| elb | I think those transfers are limited to 64 bytes by protocol | 00:11 |
|---|---|---|
| elb | you have to use a different mode for things like block transfers | 00:11 |
| vkoskiv | minute: yeah, the atmel one. I've been working on the OLED project again recently, I really would benefit from a firm deadline, otherwise it'll never be published :D | 00:38 |
| - bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@h69.128.19.98.dynamic.ip.windstream.net) | 01:36 | |
| + bkeys1 (~Thunderbi@98.19.128.69) | 01:36 | |
| * bkeys1 -> bkeys | 01:38 | |
| - chomwitt (QUIT: Ping timeout: 260 seconds) (~chomwitt@2a02:85f:9a0a:f200:1ac0:4dff:fedb:a3f1) | 01:47 | |
| + ohnoitsnoah (~ohnoitsno@23.112.150.169) | 01:54 | |
| - bkeys (QUIT: Ping timeout: 240 seconds) (~Thunderbi@98.19.128.69) | 02:15 | |
| - Svp (QUIT: Ping timeout: 240 seconds) (~svp@host-79-7-240-189.business.telecomitalia.it) | 02:15 | |
| - cobra (QUIT: Ping timeout: 240 seconds) (~cobra@user/Cobra) | 02:15 | |
| - kfx (QUIT: Ping timeout: 240 seconds) (~kfx@grendel.sciops.net) | 02:15 | |
| - gordon1 (QUIT: Ping timeout: 240 seconds) (~gordon1@citadel.niflheim.info) | 02:15 | |
| - sknebel (QUIT: Remote host closed the connection) (~quassel@v22016013254630973.happysrv.de) | 02:15 | |
| + cobra (~cobra@user/Cobra) | 02:16 | |
| + sknebel (~quassel@v22016013254630973.happysrv.de) | 02:17 | |
| + gordon1 (~gordon1@citadel.niflheim.info) | 02:28 | |
| - paperManu (QUIT: Ping timeout: 260 seconds) (~paperManu@146.71.9.156) | 02:30 | |
| + Svp (~svp@2002:4f07:f0bd:0:95e7:dc62:c203:a24) | 02:30 | |
| - angelwood (QUIT: Remote host closed the connection) (~angelwood@user/angelwood) | 02:41 | |
| + angelwood (~angelwood@user/angelwood) | 02:42 | |
| + bkeys (~Thunderbi@98.19.128.69) | 02:50 | |
| - mjw (QUIT: Ping timeout: 244 seconds) (~mjw@gnu.wildebeest.org) | 02:56 | |
| + paperManu (~paperManu@146.71.9.156) | 03:06 | |
| - ohnoitsnoah (QUIT: Ping timeout: 240 seconds) (~ohnoitsno@23.112.150.169) | 03:12 | |
| + ohnoitsnoah (~ohnoitsno@static-68-235-46-207.cust.tzulo.com) | 03:12 | |
| + ohnoitsnoah_ (~ohnoitsno@2600:1702:51f0:5df0::d) | 03:54 | |
| - ohnoitsnoah (QUIT: Ping timeout: 264 seconds) (~ohnoitsno@static-68-235-46-207.cust.tzulo.com) | 03:54 | |
| - ohnoitsnoah_ (QUIT: Quit: Konversation terminated!) (~ohnoitsno@2600:1702:51f0:5df0::d) | 04:21 | |
| - paperManu (QUIT: Ping timeout: 264 seconds) (~paperManu@146.71.9.156) | 04:29 | |
| - bkeys (QUIT: Read error: Connection reset by peer) (~Thunderbi@98.19.128.69) | 04:31 | |
| + bkeys1 (~Thunderbi@98.19.128.69) | 04:31 | |
| - aloo_shu (QUIT: Ping timeout: 240 seconds) (~aloo_shu@85.51.17.86) | 04:32 | |
| * bkeys1 -> bkeys | 04:34 | |
| + aloo_shu (~aloo_shu@85.51.18.67) | 04:36 | |
| + reform6118 (~ryan@user/wirelessrpm) | 04:39 | |
| - vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:50) | 05:08 | |
| - Ar|stote|is (QUIT: Quit: https://quassel-irc.org - Chat comfortably. Anywhere.) (~linx@149.210.3.83) | 05:10 | |
| + Ar|stote|is (~linx@149.210.3.83) | 05:19 | |
| - reform6118 (QUIT: Quit: Leaving) (~ryan@user/wirelessrpm) | 05:22 | |
| - lidstah (QUIT: Remote host closed the connection) (~lidstah@gateway/tor-sasl/lidstah) | 06:56 | |
| + lidstah (~lidstah@gateway/tor-sasl/lidstah) | 06:59 | |
| + pickybear (~pickybear@user/pickybear) | 07:49 | |
| pickybear | Test recap changing CFG_TUD_HID_EP_BUFSIZE (tusb_config.h) to 640: keyboard does not work, i.e. normal keypresses do not register. | 08:01 |
| - lanodan (QUIT: Ping timeout: 256 seconds) (~lanodan@2a01:e0a:d6:9930::35) | 08:27 | |
