Ethernet Layer 2 Framing and Addressing
In this book, as in many Cisco courses and documents, the word frame refers to the bits and bytes that include the Layer 2 header and trailer, along with the data encapsulated by that header and trailer. The term packet is most often used to describe the Layer 3 header and data, without a Layer 2 header or trailer. Ethernet's Layer 2 specifications relate to the creation, forwarding, reception, and interpretation of Ethernet frames.
The original Ethernet specifications were owned by the combination of Digital Equipment Corp., Intel, and Xerox—hence the name "Ethernet (DIX)." Later, in the early 1980s, the IEEE standardized Ethernet, defining parts (Layer 1 and some of Layer 2) in the 802.3 Media Access Control (MAC) standard, and other parts of Layer 2 in the 802.2 Logical Link Control (LLC) standard. Later, the IEEE realized that the
1-byte DSAP field in the 802.2 LLC header was too small. As a result, the IEEE introduced a new frame format with a Sub-Network Access Protocol (SNAP) header after the 802.2 header, as shown in the third style of header in Figure 1-4. Finally, in 1997, the IEEE added the original DIX V2 framing to the 802.3 standard as well as shown in the top frame in Figure 1-40.
Table 1-3 lists the header fields, along with a brief explanation. The more important fields are explained in more detail after the table.
Ethernet (DIX) and Revised (1997) IEEE 802.3
Ethernet (DIX) and Revised (1997) IEEE 802.3
|
Topic
|
|
|
|
|
|
|
|
|
Preamble
|
Dest.
|
Source
|
Type/
|
Data
|
FCS
|
|
|
Address
|
Address
|
Length
|
Original IEEE Ethernet (802.3)
7 16 6
11 1-2 Variable 4
Original IEEE Ethernet (802.3)
7 16 6
11 1-2 Variable 4
|
|
|
|
|
|
D
|
S
|
|
|
|
|
Preamble
|
SD
|
Dest. address
|
Source address
|
Length
|
S A P
|
S A P
|
Control
|
Data
|
802.2
802.3
802.3
802.2
802.3
IEEE 802.3 with SNAP Header
2 Variable 4
IEEE 802.3 with SNAP Header
2 Variable 4
Preamble
|
|
|
|
|
D
|
S
|
|
|
|
|
|
|
SD
|
Dest. address
|
Source address
|
Length
|
S A P
|
S A P
|
Control
|
OUI
|
TYPE
|
Data
|
802.2
SNAP
802.3
Preamble
802.3
802.2
SNAP
802.3
Table 1-3 Ethernet Header Fields
|
Field
|
Description
|
|
Preamble (DIX)
|
Provides synchronization and signal transitions to allow proper clocking of the transmitted signal. Consists of 62 alternating 1s and 0s, and ends with a pair of 1s.
|
|
Preamble and Start of Frame Delimiter (802.3)
|
Same purpose and binary value as DIX preamble; 802.3 simply renames the 8-byte DIX preamble as a 7-byte preamble and a 1-byte Start of Frame Delimiter (SFD).
|
|
Type (or Protocol Type) (DIX)
|
2-byte field that identifies the type of protocol or protocol header that follows the header. Allows the receiver of the frame to know how to process a received frame.
|
|
Length (802.3)
|
Describes the length, in bytes, of the data following the Length field, up to the Ethernet trailer. Allows an Ethernet receiver to predict the end of the received frame.
|
|
Destination Service Access Point (802.2)
|
DSAP; 1-byte protocol type field. The size limitations, along with other uses of the low-order bits, required the later addition of SNAP headers.
|
|
Source Service Access Point (802.2)
|
SSAP; 1-byte protocol type field that describes the upper-layer protocol that created the frame.
|
continues continues
Table 1-3 Ethernet Header Fields (Continued)
|
Field
|
Description
|
|
Control (802.2)
|
1- or 2-byte field that provides mechanisms for both connectionless and connection-oriented operation. Generally used only for connectionless operation by modern protocols, with a 1-byte value of 0x03.
|
|
Organizationally Unique Identifier (SNAP)
|
OUI; 3-byte field, generally unused today, providing a place for the sender of the frame to code the OUI representing the manufacturer of the Ethernet NIC.
|
|
Type (SNAP)
|
2-byte Type field, using same values as the DIX Type field, overcoming deficiencies with size and use of the DSAP field.
|
Types of Ethernet Addresses
Ethernet addresses, also frequently called MAC addresses, are 6 bytes in length, typically listed in hexadecimal form. There are three main types of Ethernet address, as listed in Table 1-4.
Table 1-4 Three Types of Ethernet/MAC Address
|
Type of Ethernet/MAC Address
|
Description and Notes
|
|
Unicast
|
Fancy term for an address that represents a single LAN interface. The I/G bit, the most significant bit in the most significant byte, is set to 0.
|
|
Broadcast
|
An address that means "all devices that reside on this LAN right now." Always a value of hex FFFFFFFFFFFF.
|
|
Multicast
|
A MAC address that implies some subset of all devices currently on the LAN. By definition, the I/G bit is set to 1.
|
Most engineers instinctively know how unicast and broadcast addresses are used in a typical network. When an Ethernet NIC needs to send a frame, it puts its own unicast address in the Source Address field of the header. If it wants to send the frame to a particular device on the LAN, the sender puts the other device's MAC address in the Ethernet header's Destination Address field. If the sender wants to send the frame to every device on the LAN, it sends the frame to the FFFF.FFFF.FFFF broadcast destination address. (A frame sent to the broadcast address is named a broadcast or broadcast frame, and frames sent to unicast MAC addresses are called unicasts or unicast frames.)
