Click here for more information on LAN testing.

View this file in pdf format.

The Data Link Layer defines how data is formatted for transmission and how access to the network is controlled. This layer has been divided by the IEEE 802 standards committee into two sublayers: media access control (MAC) and logical link control (LLC).

The following data link layer protocols are described:

• Ethernet.
• Token Ring.
• FDDI.
• LLC.
• CIF
• SRP.
• GARP.
• GMRP.
• GVRP.
• SRP
• VLAN.

FDDI, Token Ring and Ethernet may be physical interfaces or may act as logical protocols encapsulated over a WAN protocol or ATM.

The following illustration represents the LAN protocols in relation to the OSI model:

Ethernet

Ethernet is a widely used data communications network standard developed by DEC, Intel, and Xerox. It uses a bus topology and CMSA/CD access method. The terms Ethernet and the IEEE 802.3 standard are often used interchangeably.

The Ethernet header structure is shown in the illustration below.

Destination	Source	Len	Data unit + pad	FCS
(6 bytes)	(6 bytes)	(2)	(46-1500 bytes)	(4 bytes)

Destination address
The address structure is as follows:

I/G

U/L

Address bits

I/G	Individual / group address may be:
	0 Individual address. 1 Group address.
U/L	Universal /local address may be:
	0 Universally administered. 1 Locally administered.

Source address
The address structure is as follows:

0

U/L

Address bits

0	The first bit is always 0.
U/L	Universal/local address may be:
	0 Universally administered. 1 Locally administered.

Length/type
In the Ethernet protocol, the value (³ 0x0600 Hex) of this field is Ethernet List, indicating the protocol inside.

In the 802.3 protocol, the value (46-1500 Dec) is the length of the inner protocol, which is the LLC encapsulated inner protocol. (The LLC header indicates the inner protocol type.)

Data unit + pad
LLC protocol.

FCS
Frame check sequence.

Token Ring

Token Ring is a LAN protocol where all stations are connected in a ring and each station can directly hear transmissions only from its immediate neighbor. Permission to transmit is granted by a message (token) that circulates around the ring.

The Token Ring header structure is shown in the illustration below.

SDEL / EDEL
Starting Delimiter / Ending Delimiter. Both the SDEL and EDEL have intentional Manchester code violations in certain bit positions so that the start and end of a frame can never be accidentally recognized in the middle of other data.

Access control
The format is as follows:

P

P

P

T

M

R

R

R

PPP	Priority bits:
	000	Lowest priority.
	111	Highest priority.
T	Token bit:
	0	Token.
	1	Frame.
M	Monitor count:
	0	Initial Value.
	1	Modified to active monitor.
RRR	Reservation bits:
	000	Lowest priority reservation.
	111	Highest priority reservation.

Frame control
The format is as follows:

Frame type (2)

0

0

Attention (4 bits)

Frame type may have the following values:

00	MAC frame.
01	LLC frame.
11 or 10	Undefined.

The second 2 bits are always zero.

Attention indicates those frames for which the adapter does special buffering and processing.

0001	Express buffer.
0010	Beacon.
0011	Claim token.
0100	Ring purge.
0101	Active monitor present.
0110	Standby monitor present.

Destination address
The address structure is as follows:

I/G

U/L

Address bits

I/G	Individual / group address may be:
	0 Individual address. 1 Group address.
U/L	Universal /local address may be:
	0 Universally administered. 1 Locally administered.

Source address
The address structure is as follows:

RII

I/G

Address bits

RII	Routing information indicator:
	0	RI absent.
	1	RI present.
I/G	Individual/group address:
	0	Group address.
	1	Individual address.

Route information
The structure is as follows:

				RI Field
		RC Field					RD Fields
RT	LTH	D	LF	r	RD1	RD2	…	RDn
3	5	1	6	1	16	16		16bits
				Length in LTH Field

RC	Routing control (16 bits).
RDn	Route descriptor (16 bits).
RT	Routing type (3 bits).
LTH	Length (5 bits).
D	Direction bit (1 bit).
LF	Largest frame (6 bits).
r	Reserved (1 bit).