| + lanodan (~lanodan@2a01:e0a:d6:9930::35) | 08:28 | |
| - pickybear (QUIT: Quit: Client closed) (~pickybear@user/pickybear) | 08:44 | |
| + pickybear (~pickybear@user/pickybear) | 08:44 | |
| - pickybear (QUIT: Client Quit) (~pickybear@user/pickybear) | 08:45 | |
| + pickybear (~pickybear@user/pickybear) | 08:48 | |
| - amospalla (QUIT: Ping timeout: 240 seconds) (~jordi@user/amospalla) | 08:52 | |
| - pickybear (QUIT: Quit: pickybear) (~pickybear@user/pickybear) | 08:53 | |
| + pickybear (~pickybear@user/pickybear) | 08:54 | |
| + lanodan_ (~lanodan@2a01:e0a:d6:9930::35) | 08:54 | |
| - pickybear (QUIT: Client Quit) (~pickybear@user/pickybear) | 08:55 | |
| - lanodan (QUIT: Ping timeout: 255 seconds) (~lanodan@2a01:e0a:d6:9930::35) | 08:55 | |
| + voltaire28_ (~jlafon@28.162.2.93.rev.sfr.net) | 09:28 | |
| rwa_ | josch, minute: regarding the "classic reform won't boot cause nvme failed initialization" (see https://paste.sr.ht/~rwa/777f5b950927f52e690d02dfb0957f2a6a663377 for kernel log) i finally managed to make it work again | 09:31 |
| rwa_ | first attempt: boot from sd card and use "reform-migrate" to move / to nvme and /boot to emmc - script was successfull, but boot still failed with the same error | 09:32 |
| rwa_ | so i used "reform-flash-uboot emmc" to update uboot, now it is successfully booting from emmc with / on nvme again | 09:33 |
| rwa_ | difference to my borked setup is that i didn't mess with manually updating reform-tools from sid, it's a Debian Trixie image with backports running reform-tools 1.82-2 | 09:34 |
| rwa_ | u-boot version is shown as 2025-11-04 | 09:35 |
| rwa_ | iirc the borked version had timestamp of 2026-01-something | 09:35 |
| josch | rwa_: so the fix was downgrading u-boot? | 09:50 |
| rwa_ | seems so | 09:59 |
| minute | that's interesting (and concerning) | 10:15 |
| - voltaire28_ (QUIT: Ping timeout: 244 seconds) (~jlafon@28.162.2.93.rev.sfr.net) | 10:16 | |
| rwa_ | i can do whatever tests you like me to do with newer uboot versions on Debian Trixie | 10:17 |
| minute | btw there might be a new banana pi module on the horizon :D | 10:22 |
| minute | currently pondering it, but i'm already quite sure we should try it (8GB rk3588 soldered-on-som) | 10:23 |
| minute | because it could be made quite affordable... | 10:23 |
| minute | (and we need a bit more supply chain options) | 10:24 |
| + voltaire28_ (~jlafon@28.162.2.93.rev.sfr.net) | 10:26 | |
| * gianmarcogg03_ -> gianmarcogg03 | 10:28 | |
| gordon1 | what about radxa soms on rk3588 btw? is there a problem with it? | 10:32 |
| - voltaire28_ (QUIT: Remote host closed the connection) (~jlafon@28.162.2.93.rev.sfr.net) | 10:35 | |
| minute | gordon1: no, i'm also considering those. they are just significantly more expensive | 10:37 |
| minute | i need to check pin assignments of radxa cm5 and bpi cm6 and rpi cm5 to see if there are some synergies there maybe | 10:37 |
| minute | (bpi cm6 is a funky risc-v thing) | 10:37 |
| gordon1 | are they? i mean are banana pies cheaper than that? | 10:38 |
| gordon1 | i need to check that | 10:38 |
| minute | gordon1: the lm7 is cheaper yes. | 10:38 |
| minute | i can't tell the exact prices because i got personal quotes, but that's the general idea | 10:39 |
| gordon1 | lga oh god | 10:39 |
| minute | usually solder on soms are cheaper than soms with connectors too, and radxa engineering is pretty high end now from my POV | 10:39 |
| minute | gordon1: yeah. | 10:39 |
| minute | a bit tricky in terms of how to produce a board, but we have to solve that for thundercomm anyway | 10:40 |
