RE: [802.3ae] FCS representation in walker_1_p0700.pdf
The MAC layer transmits bytes LSB first.
It computes the CRC in transmited bit order (first bit transmitted
is the most significant coefficient) and it puts the CRC into bytes
such that the CRC is transmitted in order from most significant
coefficient to least significant coefficient. That is, the x^31
coefficient of the CRC is placed in the LSB of the first byte and
the x^0 coefficient is placed in the MSB of the fourth CRC byte.
From: David Barral [mailto:David.Barral@xxxxxxxxxxxx]
Sent: Thursday, May 30, 2002 12:03 AM
Subject: [802.3ae] FCS representation in walker_1_p0700.pdf
I will need some clarification about the FCS computation and its
representation at the MAC layer in the 64b/66b test vector given in
As mentioned in the section 3.3 of the IEEE 302.3 standard, "Each octet
of the MAC frame, with the exception of the
FCS, is transmitted low-order bit first".
Given that, in the 64b/66b test vector given in walker_1_p0700.pdf, the
four bytes corresponding to the FCS position
are 93 eb f7 79.
In the same 64b/66b test vector given in walker_1_p0700.pdf, looking at
the PCS scrambled "frame", it seems that the FCS values at the input of
the scrambler are c9 d7 ef 9e. That would mean that the FCS value is c9
d7 ef 9e
and that the FCS is represented LSBit first at the MAC layer level.
Hence, I simply would like to know the following:
- Considering the 93 eb f7 79 byte sequence, does the MAC layer transmit
each byte LSBit or MSBit first?
- If the MAC layer transmit them LSBit first, can anybody confirm that
the FCS value is then c9 d7 ef 9e.
Thanks very much,