drivers/uart: stm32: Report 9bits transactions as not supported

[erwango]

STM32 uart driver doesn't support 9bits transactions in any case,
so remove case were it was declared as supported.

Fixes #31799

Signed-off-by: Erwan Gouriou [email protected]

[erwango]

Reported error is the following:

-:13: WARNING:CONSTANT_COMPARISON: Comparisons should place the constant on the right side of the test
#13: FILE: drivers/serial/uart_stm32.c:342:
+	    || (UART_CFG_DATA_BITS_9 == cfg->data_bits)) {

Need to fix the whole file...

[ABOSTM]

LGTM

[hwilmers]

This removes support for following configuration:
UART_CFG_PARITY_EVEN,
UART_CFG_DATA_BITS_9
I found this configuration to result in the uart performing "8 bits, even parity" (on stm32l475) with good success.
Would there be any other way to achieve this configuration after above PR?

[erwango]

I found this configuration to result in the uart performing "8 bits, even parity" (on stm32l475) with good success.

@hwilmers, do you mean "9 bits, even parity"? Or maybe I'm not understanding correctly your point.

[hwilmers]

I mean "8 bits, even parity", like in "19200 8E1".
It may be the result of an unintended side effect that this works.

[erwango]

It may be the result of an unintended side effect that this works.

Probably

@dcpleung, do you have any hint for this question: #32204 (comment) ? Should we find a way to let it work "unintentionally" ?

[hwilmers]

Apparently, the word length as it is used in the stm32 usart includes the optional parity bit. This is what I found in this official reference manual, on page 1345:
https://www.st.com/content/ccc/resource/technical/document/reference_manual/02/35/09/0c/4f/f7/40/03/DM00083560.pdf/files/DM00083560.pdf/jcr:content/translations/en.DM00083560.pdf

A straight forward translation between usual notation and the implementation on stm32 would be to add one to the word length if a parity bit is appended.