Digital PDFs

EK-SCSIS-SP-001

December 1989

103 pages

Original

0.4MB

Document:	Small Computer System Interface A Developer's Guide
Order Number:	EK-SCSIS-SP
Revision:	001
Pages:	103
Original Filename:	https://web-docs.gsi.de/~kraemer/COLLECTION/DEC/ds-scsi-developer-guide.pdf

OCR Text

EK–SCSIS–SP–001

Small Computer
System Interface
A Developer’s Guide
Revision/Update Information: Version 1.0

digital equipment corporation
maynard, massachusetts

First Printing, December 1989
The information in this document is subject to change without notice
and should not be construed as a commitment by Digital Equipment
Corporation.
Digital Equipment Corporation assumes no responsibility for any errors
that may appear in this document.
Any software described in this document is furnished under a license
and may be used or copied only in accordance with the terms of such
license. No responsibility is assumed for the use or reliability of software
or equipment that is not supplied by Digital Equipment Corporation or
its affiliated companies.
Restricted Rights: Use, duplication, or disclosure by the U.S.
Government is subject to restrictions as set forth in subparagraph
(c) (1) (ii) of the Rights in Technical Data and Computer Software clause
at DFARS 252.227-7013.
© Digital Equipment Corporation 1989.
All rights reserved.
Printed in U.S.A.
The postpaid Reader’s Comments forms at the end of this document
request your critical evaluation to assist in preparing future
documentation.

The following are trademarks of Digital Equipment Corporation:
DEC
DECserver
DECstation
DECsystem
DECUS
DECwriter
DIBOL
DIGITAL

MicroVAX
PDP
RSTS
RSX
RT
ULTRIX
UNIBUS
VAX

The following are third-party trademarks:

VAX DOCUMENT
VAXELN
VAXserver
VAXstation
VMS

NCR is a registered trademark of NCR Corporation. ANSI is a registered
trademark of the American National Standards Institute, Inc. IEEE is
a registered trademark of the Institute of Electrical and Electronics
Engineers, Inc.

S1260

This document was prepared with VAX DOCUMENT, Version 1.2.

Contents
About This Guide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

vii

1 Physical Characteristics
Physical Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Cable Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.2.1 Single-Ended Cable . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.2.2 Ordering Information for Digital Cables . . . . . . . . . . . . . . . . .
1.3
Connector Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.3.1 50-Pin Nonshielded Device and Cable Connector . . . . . . . . . . .
1.3.2 50-Pin Shielded Device and Cable Connector . . . . . . . . . . . . . .
1.3.3 50-Pin Connector Pin Assignments . . . . . . . . . . . . . . . . . . . . .
1.3.4 68-Pin Connector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.3.5 68-Pin Connector Pin Assignments . . . . . . . . . . . . . . . . . . . . .
1.4
Electrical Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.4.1 Single-Ended Signals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.4.2 Terminator Power . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.4.3 Signal Termination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.4.4 Reserved Lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.5
SCSI Bus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.6
SCSI Bus Signals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.6.1 Signal Values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.6.2 Signal Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.7
SCSI Bus Timing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1.1
1.2

1–2
1–2
1–3
1–4
1–5
1–6
1–8
1–10
1–12
1–15
1–16
1–16
1–17
1–18
1–19
1–19
1–22
1–23
1–23
1–25

iii

2 Logical Characteristics
SCSI Bus Phases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.1.1 Bus Free Phase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.1.2 Arbitration Phase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.1.3 Selection Phase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.1.4 Reselection Phase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.1.5 Information Transfer Phases . . . . . . . . . . . . . . . . . . . . . . . . . .
2.1.6 Command Phase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.1.7 Data Phase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.1.8 Status Phase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.1.9 Message Phase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.1.10 Signal Restrictions Between Phases . . . . . . . . . . . . . . . . . . . . .
2.2
SCSI Bus Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.2.1 Attention Condition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.2.2 Reset Condition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.3
SCSI Bus Phase Sequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.4
SCSI Pointers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.5
SCSI Message System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.5.1 Messages—Format and Codes . . . . . . . . . . . . . . . . . . . . . . . . .
2.5.2 Messages—Implemented by the VMS Operating System . . . . .
2.6
Messages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.6.1 ABORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.6.2 COMMAND COMPLETE . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.6.3 DISCONNECT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.6.4 RESTORE POINTERS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.6.5 SAVE DATA POINTER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2.1

2–2
2–3
2–3
2–3
2–4
2–5
2–6
2–7
2–7
2–7
2–7
2–8
2–8
2–8
2–10
2–14
2–16
2–16
2–19
2–21
2–22
2–22
2–23
2–23
2–23

3 SCSI Commands and Status
Command Implementation Requirements . . . . . . . . . . . . . . . . . . .
3.1.1 Reserved . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.1.2 Operation Code Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.2
Command Descriptor Block . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.2.1 Operation Code . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.2.2 Logical Unit Number . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.2.3 Logical Block Address . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.2.4 Transfer Length . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.2.5 Parameter List Length . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.2.6 Allocation Length . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.2.7 Control Byte . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.3
Status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.4
Command Processing Considerations and Exception Conditions . .
3.4.1 Programmable Operating Definition . . . . . . . . . . . . . . . . . . . .
3.4.2 Incorrect Initiator Connection . . . . . . . . . . . . . . . . . . . . . . . . .

3.1

3–2
3–2
3–2
3–3
3–5
3–5
3–6
3–6
3–7
3–7
3–7
3–8
3–10
3–10
3–10

3.4.3 Selection of an Invalid Logical Unit . . . . . . . . . . . . . . . . . . . . .
3.4.4 Parameter Rounding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.5
Asynchronous Event Notification . . . . . . . . . . . . . . . . . . . . . . . . . .
3.6
Contingent Allegiance Condition . . . . . . . . . . . . . . . . . . . . . . . . . .
3.7
Extended Contingent Allegiance Condition . . . . . . . . . . . . . . . . . .
3.8
Unit Attention Condition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3–10
3–11
3–12
3–14
3–14
3–14

4 Characteristics of Most Device Types
Model for Most Device Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.1.1 SCSI Addresses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.1.2 Commands Implemented by SCSI Targets . . . . . . . . . . . . . . . .
4.2
Commands for Most Device Types . . . . . . . . . . . . . . . . . . . . . . . . .

4.1

4–1
4–1
4–2
4–5

5 Power-On Protocols and Initialization Procedures
Dynamic Reconfiguration of Devices . . . . . . . . . . . . . . . . . . . . . . .
External Boxes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Device Behavior Following Power-On . . . . . . . . . . . . . . . . . . . . . . .
Device Behavior Following Bus Reset . . . . . . . . . . . . . . . . . . . . . .

5.1
5.2
5.3
5.4

5–1
5–1
5–2
5–3

A Associated Documents
B ANSI SCSI-2 Specification Cross-Reference
Glossary
Index
Figures
1–1
1–2
1–3
1–4
1–5
1–6

Digital SCSI Connector Cables . . . . . . . . . . . . . . . . . . . . . . . . . . . .
50-Pin Nonshielded SCSI Device Connector . . . . . . . . . . . . . . . . . . .
50-Pin Nonshielded SCSI Cable Connector . . . . . . . . . . . . . . . . . . .
50-Pin Shielded SCSI Device Connector . . . . . . . . . . . . . . . . . . . . . .
50-Pin Shielded SCSI Cable Connector . . . . . . . . . . . . . . . . . . . . . .
Nonshielded Pin Assignments . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1–4
1–6
1–7
1–8
1–9
1–10

1–7
1–8
1–9
1–10
1–11
1–12
1–13
1–14
1–15
2–1
2–2
2–3
2–4
2–5
3–1
3–2
3–3
3–4
3–5

Shielded Pin Assignments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
68-Pin Connector Assembly . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Configuration/Dimensions for 68-Pin Connector . . . . . . . . . . . . . . . .
Cover and Strain Relief for 68-Pin Connector . . . . . . . . . . . . . . . . .
68-Pin Connector Pin Assignments . . . . . . . . . . . . . . . . . . . . . . . . .
Termination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
SCSI ID Bits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Sample SCSI Configurations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Signal Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Information Transfer Phases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
SCSI Bus Phase Sequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Sequence of Signals in an INQUIRY Command . . . . . . . . . . . . . . . .
SCSI Pointers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Message Codes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
CDB for Six-Byte Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
CDB for Ten-Byte Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Operation Code . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Control Byte . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Status Byte . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1–11
1–12
1–13
1–14
1–15
1–18
1–19
1–21
1–24
2–5
2–11
2–12
2–15
2–17
3–3
3–4
3–5
3–7
3–8

Digital’s Implementation of SCSI . . . . . . . . . . . . . . . . . . . . . . . . . . .
Cable Ordering Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Terminator Ordering Information . . . . . . . . . . . . . . . . . . . . . . . . . .
SCSI ID Assignments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
SCSI Bus Signals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
SCSI Bus Timing Values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
SCSI Bus Phases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Digital Systems and Associated Host Adapter . . . . . . . . . . . . . . . . .
Mnemonics Used in Figure 2–5 . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Extended Message Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Extended Message Codes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
SCSI Messages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Operation Code Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Status Byte Codes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Commands for Most Device Types . . . . . . . . . . . . . . . . . . . . . . . . . .
Cross-Reference to ANSI SCSI-2 Specification . . . . . . . . . . . . . . . . .

viii
1–5
1–16
1–20
1–22
1–26
2–2
2–6
2–18
2–19
2–19
2–21
3–2
3–9
4–5
B–1

Tables
1
1–1
1–2
1–3
1–4
1–5
2–1
2–2
2–3
2–4
2–5
2–6
3–1
3–2
4–1
B–1

About This Guide
Digital encourages third-party vendors to develop complementary
SCSI peripheral devices that are not currently available through
Digital. Confirming its commitment to open systems, Digital
has chosen to provide developers with the technical information
they need to interface peripheral devices to Digital’s SCSI-based
systems.

Purpose of This Guide
This guide defines Digital Equipment Corporation’s
implementation of the American National Standard for
Information Systems—Small Computer System Interface-2
(SCSI-2) specification. It defines the mechanical, electrical, and
functional requirements for interconnecting small computers and
intelligent peripheral devices.
This guide is not intended to replace the SCSI-2 specification,
but rather to specify which options of the SCSI-2 specification
must be implemented, and to impose additional restrictions when
necessary.
Note At this writing, the Small Computer System Interface (SCSI-2)
is under development. The draft SCSI-2 specification (Revision
10b) should be the official guide to what is implemented by a
third-party device.
Table 1 summarizes Digital’s implementation of SCSI.

vii

Table 1

Digital’s Implementation of SCSI

FEATURE

Currently
Not implemented

IMPLEMENTED

Data transfer width

16 and 32 bit-wide
SCSI

8-bit SCSI

Transfer mode

Fast data transfer

1.5 MB/s (async), 4+
MB/s (synch)

Drivers

Differential

Single-ended drivers and
receivers

Termination

Internal termination

External termination of
last peripheral device on
bus

Cable

25 meters long

6 meters long

68-conductor B cable

50-conductor A cable

Initiators

Multiple initiators

Single initiator with
multiple targets

Performance

Command queuing
(tagged commands)
Disconnect and
reconnect
Soft Reset to clear
devices from bus

Hard Reset to clear
devices from bus. Also,
ABORT, BUS DEVICE
RESET.

