07-30-2008 10:35 PM
does any body know how is the CRC of jumbo frames (in Gigabit Ethernet)of size greater than 9K is found? What algorithm is used for that?
08-01-2008 02:31 PM
I believe CRC for jumbo frames is calculated the same way as a standard frame (CRC32). Specifically, using the method described in sec 3.2.8 of IEEE 802.3-2002. As paraphrased in UG194:
"The value of the FCS field is calculated over the destination address, source address,
length/type, data, and pad fields using a 32-bit Cyclic Redundancy Check (CRC), as
defined in IEEE Std 802.3-2002 para. 3.2.8:
G(x) = x32 + x26 + x23 + x22 + x16 + x12 + x11 + x10 + x8 + x7 + x5 + x4 + x2 + x1 + x0
The CRC bits are placed in the FCS field with the x31 term in the left-most bit of the first
byte, and the x0 term is the right-most bit of the last byte (i.e., the bits of the CRC are
transmitted in the order x31, x30,..., x1, x0)."
08-13-2008 11:57 PM
It has been found in studies that the efficiency of CRC-32 remains same for frame sizes from 1500 to 9016 (approx). However, implementation of frame sizes bigger than this doesnt make sense, because the effciency of CRC-32 decreases and errors creep in. Hence, using conventional CRC-32 for frame sizes bigger than 9K wont work.
So incase u are looking in terms of experimenting with jumbo frames greater than 9000 which is presently being pushed through, then u will also have to think in terms of implementing a bigger CRC-32 genrator polynomiial, for which yu will again voilate the frame format, which seems unlikely in near future.
10-05-2008 04:27 PM