Phil Elwell

Yes, that's all fine. For context, is this a configuration that has worked for you before?

Note that the explicit core_freq and core_freq_min are redundant - the firmware should spot when UART1 is in use and lock the core clock.

It's also failing here - leave it with me.

So far I can only make it work by actually _enabling_ the flow control pins on the mini-UART - something I failed to at the birth of Pi 3. Try...

Yes - to avoid anything in miniuart-bt overwriting something it has also changed (although I didn't look to see if that was a possibility).

Not really - having the modem signals on hidden GPIOs makes it hard to debug.

In what way is it not working for you? What are you attempting to do with Bluetooth, and with what kind of bandwidth? When using the mini-UART there is no...

When you say "latest Bluez release", am I right in thinking this isn't an RPiOS image? (Not that that's bad or unsupported, just trying to understanding what I'm dealing with).

Is that the mainline/upstream kernel or ours?

I think @l1k has previously done a lot of work on SPI optimisation - I haven't - so may be able to offer some suggestions if he has the time...