Dynamic reconfiguration of peripheral
devices on bus
Command set (class
driver)
Operating system
support for SCSI
devices not supplied
by Digital
Operating system
support for SCSI
disk and tape drives
supplied by Digital

Common Command Set
for disks
ULTRIX application
level interface.
VAXELN application
level interface.

VMS Version 5.3
third-party device
development support
ULTRIX Workstations
Software V2.1 for
MicroVAX 3100, DS2100,
DS3100, VS3100, and
VS3520/3540 systems
VMS driver support on
MicroVAX 3100, VS3100,
and VS3520/3540
systems

viii

Digital computer systems support multiple operating systems.
Some operating systems may have a different set of SCSI
requirements and restrictions. Restrictions specific to an
operating system are indicated by a delta symbol as defined
in the Conventions section of this guide.

Who Should Use This Guide
This guide is intended for developers of peripheral devices
who need information about Digital’s SCSI requirements. The
developers should have knowledge of the draft ANSI SCSI-2
specification.
VMS Development Considerations
The VMS operating system provides two mechanisms for the
attachment of third-party devices to Digital systems. Depending
upon the peripheral device to be supported, the requirements
of the application programs that will access the device, and
programming resources available, a third-party vendor may elect
to use either the VMS generic SCSI class driver or to write a
device-specific class driver.
To use the VMS generic SCSI class driver, programmers should
understand how to use the VMS Queue I/O Request ($QIO)
system service interface with a high-level language. To write
a third-party SCSI class device driver, programmers should
have detailed knowledge of the VMS operating system and the
VAX MACRO assembly language. See Small Computer System
Interface: An Overview for a technical summary of the VMS
Version 5.3 SCSI Device Support product, and Appendix A for a
list of associated documents.

Structure of This Guide
This guide contains five chapters and two appendixes.
Chapter 1 contains the physical description of Digital’s
implementation of the Small Computer System Interface (SCSI-2)
specification.
Chapter 2 describes the eight bus phases that define the logical
characteristics of the SCSI bus.
Chapter 3 defines the SCSI command and status structures.
Chapter 4 describes the characteristics expected of most SCSI
devices.
Chapter 5 describes Digital’s requirements for system
initialization of SCSI devices.
Appendix A lists the manuals associated with Digital’s
implementation of the Small Computer System Interface.
Appendix B contains a cross-reference between the ANSI
specification and this guide. See Table B–1.
The Glossary explains technical terms used in the manual.

ANSI Cross-References
ANSI SCSI-2
specification,
Section n.n.n.

Chapters, sections, and paragraphs in this manual are crossreferenced to the American National Standard for Information
Systems—Small Computer System Interface-2 (SCSI-2)
specification (X3T9/89-042), Revision 10b. In the printed version
of the manual, the cross-references are adjacent to the text; in the
online version they appear above the text.
The format and content of this manual are similar to the
referenced specification. However, this guide contains only the
information applicable to Digital’s implementation of SCSI.
Refer to the ANSI SCSI-2 specification for information about
numeric order codes and vendor identification.

Conformance to Standards
The ANSI SCSI-2 specification (revision 10b) should be the official
guide to what features should be implemented on the device and
how to implement them. Digital’s requirements for third-party
SCSI devices are based on the evolving SCSI-2 specification.
SCSI-2 is compatible with the SCSI-1 standard; SCSI-2 clarifies
much of the material in SCSI-1.
Digital has designed its hardware and software so that it adheres
to the draft SCSI-2 specification and yet provides as much
flexibility as possible. To work with Digital systems, third-party
devices must implement all the mandatory features of the SCSI-2
interface. Third-party devices may implement optional features,
as long as the implementation follows the SCSI-2 specification.
Devices may implement vendor-specific features as long as the
features are implemented in areas clearly designated as such by
the SCSI-2 specification.

Development Considerations
Products licensed or sold by Digital Equipment Corporation
have been tested by Digital. Products should operate with
Digital computer systems if they are developed using Digital
SCSI documentation and implemented according to Digital
requirements as specified in this guide. The proper operation
of products that have been developed by third-party vendors
cannot be assured by Digital. Many third-party products require
additional features that are not part of Digital’s software. These
products may not be fully functional with Digital’s software
as licensed and sold. Digital customers who use unsupported
products may not qualify for Digital customer service programs.
Third-party peripheral devices may be attached to Digital
SCSI-based systems; however:
Performance and functionality with the operating system may
not be optimized.
Data integrity and reliability are potentially at risk due to loss
of data, error rate, data corruption, and disabling of the bus.
Quality of peripheral devices may not meet Digital standards.
A product that is not supplied by Digital may appear to work in
a specific configuration; it may not work with future updates of
Digital hardware and software.

Conventions
The following conventions are used in this guide:
boldface

Boldface type represents the introduction of a new term. New terms are
defined in the Glossary.

Caution

Cautions provide information to prevent damage to equipment or
software. Read these carefully.

Note

Notes provide information about the current topic and are of special
interest to vendors of peripherals.

ANSI SCSI-2
specification,
Section n.n.n.

This notation indicates that the section, paragraph, or table is crossreferenced to a numbered section in the draft ANSI Small Computer
System Interface-2 (SCSI-2) specification (X3T9/89-042), Revision 10b.

SCSI

The term SCSI is used wherever it is not necessary to distinguish
between SCSI-1 and SCSI-2.
The delta symbol flags information in Digital’s implementation
of SCSI that differs from the draft SCSI-2 specification. In the printed
version of the manual, the delta symbols are adjacent to the text; in the
online version, they appear above the text.
The delta symbol with the letters VMS below signifies a
restriction, selection, or a change from the draft SCSI-2 specification that
is specific to the VMS operating system.

VMS

UPPERCASE

Uppercase text represents the names of signals, messages, and
commands; for example, INQUIRY.

Numbers that are not immediately followed by lowercase ‘‘b’’ or ‘‘h’’ are
decimal values.

xxb

Numbers immediately followed by lowercase ‘‘b’’ are binary values.

xxh

Numbers immediately followed by lowercase ‘‘h’’ are hexadecimal values.

xii

1
Physical Characteristics
This chapter contains the physical definition of Digital Equipment
Corporation’s implementation of the draft proposed ANSI Small
Computer System Interface-2 (SCSI-2) specification. Cables,
connectors, terminators, signals, and bus timing are described in
the following sections:
Physical description, Section 1.1
Cable requirements, Section 1.2
Connector requirements, Section 1.3
Electrical description, Section 1.4
SCSI bus, Section 1.5
SCSI bus signals, Section 1.6
SCSI bus timing, Section 1.7

Physical Characteristics 1–1

1.1 Physical Description
ANSI SCSI-2
specification,
Section 4.1.

SCSI devices are daisy-chained together using 50-pin connectors
on the A cable. Both ends of the cable must be terminated. All
signals are common between all SCSI devices on the cable.
Single-ended drivers and receivers, with a maximum cable length
of 6 meters, are specified.
Single-ended and differential devices are mutually exclusive
within a system. They are electrically different and must not
be used on the same bus. Digital supports only the single-ended
implementation.
The fast synchronous data transfer option is not supported.
Caution Devices cannot be added to the SCSI bus, removed from the bus,
or re-cabled while equipment is powered on. Failure to meet this
requirement may cause loss of user data or system failure.

1.2 Cable Requirements
ANSI SCSI-2
specification,
Section 4.2.

An ideal impedance match with cable terminators implies a cable
characteristic impedance of 132 ohms (a lower impedance is
satisfactory). A characteristic impedance of 100 ohms ±10 percent
is recommended for all cable types.

2 (28 AWG) is required
A minimum conductor size of 0.08042
for the TERMPWR line. This wire size minimizes the effect of
noise and ensures proper distribution of terminator power.

Note To minimize discontinuities and signal reflections, do not use
cables of different impedances on the same bus. Implementations
may require trade-offs in shielding effectiveness, cable length, the
number of loads, transfer rates, and cost to achieve satisfactory
system operation.
To minimize discontinuities due to local impedance variation,
space a flat cable at least 1.27 mm (0.050 in) from other cables,
any other conductor, or the cable itself when the cable is folded.
Regulatory agencies may require you to use a larger wire size.

1–2 Physical Characteristics

For proper operation of the SCSI bus, use high-quality, properly
configured cables and connectors to connect all devices. It is
recommended that you use only cable assemblies supplied by
Digital intended for interconnecting SCSI devices. This ensures
that the impedance characteristics, signal propagation, velocity,
inductance, capacitance, cross-talk, grounding, conductor pairing,
and shielding meet the requirements for proper operation of the
bus.
In addition, it is recommended that all units on the SCSI bus be
powered from a common ac power source.
The proper operation of any SCSI bus that uses cable assemblies
not supplied by Digital, or not configured in accordance with
Digital’s recommendations, is not guaranteed.
Section 1.2.1 explains the requirements for single-ended cable.
Section 1.2.2 contains ordering information for Digital cables.

1.2.1 Single-Ended Cable
ANSI SCSI-2
specification,
Section 4.2.1.

Use a 50-conductor flat cable or 25-signal twisted-pair cable as the
A cable. Digital recommends a maximum cable length of 4 meters
for the best signal integrity. If you use twisted-pair cable, the
twisted pairs in the cable must be wired to opposing contacts in
the connector. The contact assignments for the shielded connector
differ from those of the nonshielded connector:
For the shielded connector, the order of the twisted pairs is
1/26, 2/27, . . . , 25/50.
For the nonshielded connector, the order of the twisted pairs
is 1/2, 3/4, . . . , 49/50.
See Section 1.3.3 for 50-pin connector pin assignments.
A stub length of no more than 0.1 meters is allowed off the
mainline interconnection within any connected equipment or from
any connected point.
Note Do not cluster stubs. Space stubs at least 0.3 meters apart.
Terminate bus devices externally with a terminator supplied by
Digital.

Physical Characteristics 1–3

1.2.2 Ordering Information for Digital Cables
Figure 1–1 shows the types of SCSI connector cables available
from Digital.
Figure 1–1

Digital SCSI Connector Cables

50-Pin to 50-Pin Straight End
Connector Cable
Order No. BC19J-xx

68-Pin Straight End to 50-Pin
Right Angle Connector Cable
Order No. BC09J-xx

50-Pin to 50-Pin Right Angle
Connector Cable
Order No. BC19K-xx

50-Pin Straight End to 50-Pin
Right Angle Connector Cable
Order No. BC13C-07

68-Pin to 50-Pin Straight End
Connector Cable
Order No. BC56H-xx

Note: See Table 1-1.
xx represents cable length in feet.
MLO-003600

1–4 Physical Characteristics

Table 1–1 lists Digital SCSI cable part numbers.
Table 1–1

Cable Ordering Information
Digital
Part
Number

Customer
Order
Number

Cable
Length
In Meters

Cable
Length
In Feet

VAXstation 3520/3540

17–02238–01

BC13C–07

2.14

7.0

DECstation 3100 and
VAXstation 31001

17–02008–01

BC56H–03

0.91

3.0

17–02008–02

BC56H–06

1.83

6.0

17–02443–03

BC09J–03

0.91

3.0

17–01351–01

BC19J–03

0.91

3.0

17–01351–02

BC19J–06

1.83

6.0

17–01351–04

BC19J–1E

0.46

1.5

17–02446–01

BC19L–01

0.46

1.5

17–02446–02

BC19K–02

0.61

2.0

System

MicroVAX 3100 or
expansion boxes

1 If your workstation is shipped with a 68–50 pin adapter, remove the adapter cover

before connecting the SCSI cable.