| minute | (i think we just have to request custom parts @ jlc) | 10:40 |
| minute | and then do consignment | 10:40 |
| minute | ok one more difference, radxa cm5 is rk3588s2 | 10:42 |
| gordon1 | you right, it's like half price | 10:43 |
| minute | gordon1: the difference is not that extreme in the quotes i got, but still very significant | 10:43 |
| minute | i wonder if rk3588s2 runs cooler or if that's a myth | 10:43 |
| minute | otherwise, rk3588s2 is worse for us because of less connectivtiy (less hdmi, less pcie) | 10:44 |
| gordon1 | oh, meh it has only 2x pci | 10:45 |
| minute | yep | 10:47 |
| minute | haha we're already so spoiled by the tons of pcie on rk3588 | 10:48 |
| josch | actually what i need more of in my reform is usb not pci-e | 10:49 |
| josch | lte is connected via usb | 10:49 |
| josch | bluetooth is connected via usb | 10:49 |
| josch | i only have pci-e for wifi and ssd | 10:50 |
| josch | rwa_: minute: i have some test data regarding the nvme issue with most recent u-boot | 10:50 |
| josch | so currently i'm running debian trixie with kernel 6.17 and old reform-tools and i have: | 10:50 |
| josch | cat /proc/device-tree/chosen/u-boot,version | 10:50 |
| josch | 2024.10 MNT Reform 2 with RCORE-DSI RK3588 Module 2026-01-11-g424c714eb247-dirty% | 10:50 |
| josch | so version 2026-01-11 *can* boot a system on nvme | 10:51 |
| josch | and i tested that before filing my MRs and my serial logs are in the MNT gitlab | 10:51 |
| josch | i was just making sure i didn't hallucinate | 10:51 |
| josch | *but* | 10:51 |
| minute | josch: right, we do have tons of usb too, just not going to enough places. so that calls for mb 4.0 | 10:51 |
| josch | yay \o/ | 10:52 |
| josch | or reform-next mobo in classic :) | 10:52 |
| minute | josch: yeah, or some mutation of that, exactly | 10:52 |
| josch | but then i flashed u-boot 2026-01-11 to my emmc -- before i only tested with u-boot on sd-card because then i cannot mess up and i didn't expect any difference in behaviour depending where rk3588 reads u-boot from | 10:52 |
| josch | and then i experienced the exact same problem rwa_ reported: my initramfs was unable to find my nvme: /dev/nvme0n1 was completely missing | 10:53 |
| josch | so i zeroed out my emmc and made it use u-boot on sd-card again: then it worked | 10:53 |
| josch | so i thought maybe it *does* make a difference whether u-boot is read from emmc or sd card? | 10:54 |
| josch | so i rebooted again with u-boot on sd-card and *then* i got the same problem | 10:54 |
| josch | so there is some randomness, some flakiness | 10:54 |
| josch | it works sometimes (like it does right now, i'm typing this from a machine booted with 2026-01-11) but not always | 10:55 |
| josch | minute: which means we should probably delete the latest artifacts so that nobody else flashes it | 10:55 |
| josch | and then get some time to investigate | 10:55 |
| gordon1 | >so that calls for mb 4.0 | 10:57 |
| gordon1 | extra usb jst header pretty please | 10:57 |
| minute | gordon1: just a third one you mean? | 10:58 |
| + AnimaInvicta (~AnimaInvi@88-120-179-216.subs.proxad.net) | 10:58 | |
| gordon1 | well, either that or kbd firmware to get trackball running? | 10:58 |
| gordon1 | but i mean if you'll have spare lines from bigger hub, just drop it to the header i guess | 10:59 |
| josch | gordon1: there is also the extra usb that you get from the latest rcore version together with pci-e | 10:59 |
| josch | i was messing around in kicad to shrink the wifi/bt board for the pocket | 11:00 |
