03:05 <tortoise> thanks for pointing that linux-firmware bbapend, so I tried just updating the SRCREV commit hash to a more recent libreelec firmware commit hash. That partially works, because the symlinks for radxa-zero are present in newer commits. So firmware loads. But it complains about not finding the expected hcd file brcm/BCM4345C5.hcd , which is present in libreelec repo but doesn't seem to copy over to /lib/firmware in recipe

03:05 <tortoise> also the clm_blob file is missing in libreelec

03:06 <tortoise> the .bin and .txt of the firmware are identical to armbian's verison, but that missing hcd file apparently is different between the libreeelc and armbian , which seems weird

03:11 <tortoise> didn't know this before, but apparently hcd file are some kind of patch from what I gather after googling and finding this other broadcom thingy

03:11 <tortoise> https://github.com/fredldotme/brcm-patchram-plus/blob/master/brcm_patchram_plus.c#L23

03:11 <tortoise> ** Description: This program downloads a patchram files in the HCD format

03:50 <steev> which radxa zero do you have? what does the wifi chip have on it

03:52 <steev> fwiw, this is the firmware i use across them - https://gitlab.com/kalilinux/build-scripts/kali-arm/-/tree/master/bsp/firmware/radxa-zero - i don't have the newer one with the 43455 chip, but someone who does tested that firmware as well, and it worked

03:53 <tortoise> the chip is the AP6256 which I think is the case for almost all the radxa-zero except the ones with 1GB or less of RAM

03:53 <steev> wifi and bluetooth both work

03:53 <tortoise> I'm able to get the wifi kind of working, wpa_supplicant associated with the AP, and I was able to lease an ip address and then it kind of acts strange

03:53 <tortoise> https://wiki.radxa.com/Zero/hardware/models

03:53 <steev> the newer models have a different wifi chip - AW-CM256SM - that's a 43455 not a 43456

03:54 <steev> https://forum.radxa.com/t/manjaro-wifi-connection-with-the-new-chipset-aw-cm256sm/12207/2

03:56 <tortoise> hrmm, I'm not sure, I bought these a while back. and the chip has AP6256 printed on it. But I'll look into this too thanks

03:56 <steev> if you grab my firmware directory, you should be good for both

03:56 <steev> i didn't include the clm_blob though, but the forum link has it

03:59 <tortoise> Interesting. Any idea why the firmware binaries differ?

03:59 <tortoise> Binary files brcmfmac43456-sdio.bin and ../../../../firmware/brcm/brcmfmac43456-sdio.bin differ

03:59 <tortoise> diff brcmfmac43456-sdio.bin ../../../../firmware/brcm/brcmfmac43456-sdio.bin

03:59 <steev> mine came from the radxa debian image

03:59 <steev> Firmware: BCM4345/9 wl0: Feb 11 2020 11:54:51 version 7.45.96.61 (be7af2d@shgit) (r745790) FWID 01-a41d86bd es7.c5.n4.a3

04:00 <tortoise> 👍

04:01 <steev> at least, i think it did

04:01 <steev> i really hate the broadcom firmware and trying to keep track of it all

04:21 <steev> unfortunately, my initial commit says the firmware came from their factory images, but the updated firmware doesn't say where i grabbed from :(

04:23 <tortoise> I ran strings on the .bin

04:23 <tortoise> CRC: 279731d4 Date: Tue 2020-02-11 11:55:51 CST Ucode Ver: 1043.20721 FWID: 01-a41d86bd

04:24 <tortoise> which is newer than the version in armbian / libreelec

04:24 <tortoise> Version: 7.45.96.0 CRC: 1d0e37c2 Date: Fri 2017-06-16 12:39:23 CST Ucode Ver: 1043.2070 FWID: 01-1813af8

04:25 <tortoise> will try your firmware tomorrow, thanks again!

04:46 <steev> 👍