1.3 Connector Requirements
ANSI SCSI-2
specification,
Section 4.3.

Nonshielded connectors are typically used for in-cabinet
applications. Shielded connectors are used for external
applications where electromagnetic compatibility (EMC) and
protection from electrostatic discharge (ESD) are required. You
can use either type of connector with single-ended drivers.
In order to support daisy-chain connections, SCSI devices that use
shielded connectors must provide two shielded device enclosures
on the device. You can wire the two connectors one-to-one with a
stub to the SCSI device’s drivers and receivers, as long as you do
not exceed the maximum stub length. Alternatively, you can run
two cables from the two shielded connectors to the drivers and
receivers to ensure that the maximum stub length is not exceeded.
When you calculate the total cable length of the SCSI bus, be sure
to include the length of the cable in the device enclosure.
Nonshielded connectors must be shrouded and keyed in order to
protect users from making cabling errors.
Section 1.3.1 through Section 1.3.5 describe cable and connector
requirements in detail.

Physical Characteristics 1–5

1.3.1 50-Pin Nonshielded Device and Cable
Connector
ANSI SCSI-2
specification,
Section 4.3.1.2.

The nonshielded, low-density, PC-board mounted, SCSI device
connector for the A cable ( Figure 1–2) is a 50-pin connector
consisting of two rows of 25 male pins with adjacent pins 2.54 mm
(0.1 in) apart. A shroud and header body must be used.
Figure 1–2
D10

50-Pin Nonshielded SCSI Device Connector
Top View

D8
D2
D6

Pin 49

Pin 1

Shrouded Header
D1

D3
See Note 1

Pin 50

Pin 2

D5
Bottom View
Dimension

Millimeters

Inches

D1
D2
D3
D4
D5
D6
D7
D8
D9
D10

2.54
82.80
2.54
8.89
72.64
78.74
13.94
_ 0.25
4.19 +
6.09
6.60

0.100
3.260
0.100
0.350
2.860
3.100
0.549
_ 0.01
0.165 +
0.240
0.260

Comments

Reference only
Reference only
Reference only
Reference only
Reference only

Reference only

Notes:
(1) Two rows of 25 contacts on 2.54-mm
(0.100-inch) spacing = 60.96 mm (2.400 inch).
_ 0.127 mm (0.005 inch) noncumulative,
(2) Tolerances +
unless specified otherwise.
MLO-003484

1–6 Physical Characteristics

ANSI SCSI-2
specification,
Section 4.3.1.2.

The nonshielded, low-density cable connector recommended
for attachment to the A cable ( Figure 1–3) is a 50-conductor
connector consisting of two rows of 25 female contacts with
adjacent contacts 2.54 mm (0.1 in) apart. The connector must
be shrouded and keyed.
Figure 1–3

50-Pin Nonshielded SCSI Cable Connector
Pin 50

Back View
Pin 1

C3
C2

Side View

See Note 1

Contacts for even
pins are not shown.

C5
Front View

Pin 1

Pin 49

Pin 2

Pin 50

Dimensions

Millimeters

Inches

Comments

C1
C2
C3
C4
C5
C6
C7
C8

2.540
60.960
2.540
3.302
32.385
68.972
6.096
7.620

0.100
2.400
0.100
0.130
1.275
2.680
0.240
0.300

Maximum

Notes:
(1) Fifty contacts on 1.27 mm (0.05 inch) staggered
spacing = 62.23 mm (2.450 inch) [reference only].
_ 0.127 mm (0.005 inch) noncumulative
(2) Tolerances +
unless specified otherwise.
(3) Connector cover and strain relief are optional.
MLO-003485

Physical Characteristics 1–7

1.3.2 50-Pin Shielded Device and Cable Connector
ANSI SCSI-2
specification,
Section 4.3.2.2.

The shielded low-density device connector for the A cable
( Figure 1–4) is a 50-pin connector consisting of two rows of
ribbon contacts spaced 2.16 mm (0.085 in) apart.
Figure 1–4

50-Pin Shielded SCSI Device Connector
E4
E7

E10

Pin 1
E6

Pin 25

E8
E12

Pin 50

Dimensions
E1
E2 (3)
E3
E4
E5
E6
E7
E8
E9
E10
E11
E12

Pin 26
E11

Millimeters

Inches

56.39 Minimum
2.62 Minimum
3.99 Minimum
5.84 Minimum
74.85
64.29 Maximum
83.06
15.24
12.04 Maximum
9.78 Maximum
2.16
_ 2
15 +

2.220 Minimum
0.103 Minimum
0.157 Minimum
0.230 Minimum
2.947
2.531 Maximum
3.270
0.600
0.474 Maximum
0.385 Maximum
0.085
_ 2
15 +

Notes:
_ 0.127 mm (0.005 inch)
(1) Tolerances +
noncumulative unless specified otherwise.
(2) Dimensions are shown for reference.
(3) Dimension E2 to accommodate
4-40 or 6-32 threaded screws.
MLO-003490

1–8 Physical Characteristics

ANSI SCSI-2
specification,
Section 4.3.2.2.

The shielded low-density cable connector for the A cable
( Figure 1–5) is a 50-conductor connector consisting of two rows of
ribbon contacts spaced 2.16 mm (0.085 in) apart.
Figure 1–5

50-Pin Shielded SCSI Cable Connector
F4
F12

Pin 1
F8

F11

Pin 25

F6
F3

F9
F7
Pin 26

F10

Pin 50
F5

F13

Dimensions

Millimeters

Inches

F1
F2
F3
F4
F5
F6
F7
F8
F9
F10
F11
F12
F13

56.26 Maximum
64.29 Minimum
15.24
7.29 Minimum
1.02
4.09 Maximum
5.08
6.10
12.04 Minimum
68.45
_2
15 +
76.71
2.16

2.215 Maximum
2.531 Minimum
0.600
0.287 Minimum
0.040
0.161 Maximum
0.200
0.240
0.474 Minimum
2.695
_2
15 +
3.020
0.085

Notes:
_ 0.127 mm (0.005 inch)
(1) Tolerances +
noncumulative unless specified otherwise.
(2) Dimensions are shown for reference.
MLO-003491

Physical Characteristics 1–9

1.3.3 50-Pin Connector Pin Assignments
The pin assignments for the single-ended A cable (with
nonshielded low-density connector) are listed in Figure 1–6.
Figure 1–6

Nonshielded Pin Assignments

Signal
Name

Pin
Number

Signal
Name

Ground
Ground
Ground
Ground
Ground
Ground

1
3
5
7
9
11

2
4
6
8
10
12

-DB(0)
-DB(1)
-DB(2)
-DB(3)
-DB(4)
-DB(5)

Ground
Ground
Ground
Ground
Ground
Reserved

13
15
17
19
21
23

14
16
18
20
22
24

-DB(6)
-DB(7)
-DB(P)
Ground
Ground
Reserved

Open
Reserved
Ground
Ground
Ground
Ground

25
27
29
31
33
35

26
28
30
32
34
36

Termpwr
Reserved
Ground
-ATN
Ground
-BSY

Ground
Ground
Ground
Ground
Ground
Ground
Ground

37
39
41
43
45
47
49

38
40
42
44
46
48
50

-ACK
-RST
SMG
-SEL
-S/D
-REQ
-I/O

Notes:
Minus sign next to a signal indicates an
active low.
Reserved indicates one of the following:
An open signal (recommended)
A signal ground in the end device
A signal ground at the bus terminator
MLO-003601

1–10 Physical Characteristics

The pin assignments for the single-ended A cable (with shielded
low-density connector) are listed in Figure 1–7.
Figure 1–7

Shielded Pin Assignments

Signal
Name

Pin
Number

Signal
Name

Ground
Ground
Ground
Ground
Ground
Ground

1
2
3
4
5
6

26
27
28
29
30
31

-DB(0)
-DB(1)
-DB(2)
-DB(3)
-DB(4)
-DB(5)

Ground
Ground
Ground
Ground
Ground
Reserved

7
8
9
10
11
12

32
33
34
35
36
37

-DB(6)
-DB(7)
-DB(P)
Ground
Ground
Reserved

Open
Reserved
Ground
Ground
Ground
Ground

13
14
15
16
17
18

38
39
40
41
42
43

Termpwr
Reserved
Ground
-ATN
Ground
-BSY

Ground
Ground
Ground
Ground
Ground
Ground
Ground

19
20
21
22
23
24
25

44
45
46
47
48
49
50

-ACK
-RST
SMG
-SEL
-S/D
-REQ
-I/O

Physical Characteristics 1–11

1.3.4 68-Pin Connector
The SCSI port on the VAXstation 3100 and DECstation 3100
and 2100 systems contains a 68-pin device connector. In order to
attach expansion boxes to the system unit SCSI port, a 68-pin to
50-pin connector cable is required.
The 68-pin to 50-pin SCSI cable assembly ( Figure 1–8) consists
of:
A 68-position (2 x 34) high density connector at one end
A 38-conductor 19 twisted-pair (#28 AWG) foil with braid
shielded cable
A 50-position shielded low-density SCSI cable connector at the
other end
Figure 1–5 illustrates the 50-position shielded cable connector.
Figure 1–8 through Figure 1–10 illustrate the 68-position cable
connector assembly. The 68-pin connector is a micro D, female,
IDC type for flat cable, with metal shell.
Figure 1–8

68-Pin Connector Assembly
38-Conductor,19 TwistedPair, #28 AWG, Foil with
Braid Shielded Cable

Die Cast Metal
Shield Cover
Cable Cover Keyed with
"1" and "A" Filled
Pin 34

Pin 1
68-Position (2x34)
High Density Connector

Pin 68

Pin 35
MLO-003588

1–12 Physical Characteristics

The shielded 68-pin connector ( Figure 1–9) for the 68-pin to 50pin adapter cable consists of two rows of contacts spaced 5.51 mm
(.217 in) apart.
Figure 1–9

Configuration/Dimensions for 68-Pin Connector
IDC Cover
for Flat Cable
G9

Position 34

Position 1

33 Spaces
@ 0.050 = 1.650

G10

G8
G7

G11
G6
Position 68

Position 35

G4
Shell
G3
G1

Housing
IDC Termination Area

G2
Dimension

Millimeters

Inches

G1
G2
G3
G4
G5
G6
G7
G8
G9
G10
G11

1.27
41.91
13.21
5.00
46.07
_ 30’
15 +
7.29
2.54
52.09
1.27
5.51

0.050 Typ
1.650
0.52
0.197
1.814
_ 30’
15 +
0.287
0.100
2.051
0.050 Typ
0.217

Notes:
_ 0.397 mm (+
_ 0.016 inch) unless
(1) Tolerances +
specified otherwise.
(2) Dimensions are shown for reference.
MLO-003587

