Pocket Reform and the display glitched out, then turned pink, then went to static o.o | 16:30 |
| jfred | Rebooted the machine and the display was then flickering for a bit after booting, seems to have stopped now | 16:30 |
| - emdevt (QUIT: Ping timeout: 248 seconds) (~emdevt@2001:7e8:fa17:7901:889:cfa1:9c37:6514) | 16:31 | |
| jfred | Wait no it's still flickering when something darker is on the screen | 16:32 |
| jfred | My monitor at home did something like this a while back too and reducing the backlight brightness for a while seemed to help there, gonna try that | 16:33 |
| + matt2 (~matt@97-122-64-98.hlrn.qwest.net) | 16:35 | |
| jfred | oh it happened again when I plugged back in, and this time I caught it on camera: https://ibb.co/wN1WpVdR | 16:49 |
| jfred | brownout or something? | 16:49 |
| + espressocheese (~espressoc@2601:8c1:c180:4630:d4b6:7e57:ea7:be8c) | 17:33 | |
| - espressocheese (QUIT: Client Quit) (~espressoc@2601:8c1:c180:4630:d4b6:7e57:ea7:be8c) | 17:37 | |
| - pastaonirc (QUIT: Quit: WeeChat 4.7.0) (~pastaonir@user/xha) | 17:49 | |
| - bkeys (QUIT: Ping timeout: 248 seconds) (~Thunderbi@h193.131.19.98.dynamic.ip.windstream.net) | 18:27 | |
| + espressocheese (~espressoc@2601:8c1:c180:4630:102c:4914:7fb:e38a) | 19:13 | |
| espressocheese | Hi, I have a question: I am trying to do some ML either on the GPU or the NPU, but I get a bus error whenever I try to run clinfo... what can i do about that? | 19:15 |
| + espressocheese_p (~espressoc@2601:8c1:c180:4630::a430) | 19:17 | |
| + AnimaInvicta (~AnimaInvi@88-120-179-216.subs.proxad.net) | 19:31 | |
| + bkeys (~Thunderbi@134.22.115.162) | 19:34 | |
| espressocheese_p | just fyi, audio does not work at all on pocket with rk3588, not speakers and not headphones | 19:40 |
| espressocheese_p | is that just me? or can others get headphone audio? | 19:41 |
| - bkeys (QUIT: Remote host closed the connection) (~Thunderbi@134.22.115.162) | 19:45 | |
| + bkeys (~Thunderbi@134.22.115.162) | 19:46 | |
| minute | jfred: the flickering will go away after 10 minutes or so. it's some oscillation effect in tfts. i'll work on this issue soon | 19:53 |
| minute | irimi1: not really, i'm sorry. we should make a warranty case | 19:53 |
| minute | espressocheese_p: someone already sent you a link about how to fix the audio, it's on the community forum | 19:54 |
| espressocheese_p | I thought they didnt have a fix yet? but ill read it over | 19:54 |
| espressocheese_p | this post, right? https://community.mnt.re/t/rk3588-upgrade-no-audio-from-built-in-speakers/3640/6 | 19:55 |
| espressocheese_p | minute: could you please help? I think I missed the irc message | 20:02 |
| + espressocheese33 (~espressoc@2601:8c1:c180:4630:102c:4914:7fb:e38a) | 20:03 | |
| + echeese_pocket (~echeese_p@2601:8c1:c180:4630::a430) | 20:04 | |
| echeese_pocket | Hi, I got kicked out of irc | 20:04 |
| echeese_pocket | but im back now | 20:05 |
| echeese_pocket | minute: could you please repost the support article? | 20:06 |
| - espressocheese_p (QUIT: Ping timeout: 250 seconds) (~espressoc@2601:8c1:c180:4630::a430) | 20:07 | |
| - espressocheese (QUIT: Ping timeout: 250 seconds) (~espressoc@2601:8c1:c180:4630:102c:4914:7fb:e38a) | 20:07 | |