Information
The Information field may be LLC or MAC. The MAC information structure is as follows:

Major vector		Subvector 1				Subvector n
VL	VI	SVL	SVI	SVV	…	SVL	SVI	SVV
2	2	1	1	n		1	1	n bytes

VL
Major vector length. Specifies the length of the vector in octets.

VI
Major vector identifier. A code point that identifies the vector. The VI format is as follows:

4	8	16 bits
Destination class	Source class	Major vector code

Destination class / source class
Class fields assure proper routing within a ring station:

0	Ring station.
4	Configuration report server.
5	Ring parameter server.
6	Ring error monitor.

Major vector code
The vector code uniquely defines the vector:

0x00	Response.
0x02	Beacon.
0x03	Claim token.
0x04	Ring purge.
0x05	Active monitor present.
0x06	Standby monitor present.
0x07	Duplicate address test.
0x08	Lobe media test.
0x09	Transmit forward.
0x0B	Remove ring station.
0x0C	Change parameters.
0x0D	Initialize ring station.
0x0E	Request station addresses.
0x0F	Request station state.
0x10	Request station attachment.
0x20	Request initialization.
0x22	Report station addresses.
0x23	Report station state.
0x24	Report station attachment.
0x25	Report new active monitor.
0x26	Report SUA change.
0x27	Report neighbor notification incomplete.
0x28	Report active monitor error.
0x29	Report error.

SVL
Sub-vector length. Specifies the length of the sub-vector in octets.

SVI
Sub-vector identifier. A code point that identifies the sub-vector:

0x01	Beacon type.
0x02	Upstream neighbor addresses next.
0x03	Local ring number.
0x04	Assign physical drop number next.
0x05	Error timer value.
0x06	Authorized function classes next.
0x07	Authorized access priority.
0x08	Authorized environment.
0x09	Correlation.
0x0A	SA of last AMP or SMP.
0x0B	Physical drop number.
0x20	Response code.
0x21	Reserved.
0x22	Product instance ID.
0x23	Ring station version number.
0x26	Wrap data.
0x27	Frame forward.
0x28	Station identifier.
0x29	Ring station status.
0x2A	Transmit status code.
0x2B	Group address(es).
0x2C	Functional address(es).
0x2D	Isolating error count.
0x2E	Non-isolating error count.
0x2F	Function request ID.
0x30	Error code.

SVV
Sub-vector value - Variable length sub-vector information.

FCS
Frame check sequence.

Frame status
Contains bits that may be set on by the recipient of the frame to signal recognition of the address and whether the frame was successfully copied.

FDDI

The Fiber Distributed Data Interface (FDDI) is a 100 Mega-bit technology using a timed token over a dual ring of trees. FDDI is standardized by the American National Standards Institute (ANSI).

The FDDI header structure is shown in the illustration below.

Frame control	Destination address	Source address	Route information	Information	FCS
2	6	6	0-30		4 bytes

Frame control
The frame control structure is as follows:

C	Class bit:
	0	Asynchronous frame.
	1	Synchronous frame/
L	Address length bit:
	0	16 bits (never).
	1	48 bits (always).
FF	Format bits.
ZZZZ	Control bits.
	The following is a description of the various Frame Control field values (CLFF ZZZZ to ZZZZ):
	0x00 0000	Void frame.
	1000 0000	Non-restricted token.
	1100 0000	Restricted token.
	0L00 0001 to 1111	Station management frame.
	1L00 1111	SMT next station addressing frame.
	1L00 0001 to 1111	MAC frame.
	1L00 0010	MAC beacon frame.
	1L00 0011	MAC claim frame.
	CL01 r000 to r111	LLC frame.
	0L01 rPPP	LLC information frame (asynchronous, PPP=frame priority).
	0L01 rrrr	LLC information frame (synchronous, r=reserved).
	CL10 r000 to r111	Reserved for implementer.
	CL11 rrrr	Reserved for future standardization.

Destination address
The address structure is as follows:

I/G

U/L

Address bits

I/G	Individual / group address may be:
	0 Individual address. 1 Group address.
U/L	Universal /local address may be:
	0 Universally administered. 1 Locally administered.

Source address
The address structure is as follows:

I/G

RII

Address bits

I/G	Individual/group address:
	0	Group address.
	1	Individual address.
RII	Routing information indicator:
	0	RI absent.
	1	RI present.