Physical Characteristics 1–13

The cover and strain relief for the 68-pin connector are shown in
Figure 1–10.
Figure 1–10

Cover and Strain Relief for 68-Pin Connector
H11

H10
H9
H7

Cable Clamp

M2.5
Pan Head
Screw
H6

Lock Spring
A

E
H4

M2.5
Flat Head
Screw

Cover
Half 1
Cover
Half 2

H3
A

H1
Key Rivet

Dimensions

Millimeters

Inches

H1
H2
H3
H4
H5
H6
H7
H8
H9
H10
H11

7.29
4.06
60.30
8.00
2.01
32.00
50.59
10.01
11.61
9.98
13.00

0.287
0.160
2.374
0.315
0.079
1.260
1.992
0.394
0.457
0.393
0.512

Notes:
_ 0.397 mm ( +
_ 0.016 inch)
(1) Tolerances +
unless specified otherwise.
(2) Dimensions are shown for reference.
MLO-003595

1–14 Physical Characteristics

1.3.5 68-Pin Connector Pin Assignments
The pin assignments for the 68-pin connector are listed in
Figure 1–11.
Figure 1–11
Signal
Name
-DB(0)
Ground
-DB(1)
Ground
-DB(2)
Ground
-DB(3)
Ground
-DB(4)
Ground
-DB(5)
Ground
-DB(6)
Ground
-DB(7)
Ground
-DB(P)
-ATN
-BSY
-ACK
Ground
-RST
-MSG
-SEL
-C/D
-REQ
-I/O
Termpwr2
Ground

68-Pin Connector Pin Assignments
Pin
Number

Pin
Number

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34

35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68

Signal
Name
Ground

Ground
Ground

Ground

Note:
Minus sign next to a signal indicates an
active low.
MLO-003608

Physical Characteristics 1–15

1.4 Electrical Description
ANSI SCSI-2
specification,
Section 4.4.

This section contains a description of Digital’s implementation of
the SCSI electrical specifications, as follows:
Single-ended signals, Section 1.4.1
Terminator Power, Section 1.4.2
Signal Termination, Section 1.4.3
Reserved Lines, Section 1.4.4
SCSI bus termination is assumed to be external to the SCSI
device for all measurements described in Section 1.4.1 and
Section 1.4.2.

1.4.1 Single-Ended Signals
ANSI SCSI-2
specification,
Section 4.4.1.

It is recommended that Digital terminating connectors be used.
See Table 1–2 for ordering information.
Table 1–2

Terminator Ordering Information
Part
Number

Description

VAXstation 3520/3540
Expansion boxes
MicroVAX 3100

12–30552–01

50-pin IEEE terminator

VAXstation 3100
DECstation 2100/3100

12–29635–01

68-pin Micro D-sub
terminator

System

All signals use open-collector or three-state drivers.
SCSI devices (with power off) must have the low-level electrical
characteristics and high-level electrical characteristics on each
signal described in the ANSI specification. The recommended
nominal switching threshold is approximately 1.4 volts.
Refer to the ANSI SCSI-2 specification for input and output
characteristics.

1–16 Physical Characteristics

1.4.2 Terminator Power
ANSI SCSI-2
specification,
Section 4.4.3.

Each SCSI device should supply terminator power to a TERMPWR
pin that is either fused or protected from a short circuit. This
power is supplied through a diode or similar semiconductor that
prevents backflow of power to the SCSI device. A 1.5 ampere
current limiter is required.
The terminating connector, regardless of its location, must be
powered from the TERMPWR pin. The use of keyed connectors is
required in SCSI devices that provide terminator power, in order
to prevent accidental grounding or misconnection of terminator
power. Digital requires that all devices on the SCSI cable be
powered on at all times.
SCSI devices must sink no more than 1 milliampere from
TERMPWR except to power an optional internal terminator.
Single-ended SCSI devices providing terminator power on cable A
have the following characteristics:
VTerm = 4.25 volts dc to 5.25 volts dc, 900 milliamperes
minimum source drive capability
Note It is recommended that the terminator power lines be decoupled to
improve signal quality.

Physical Characteristics 1–17

1.4.3 Signal Termination
Figure 1–12 shows one termination scheme for single-ended buses.
Note Signal termination for SCSI devices is changing. Refer to
the ANSI SCSI-2 specification for information on alternative
termination for single-ended buses.
Figure 1–12

Termination

220
SIGNAL

330
GROUND

MLO-003486

Near-end termination is provided within the system enclosure.
The far-end termination for the external bus is provided by a
terminator plugged into the last SCSI device connector. There
must be no other terminators on the bus. The bus must be
terminated externally. If you follow these rules for all devices, you
will be able to configure the bus with proper termination.
Caution Failure to terminate the bus properly may cause a system crash or
data corruption.

1–18 Physical Characteristics

1.4.4 Reserved Lines
ANSI SCSI-2
specification,
Section 4.4.4.

Lines labeled reserved in the pin assignments shown in
Figure 1–6 and Figure 1–7 must be connected to ground in
the bus terminator assemblies. Reserved lines may not be driven
by SCSI devices; they should be connected to ground.

1.5 SCSI Bus
ANSI SCSI-2
specification,
Section 4.5.

Communication on the SCSI bus can only occur between two SCSI
devices at any given time. The SCSI bus can have a maximum
of eight SCSI devices; the devices can be any combination of
initiator and target. Each SCSI device has a SCSI ID bit
assigned, as shown in Figure 1–13.
Figure 1–13

SCSI ID Bits

DB(7) DB(6) DB(5) DB(4) DB(3) DB(2) DB(1) DB(0)

SCSI ID 0
SCSI ID 1
SCSI ID 2
SCSI ID 3
SCSI ID 4
SCSI ID 5
SCSI ID 6 (VAXstation 3100, MicroVAX 3100)
SCSI ID 7 (VAXstation 3520/3540)
Note: DB = data bus

MLO-003578

Physical Characteristics 1–19

The SCSI ID is a value in the range 7–0. The SCSI ID determines
the logical position of the device controller on the SCSI bus.
You determine the SCSI ID you wish to use before configuring a
SCSI device in a system. Each device must have user-accessible
switches or jumpers so that the SCSI ID can be changed easily.
SCSI ID assignments vary from system to system. A third-party
device may use any SCSI ID that does not conflict with another
SCSI ID on the same bus. Suggested SCSI ID assignments are
shown in Table 1–3.
Table 1–3

SCSI ID Assignments

SCSI ID

Assignment

VAX CPU on the VAXstation 3520/3540

VAX CPU on the VAXstation 3100 and MicroVAX 3100

Tape drive

4, 3, 2

Disks

1, 0

Open

Note The device logical unit number (LUN) identifies the different
units attached to a single SCSI controller.
When two SCSI devices communicate on the SCSI bus, one acts as
an initiator and the other acts as a target. The initiator originates
an operation and the target performs the operation. A SCSI device
usually has a fixed role as an initiator or target, but some devices
may be able to assume either role.
An initiator may address up to seven peripheral devices
connected to a target. If the target is physically housed within
the peripheral device, then the device is referred to as an
embedded SCSI device. Sample system configurations are shown
in Figure 1–14.

1–20 Physical Characteristics

SCSI Bus

Terminator

Computer

Sample SCSI Configurations
Host
Adapter

Figure 1–14

Controller

SCSI Bus

Controller

Terminator

Computer

Host
Adapter

Single Initiator, Single Target

Controller

Single Initiator, Multiple Target

Peripheral devices such as magnetic disks, printers,
scanners, optical disks, and magnetic tapes.
MLO-003487

Certain SCSI bus functions are assigned to the initiator and
certain SCSI bus functions are assigned to the target. The
initiator may arbitrate for the SCSI bus and select a particular
target. The target may request the transfer of command, data,
status, or other information on the data bus, and in some cases
it may arbitrate for the SCSI bus and reselect an initiator for the
purpose of continuing an operation.
Information transfers on the data bus are asynchronous and follow
a defined REQ/ACK (Request/Acknowledge) handshake protocol.
One byte of information can be transferred with each handshake
on the A cable. The SCSI-2 specification defines an option for
synchronous data transfer.

Physical Characteristics 1–21

1.6 SCSI Bus Signals
ANSI SCSI-2
specification,
Section 4.6.

Eighteen signals are used on the A cable: nine signals are used for
control and nine signals are used for data (messages, commands,
status, and data), including parity. Table 1–4 describes the
signals. Section 1.6.1 describes signal values, and Section 1.6.2
describes signal sources.
Table 1–4

SCSI Bus Signals

Signal

Description

BSY (BUSY)

An OR-tied signal that indicates that the bus
is being used.

SEL (SELECT)1

An OR-tied signal used by an initiator to
select a target or by a target to reselect an
initiator.

C/D (CONTROL/DATA)

A signal driven by a target that indicates
whether control or data information is on the
data bus. True2 indicates control.

I/O (INPUT/OUTPUT)

A signal driven by a target that controls the
direction of data movement on the data bus
with respect to an initiator. True indicates
input to the initiator. This signal is also
used to distinguish between Selection and
Reselection phases.

MSG (MESSAGE)

A signal driven by a target during the
Message phase.

REQ (REQUEST)

A signal driven by a target on A cable to
indicate a request for a REQ/ACK data
transfer handshake.

ACK (ACKNOWLEDGE)

A signal driven by an initiator on A cable to
indicate an acknowledgment for a REQ/ACK
data transfer handshake.

ATN (ATTENTION)

A signal driven by an initiator to indicate the
Attention condition.

1
Digital’s SCSI System Interface Interconnect (SII) chip does not have the
SEL line OR-tied. The SII chip does not directly drive the SCSI bus. The SII chip
together with the NCR 8310 SCSI transceiver chip combine the SII’s SELIN and
SELOUT lines into the SCSI SEL line.
2

Signal assertion is the process of driving the signal to the true state (low),
0–.5V. Signal negation is the process of driving the signal to the false state (high),
2.5–5.5V.

(continued on next page)

1–22 Physical Characteristics

Table 1–4 (Cont.)

SCSI Bus Signals

Signal

Description

RST (RESET)

An OR-tied signal that indicates the reset
condition. Only the Digital host adapter can
assert reset.

DB(7–0,P) (DATA BUS)

Eight data-bit signals, plus a parity-bit signal,
that form a data bus. DB(7) is the most
significant bit and has the highest priority
during the Arbitration phase. Bit number,
significance, and priority decrease downward
to DB(0).

A data bit is defined as one when the signal value is true and
is defined as zero when the signal value is false. A device must
generate odd parity during all information transfer phases in
which the device writes data to the SCSI bus. The device must
check for odd parity during all Information Transfer phases in
which it reads data from the bus. Parity is undefined during the
Arbitration phase.

