I also get this error when trying to flash/verify, this has an error result, might be helpful.

Updating C:/Users/Public/Documents/EVE Asset Builder/work_space/output/Image400x400.bin to flash, erasing if necessary ...
WARNING: Blob driver will be overwriten!
Updating "C:/Users/Public/Documents/EVE Asset Builder/work_space/output/Image400x400.bin" to Flash Storage.
Switch flash status from FT810 to FULL. Result: ERROR
Unknown error code: 0xCD36DCD5
Flash status: FULL.
Verifying Flash...
Verifying flash fail at flash address 0, file address 0
Verifying FAIL!

Kindest regards

The jumpers I'm using are the breadboard wires, maybe 10-12in long.  I'll try the short ones and see if that makes a difference.  Thanks for the response.

Shortest breadboard wires I have still produce the same results.

I also tried changing the clock speed from 1-20 in steps of 1 and it didn't yield any different results, still fails to detect most of the time, never recognizes the blob.  I also tried to erase the chip and updated, it will sometimes say it's successful but running detect again shows same result, never detecting the blob and failing to detect.

Kindest regards

LL_SPI_Enable() should be at the end of MX_SPI1_Init() in main.c.

What's worse for me, it didn't include the stm32wbxx_ll_spi.h file, so I had to add that to my main.h file as well, ugh.  This has solved the issue though and is a much better solution that what I was doing.

I just checked what STM32CubeIDE is generating for me and I also do not have this line,
whatever the idea behind fully configuring a unit and then not activating it in the end is.

Yeah, I'm beginning to question my choice here.

Well, STM32CubeIDE, I still have the Code that it generated for the STM32G0B1CET6, took me a bit to find out that the
generated MX_SPI1_Init() neither activated the clock or configured the SPI pins.
And the one working version I have in the other folder has __HAL_SPI_ENABLE(&hspi1); at the end, the generated one does not.

I apologize for this consuming so much of your time, I really do appreciate your assistance with this, it was driving me crazy.

If only you had pick one of the few STM32 controllers I have a Nucleo board with here... :-)

Ha, I was getting to a place that I would have sent you one .

Thanks again for all of your assistance.

Kindest regards

Hello,

I just came across it while I was searching for why the RXNE was always return 0, there are two methods the LL SPI uses to initialize itself, LL_SPI_Init() and LL_SPI_Enable().  I searched the STM32 documentation and couldn't find exactly what it does but this is the source:

/**
  * @brief  Enable SPI peripheral
  * @rmtoll CR1          SPE           LL_SPI_Enable
  * @param  SPIx SPI Instance
  * @retval None
  */
__STATIC_INLINE void LL_SPI_Enable(SPI_TypeDef *SPIx)
{
  SET_BIT(SPIx->CR1, SPI_CR1_SPE);
}

There's also a way to ensure the LL_SPI is initialized:

/**
  * @brief  Check if SPI peripheral is enabled
  * @rmtoll CR1          SPE           LL_SPI_IsEnabled
  * @param  SPIx SPI Instance
  * @retval State of bit (1 or 0).
  */
__STATIC_INLINE uint32_t LL_SPI_IsEnabled(SPI_TypeDef *SPIx)
{
  return ((READ_BIT(SPIx->CR1, SPI_CR1_SPE) == (SPI_CR1_SPE)) ? 1UL : 0UL);
}

I changed Rudolph's spi_transmit method to look like this:

static inline void spi_transmit(uint8_t data)
{
   if(!LL_SPI_IsEnabled(EVE_SPI))
   {
      LL_SPI_Enable(EVE_SPI);
   }

   LL_SPI_TransmitData8(EVE_SPI, data);
   while (!LL_SPI_IsActiveFlag_TXE(EVE_SPI)) {}
   while (!LL_SPI_IsActiveFlag_RXNE(EVE_SPI)) {}
   LL_SPI_ReceiveData8(EVE_SPI); /* dummy read-access to clear SPI_SR_RXNE */
}

I still have an icky feeling putting that code in the spi_transmit(), just doesn't feel right and I doubt Rudolph has anyone complaining about it.  Could be just something in my CUBE IDE configuration I missed too.

I don't know if it needs to call, LL_SPI_DISABLE() at some point or not.  The above source came from my, stm32wbxx_ll_spi.h.

If there is anything I can look at for you let me know.

Kindest regards

Wow!  Thank you so much Rudolph!  I'm a newbie to CUBE IDE and trying to figure all of this out.  Guess I shouldn't have started with a display, LOL.

I have it building without errors or warnings using your very detailed steps, very cool!!!

This is light years ahead of where I've been for the last week.  Thanks again for your time and effort helping a newbie and putting this library together.

Currently stuck inside:

EVE_target_STM32.h->spi_transmit(...)->while (!LL_SPI_IsActiveFlag_RXNE(EVE_SPI)) {}

It never comes out of the while loop and the return always = 0, ugh. 

Kindest regards

Hello All,

New here, I've been writing Windows based software for longer than I'd like to admit and I'm having the hardest time trying to integrate any of the EVE libraries out on GitHub into an STM32CubeIDE project.  I've tried Riverdi (that's the display I'm trying to get working), Rudolph Riedel's and MatrixOrbital's libraries without much luck.  The steps I've tried:

1. Create a new project in Stm32CubeIDE using the STM32Wb55RG Nucleo board (I'm creating a custom board based off the MCU so bought the Nucleo for prototyping).
2. Dragged the cooresponding .c & .h files from the perspective library into the STM32CubeIDE file tree, .c to Src folder and .h to Inc folder.  Trying to keep it simple instead of creating specific folders like in the libraries for now to try and get something to compile.
3. Copied the contents of the host_layer/Stm32/platform.h and platform.c and copied into my main.h and main.c files.

Then I spend a few hours trying to get it to build.  Fail miserably and moved onto the next library trying the same steps (I am the definition of insanity, lol).  Can anyone provide some guidance to getting one of the libraries to build in an STM32CubeIDE project?

I am very new to hardware and STM32CubeIDE, which is why I think I'm having so much difficulty.

Kindest regards