question about race condition in fat32lib SPI functions

[steve-chamberlin]

In the recvspiblock() function in fat32lib's sdhc.c, there's a comment saying "there is actually a race condition if an interrupt occurs here". Could you explain more about the potential race condition? I've looked at the code and don't see a problem - an interrupt would add some delay but wouldn't harm the SPI transfer as far as I can see. There's also a second comment about "double buffered" that I don't understand - is this referring to the hardware's automatic read data buffer? I started an AVR Freaks thread about these questions here: https://www.avrfreaks.net/s/topic/a5C3l000000UlCuEAK/t193130 Thanks!