1.6.1 Signal Values
ANSI SCSI-2
specification,
Section 4.6.1.

Signals can assume true or false values during information
transfer. When a value is true, the signal is driven true, or
asserted. When a value is false, the non-OR-tied signals are
driven false, or negated, and OR-tied signals are released.
In this guide, whenever the term negated is used, it means that
the signal is actively driven false, or simply released at the option
of the implementor.
The advantage of actively driving signals false during information
transfer is that the transition from true to false occurs more
quickly and the noise margin is much higher than if the signal
is simply released; this facilitates reliable data transfer at high
rates.
See the ANSI SCSI-2 specification for additional information.

1.6.2 Signal Sources
ANSI SCSI-2
specification,
Section 4.6.3.

Figure 1–15 indicates which type of SCSI device (target or
initiator) is allowed to source each signal. The figure does not
show whether the source is driving asserted, driving negated,
or passive. All SCSI device drivers that are not active sources
must be in the passive state. The RST signal may be asserted by
any SCSI device at any time.

Physical Characteristics 1–23

Figure 1–15

Signal Sources
A Cable Signals

Bus Phase

BSY

SEL

C/D,
I/O,
MSG,
REQ

Bus Free
Arbitration
Selection
Reselection
Command
Data in
Data out
Status
Message in
Message out

None
All
I&T
I&T
Targ
Targ
Targ
Targ
Targ
Targ

None
Win
Init
Targ
None
None
None
None
None
None

None
None
None
Targ
Targ
Targ
Targ
Targ
Targ
Targ

ACK,
ATN

DB(7-0)
DB(P)

None
None
Init
Init
Init
Init
Init
Init
Init
Init

None
S ID
Init
Targ
Init
Targ
Init
Targ
Targ
Init
MLO-003599

The following table describes the signal sources shown in
Figure 1–15.
Source

Description

All

The signal must be driven by all SCSI devices that are actively
arbitrating.

S ID

A unique data bit (the SCSI ID) must be driven by each SCSI
device that is actively arbitrating; the other seven data bits
must be released (that is, not driven) by this SCSI device. The
parity bit, DB(P), may be undriven or driven to the true state,
but must never be driven to the false state during this phase.

I&T

The signal must be driven by the initiator, target, or both, as
specified in the Selection and Reselection phases.

Init

If driven, this signal must be driven only by the active initiator.

None

The signal must be released; that is, not be driven by any SCSI
device. The bias circuitry of the bus terminators pulls the signal
to the false state.

Win

The signal must be driven by the one SCSI device that wins
arbitration.

Targ

If the signal is driven, it must be driven only by the active
target.

1–24 Physical Characteristics

1.7 SCSI Bus Timing
ANSI SCSI-2
specification,
Section 4.7
through 4.7.16.

VMS

Unless otherwise indicated, you must calculate the delay-time
measurement for each SCSI device from the signal conditions
that exist at that SCSI device’s SCSI bus connection. You can
make these measurements (except Cable Skew delay) without
considering delays in the cable. Table 1–5 describes the timing
characteristics of each signal.
The device should not hold the bus ‘‘busy’’ in a single state before
disconnecting for longer than the following periods of time:
500 µs is recommended for optimum system performance.
3 ms is required for performance reasons.

Physical Characteristics 1–25

Table 1–5

SCSI Bus Timing Values

Signal
Condition

Delay
Time

Arbitration
delay

2.4 µs

The minimum time a SCSI device waits
(from asserting BSY for arbitration to
examining the data bus) to see if arbitration
has been won. There is no maximum time.

Assertion
period

90 ns

The minimum time that a target must
assert REQ while using synchronous data
transfers. Also, the minimum time that
an initiator must assert ACK while using
synchronous data transfers.

Bus Clear
delay

800 ns

The maximum time for a SCSI device to stop
driving all bus signals after:

Explanation

The Bus Free phase is detected (the
BSY and SEL signals are both false for
a Bus Settle delay).
The SEL signal is received from another
SCSI device during the Arbitration
phase.
The transition of RST to true.
The maximum time for a SCSI device to
clear the bus is 1200 nanoseconds from the
time BSY and SEL first both become false.
If a SCSI device requires more than a Bus
Settle delay to detect the Bus Free phase, it
must clear the bus within a Bus Clear delay
minus the excess time.
Bus Free
delay

800 ns

The minimum time that a SCSI device must
wait from its detection of the Bus Free phase
(BSY and SEL both become false for a Bus
Settle delay) until its assertion of BSY when
going to the Arbitration phase.

Bus Set
delay

1.8 µs

The maximum time for a SCSI device to
assert BSY and its SCSI ID bit on the data
bus after it detects the Bus Free phase
(BSY and SEL both become false for a Bus
Settle delay) for the purpose of entering the
Arbitration phase.
(continued on next page)

1–26 Physical Characteristics

Table 1–5 (Cont.)

SCSI Bus Timing Values

Signal
Condition

Delay
Time

Bus Settle
delay

400 ns

The minimum time to wait for the bus to
settle after changing certain control signals
as called out in the protocol definitions.

Cable Skew
delay

10 ns

The maximum difference in propagation
time allowed between any two SCSI bus
signals measured between any two SCSI
devices.

Data Release
delay

400 ns

The maximum time for an initiator to
release the DATA BUS signals following
the transition of the I/O signal from false to
true.

Deskew
delay

45 ns

The minimum time required for deskew of
certain signals.

Disconnection
delay

200 µs

The minimum time that a target must wait
after releasing BSY before participating
in an Arbitration phase when honoring a
DISCONNECT message from the initiator.

Hold time

45 ns

The minimum time added between the
assertion of REQ and the changing of
the data lines to provide hold time in the
initiator or target during synchronous data
transfers.

Negation
period

90 ns

The minimum time that a target must
negate REQ during synchronous data
transfers. Also, the minimum time that
an initiator must negate ACK during
synchronous data transfers.

Power-on to
Selection
time

15 sec
recommended

The maximum time from power application
until a SCSI target is able to respond
with appropriate status and sense data
to the TEST UNIT READY, INQUIRY, and
REQUEST SENSE commands.

Reset to
Selection
time

250 ms
recommended

The maximum time after a hard RESET
condition until a SCSI target is able to
respond with appropriate status and sense
data to the TEST UNIT READY, INQUIRY,
and REQUEST SENSE commands.

Reset Hold
time

25 µs

The minimum time for which RST is
asserted. There is no maximum time.

Explanation

(continued on next page)

Physical Characteristics 1–27

Table 1–5 (Cont.)

SCSI Bus Timing Values

Signal
Condition

Delay
Time

Selection
Abort
time

200 µs

The maximum time that a target (or
initiator) may take from its most recent
detection of being selected (or reselected),
until asserting a BSY response. This timeout is required to ensure that a target
(or initiator) does not assert BSY after a
Selection (or Reselection) phase has been
aborted. This is not the selection time-out
period. See Section 2.1.3 and Section 2.1.4
for additional information.

Selection
Time-out
delay

250 ms
minimum

The minimum time that an initiator (or
target) must wait for a BSY response during
the Selection (or Reselection) phase before
starting the time-out procedure. Consult the
peripheral device’s specifications for actual
timing requirements.

Transfer
period

Set
during
an SDTR
message

The minimum time allowed between
the leading edges of successive REQ
pulses and of successive ACK pulses
during synchronous data transfers. (See
Section 2.1.5 for additional information.)

1–28 Physical Characteristics

Explanation

2
Logical Characteristics
This chapter defines the logical characteristics of the SCSI bus in
the following sections:
SCSI bus phases, Section 2.1
SCSI bus conditions, Section 2.2
SCSI bus phase sequences, Section 2.3
SCSI pointers, Section 2.4
Message system, Section 2.5
Messages, Section 2.6

Logical Characteristics 2–1

2.1 SCSI Bus Phases
ANSI SCSI-2
specification,
Section 5.1.

The SCSI architecture provides eight bus phases, as shown in
Table 2–1:
Table 2–1

SCSI Bus Phases

Bus Phase

Description
Bus control phases:

Bus Free

Indicates that no SCSI device is actively using the bus

Arbitration

Allows one SCSI device to gain control of the bus

Selection

Allows an initiator to select a target

Reselection

Allows the target to reconnect to an initiator
Information transfer phases:

Command

Allows the target to request command information
from the initiator

Data

Provides data transfer between the initiator and the
target

Status

Provides status information from the target to the
initiator

Message

Allows the transfer of messages between the initiator
and the target

The SCSI bus is in only one phase at a time. Unless otherwise
noted in the following sections, signals that are not mentioned are
not asserted.
Bus Free phase, Section 2.1.1
Arbitration phase, Section 2.1.2
Selection phase, Section 2.1.3
Reselection phase, Section 2.1.4
Information transfer phase, Section 2.1.5
Command phase, Section 2.1.6
Data phase, Section 2.1.7
Status phase, Section 2.1.8
Message phase, Section 2.1.9

2–2 Logical Characteristics

The target must always wait for the initiator to act. The
device should proceed quickly through all phases of a SCSI
command, and should not hold the bus for long periods without
disconnecting.

2.1.1 Bus Free Phase
ANSI SCSI-2
specification,
Section 5.1.1.

The Bus Free phase indicates that a SCSI device is not actively
using the SCSI bus and whether or not it is available for users.
A target may revert to the Bus Free phase to indicate an error
condition that it has no other way to handle. This is called an
unexpected disconnect.
SCSI devices detect the Bus Free phase after the SEL and BSY
signals are both false for at least a bus settle delay.
SCSI devices release all SCSI bus signals within a bus clear delay
after BSY and SEL become continuously false for a bus settle
delay. If a SCSI device requires more than a bus settle delay
to detect the Bus Free phase, then it must release all SCSI bus
signals within a bus clear delay minus the excess time to detect
the Bus Free phase. The total time to clear the SCSI bus must
not exceed a bus settle delay plus a bus clear delay.
The target must not drive any SCSI bus lines after the deassertion
of BSY, except during selection or reselection, as outlined in the
SCSI-2 specification.

2.1.2 Arbitration Phase
ANSI SCSI-2
specification,
Section 5.1.2.

The Arbitration phase allows one SCSI device to gain control of
the SCSI bus so that it can initiate or resume an I/O process.
Implementation of the Arbitration phase is a requirement of the
SCSI-2 specification.

2.1.3 Selection Phase
ANSI SCSI-2
specification,
Section 5.1.3.

The Selection phase allows an initiator to select a target for the
purpose of initiating some target function, for example, a READ
or WRITE command.
Note During the Selection phase the I/O signal is negated so that this
phase can be distinguished from the Reselection phase.
The SCSI device that won the arbitration has both BSY and SEL
asserted and has delayed at least a bus clear delay plus a bus
settle delay before ending the Arbitration phase. The SCSI device
that won the arbitration becomes an initiator by not asserting the
I/O signal.

