You can commit to your local branches -- which is why I suggested that you make your own and commit to it. That way you can later "rebase" your changes on top of an updated master from OpenWrt. You can even have a "last good" and "devel" branch, or the like.
all commands up to "git commit" completed successfully. Unfortunately I can't get back to the console output as I startet the build in the same shell.
If it works, I might just redo all steps to get the output you need.
EDIT:
First build finished with errors, will try again with "-j1" as advised by make. "git commit" didn't work because I forgot to provide an identity.
EDIT2:
Building with "-j1" is awefully slow and still hasn't finished. Realized the build process involves a lot of downloading and the machine was offline during the first run.
Thanks to everyone. Expect some (days?) delay on further updates.
It helps if you let it fail a couple of times at high build concurrency, before falling back to -j1 V=s. Each time the 'unrelated' parts of the build will get a bit further, leaving less for the final -serialized- single-threaded build to do.
As it turned out, everything was ok-ish before and building only failed because there was no internet connection after like 10 minutes into the build. So the second try produced the initramfs and squashfs just fine.
I finally found some time to boot the initramfs and it turns out somewhat different than yesterday's openwrt-ath79-generic-devolo_dvl1750e-initramfs-kernel.bin.
ifconfig has only br-lan and eth0 (same MAC) plus "lo" of course, but no eth1 like with the devolo image.
On the upside, USB seems to be working:
[ 1.628138] usbcore: registered new interface driver usbfs
[ 1.633829] usbcore: registered new interface driver hub
[ 1.639294] usbcore: registered new device driver usb
[ 1.649862] ehci_hcd: USB 2.0 'Enhanced' Host Controller (EHCI) Driver
[ 1.657942] ehci-platform: EHCI generic platform driver
[ 1.663510] ehci-platform 1b000000.usb: EHCI Host Controller
[ 1.669304] ehci-platform 1b000000.usb: new USB bus registered, assigned bus number 1
[ 1.677352] ehci-platform 1b000000.usb: irq 14, io mem 0x1b000000
[ 1.712819] ehci-platform 1b000000.usb: USB 2.0 started, EHCI 1.00
[ 1.719850] hub 1-0:1.0: USB hub found
[ 1.724041] hub 1-0:1.0: 1 port detected
connecting some USB drive will generate a detection message on the console but i couldn't find it in /dev and the commands I know (lsblk, etc.) don't work in OpenWrt.
Thank you all for the tips about the build, I'm sure it will be faster next time, returning to -j12.
Now that I finally got around to looking at the logs and configs again, from my understanding the wifi drivers/firmware should be the same as "openwrt-ath79-generic-devolo_dvl1750e-initramfs-kernel.bin" as the dvl1750e config was the base for this. Yet they fail to load in my build.
Looking at "make menuconfig" again, I can't really tell what else to chose in the firmware menu, other than the already selected "ath10k-firmware-qca988x-ct".
The "target/linux/ath79/dts/qca9558_sophos_ap55.dts" file doesn't seem to contain anything related either. There's just the hardware buttons and speaker that will eventually be removed.
Sorry, jeff, I think you lost me there. Okay, "set -vx" I understand but the other line wouldn't output anything useful, right?
I still executed both commands after booting with the expected result of "nothing useful" inside the debug.log. How I would "set -vx" earlier, during boot maybe, is beyond me.
Anyway: On the last boot, I set debug level to 4 at
Press the [1], [2], [3] or [4] key and hit [enter] to select the debug level
Yet the output didn't include anything new on the network part.
Like I said earlier, I'm no developer... so without some more details, I'm afraid I'm not going to be able to get this done.
Sorry for being such a pain here. Seeing so many similar devices supported already and the AP55 running OpenWRT ("from Sophos") already, I mistakenly thought this couldn't be too much of a challenge, even with very limited knowledge.
The other line is "printf debugging" -- print out what you need to "see" the script being executed and what it is being called with. I don't think it is the network part, but that the hotplug script isn't doing its job, at least from what I've seen here. If you're not seeing anything from those, then the next step would be figuring out why not.
I did some more reading on the hotplug/procd part and from what I understand I should see something "hotplug" after enabling debug via [4] (see above) as procd clearly runs afterwards. Can't identify anything useful here, though:
There's not much info for a "openwrt hotplug" search, the only remotely interesting thing I found is:
I solved my problem. After inspecting carefully the boot sequence , i created my script in /etc/init.d and made a symlink in /etc/rc.d after that my script ran at boot.