Split out PHY init into esp-phy crate.

[Frostie314159]

Thank you for your contribution!

We appreciate the time and effort you've put into this pull request.
To help us review it efficiently, please ensure you've gone through the following checklist:

Submission Checklist 📝

• I have updated existing examples or added new ones (if applicable).
• I have used cargo xtask fmt-packages command to ensure that all changed code is formatted correctly.
• My changes were added to the CHANGELOG.md in the proper section.
• I have added necessary changes to user code to the Migration Guide.
• My changes are in accordance to the esp-rs developer guidelines

Extra:

• I have read the CONTRIBUTING.md guide and followed its instructions.

Pull Request Details 📖

Description

This is a follow-up to #3687, that extracts the PHY init into a separate esp-phy crate, to make it generically usable for both esp-radio and esp-wifi-hal. The PHY init is now tied to the peripheral singleton, just like the PHY clock initialization. LMK if this architecture looks good to you.

I think, that it would be good to add this PR to #3814.

Testing

Describe how you tested your changes.

closes #2469

[bjoernQ]

I think this is going into a good direction.

As a matter of taste I would probably move the common_adapter_XXX and phy_init_data_XXX files into own its modules. (i.e. move the common_adapter_XXX into it's own module)

We can also take the opportunity to turn a few static muts into atomics

[Frostie314159]

@bjoernQ I may need some help with the linker files for esp-radio and esp-phy.

[bjoernQ]

A quick look at libphy.a shows these undefined functions (used the ESP32 library - will slightly vary across chips):

To be defined:

• esp_dport_access_reg_read
• phy_enter_critical
• phy_exit_critical
• rtc_get_xtal

Undefined - no idea if needed

• temprature_sens_read
• phy_tx_pwr_track_en

esp-wifi-sys

• sprint
• phy_printf

esp-rom-sys

• __divsf3
• __floatunsidf
• __divdi3
• memcpy
• __truncdfsf2
• ets_delay_us
• __divdf3
• phy_get_romfuncs

Things defined in esp-*-sys you don't need to worry about.

No idea about the second category - we don't define them anywhere. Maybe they come from one of the other static libraries but haven't seen anything there

Things from the first category should be implemented following the pattern in #3890

[Frostie314159]

Alright, I will implement it like that. Thanks for the detailed response

[Frostie314159]

Hi, sorry for the delay. I'll work on this tomorrow.

[bjoernQ]

There were some changes around the moved code - just FYI

[Frostie314159]

@bjoernQ I'm not quite sure, what's causing the CI failure on the S3. Do you have any ideas?

[Frostie314159]

Oh crap, I forgot about the other uses of critical-section. Sorry

[bugadani]

I think this is now ready to go, thanks very much :)

[SergioGasquez]

Can you modify https://github.com/esp-rs/esp-hal/blob/main/.github/workflows/changelog.yml to check the new esp-phy changelog too?

[bjoernQ]

LGTM

[bjoernQ]

@SergioGasquez I guess we need a thumbs up from your side, too

stale

[bugadani]

This config is unused. Was this an intentional change?

[Frostie314159]

Crap, I forgot to implement that. I can create a new PR for that, if you want.

[bugadani]

That would be awesome, if you don't mind