Logical Characteristics 2–3

2.1.4 Reselection Phase
ANSI SCSI-2
specification,
Section 5.1.4
through 5.1.4.2.

Reselection is an optional phase that allows a target to reconnect
to an initiator for the purpose of continuing some operation that
was previously started by the initiator but was suspended by the
target; that is, the target disconnected by allowing a Bus Free
phase to occur before the operation was complete.
Upon completing the Arbitration phase, the winning SCSI device
asserts both the BSY and SEL signals. The device delays at
least a bus clear delay plus a bus settle delay before entering the
Reselection phase. The winning SCSI device becomes a target by
asserting the I/O signal.
When the host responds to a target reselection by asserting
the BSY signal, the target deasserts the SEL signal within 500
microseconds of the BSY signal assertion. The target also
asserts REQ for the first command or message byte within 500
microseconds of the BSY assertion.
When a device is disconnected from the bus, it must respond
to a selection with ATN by entering the Message Out phase and
accepting the message from the CPU. The device must be prepared
to properly handle either the ABORT or the BUS DEVICE RESET
message at this time.
If the initiator fails to respond to reselection, the device must time
out the reselection attempt after 250 milliseconds as described in
Section 5.1.4.2 of the SCSI-2 specification: Reselection Time-out
Procedure. The device must employ the second option described in
this section of the specification to release the bus; the device may
not assert the RST signal.
After a Reselection time-out, the target should retry the host at
least four times, but not more than ten times. If the device retries
more than ten times, it can unnecessarily tie up the bus in an
error condition. In order to prevent the retries from timing out
other devices, the device should delay at least 20 microseconds
between retry attempts.

2–4 Logical Characteristics

2.1.5 Information Transfer Phases
ANSI SCSI-2
specification,
Section 5.1.5.

The Command, Data, Status, and Message phases are grouped
together as the Information Transfer phases because they are
used to transfer data or control information on the data bus.
The C/D, I/O, and MSG signals are used to distinguish between
the different Information Transfer phases (see Figure 2–1). The
target drives these three signals and therefore controls all changes
from one phase to another. The initiator requests a Message Out
phase by asserting ATN; the target causes the Bus Free phase by
releasing all SCSI signals.
Figure 2–1

Information Transfer Phases

MSG

Signal
C/D

I/O

Phase Name

Direction of Transfer

Comment

0
0
0
0
1
1
1
1

0
0
1
1
0
0
1
1

0
1
0
1
0
1
0
1

Data Out
Data In
Command
Status
*
*
Message Out
Message In

Initiator to Target
Initiator from Target
Initiator to Target
Initiator from Target

Data Phase
Data Phase

Initiator to Target
Initiator from Target

Message Phase
Message Phase

Key:
0 = False
1 = True
* = Reserved for future standardization.
MLO-003598

Logical Characteristics 2–5

Table 2–2 shows the modes of data transfer that are supported on
Digital systems:
Table 2–2

Digital Systems and Associated Host Adapter
Host
Adapter

Maximum
Asynchronous
Transfer Rate1

Maximum
Synchronous
Transfer Rate1

VAXstation 3100
and MicroVAX 3100

NCR 5380

1.5 MB/s

Not Implemented

DECstation 2100/3100

DEC SII

2 MB/s

4+ MB/s

VAXstation 3520/3540

DEC SII

2 MB/s

4+ MB/s

System

Optimized rates are shown.

Note Maximum asynchronous transfer rate is a function of cable
length, number of devices, and device propagation delay. The fast
synchronous transfer option is not supported on Digital systems.
Asynchronous Information Transfer
ANSI SCSI-2
specification,
Section 5.1.5.1.

The target controls the direction of information transfer by means
of the I/O signal. When the I/O signal is true, information is
transferred from the target to the initiator. When I/O is false,
information is transferred from the initiator to the target.
Synchronous Data Transfer

ANSI SCSI-2
specification,
Section 5.1.5.2.

Synchronous data transfer is optional. It is used only in Data
phases. It must be used in a Data phase if a synchronous data
transfer agreement has been established. The agreement specifies
the REQ/ACK offset and the minimum transfer period.

2.1.6 Command Phase
ANSI SCSI-2
specification,
Section 5.1.6.

VMS

The Command phase allows the target to request command
information from the initiator.
The target asserts the C/D signal and negates the I/O and MSG
signals during the REQ/ACK handshakes in this phase.
The device can enter the Command phase only once per command;
for example, linked commands cannot be issued to the device.

2–6 Logical Characteristics

2.1.7 Data Phase
ANSI SCSI-2
specification,
Sections 5.1.7
through 5.1.7.2.

The Data phase is a term that encompasses both the Data In
phase and the Data Out phase.
Data In phase
The Data In phase allows the target to request that data be
sent to the initiator from the target.
The target must assert the I/O signal and negate the C/D and
MSG signals during the REQ/ACK handshakes of this phase.
Data Out phase
The Data Out phase allows the target to request that data be
sent from the initiator to the target.
The target must negate the C/D, I/O, and MSG signals during
the REQ/ACK handshakes of this phase.

2.1.8 Status Phase
ANSI SCSI-2
specification,
Section 5.1.8.

VMS

The Status phase allows the target to request that status
information be sent from the target to the initiator.
The target must assert C/D and I/O and negate the MSG signal
during the REQ/ACK handshake of this phase.
A device may enter the Status phase only once per command. The
one exception is during an error condition when the device goes
immediately to the Bus Free phase. Storage devices should not
enter the Status phase following a WRITE command until the
data is written to nonvolatile media, in order to avoid accidental
loss of user data.

2.1.9 Message Phase
ANSI SCSI-2
specification,
Section 5.1.9.

The Message phase is a term that references either a Message In
or a Message Out phase. Multiple messages may be sent during
either phase. The first byte transferred in either phase must be a
single-byte message or the first byte of a multiple-byte message.
Multiple-byte messages must be contained wholly within a single
Message phase.

2.1.10 Signal Restrictions Between Phases
ANSI SCSI-2
specification,
Section 5.1.10.

See the draft SCSI-2 specification for restrictions that apply to
SCSI bus signals, when the SCSI bus is between two Information
Transfer phases.

Logical Characteristics 2–7

2.2 SCSI Bus Conditions
ANSI SCSI-2
specification,
Section 5.2.

The SCSI bus has two asynchronous conditions:
The Attention condition, Section 2.2.1
The Reset condition, Section 2.2.2
These conditions cause the SCSI device to perform certain actions
and alter the phase sequence.
Each SCSI device, as it is powered on, must perform appropriate
internal reset operations and internal test operations. It is
recommended that within 15 seconds of Power-on to Selection
time, SCSI targets be able to respond with appropriate status and
sense data to the TEST UNIT READY, INQUIRY, and REQUEST
SENSE commands.

2.2.1 Attention Condition
ANSI SCSI-2
specification,
Section 5.2.1.

VMS

The Attention (ATN) condition allows an initiator to inform a
target that the initiator has a message ready. The target may get
this message by performing a Message Out phase.
The device must respond to an ATN condition at every phase
transition, as long as the VAX CPU running the VMS operating
system sets the ATN signal before it asserts the ACK signal for
the last byte of the previous phase. During the Data phase, the
device should also respond to an ATN condition at least once per
millisecond.

2.2.2 Reset Condition
ANSI SCSI-2
specification,
Sections 5.2.2
and 5.2.2.1.

The Reset (RST) condition is used to immediately clear all SCSI
devices from the bus. This condition takes precedence over all
other phases and conditions. During the Reset condition, the state
of all SCSI bus signals other than RST is not defined.

2–8 Logical Characteristics

All SCSI devices must release all SCSI bus signals (except RST)
within a bus clear delay of the transition of RST to true. The Bus
Free phase always follows the Reset condition.
SCSI devices must implement a hard Reset as defined in the
draft ANSI SCSI-2 specification. The device, upon detection of the
Reset condition, performs the following:
Clears all I/O processes including queued I/O processes.
Releases all SCSI device reservations.
Returns any SCSI device operating modes to their appropriate
initial conditions, similar to those conditions that would be
found after a normal Power-on Reset. Mode Select conditions
must be restored to their last saved values if saved values
have been established. Mode Select conditions for which no
values have been saved must be returned to their default
values.
Sets the Unit Attention condition (see Section 3.8).
No peripheral device on the SCSI bus can assert Reset. Only the
Digital host adapter can assert a Reset when appropriate.
If a data storage device receives a hard Reset condition, it must
be able to complete writing the current block of data and not leave
the medium in a corrupted state before responding to the hard
Reset.
It is recommended that following a Reset to Selection Time after
a hard Reset condition ends, SCSI targets be able to respond with
appropriate status and sense data to the TEST UNIT READY,
INQUIRY, and REQUEST SENSE commands.

Logical Characteristics 2–9

2.3 SCSI Bus Phase Sequences
ANSI SCSI-2
specification,
Section 5.3.

The order in which phases are used on the SCSI bus follows a
prescribed sequence.
The Reset condition can abort any phase and is always followed
by the Bus Free phase. Also, any other phase can be followed
by the Bus Free phase, but many such instances indicate error
conditions (see Section 2.1.1).
All devices must conform to the bus state transition sequences
shown in Figure 2–2. The normal progression is as follows:
From the Bus Free phase to Arbitration phase
From Arbitration phase to Selection phase or Reselection
phase
From Selection phase or Reselection phase to one or more of
the Information Transfer phases (Command, Data, Status, or
Message)
The final information transfer phase is normally the Message In
phase in which a DISCONNECT and a COMMAND COMPLETE
message is transferred, followed by the Bus Free phase.

2–10 Logical Characteristics

Figure 2–2

SCSI Bus Phase Sequences

Reset or
Protocol
Error

Bus Free

Message Out

Selection

Command

Arbitration

Data In or
Data Out

Reselection

Status

Message In

MLO-003596

During the bus phases, devices must first contend for access to the
bus. Then a physical path is established between the initiator and
target. One way of viewing bus phase operation is by looking at a
signal sequence diagram ( Figure 2–3) for the INQUIRY command.
Note that in this example the target does not disconnect from the
SCSI bus before the I/O process has completed.

Logical Characteristics 2–11

Figure 2–3

Bus Free Delay
Bus Set Delay
Bus Clear Delay
Arbitration Delay
Bus Clear Delay plus Bus Settle Delay

Initiator
asserts BSY

Target
asserts BSY

Sequence of Signals in an INQUIRY Command

2–12 Logical Characteristics

Bus Settle
Delay

+ BSY
+ SEL
+ C/D

+ I/O
+ MSG
+ REQ
+ ACK
+ ATN
+ RST
DB(7-0,P)
(DATA BUS)

BUS
FREE

Arb.
Id’s

Targ. &
Init. ID’s

ARBI- SELECTION
TRATION

IDENTIFY

MESSAGE OUT

1st CMD
Byte

Last CMD
Byte
COMMAND

Data Bytes

DATA IN

Status Byte

COMMAND
COMPLETE

STATUS

MESSAGE IN
BUS FREE
MLO-003492

Compare the bus phases for the INQUIRY command ( Figure 2–3)
with the following information:
Bus Free Phase
The Bus Free phase begins when the SEL and BSY signals
are false for one Bus Settle delay.
Arbitration Phase
After the initiator sees that the bus is free, the BSY signal is
asserted, the initiator arbitrates for the bus, and its assigned
SCSI device ID is put on the data bus.
Selection Phase
During the Selection phase, the initiator asserts the ATN
signal and tells the target that it wants to send a message.
Message Out Phase
The target enters the Message Out phase and the initiator
sends a 1-byte message to enable reselection for the command.
In Figure 2–3, the initiator sends an IDENTIFY message to
the target.
Command Phase
The device enters the Command phase and the device
controller sends the 6-byte command.
Data In Phase
During the Data In phase, the target requests a data transfer
with the initiator. The device transfers the data. The device
now determines what to do next.
Status
The device controller reads the (1-byte) Status code from the
target device. The data transfer is from target to initiator.
Message In Phase
During the Message In phase, the target requests that the
initiator take the 1-byte COMMAND COMPLETE message.
Bus Free Phase
The target device releases the BSY signal and returns to the
Bus Free phase. The bus is now available.

