spi_slave_api.c: Allow to use GPIO 32 and 33 for interface signals.

[robert-hh]

Using consistently the IDF API instead of direct port access. There is a small performace drop compared to direct port access for setting the pin, but that can be accepted in favoer of consistence.

GPIO33 for instance is used by Adafruit and NINA modules for the handshake signal, and these modules cannot be rewired.

[CLAassistant]

All committers have signed the CLA.

[mantriyogesh]

There are some issues in the PR. can you please look into?

[robert-hh]

As far as I see, the conflict is just a more literal one, not finally affecting the code's operation. It seems that some changes have been made to the master since I made the PR. I'll resolve it.

[robert-hh]

The PR is updated now. It has no conflict with the master branch and a test build works with Adafruit Airlft modules.

[mantriyogesh]

I understand the rework pain. But this MR is important to handle GPIOs > 32..

[mantriyogesh]

Change looks good. We will pull this in next week.

[mantriyogesh]

Thank you @robert-hh @JAndrassy for all your efforts and time ..

[robert-hh]

@JAndrassy The actual PR does the same with the defines a few lines before defining the masks. I know that this is a kind of additional step. but GPIO_HS amd GPIO_DR are also used another place in the code.

#define GPIO_HS                    CONFIG_ESP_SPI_GPIO_HANDSHAKE
#define GPIO_DR                    CONFIG_ESP_SPI_GPIO_DATA_READY

#define GPIO_MASK_DATA_READY (1ull << GPIO_DR)
#define GPIO_MASK_HANDSHAKE (1ull << GPIO_HS)

[robert-hh]

Change looks good. We will pull this in next week.

@mantriyogesh Thanks.

[JAndrassy]

@robert-hh yes sorry. I have my changes in the 0.0.5 tag sources