c0rnelius | fe.dai-link-2 is failing, so I'll change it to 2 and see if that defaults it. | 00:05 |
---|---|---|
steev | when you say that, you mean the u200? | 00:08 |
c0rnelius | fe.dai-link-2: ASoC: no backend DAIs enabled for fe.dai-link-2 | 00:17 |
c0rnelius | So I'm gonna open more channels in the frddr file | 00:18 |
c0rnelius | long shot at best... but its the only thing sticking out at me after seeing aplay -l | 00:19 |
c0rnelius | One other option would be "the dts is wrong" and that sound bit needs to be changed. | 00:21 |
c0rnelius | I wish they would just mail this board already :\ Stupid Ameridroid | 00:25 |
c0rnelius | steev: Not a betting man, but see if these have any audio change. | 01:09 |
c0rnelius | https://github.com/pyavitz/binary/releases/download/060420/meson64-linux-headers_5.15.6-3_arm64.deb | 01:09 |
c0rnelius | https://github.com/pyavitz/binary/releases/download/060420/meson64-linux-image_5.15.6-3_arm64.deb | 01:09 |
steev | c0rnelius: ah, i bought mine off allchina or something... i bought it saturday, it shipped monday and was at my door on fridday | 01:21 |
steev | https://shop.allnetchina.cn/ | 01:21 |
steev | will test those kernels shortly | 01:21 |
steev | sadly, no change | 01:26 |
c0rnelius | steev: thanx for testing | 02:01 |
steev | no problem, helping you helps me :) | 02:13 |
c0rnelius | Well when I ordered the C4 it was kinda the same and I had it fast, but that was from Hardkernel directly which is in SK. I don't really trust those China sellers all that much. | 02:39 |
c0rnelius | Same deal with aliexpress, I don't order from them. | 02:40 |
c0rnelius | Doesn't seem like Radxa has its own in house retail store. Its just sellers and whatnot. | 02:42 |
c0rnelius | steev: saw this in ur dmesg -> [ 1.794680] soc soc0: Amlogic Meson G12A (Unknown) Revision 28:b (30:2) Detected | 18:10 |
c0rnelius | thinking this will fix it -> https://github.com/pyavitz/debian-image-builder/blob/feature/patches/amlogic/radxazero/006-zadxa-zero-meson-gx-socinfo.patch | 18:11 |
steev | seems like it would yeah | 18:12 |
steev | the same occurs on the vendor 5.10 btw | 18:12 |
c0rnelius | maybe thats related to the audio problem? | 18:12 |
c0rnelius | board doesn't know what it is? | 18:12 |
c0rnelius | I did some reading on lore this morning and yeah, what they did was just edit the u200 dts and essentially made some edits and renamed it to the radxa dts | 18:13 |
c0rnelius | I thought the two looked to close for it not to be the case. | 18:14 |
steev | hm, that could be possible | 19:01 |
steev | c0rnelius: although, the audio driver should be using the compatible | 20:01 |
c0rnelius | Yeah. Something isn't right in the dts I suspect. | 20:02 |
c0rnelius | for example | 20:03 |
c0rnelius | **** List of PLAYBACK Hardware Devices **** | 20:03 |
c0rnelius | card 0: G12BODROIDN2 [G12B-ODROID-N2], device 0: fe.dai-link-0 (*) [] | 20:03 |
c0rnelius | Subdevices: 1/1 | 20:03 |
c0rnelius | Subdevice #0: subdevice #0 | 20:03 |
c0rnelius | card 0: G12BODROIDN2 [G12B-ODROID-N2], device 1: fe.dai-link-1 (*) [] | 20:03 |
c0rnelius | Subdevices: 1/1 | 20:03 |
c0rnelius | Subdevice #0: subdevice #0 | 20:03 |
c0rnelius | card 0: G12BODROIDN2 [G12B-ODROID-N2], device 2: fe.dai-link-2 (*) [] | 20:03 |
c0rnelius | Subdevices: 1/1 | 20:03 |
c0rnelius | Subdevice #0: subdevice #0 | 20:03 |
c0rnelius | Each one is linked to its own device. I can't remember exactly but I thought before the aplay on zero was all same card and same device. | 20:04 |
c0rnelius | anyway... removing the fe.dai-link-0 like they did isn't gonna change anything I believe. | 20:05 |
steev | no, i don't think it will | 20:08 |
steev | c0rnelius: where is that soc info patch originally from? | 20:50 |
c0rnelius | Found it this morning on lore.kernel.org | 20:52 |