Logical Characteristics 2–13

2.4 SCSI Pointers
ANSI SCSI-2
specification,
Section 5.4.

Consider the system shown in Figure 2–4 in which the initiator
(the SCSI host adapter) and the target (the SCSI controller)
communicate on the SCSI bus in order to execute an I/O process.
The SCSI architecture provides three pointers, called saved
pointers, for each I/O process. One pointer is for the command
(CMD), one is for the data (DATA), and one is for the status byte
(STAT). When an I/O process becomes active, its three saved
pointers are copied into the initiator’s set of three active pointers.
There is only one set of active pointers in each initiator. The
active pointers point to the next command, data, or status byte to
be transferred between the initiator’s memory and the target. The
saved and active pointers reside in the initiator.
The Saved Command pointer always points to the start of the
command descriptor block. (See Section 3.2 for a discussion of
the I/O process.)
The Saved Status pointer always points to the start of the
status area for the I/O process.
The Saved Data pointer points to the start of the data area
until the target sends a SAVE DATA POINTER message for
the I/O process.
In response to the SAVE DATA POINTER message, the initiator
stores the value of the Active Data Pointer in the Saved Data
Pointer for that I/O process. The target may restore the active
pointers to the saved pointer values for the active I/O process by
sending a RESTORE POINTERS message to the initiator (see
Section 2.6.4). The initiator then copies the set of saved pointers
into the set of active pointers. Whenever a target disconnects
from the bus, only the set of saved pointers is retained. The set
of active pointers is restored from the set of saved pointers upon
reconnection of the I/O process.
Note Since the data pointer value may be modified by the target before
the I/O process ends, do not use it to test for actual transfer length.

2–14 Logical Characteristics

Figure 2–4

SCSI Pointers
Host Memory
Data Block (DATA)
Command Descriptor
Block (CDB)
Status Byte (STAT)

SCSI
Host Adapter
Access
to host
memory

Active
Pointers
CMD
STAT
DATA

SCSI
Controller

SCSI
Devices
I/O Device

Device
Specific
I/O Device
(LUN #1)

SCSI
Bus

I/O Channel

I/O Device
CMD
STAT
DATA

SCSI
Controller
I/O Device

Saved
Pointers
Device
Specific

I/O Device
(LUN #1)

I/O Device

To other
SCSI devices
MLO-003554

VMS

The VMS operating system always performs an implicit
RESTORE POINTERS message when the target reselects. The
target should not use this message because it adds unnecessary
overhead to the CPU at each I/O operation. Do not use the

Logical Characteristics 2–15

RESTORE POINTERS message to restart the Command phase or
Status phase.

2.5 SCSI Message System
ANSI SCSI-2
specification,
Section 5.5.

The SCSI message system allows communication between an
initiator and a target for the purpose of interface management.
This section covers the following topics:
Messages—Format and Codes, Section 2.5.1
Messages—Implemented by the VMS Operating System,
Section 2.5.2
A message may be one, two, or multiple bytes in length. One or
more messages may be sent during a single Message phase, but
a message may not be split over Message phases. The initiator is
required to end the Message Out phase (by negating ATN) when
it sends certain messages identified in Figure 2–5.

2.5.1 Messages—Format and Codes
One-byte, two-byte, and extended message formats are defined.
The first byte of the message determines the format as follows:
Value

Message Format

00h

One-byte message (COMMAND COMPLETE)

01h

Extended messages

02h–1Fh

One-byte messages

20h–2Fh

Two-byte messages

30h–7Fh

Reserved for future standardization

80h–FFh

One-byte message (IDENTIFY)

One-byte messages consist of a single byte transferred during a
Message phase. The value of the byte determines which message
is to be defined. See Figure 2–5.
Note Digital has designed its hardware and software so that it adheres
to the SCSI-2 standard and yet provides as much flexibility as
possible. To work with Digital’s systems, third-party devices must
implement all the mandatory features of the SCSI-2 interface.
Third-party devices may implement optional features, as long
as they are implemented according to the SCSI-2 specification.
Devices may implement vendor-specific features as long as the
features are implemented in areas clearly designated as such by
the SCSI-2 specification.

2–16 Logical Characteristics

Figure 2–5

Code

Message Codes

VMS
Implementation*

Support
Init Targ

06h
0Dh
0Ch
0Eh
00h
04h
04h
80h
80h
24h
0Fh
0Fh
05h
0Ah
0Bh
09h
07h
***
08h

A, C
N
A
N
A, C
A
N
A
A
N
N
N
A
N
N
A
A
N
A

O
O
O
O
M
O
O
M
M
O
O
O
M
O
O
M
M
O
M

M
O
M
O
M
O
O
O
M
O
O
O
M
O
O
M
M
O
M

21h
22h
20h
10h
03h
02h
***
***
11h
12h -1Fh
24h - 2Fh
30h -7Fh

N
N
N
N
A, C
A, C
A**
N
N

O
O
O
O
O
O
O
O
O

Message
Direction

Message Name

ABORT
ABORT TAG
BUS DEVICE RESET
CLEAR QUEUE
COMMAND COMPLETE
DISCONNECT
DISCONNECT
IDENTIFY
IDENTIFY
IGNORE WIDE RESIDUE (two bytes)
INITIATE RECOVERY
INITIATE RECOVERY (see note)
INITIATOR DETECTED ERROR
LINKED COMMAND COMPLETE
LINKED COMMAND COMPLETE (with flag)
MESSAGE PARITY ERROR
MESSAGE REJECT
MODIFY DATA POINTER
NO OPERATION
Queue Tag Messages (two bytes)
HEAD OF QUEUE TAG
ORDERED QUEUE TAG
SIMPLE QUEUE TAG
RELEASE RECOVERY
RESTORE POINTERS
SAVE DATA POINTER
SYNCHRONOUS DATA TRANSFER REQUEST
WIDE DATA TRANSFER REQUEST
TERMINATE I/O PROCESS
Reserved
Reserved for two-byte messages
Reserved

Negate
ATN
Before
Last
ACK
Yes
Yes
Yes
Yes
_
_
Yes
_

Out
Out
Out
Out
In
In
Out
In

Out
In Out
In
Out

No
_
_
Yes
Yes
_
_
Yes
Yes
_
Yes

Out
Out
In Out
Out
In
In
In Out
In Out
Out

No
No
No
Yes
_
_
Yes
Yes
Yes

Out
In
In
Out
Out
In
In

Note:
Outbound INITIATE RECOVERY messages are valid only during asynchronous
event notification protocol.
The ABORT TAG and CLEAR QUEUE messages are required if tagged queuing is implemented.
* Use of messages other than those implemented by Digital may result in an error message.
** Implemented on the VAXstation 3520/3540.
MLO-003597

Logical Characteristics 2–17

Table 2–3 describes the mnemonics used in Figure 2–5:
Table 2–3

Mnemonics Used in Figure 2–5

Mnemonic Description1
A

Used by Digital as described in the draft ANSI SCSI-2
specification

Digital change to ANSI SCSI-2 specification

Mandatory support

Not implemented by Digital

Optional; implement according to the draft ANSI SCSI-2
specification

Target to initiator

Out

Initiator to target

Yes

Initiator must negate ATN before last acknowledgment
(ACK) of message

Initiator may or may not negate ACK before last ACK of
message (see Section 2.2.1)

–

Not applicable

***

Extended message (see Table 2–4 and Table 2–5)

80h+

Codes 80h through FFh are used for IDENTIFY messages

Use of messages other than those implemented by Digital may result in an error
message.

Two-byte messages consist of two consecutive bytes transferred
during a Message phase. The value of the first byte determines
which message is to be performed, as defined in Figure 2–5. The
second byte is a parameter byte, which is used as defined in the
message description (see Section 2.6).
A value of one in the first byte of a message indicates the
beginning of a multiple-byte extended message. Extended
messages send a minimum of three bytes. The extended message
format and the extended message codes are shown in Table 2–4
and Table 2–5, respectively.

2–18 Logical Characteristics

Table 2–4

Extended Message Format

Byte

Value

Description

01h

Extended message

Extended message length

Extended message code

3 to n+1

Extended message arguments

The extended message length specifies the length in bytes of the
extended message code plus the extended message arguments
to follow. Therefore, the total length of the message is equal to
the extended message length plus two. A value of zero for the
extended message length indicates that 128 bytes follow.
The extended message codes are listed in Table 2–5. The extended
message arguments are specified within the extended message
descriptions (see Section 2.6).
Table 2–5

Extended Message Codes

Code (y)

Description

02h

Reserved

00h

MODIFY DATA POINTER

01h

SYNCHRONOUS DATA TRANSFER REQUEST

03h

WIDE DATA TRANSFER REQUEST

04h–7Fh

Reserved

80h–FFh

Vendor unique

The first message sent by the initiator after the Selection phase
must be the IDENTIFY, ABORT, or BUS DEVICE RESET
message. If a target receives any other message, it must go to
the Bus Free phase (see Section 2.1.1).

