Hello!

I am trying to use/understand the Ethernet CRC module but unable to understand the generation of magic number for nibbles when 32 bit CRC is generated. I found the following link http://goo.gl/stlQv where Illan Glasner has uploaded a verilog file for the CRC generation and verification using magic number. I am unable to understand that how did he get the values for cnt during the second run when CRC is being verified.

I anybody knows how the values of cnt were selected [I am assuming some swap operation on original values from first run but couldn't figure it out myself] please explain to me.

Thank you !

P.S : I am attaching the file here for your convenience.

Ethernet CRC has some slightly odd properties that people address in different ways; it's been a while since I looked at the 802.3 spec but as I recall it gives the CRC polynomial and then specifies it needs to be inverted and sent bit-reversed. Some generators use a different polynomial to effectively incorporate these steps into the computation.

If your answer is off, check it see if the result is similar by some regular pattern (byte swap, bit swap, invert).