A quick search harvested this webpage. I wasn't able to find the license for these code snippets.
The following should do the job:
// ----------------------------- crc32b --------------------------------
/* This is the basic CRC-32 calculation with some optimization but no
table lookup. The the byte reversal is avoided by shifting the crc reg
right instead of left and by using a reversed 32-bit word to represent
the polynomial.
When compiled to Cyclops with GCC, this function executes in 8 + 72n
instructions, where n is the number of bytes in the input message. It
should be doable in 4 + 61n instructions.
If the inner loop is strung out (approx. 5*8 = 40 instructions),
it would take about 6 + 46n instructions. */
unsigned int crc32b(unsigned char *message) {
int i, j;
unsigned int byte, crc, mask;
i = 0;
crc = 0xFFFFFFFF;
while (message[i] != 0) {
byte = message[i]; // Get next byte.
crc = crc ^ byte;
for (j = 7; j >= 0; j--) { // Do eight times.
mask = -(crc & 1);
crc = (crc >> 1) ^ (0xEDB88320 & mask);
}
i = i + 1;
}
return ~crc;
}