Björn Quentin

`MCPWM` support added in #255

> The simplest solutions I would come up with is to call self.flush() before doing anything in [Instance::write_bytes](https://github.com/esp-rs/esp-hal/blob/b8b041b51a0d9ad31c45a508337a4bd1f4158d48/esp-hal/src/spi/master.rs#L2619), but I don't know the inner workings of the SPI driver well...

I guess the problem is that we compile `wpa_supplicant` with TinyCrypt - not mbed-tls: https://github.com/esp-rs/esp-wireless-drivers-3rdparty/blob/master/patch/esp32c3/sdkconfig.defaults The reason for that is mbed-tls is quite huge and pulls in a lot of...

TinyCrypt only includes support for a subset of crypto that is needed for more advanced things: https://github.com/intel/tinycrypt/tree/master/lib/source Probably the missing parts are no-ops then Ah seems it's actually not TinyCrypt...

But the thing about the crypto is just guessing currently - probably we need to test that with an ESP-IDF example and see what config options are really needed

I'd say option one (rustup as a requirement) is a very good option. As you said most users will already have it. Also, I'd say that if someone is willing...

Maybe we could even avoid those functions and use something like https://crates.io/crates/tock-registers But I think such code originates from 1:1 translation of esp-idf code - but once such code is...

> Maybe I'm missing something, but why are we going to all this effort instead of just fixing the PACs? That would certainly be the best solution for registers! I...

Ah we actually configure something for S3, didn't knew that 🤔 Oh - that was me

I _think_ for most targets the second stage bootloader will enable it and use whatever was configured when building it For S3 (something I re-learned) we always configure it in...