| - chomwitt (QUIT: Ping timeout: 248 seconds) (~chomwitt@2a02:85f:9a74:ce00:1ac0:4dff:fedb:a3f1) | 20:08 | |
| echeese_pocket | I should find a better irc solution | 20:09 |
| irimi1 | minute: sorry, i tried ^^" should i just use the form at support.mnt.re for that? | 20:10 |
| echeese_pocket | minute; | 20:11 |
| echeese_pocket | if you posted the support link, then I missed it | 20:11 |
| echeese_pocket | I dont think it even detects my headphones | 20:19 |
| echeese_pocket | so I cannot get ANY audio from my pocket reform atm | 20:20 |
| * sadplan -> sad_plan | 20:22 | |
| - espressocheese33 (QUIT: Ping timeout: 250 seconds) (~espressoc@2601:8c1:c180:4630:102c:4914:7fb:e38a) | 20:24 | |
| - echeese_pocket (QUIT: Ping timeout: 250 seconds) (~echeese_p@2601:8c1:c180:4630::a430) | 20:24 | |
| + espressocheese (~espressoc@2601:8c1:c180:4630:102c:4914:7fb:e38a) | 20:25 | |
| + echeese_pocket (~echeese_p@2601:8c1:c180:4630::a430) | 20:25 | |
| - echeese_pocket (QUIT: Client Quit) (~echeese_p@2601:8c1:c180:4630::a430) | 20:26 | |
| + echeese_pocket (~echeese_p@2601:8c1:c180:4630::a430) | 20:28 | |
| echeese_pocket | I figured it out | 20:28 |
| echeese_pocket | updated the libraries and chose the audio source | 20:28 |
| - bkeys (QUIT: Ping timeout: 256 seconds) (~Thunderbi@134.22.115.162) | 20:33 | |
| - echeese_pocket (QUIT: Ping timeout: 250 seconds) (~echeese_p@2601:8c1:c180:4630::a430) | 20:41 | |
| + bkeys (~Thunderbi@134.22.115.162) | 20:46 | |
| - bkeys (QUIT: Ping timeout: 255 seconds) (~Thunderbi@134.22.115.162) | 20:59 | |
| + echeese_pocket (~echeese_p@2601:8c1:c180:4630::a430) | 21:05 | |
| - espressocheese (QUIT: Ping timeout: 250 seconds) (~espressoc@2601:8c1:c180:4630:102c:4914:7fb:e38a) | 21:06 | |
| + chomwitt (~chomwitt@2a02:85f:9a74:ce00:1ac0:4dff:fedb:a3f1) | 21:22 | |
| - echeese_pocket (QUIT: Ping timeout: 250 seconds) (~echeese_p@2601:8c1:c180:4630::a430) | 21:24 | |
| + bkeys (~Thunderbi@134.22.115.162) | 21:32 | |
| - chomwitt (QUIT: Ping timeout: 248 seconds) (~chomwitt@2a02:85f:9a74:ce00:1ac0:4dff:fedb:a3f1) | 21:43 | |
| + erle (~erle@user/erle) | 22:03 | |
| - bkeys (QUIT: Ping timeout: 248 seconds) (~Thunderbi@134.22.115.162) | 22:05 | |
| + echeese_pocket (~echeese_p@2601:8c1:c180:4630::a430) | 22:06 | |
| minute | echeese_pocket: glad you found it | 22:08 |
| - plomlompom (QUIT: Server closed connection) (~plom@play.plomlompom.com) | 22:13 | |
| + plomlompom (~plom@oldplay.plomlompom.com) | 22:14 | |
| - echeese_pocket (QUIT: Ping timeout: 250 seconds) (~echeese_p@2601:8c1:c180:4630::a430) | 22:14 | |
| - gustav28 (QUIT: Quit: Quit) (~gustav@c-92-32-80-134.bbcust.telenor.se) | 22:15 | |
| - matt2 (QUIT: Ping timeout: 258 seconds) (~matt@97-122-64-98.hlrn.qwest.net) | 22:31 | |
| - antti (QUIT: Ping timeout: 258 seconds) (~antti@user/antti) | 22:31 | |
| + antti (~antti@user/antti) | 22:32 | |
| + matt2 (~matt@97-122-64-98.hlrn.qwest.net) | 22:32 | |
| + echeese_pocket (~echeese_p@2601:8c1:c180:4630::a430) | 22:35 | |
| gordon1 | yeah with new kernel and new patches i cannot get past "phy link never came up" with hibernate testing, i need to dig up some logs and figure out how i managed to get past it last time | 22:43 |
| gordon1 | there is a small piece of the puzzle missing | 22:44 |