Multicast Ethernet frames are used to communicate with a possibly dynamic subset of the devices on a LAN. The most common use for Ethernet multicast addresses involves the use of IP multicast. For example, if only 3 of 100 users on a LAN want to watch the same video stream using an IP multicast-based video application, the application can send a single multicast frame. The three interested devices prepare by listening for frames sent to a particular multicast Ethernet address, processing frames destined for that address. Other devices may receive the frame, but they ignore its contents. Because the concept of Ethernet multicast is most often used today with IP multicast, most of the rest of the details of Ethernet multicast will be covered in Chapter 16, "Introduction to IP Multicasting."
Ethernet Address Formats
The IEEE intends for unicast addresses to be unique in the universe by administering the assignment of MAC addresses. The IEEE assigns each vendor a code to use as the first 3 bytes of its MAC addresses; that first half of the addresses is called the Organizationally Unique Identifier (OUI). The IEEE expects each manufacturer to use its OUI for the first 3 bytes of the MAC assigned to any Ethernet product created by that vendor. The vendor then assigns a unique value in the low-order 3 bytes for each Ethernet card that it manufactures—thereby ensuring global uniqueness of MAC addresses. Figure 1-5 shows the basic Ethernet address format, along with some additional details.
Figure 1-5 Ethernet Address Format
Most Significant Byte
Least Significant Byte
Most Significant Byte
Least Significant Byte
|
1st Byte
|
2nd Byte
|
3rd Byte
|
4th Byte
|
5th Byte
|
6th Byte
|
Vendor-Assigned
Vendor-Assigned
Least Most
Significant Significant Bit Bit
Note that Figure 1-5 shows the location of the most significant byte and most significant bit in each byte. IEEE documentation lists Ethernet addresses with the most significant byte on the left. However, inside each byte, the leftmost bit is the least significant bit, and the rightmost bit is the most significant bit. Many documents refer to the bit order as canonical; other documents refer to it as little-endian. Regardless of the term, the bit order inside each byte is important for understanding the meaning of the two most significant bits in an Ethernet address:
Table 1-5 summarizes the meaning of each bit.
18 Chapter 1: Ethernet Basics Table 1-5 I/G and U/L Bits
|
Field
|
Meaning
|
|
I/G
|
Binary 0 means the address is a unicast; Binary 1 means the address is a multicast or broadcast.
|
|
U/L
|
Binary 0 means the address is vendor assigned; Binary 1 means the address has been administratively assigned, overriding the vendor-assigned address.
|
The I/G bit signifies whether the address represents an individual device or a group of devices, and the U/L bit identifies locally configured addresses. For instance, the Ethernet multicast addresses used by IP multicast implementations always start with 0x01005E. Hex 01 (the first byte of the address) converts to binary 00000001, with the most significant bit being 1, confirming the use of the I/G bit.
NOTE Often, when overriding the MAC address to use a local address, the device or device driver does not enforce the setting of the U/L bit to a value of 1.
Protocol Types and the 802.3 Length Field
Each of the three types of Ethernet header shown in Figure 1-4 has a field identifying the format of the Data field in the frame. Generically called a Type field, these fields allow the receiver of an Ethernet frame to know how to interpret the data in the received frame. For instance, a router might want to know whether the frame contains an IP packet, an IPX packet, and so on.
DIX and the revised IEEE framing use the Type field, also called the Protocol Type field. The originally-defined IEEE framing uses those same 2 bytes as a Length field. To distinguish the style of Ethernet header, the Ethernet Type field values begin at 1536, and the length of the Data field in an IEEE frame is limited to decimal 1500 or less. That way, an Ethernet NIC can easily determine whether the frame follows the DIX or original IEEE format.
The original IEEE frame used a 1-byte Protocol Type field (DSAP) for the 802.2 LLC standard type field. It also reserved the high-order 2 bits for other uses, similar to the I/G and U/L bits in MAC addresses. As a result, there were not enough possible combinations in the DSAP field for the needs of the market—so the IEEE had to define yet another type field, this one inside an additional IEEE SNAP header. Table 1-6 summarizes the meaning of the three main Type field options with Ethernet.
Table 1-6 \ Topic
Responses
Ethernet Type Fields
|
Type Field
|
Description
|
|
Protocol Type
|
DIX V2 Type field; 2 bytes; registered values now administered by the IEEE
|
|
DSAP
|
802.2 LLC; 1 byte, with 2 high-order bits reserved for other purposes; registered values now administered by the IEEE
|
|
SNAP
|
SNAP header; 2 bytes; uses same values as Ethernet Protocol Type; signified by an 802.2 DSAP of 0xAA
|
Continue reading here: Virtual LANs
Was this article helpful?
|
|