All,
Some of my colleagues, who haven’t been able to participate in the task force work directly, passed me some questions that they would like to discuss with the IET experts in IEEE 802.3br task force.
I’ve asked them to call in to the conference calls so that they can contribute. In the meantime, here are the questions (all based on the adopted baseline presentation):
-
Page 12: The proposal replaces the SFD delimiter of Express frames by SMD-E.
-
If a standard preamble is: 0xFB 0x55 0x55 0x55 0x55 0x55 0x55 0xD5, is SMD-E = 0xAA?
-
Alternative: keep the preamble of Express frames unmodified because non-express or normal frames can be unambiguously distinguished by SMD-Ix and SMD-Cx.
-
Page 14: alignment size octet is shown as TBD.
-
Page 12: Since Intermediate Fragments are not Ethernet frames (no DA, SA, EtherType), the proposal appears to indicate hop-to -hop operation. For such an implementation, do we need to have 2 bytes of preamble — only 1 byte (SOP)
should be sufficient?
-
Page 12 and Page 16: What is the value of octet Frag Count in Intermediate Fragments? Is it a rolling value of 1, 2, 3, 4 or the Frag value shown on page 16?
-
For ease of implementation, can we consider an alternative to SMD-Cx + Frag Count: incrementing Fragment Count + CRC protection
-
Page 12: Is MFCS = (running FCS of bytes transmitted) XOR 0x0000_FF00?
-
Page 12: Is the alternative of MFCS, a 1-byte fragmentation trailer, under consideration? It may be easier to implement.
We can talk tomorrow.
Hugh.
|