| josch | (after my last miserably failed attempt to fit that board into the classic...) | 11:00 |
| gordon1 | actually i need to see what needs to be done to get double sensor trackball going | 11:01 |
| josch | minute: or instead of pulling the gitlab artifacts i can also upload a reform-tools version which reverts u-boot to the last-known version, what do you prefer? | 11:03 |
| josch | *last-known working version | 11:03 |
| rwa_ | josch: wow, this sounds weird | 11:10 |
| minute | josch: last known good sounds good | 11:11 |
| rwa_ | but iirc i had the same thing that it worked once, but mostly it didn't work via emmc, but sd card was always fine | 11:11 |
| minute | josch: i added you as maintainer of reform-rk3588-uboot, is that enough so you can delete the artifacts? | 11:13 |
| josch | minute: yes, done. Sorry for the mess. | 11:14 |
| minute | josch: no problem! mess happens | 11:16 |
| minute | there are many moving parts :D | 11:16 |
| josch | but i'm happy that i posted my u-boot uart output in the gitlab issues -- at least that way i know i didn't hallucinate that it didn't work when i tested the artifacts | 11:17 |
| josch | i just should've tested more than once or twice... | 11:18 |
| josch | now it will be interesting to see what the problem was... | 11:18 |
| minute | pasting logs on the internet is always good | 11:23 |
| rwa_ | josch: not sure if we had this topic before - is there a reason nfs-common is not included in the Debian Trixie images? | 11:23 |
| minute | btw i'm open for ideas on how to most cleanly set the default governors to powersave / limit the gpu frequency _at boot_ on rk3588 pocket | 11:32 |
| minute | i'm using some 5000mAh amazon batteries that seem to have weakened quicker than the default eremits... and if i'm on like 50% bat i can't boot without power adapter | 11:32 |
| minute | (before reform-power-daemon has started) | 11:33 |
| josch | rwa_: the usual answer is "everything we include increases the image size" | 11:34 |
| rwa_ | ok, thats a reasonable point | 11:38 |
| rwa_ | was just irritated that mounting nfs volumes didn't work ootb | 11:39 |
| minute | rwa_: we like to add stuff like that when someone reports an irritation like that | 11:41 |
| josch | minute: i gathered some data so that you can better decide. Installing nfs-common would increase the uncompressed image size by 2.4 MiB and add four extra packages overall. Now it's your call. :) | 12:02 |
| minute | josch: awesome, thanks! i say go for it | 12:06 |
| - lidstah (QUIT: Remote host closed the connection) (~lidstah@gateway/tor-sasl/lidstah) | 12:09 | |
| + lidstah (~lidstah@gateway/tor-sasl/lidstah) | 12:10 | |
| josch | thank you, i added that to the MR which reverts rk3588 u-boot to the last known working version: https://source.mnt.re/reform/reform-debian-packages/-/merge_requests/160 | 12:12 |
| josch | rwa_: sorry for the mess and thanks a lot for reporting it! | 12:12 |
| + rick_ (~m-x42ph2@m.n0id.space) | 12:21 | |
| minute | ch: what is the cut-off (low end) voltage you configured the max chip to in the experimental fw? my pocket turned off at 3.5V under load, battery gauge showed 32% or so | 12:22 |
| minute | ch: but it also turned on again after a few seconds (and then couldn't boot anymore), like as if there's not much hysteresis | 12:22 |
| minute | ch: but could also be unrelated, so wanted to confirm what's expected? | 12:22 |
| ch | minute: i think 3.25 or 3.22 | 12:27 |
| ch | minute: the lipo spec says 3.20 i think | 12:27 |
