LPSPI master - How to perform an LPSPI continuous transfer

[KenG]

I would like to perform a SPI master transaction using the lpspi driver.

Transaction 1:

1. Set chip select to low

2. Write 4 bytes to target device. Write command + 3 bytes of address.

3. Keep chip select low.

Traction 2 (continued transaction):

1. Write payload bytes

3. Release chip select

Chip select must be held low over the two transactions.

I have worked with LPSPI_MasterTransferBlocking  and the kLPSPI_MasterPcsContinuous option, but so far I am only able to do this by creating a second larger buffer where the buffer includes both overhead + payload.  How do you keep chip select active while the payload is written?

[Pavel_Hernandez]

Hello, my name is Pavel, and I will be supporting your case, could you tell me about the MCU are you using?

Best regards,
Pavel

[KenG]

I am using IMXRT 1171, Cortex M7 processor.  

To clarify, I would like to implement one function that starts the transfer and have another function continue and complete the transfer.  

Here is an example:

The user passes a buffer and asks SPI memory to space to be read.  Read back data should be placed in the buffer.  The buffer is allocated just enough to hold the memory read, no space for any overhead.

The read implementation needs to

1. Setup memory device pointer (function 1) within the target SPI memory read device. CS must be held low.

2. Perform the actual read of memory (function 2).  Continue to read and complete read.

The issue I have is that SPI master read function does an atomic read where step 1 and 2 is performed in one step using CONT and CONTC bits.  This means a second buffer is needed to hold both "setup data" and "read data".  I would like to avoid the second second buffer and do an optimal read. 

Can this be done with current drivers?

[Pavel_Hernandez]

Hello,

Thanks for being clear about your information, unfortunately, the examples available are this, if the examples do not work for you, I suggest modifying to your application.

Best regards,
Pavel