STM32F4 does not require the HSI to erase or program FLASH
The HSI needs to be left on until a new clock source is chosen.
(we leave it on all the time)
NuttX had the CRTSCTS define incorrectly set for only output flow control, which broke our flow control logic. This commit patches NuttX and puts in addition a guard in place to prevent any future issue with the non-POSIX define being incorrect.
This has been debugged and identified by @ecmnet, which was the main contribution for this patch.
This hot fix essentialy revert commit 265af481209d60033f7cd4c4216048b1ce3eb435
in NuttX/nuttx. The commit STM32 serial: Make input hardware flow-control work with RX DMA.
From Jussi Kivilinna has broken the DMA on an STM32F4 in a yet TBD way.
The symptoms are lost data on RX, the DMA count decrements but
the data ia not written to memory. This looks to be introduced but the
non circular DMA settings.
1) Ensure if CONFIG_SERIAL_DMA is set that cdcacm uart_ops is initalized
with correct functions in correct slots.
This was detected only with PX4 build flags
2) C&P error from F7 would prevent CONFIG_STM32_SERIALBRK_BSDCOMPAT ifdefed
code from being included.
David Sidrane
Lorenz Meier