steev | https://github.com/chewitt/LibreELEC.tv/blob/amlogic/packages/audio/alsa-utils/scripts/soundconfig | 20:55 |
steev | gonna try that | 20:55 |
c0rnelius | https://lore.kernel.org/linux-amlogic/06547fcd-2881-3601-d02f-77df56b9a258@baylibre.com/ | 21:02 |
c0rnelius | You can just edit the config and leave only the amlogic bits. | 21:02 |
c0rnelius | That is kind of all encompassing | 21:03 |
c0rnelius | here it is that first link was for uboot -> https://lore.kernel.org/linux-amlogic/20210820012718.10761-1-christianshewitt@gmail.com/ | 21:05 |
c0rnelius | this one is for the actual kernel builds | 21:05 |
c0rnelius | I had already the one for u-boot included, but was missing that one as I didn't know it was needed. | 21:07 |
c0rnelius | The SoC is like 4 years old or something? who would have thought. | 21:07 |
steev | i mean, cortex-a53 is from 2012, but you'd never guess that since people still coming out with processors using cortex-a53 | 21:29 |
steev | c0rnelius: so, on the 5.10 radxa zero, which has the dai 0 removed | 21:33 |
steev | i don't get the dai spam, and i get | 21:33 |
steev | [ 7.362328] axg-sound-card sound: ASoC: no source widget found for FRDDR_A OUT 1 | 21:33 |
steev | [ 7.364678] axg-sound-card sound: ASoC: Failed to add route FRDDR_A OUT 1 -> direct -> TDMOUT_B IN 0 | 21:33 |
steev | so maybe that patch *is* needed? | 21:33 |
steev | fwiw, here's the 5.10 dmesg output https://paste.debian.net/1221736/ | 21:35 |
c0rnelius | steev: If you actually read the DTS its referencing the above -> | 23:19 |
c0rnelius | https://github.com/pyavitz/debian-image-builder/blob/feature/patches/amlogic/radxazero/001-arm64-dts-amlogic-add-support-for-radxa-zero.patch#L173 | 23:19 |
c0rnelius | Them simply removing that line also doesn't change the following -> https://github.com/pyavitz/debian-image-builder/blob/feature/patches/amlogic/radxazero/001-arm64-dts-amlogic-add-support-for-radxa-zero.patch#L310 | 23:20 |
steev | ohh, hm | 23:20 |
steev | i should definitely read through their setup more | 23:20 |
c0rnelius | So its like they just removed one line hoping it would stop an error thats referenced three times in dts file. | 23:21 |
c0rnelius | Pretty sure yesterday you said that last kernel or one of them actually gave us an error on fe.dai-link-2: ASoC: no backend DAIs enabled for fe.dai-link-2 | 23:22 |
steev | yeh, that happens on the 5.15.6 | 23:23 |
c0rnelius | Which would be different then the one you said happens on theirs if I'm accurate. | 23:23 |
steev | also correct | 23:23 |
c0rnelius | Thats why I think there is something a miss in the dts. | 23:24 |
c0rnelius | Unfortunately w/o the hardware on me I can't play with it yet. | 23:24 |
steev | yeah, i'll have some free time soon, however.... i don't think i'll have enough room to actually play with it | 23:25 |
c0rnelius | If you did you, I would just start with a kernel, pull all the audio out of the dts, build and boot and then slowly start adding it back. Wouldn't even need to build the whole thing after boot, just the dts file. Keep replacing the dtb. | 23:29 |
c0rnelius | If its not resolved by the time I get my hands on it, I'll mess with it. I can't stand errors :) | 23:30 |
c0rnelius | I don't think tons of people even have the board which is probs part of the problem. | 23:31 |
steev | even when lots of people do have the board.... very few are willing to put in the work | 23:39 |
c0rnelius | true | 23:39 |
c0rnelius | I haven't looked at the data sheet but my guess is. Currently one is referencing an analog audio port (which it doesn't have), one the hdmi and the other GPIO. | 23:41 |
c0rnelius | Of course thats just a guess | 23:41 |
steev | quite possibly - based on your guess, i removed the frddr_a... and now it spams about no backend dais for -1 and -2 | 23:58 |
steev | i'll look closer later | 23:58 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!