| ch | minute: but can certainly change it. for me it ran quite long below 3.5 | 12:28 |
| ch | minute: but you have the other lipos? | 12:28 |
| minute | ch: ok, then it's probably unrelated | 12:35 |
| minute | ch: i will switch back to the default 4000mAh when i'm back in berlin | 12:35 |
| minute | ch: thanks! | 12:35 |
| rwa_ | josch: not for this, glad a can help at least a little bit | 12:36 |
| * lanodan_ -> lanodan | 13:25 | |
| + pickybear (~pickybear@user/pickybear) | 13:28 | |
| - pickybear (QUIT: Ping timeout: 272 seconds) (~pickybear@user/pickybear) | 13:36 | |
| - sterni (QUIT: ) (~quassel@user/sterni) | 13:47 | |
| + sterni (~quassel@user/sterni) | 13:47 | |
| + mjw (~mjw@gnu.wildebeest.org) | 14:18 | |
| - pomel0 (QUIT: Ping timeout: 256 seconds) (~pomel0@user/pomel0) | 14:20 | |
| + pomel0 (~pomel0@user/pomel0) | 14:21 | |
| + paperManu (~paperManu@146.71.9.156) | 14:33 | |
| - pomel0 (QUIT: Read error: Connection reset by peer) (~pomel0@user/pomel0) | 14:33 | |
| + pomel0 (~pomel0@user/pomel0) | 14:38 | |
| - pomel0 (QUIT: Ping timeout: 240 seconds) (~pomel0@user/pomel0) | 14:45 | |
| - bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@98.19.128.69) | 15:02 | |
| + bkeys (~Thunderbi@98.19.128.69) | 15:02 | |
| + spew (~spew@user/spew) | 15:37 | |
| + pickybear (~pickybear@user/pickybear) | 15:40 | |
| - pickybear (QUIT: Client Quit) (~pickybear@user/pickybear) | 15:42 | |
| - spew (QUIT: Quit: brb) (~spew@user/spew) | 15:44 | |
| + spew (~spew@user/spew) | 15:44 | |
| + siviq (~siviq@user/siviq) | 16:02 | |
| - siviq (QUIT: Quit: Client closed) (~siviq@user/siviq) | 16:42 | |
| jfred | I just booted up my Pocket on low battery because I didn't have an outlet available, the gauge showed 20%, but under load one of the cells was at 3V. I remember there was some discussion previously about cutting power off in firmware at a slightly higher voltage, is that the experimental fw that was mentioned earlier? | 16:47 |
| jfred | I powered it off after that because I didn't want to damage that cell ^^; | 16:47 |
| - gianmarcogg03 (QUIT: Quit: https://quassel-irc.org - La chat diventa comoda. Ovunque.) (~quassel@user/gianmarcogg03) | 16:51 | |
| josch | :( i soldered the new diode but as soon as i connect the batteries it blows again... what could i test? | 16:55 |
| + siviq (~siviq@user/siviq) | 17:19 | |
| + gianmarcogg03 (~quassel@user/gianmarcogg03) | 17:30 | |
| - bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@98.19.128.69) | 17:35 | |
| minute | josch: that will be hard to debug | 17:37 |
| minute | josch: first, check all power rails for shorts | 17:38 |
| minute | josch: or actually first check all the fuses | 17:38 |
| minute | josch: maybe a fuse is blown | 17:38 |
| minute | jfred: yes, ch has a branch | 17:39 |
| gordon1 | why does it blow the diode if the fuse is blown? | 17:44 |
| gordon1 | josch: which mb revision is that? 3.0? | 17:47 |
| minute | gordon1: just an idea, maybe power wants to route around the fuse through that way | 17:51 |
| gordon1 | hmm | 17:51 |
| minute | but i don't have the schematics in front of me | 17:51 |
| minute | i repaired many very broken reform mbs though... so i have some instincts | 17:51 |
| minute | anyway, first thing is always to check fuses, and check all power rails | 17:52 |
| gordon1 | that's a good starting point, yes | 17:53 |
| gordon1 | i just vaguely remember josch mentioning that board works without batteries or something | 17:53 |
| pinsl | I designed a custom split keyboard, while waiting for the reform next to arrive: https://codeberg.org/funnes/reform-split-keyboard#readme | 17:58 |