| bremner | gordon1: OK, thanks for confirming it's not just me ;) | 22:44 |
| gordon1 | bremner: it was pretty much the case before, then i took some april patch that briefly fixed it, but now it is broken again | 22:45 |
| gordon1 | i'm saying "fixed", it wasn't really fixed, but at least it was resuming with PCIe broken but you can recover it by resetting the interface and re-adding the devices | 22:46 |
| + mhoye (~mhoye@li319-32.members.linode.com) | 22:46 | |
| bremner | aha. So maybe the more recent pcie reset patches would help with the post-resume step | 22:46 |
| gordon1 | also throttling has been observed in the wild https://tinystash.undef.im/il/S7M58hqvX6rVGpNFGgfZLpTeQV9q8JkMdexny18Xw1p1WftSmvbvrocW3YsKDnbvq2LpodjYSJATUg7jGLwkySE.png (green arrow is the cpu 6 and 7 core freq dipping, blue arrow is thermal throttling notch, notch 3 out of 6 or so), the reason it is too hard to notice is that it activates for a brief second and then goes away, i guess for it to | 22:46 |
| gordon1 | stick whole reform's case needs to warm up so thermal equilibrium somewhere in the future many hours from now | 22:46 |
| gordon1 | so i think rightnow i took patches from HEAD of reform-debian-packages, it does include some pcie reset patch | 22:48 |
| gordon1 | bremner: are you talking about that one? https://patchwork.kernel.org/project/linux-rockchip/list/?series=982601 | 22:49 |
| bremner | gordon1: yes. It doesn't help with the PHY link not coming up, to be clear | 22:51 |
| gordon1 | ah, nevermind, it was PM support patch that i noticed, not link reset | 22:51 |
| gordon1 | ffs i think the main obstacle of me testing the hibernate now is absence of working usb-c cable >.< | 22:58 |
| bremner | like a normal usb-c cable, not a fancy serial console thing? | 23:01 |
| + bkeys (~Thunderbi@134.22.115.162) | 23:03 | |
| gordon1 | yep patch didn't help | 23:11 |
| gordon1 | yeah, mine is broken, it works only on particular alignment, just need to get another one | 23:12 |
| bremner | gordon1: you mean the serial console cable is broken? there bugs in the docs, fwiw | 23:14 |
| bremner | or normal usb c cable is broken, that is boring but likely ;) | 23:14 |
| gordon1 | i mean cheap shitty usb-a to usb-c cable i use to connect reform to another machine is falling apart and i don't have another one | 23:15 |
| gordon1 | it's fine, i'm just ranting for no reason | 23:15 |
| bremner | sure, understood. | 23:15 |
| - bkeys (QUIT: Ping timeout: 252 seconds) (~Thunderbi@134.22.115.162) | 23:16 | |
| + bkeys (~Thunderbi@134.22.115.162) | 23:19 | |
| gordon1 | scary https://github.com/orangepi-xunlong/orangepi-build/issues/250 | 23:22 |
| gordon1 | but probably not related | 23:22 |
| gordon1 | should i grab hairdryer just in case? | 23:26 |
| + chomwitt (~chomwitt@2a02:85f:9a74:ce00:1ac0:4dff:fedb:a3f1) | 23:37 | |
| - bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@134.22.115.162) | 23:46 | |
| + bkeys1 (~Thunderbi@134.22.115.162) | 23:46 | |
| - bkeys1 (QUIT: Ping timeout: 256 seconds) (~Thunderbi@134.22.115.162) | 23:50 | |
| + espressocheese (~espressoc@2601:8c1:c180:4630:102c:4914:7fb:e38a) | 23:58 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