Route Information
The structure of the route information is as follows:

				RI Field
		RC Field					RD Fields
RT	LTH	D	LF	r	RD1	RD2	…	RDn
3	5	1	6	1	16	16		16bits
				Length in LTH Field

RC	Routing control (16 bits).
RDn	Route descriptor (16 bits).
RT	Routing type (3 bits).
LTH	Length (5 bits).
D	Direction bit (1 bit).
LF	Largest frame (6 bits).
r	reserved (1 bit).

Information
The Information field may be LLC, MAC or SMT protocol.

FCS
Frame check sequence.

LLC

The IEEE 802.2 Logical Link Control (LLC) protocol provides a link mechanism for upper layer protocols. LLC type I service provides a datalink connectionless mode service, while LLC type II provides a connection-oriented service at the datalink layer.

The LLC header structure is shown in the illustration below.

DSAP	SSAP	Control	LLC information
1 byte	1 bytes	1 or 2 bytes

DSAP
The destination service access point structure is as follows:

I/G	Address bits

I/G	Individual/group address may be:
	0	Individual DSAP.
	1	Group DSAP.

SSAP
The source service access point structure is as follows:

C/R	Address bits

C/R	Command/response:
	0	Command.
	1	Response.

Control
The structure of the control field is as follows:

	1	8				9	16 bits
Information	0	N(S)				P/F	N(R)
Supervisory	1	0	SS	XXXX		P/F	N(R)
Unnumbered	1	1	MM	P/F	MMM

N(S)	Transmitter send sequence number.
N(R)	Transmitter receive sequence number.
P/F	Poll/final bit. Command LLC PDU transmission/ response LLC PDU transmission.
	S	Supervisory function bits:
	00	RR (receive ready).
	01	REJ (reject).
	10	RNR (receive not ready).
X	Reserved and set to zero.
M	Modifier function bits.

LLC information
LLC data or higher layer protocols.

CIF

CIF (Cells In Frames) describes the mechanism by which ATM traffic is carried across a media segment and a network interface card conforming to the specification for Ethernet Version 2, IEEE 802.5 Token Ring, or IEEE 802.3. ATM cells can be carried over many different physical media, from optical fiber to spread spectrum radio. ATM is not coupled to any particular physical layer. CIF defines a new pseudo-physical layer over which ATM traffic can be carried. It is not simply a mechanism for translation between frames and cells; neither is it simple encapsulation. CIF carries ATM cells in legacy LAN frames. This defines a protocol between CIF end system software and CIF attachment devices ("CIF-AD") which makes it possible to support ATM services, including multiple classes of service, over an existing LAN NIC just as if an ATM NIC were in use. CIF specifies how the ATM layer protocols can be made to work over the existing LAN framing protocols in such a way that the operation is transparent to an application written to an ATM compliant API. Over Ethernet, CIF frames have an Ethernet header and trailer. CIF frames are encapsulated in Token Ring and LLC by use of a SNAP header.
(Compliant with Cells in Frames Version 1.0 Specification, Analysis and discussion.)

The format of the header is shown in the following illustration:

P
Even Parity bit for an octet.

CIF Format
CIF Format Identifier. Only three format types are defined. Formats 0 and 1 are used for CIF signalling. Format 2 is the default format for carrying user traffic. Formats 112-127 are reserved for use in experimentation and for pre-standard CIF implementations.

FF
CIF format independent flags. These bits contain flags that are independent of any CIF format type. These CIF format independent flags are reserved. They are set to 0 when sent and are ignored when received.

Format flags
CIF format dependent flags. The CIF format dependent flags differ depending on the CIF format type.

GFC
Generic Flow Control. The structure and semantics of octets 3-7 in the CIF header are the same as those of an ATM UNI cell header. These octets are collectively known as the "CIF cell header template".

VPI
Virtual Path Identifier.

VCI
Virtual Channel Identifier.

PT
Payload Type.

C
Cell Loss Priority.

HEC
Header Error Check. The sender of a LAN frame always calculates and fills in the HEC field. The receiver may either rely on the LAN CRC to detect errors in the frame (i.e., not validate the received HECs), or it may check the correctness of the HEC.