| jfred | pinsl: ooh very cool! :D | 18:00 |
| minute | gordon1: aha, right. i suspect mostly the fuse that goes to the batteries | 18:02 |
| + bkeys (~Thunderbi@98.19.128.69) | 18:08 | |
| + pickybear (~pickybear@user/pickybear) | 18:19 | |
| - pickybear (QUIT: Quit: Client closed) (~pickybear@user/pickybear) | 18:31 | |
| + mark_ (~mjw@gnu.wildebeest.org) | 18:39 | |
| - schalken (QUIT: Read error: Connection reset by peer) (~schalken@117-118-178-69.gci.net) | 18:39 | |
| - mjw (QUIT: Ping timeout: 240 seconds) (~mjw@gnu.wildebeest.org) | 18:41 | |
| + liberto (~liberto@user/libercv) | 18:43 | |
| + schalken (~schalken@117-118-178-69.gci.net) | 18:45 | |
| + pickybear (~pickybear@user/pickybear) | 18:47 | |
| josch | fuse f2 seems to be fine | 18:47 |
| josch | yes board works fine -- even off the batteries. But it cannot charge them. | 18:48 |
| + S0rin (~S0rin@user/s0rin) | 18:48 | |
| + amospalla (~jordi@user/amospalla) | 18:55 | |
| josch | meh... i think a311d has the same problem with nvme and new u-boot | 19:02 |
| josch | i downgraded u-boot on a311d and now nvme comes up reliably again | 19:03 |
| josch | anyways, i'm back on a311d now | 19:03 |
| josch | the speed of rk3588 was really awesome but i need the connectivity options i have on a311d more than i need the speed of rk3588 | 19:05 |
| + paperManu_ (~paperManu@142.169.16.197) | 19:06 | |
| - bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@98.19.128.69) | 19:10 | |
| + bkeys (~Thunderbi@98.19.128.69) | 19:11 | |
| - bkeys (QUIT: Read error: Connection reset by peer) (~Thunderbi@98.19.128.69) | 19:12 | |
| + bkeys1 (~Thunderbi@98.19.128.69) | 19:12 | |
| - bkeys1 (QUIT: Client Quit) (~Thunderbi@98.19.128.69) | 19:12 | |
| + bkeys (~Thunderbi@98.19.128.69) | 19:13 | |
| - pickybear (QUIT: Ping timeout: 264 seconds) (~pickybear@user/pickybear) | 19:13 | |
| - bkeys (QUIT: Client Quit) (~Thunderbi@98.19.128.69) | 19:16 | |
| + bkeys (~Thunderbi@98.19.128.69) | 19:16 | |
| + pomel0 (~pomel0@user/pomel0) | 19:21 | |
| - paperManu_ (QUIT: Read error: Connection reset by peer) (~paperManu@142.169.16.197) | 19:24 | |
| + pickybear (~pickybear@user/pickybear) | 19:30 | |
| - pickybear (QUIT: Client Quit) (~pickybear@user/pickybear) | 19:30 | |
| - pomel0 (QUIT: Ping timeout: 246 seconds) (~pomel0@user/pomel0) | 19:34 | |
| + pomel0 (~pomel0@user/pomel0) | 19:34 | |
| - bkeys (QUIT: Remote host closed the connection) (~Thunderbi@98.19.128.69) | 19:37 | |
| gordon1 | josch: maybe ltc6803 is bust? can you check if cathode of the diode (the pad where diode line is facing) has short to ground with a multimeter? | 19:49 |
| gordon1 | can you also check if individual cell voltages seem healthy? | 19:56 |
| - Ar|stote|is (QUIT: Quit: https://quassel-irc.org - Chat comfortably. Anywhere.) (~linx@149.210.3.83) | 20:10 | |
| + Ar|stote|is (~linx@149.210.3.83) | 20:12 | |
| rwa_ | @gordon1: josch and myself noticed that screen backlight flickering is back on low brightness settings - you were involved in the investigation a few months ago. Do you experience the same? | 20:28 |
| gordon1 | rwa_: i think minute increased the frequency a bit to combat interferece with usb audio, and i, since the only usb audio i use has no issues with this interference, opted out of keeping the old value as extra patch i add to my own kernel | 20:30 |
| - spew (QUIT: Ping timeout: 240 seconds) (~spew@user/spew) | 20:33 | |
| + spew (~spew@user/spew) | 20:35 | |
| + johl (~johl@wikidata/Jens-Ohlig) | 20:40 | |
