Ethernet Autonegotiation – How Auto-Negotiate Determines Link Speeds

autonegotiationethernet

I have two Linux hosts connected via category 6 wire. How do the LAN cards negotiate a link speed?

Essentially, I am trying to solidify if I am actually hammering a 1GbE NIC with traffic from a 10Gbe NIC or if they negotiated a 1GbE link speed.

Best Answer

A 10 Gb NIC connected directly to a 1 Gb either negotiates a 1 Gb connection, or it doesn't work at all. You cannot have the situation you describe unless you have something like a switch in between, where the 10 Gb NIC connects to a 10 Gb switch port.

It is very easy to search and find ethernet speed negotiation on the Internet, such as the Wikipedia article:

Auto-negotiation (formerly "NWay") is based on pulses similar to those used by 10BASE-T devices to detect the presence of a connection to another device. These connection present pulses are sent by Ethernet devices when they are not sending or receiving any frames. They are unipolar positive-only electrical pulses of a nominal duration of 100 ns, with a maximum pulse width of 200 ns,[5] generated at a 16 ms time interval (with a timing variation tolerance of 8 ms). These pulses are called link integrity test (LIT) pulses in the 10BASE-T terminology, and are referred to as normal link pulses (NLP) in the auto-negotiation specification.

A device detects the failure of a link if neither a frame nor two of the LIT pulses is received for 50-150 ms. For this scheme to work, devices must send LIT pulses regardless of receiving any. Three bursts of Fast Link Pulses, used by autonegotiating devices to declare their capabilities.

Auto-negotiation uses similar pulses labeled as NLP. NLP are still unipolar, positive-only, and of the nominal duration of 100 ns; but each LIT is replaced by a pulse burst consisting of 17 to 33 pulses sent 125 µs apart. Each such pulse burst is called a fast link pulse (FLP) burst. The time interval between the start of each FLP burst is the same 16 milliseconds as between normal link pulses (variation tolerance of 8 ms). How a link code word (a 16 bit word) is encoded in a fast link pulse burst

The FLP burst consists of 17 NLP at a 125 µs time interval (with a tolerance of 14 µs). Between each pair of two consecutive NLP (i.e. at 62.5 µs after first NLP of the pulse pair) an additional positive pulse may be present. The presence of this additional pulse indicates a logical 1, its absence a logical 0. As a result, every FLP contains a data word of 16 bits. This data word is called a link code word (LCW). The bits of the link code word are numbered from 0 to 15, where bit 0 corresponds to the first possible pulse in time and bit 15 to the last.

Edit for the dumbed-down, plain-English version from Fluke, per your comment:

In regular English, auto-negotiation is simply the method used by two devices to achieve the best rate possible of transmission. It allows them to discuss in a sense, the possible rates of transmission, then pick the best one that they both share. They do this by swapping advertisements of their own abilities using pulses called Fast Link Pulses (FLPs). The FLP lets one link partner know what the other is capable of. As they swap FLPs, the two stations detect the highest common denominator between them, according to the following:

  • 1000BASE-T
  • 100BASE-T2 Full Duplex
  • 100BASE-TX Full Duplex
  • 100BASE-T2
  • 100BASE-T4
  • 100BASE-TX
  • 10BASE-T Full Duplex
  • 10BASE-T

For example, if station A and B are auto-negotiating with each other, and A is capable of 10/100/1000 full or half duplex, but B is only capable of 10/100 full or half duplex, then the highest common link rate is 100, at full duplex. As long as both sides are negotiating, the link will operate at the highest possible connection rate between them.

Related Topic