| - johl (QUIT: Client Quit) (~johl@wikidata/Jens-Ohlig) | 20:43 | |
| minute | gordon1: rwa_: yeah that is a bit unfortunate. next rev should get filtering on the 5v rail for these freqs | 20:43 |
| + johl (~johl@wikidata/Jens-Ohlig) | 20:44 | |
| gordon1 | can you just hack in a cap somewhere for that btw? | 20:44 |
| gordon1 | probably worth adding a note or something | 20:45 |
| - liberto (QUIT: Quit: WeeChat 4.8.1) (~liberto@user/libercv) | 20:51 | |
| - spew (QUIT: Quit: nyaa~) (~spew@user/spew) | 21:49 | |
| rwa_ | minute: is there a way to set the pwm to a supported value for peeps who don't use usb audio? | 22:11 |
| gordon1 | rwa_: you can modify .dts/.dtb file, you need kernel sources but you don't have to compile your own kernel | 22:13 |
| gordon1 | i mean there's probably a way to do that with overlay, but i'm not 100% sure how overlays are handled by default u-boot config | 22:16 |
| gordon1 | maybe josch knows the answer | 22:16 |
| gordon1 | ACTION struggling to find a script that build reform kernel in source.mnt.re | 22:19 |
| gordon1 | rwa_: sorry, i can guide you how to do that manually, but there has to be a better way with a script | 22:20 |
| gordon1 | ah, found it | 22:22 |
| gordon1 | rwa_: try this script, it should check out kernel git for your particular kernel version https://source.mnt.re/reform/reform-kernel-dev/-/blob/main/prepare.sh?ref_type=heads | 22:22 |
| rwa_ | gordon1: would something like this work | 22:24 |
| rwa_ | https://www.jeffgeerling.com/blog/2023/how-customize-dtb-device-tree-binary-on-raspberry-pi/ | 22:24 |
| gordon1 | if it succeeds, you can apply following patch https://bpa.st/GOEUM/raw and do make dtbs - this will produce new rk3588-mnt-reform2-dsi.dtb that you can place in /boot | 22:24 |
| gordon1 | i mean it might | 22:24 |
| gordon1 | but i won't guarantee the result | 22:24 |
| gordon1 | actually let me try to decompile dtb like that | 22:24 |
| gordon1 | stand by | 22:25 |
| gordon1 | rwa_: .dts produced is ugly a/f but it compiled back to byte-to-byte the same .dtb so i guess it must be safe to try | 22:28 |
| gordon1 | rwa_: that is my diff https://bpa.st/EJSRO/raw | 22:29 |
| gordon1 | i mean that is your patch, opposite to my diff | 22:30 |
| gordon1 | you have to replace 0xc350 with 0x4c4b40 | 22:30 |
| rwa_ | yep | 22:30 |
| gordon1 | rwa_: you can try to verify newly made dtb with vbindiff tool, it should show the difference only in 3 bytes | 22:32 |
| gordon1 | https://tinystash.undef.im/il/5KAfciMA1XV7FSEXY3E1Vfuu9bQKJ1HQS1sHZbSfrKYaz2rn2caqhUfAFkDouJZmZKy6g4VhQecoj6npJ1butjEy.png | 22:33 |
| + ohnoitsnoah_ (~ohnoitsno@23.112.150.169) | 22:35 | |
| rwa_ | thanks, will give it a try | 22:35 |
| + voltaire28_ (~jlafon@28.162.2.93.rev.sfr.net) | 22:43 | |
| rwa_ | currently reading a bit about dtos, might be worth a try as well | 22:50 |
| gordon1 | rwa_: that is a _proper_ way, however u-boot needs to load those | 22:54 |
| gordon1 | if it can be specified through extlinux.conf then i would strongly recommend using dtos | 22:54 |
| - pomel0 (QUIT: Ping timeout: 240 seconds) (~pomel0@user/pomel0) | 22:55 | |
| + pomel0 (~pomel0@user/pomel0) | 22:56 | |
| + pickybear (~pickybear@user/pickybear) | 23:04 | |
| - pickybear (QUIT: Client Quit) (~pickybear@user/pickybear) | 23:05 | |
| - siviq (QUIT: Quit: Client closed) (~siviq@user/siviq) | 23:38 | |
| - pomel0 (QUIT: Ping timeout: 240 seconds) (~pomel0@user/pomel0) | 23:42 | |
| + pomel0 (~pomel0@user/pomel0) | 23:43 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