2.5.2 Messages—Implemented by the VMS
Operating System
VMS

The VMS operating system accepts the following messages during
the Message In phase. The target device must not generate any
messages except the following:
COMMAND COMPLETE
DISCONNECT

Logical Characteristics 2–19

IDENTIFY
RESTORE POINTERS
MESSAGE REJECT
SAVE DATA POINTER
SYNCHRONOUS DATA TRANSFER REQUEST (implemented
on the VAXstation 3520/3540 systems)
The VMS operating system generates the following messages,
which the target must accept and process as described in the
SCSI-2 standard.
ABORT
VMS

BUS DEVICE RESET
IDENTIFY
INITIATOR DETECTED ERROR
MESSAGE PARITY ERROR
MESSAGE REJECT
NO OPERATION
SYNCHRONOUS DATA TRANSFER REQUEST (VAXstation
3520/3540 systems only)
When a device selects the host for the purpose of performing an
asynchronous event notification (AEN), the host accepts the
following messages in addition to those stated above. When one of
these messages is received, the VMS operating system notifies the
third-party class driver of the received message.
ABORT

VMS

BUS DEVICE RESET
INITIATOR DETECTED ERROR
MESSAGE PARITY ERROR
MESSAGE REJECT
NO OPERATION

2–20 Logical Characteristics

2.6 Messages
ANSI SCSI-2
specification,
Section 5.6.

Table 2–6 lists the SCSI messages implemented by Digital. The
following messages are described in this guide:
ABORT, Section 2.6.1
COMMAND COMPLETE, Section 2.6.2
DISCONNECT, Section 2.6.3
RESTORE POINTERS, Section 2.6.4
SAVE DATA POINTER, Section 2.6.5
Table 2–6

SCSI Messages

Message1

Reference

ABORT

Section 2.6.1 of this guide

BUS DEVICE RESET

ANSI SCSI-2 spec, Section 5.6.3

COMMAND COMPLETE

Section 2.6.2 of this guide

DISCONNECT

ANSI SCSI-2 spec, Section 5.6.6

IDENTIFY

ANSI SCSI-2 spec, Section 5.6.7

INITIATOR DETECTED ERROR

ANSI SCSI-2 spec, Section 5.6.10

MESSAGE PARITY ERROR

ANSI SCSI-2 spec, Section 5.6.13

MESSAGE REJECT

ANSI SCSI-2 spec, Section 5.6.14

MODIFY DATA POINTER

ANSI SCSI-2 spec, Section 5.6.15

NO OPERATION

ANSI SCSI-2 spec, Section 5.6.16

RELEASE RECOVERY

ANSI SCSI-2 spec, Section 5.6.18

RESTORE POINTERS

Section 2.6.4 of this guide

SAVE DATA POINTER

Section 2.6.5 of this guide

SYNCHRONOUS DATA TRANSFER
REQUEST

ANSI SCSI-2 spec, Section 5.6.21

TERMINATE I/O PROCESS

ANSI SCSI-2 spec, Section 5.6.22

Use of messages other than those implemented by Digital may result in an error
message.

Logical Characteristics 2–21

2.6.1 ABORT
ANSI SCSI-2
specification,
Section 5.6.1.

The ABORT message is sent from the initiator to the target to
clear the present I/O process plus any queued I/O process for the
I_T_x nexus. The target goes to the Bus Free phase following
successful receipt of this message. Pending data, status, and
queued I/O processes for any other I_T_x nexus are not cleared.
If only an I_T nexus is established, the target goes to the Bus
Free phase. No status or message is sent for the I/O process and
the I/O process queue is not affected.
No errors occur if the initiator issues this message to an
I_T_x nexus that does not currently have an active or queued I/O
process. Transmission of this message terminates any extended
contingent allegiance condition that exists between the I_T_x
nexus.
Note The BUS DEVICE RESET and ABORT messages end one or
more active I/O processes prior to normal termination. The BUS
DEVICE RESET message ends all I/O processes for all initiators
on all logical units of the target. The ABORT message clears all
I/O processes for the selecting initiator on the specified logical unit
or target routine of the target.

VMS

On VMS systems, the ABORT message must be implemented as
described in the SCSI-2 standard. When the target receives an
ABORT, it must terminate the current operation. The device must
stop in a manner that does not cause side effects. For example, no
parity errors should be written to the media, and devices such as
tape drives that store user data in nonvolatile memory must write
the data back to nonvolatile media before honoring the ABORT.
Following the ABORT, the device must use the recovery guidelines
described in Chapter 5.

2.6.2 COMMAND COMPLETE
ANSI SCSI-2
specification,
Section 5.6.5.

The COMMAND COMPLETE message is sent from a target to
an initiator to indicate that the execution of a command of an
I/O process has completed and that valid status has been sent
to the initiator. After successfully sending this message, the
target goes to the Bus Free phase by releasing a BSY signal. The
target considers the message transmission to be successful when
it detects the negation of ACK for the COMMAND COMPLETE
message with the ATN signal false.
Note The command executes successfully or unsuccessfully as indicated
in the status.

2–22 Logical Characteristics

2.6.3 DISCONNECT
ANSI SCSI-2
specification,
Section 5.6.6.

The DISCONNECT message is sent from the target to inform an
initiator that the present connection is to be broken (the target
plans to disconnect by releasing BSY) and that a later reconnect
is required in order to complete the current I/O process.
This message must not cause the initiator to save the data pointer.
After successfully sending this message, the target goes to the Bus
Free phase by releasing the BSY signal after 10 microseconds.
The target should consider the message transmission to be
successful when it detects the negation of the ACK signal for the
DISCONNECT message with the ATN signal false.

VMS

On VMS systems, the device may not hold the BSY signal for more
than 10 microseconds after sending the COMMAND COMPLETE
or DISCONNECT message.

2.6.4 RESTORE POINTERS
ANSI SCSI-2
specification,
Section 5.6.19.

VMS

The RESTORE POINTERS message is sent from a target to direct
the initiator to copy the most recently saved command, data, and
status pointers for the I/O process to the corresponding active
pointers. The command and status pointers are restored to the
beginning of the present command and status areas. The data
pointer is restored to the value at the beginning of the data area
in the absence of a SAVE DATA POINTER message or to the value
at the point at which the last SAVE DATA POINTER message
occurred for that nexus.
The VMS operating system always performs an implicit
RESTORE POINTERS when the target reselects. The target
should not use this message because it adds unnecessary overhead
to the CPU at each I/O operation. Note that the RESTORE
POINTERS message cannot be used to retry the Command or
Status phases. (See Section 2.4 for a definition of pointers.)

2.6.5 SAVE DATA POINTER
ANSI SCSI-2
specification,
Section 5.6.20

The SAVE DATA POINTER message is sent from a target to direct
the initiator to save a copy of the present active data pointer for
the current nexus. (See Section 2.4 for a definition of pointers.)

Logical Characteristics 2–23

VMS

The VMS operating system uses this message to save the context
of an operation while a device is disconnected from the bus.
However, the message is not necessary or useful when the target
disconnects before data transfer begins. In this case, the target
should not use the message because it adds unnecessary overhead
to the CPU at each I/O operation. Once the device transfers
data, it must issue the SAVE DATA POINTER message before
disconnecting.

2–24 Logical Characteristics

3
SCSI Commands and Status
This chapter defines and provides examples of the SCSI command
and status structures in the following sections:
Command implementation requirements, Section 3.1
Command descriptor block, Section 3.2
Status, Section 3.3
Command processing exception conditions, Section 3.4
Asynchronous event notification, Section 3.5
Contingent allegiance condition, Section 3.6
Extended contingent allegiance condition, Section 3.7
Unit attention condition, Section 3.8
If you keep the functions essential to communicate by means of
this protocol to a minimum, a wide range of peripheral devices
can operate in the same environment. Because subsets of the full
architecture may be implemented, optional functions are noted in
the following sections.

SCSI Commands and Status 3–1

3.1 Command Implementation Requirements
ANSI SCSI-2
specification,
Section 6.1.

VMS

The first byte of all SCSI commands must contain an operation
code as defined in this document. Three bits (bits 7–5 in the
command descriptor block) of the second byte of each SCSI
command specify the logical unit if it is not specified using the
IDENTIFY message (see Section 2.5). The last byte of all SCSI
commands must contain a control byte as defined in Section 3.2.7.
The VMS operating system allows commands up to 128 bytes
in length. There are no restrictions on the size or type of
commands that may be sent, as long as they conform to the
SCSI-2 specification.

3.1.1 Reserved
ANSI SCSI-2
specification,
Section 6.1.1.

Reserved bits, fields, bytes, and code values are set aside for
future standardization. Their use and interpretation may be
specified by future extensions to this specification. A reserved bit,
field, or byte must be set to zero or in accordance with a future
extension to this specification. See the ANSI SCSI-2 specification
for additional information.

3.1.2 Operation Code Types
ANSI SCSI-2
specification,
Section 6.1.2.

Table 3–1 describes the operation codes used in the first byte of a
SCSI command.
Table 3–1

Operation Code Types

Opcode Type

Description

M (Mandatory)

These commands must be implemented in order
to meet the minimum requirement of the SCSI-2
specification.

O (Optional)

Use optional commands in accordance with the
SCSI-2 specification.

V (Vendor Specific)

Operation codes are unique to the vendor’s
product.

R (Reserved)

These operation codes must not be used. They
are reserved for future extensions to the SCSI-2
specification.

3–2 SCSI Commands and Status

3.2 Command Descriptor Block
ANSI SCSI-2
specification,
Section 6.2

A command is communicated by sending a command descriptor
block (CDB) to the target. For several commands, the request
is accompanied by a list of parameters sent during the Data Out
phase.
The CDB always has an operation code as the first byte of the
command. This is followed by an optional logical unit number,
command parameters (if any), and a control byte.
For all commands, if there is an invalid parameter in the CDB,
then the target terminates the command without altering the
medium. Figure 3–1 illustrates the command descriptor block for
a 6-byte command.
Figure 3–1
Bit
Byte

CDB for Six-Byte Commands
6

0
1
2

Operation Code
Logical Unit Number

(MSB)

Logical Block Address (if required)
(LSB)

3
4

Transfer Length (if required)
Parameter List Length (if required)
Allocation Length (if required)

Control
MLO−003593

SCSI Commands and Status 3–3

Figure 3–2 illustrates the command descriptor block for a 10-byte
command.
Figure 3–2
Bit
Byte

CDB for Ten-Byte Commands
6

0
1
2

Operation Code
Logical Unit Number

Reserved

(MSB)

3
Logical Block Address (if required)
4
(LSB)

5
Reserved

Transfer Length (if required)
Parameter List Length (if required)
Allocation Length (if required)

Control

(MSB)

(LSB)

MLO−003590

The following sections explain the elements of the command
descriptor block:
Operation Code, Section 3.2.1
Logical Unit Number, Section 3.2.2
Logical Block Address, Section 3.2.3
Transfer Length, Section 3.2.4
Parameter List Length, Section 3.2.5
Allocation Length, Section 3.2.6
Control Byte, Section 3.2.7

3–4 SCSI Commands and Status

3.2.1 Operation Code
ANSI SCSI-2
specification,
Section 6.2.1.

The operation code of the command descriptor block has a group
code field and a command code field (see Figure 3–3). The 3-bit
group code field provides for eight groups of command codes. The
5-bit command code field provides for 32 command codes in each
group. Thus, there are 256 possible operation codes. See the
ANSI SCSI-2 specification for additional information.
The group code specifies one of the following groups:
Group 0: Six-byte commands (see Figure 3–1)
Group 1: Ten-byte commands (see Figure 3–2)
Group 2: Not implemented
Group 3: Reserved
Group 4: Reserved
Group 5: Not implemented
Group 6: Vendor specific
Group 7: Vendor specific

Figure 3–3
Bit

Operation Code
6

Group Code

Command Code
MLO-003591

3.2.2 Logical Unit Number
ANSI SCSI-2
specification,
Section 6.2.2.

The device logical unit number (LUN) distinguishes between the
multiple units attached to a single SCSI controller. Set the logical
unit number to zero in the CDB.
Digital supports logical unit numbers in the range 0–7 and stores
the logical unit number in the IDENTIFY message that is sent
prior to every command. Digital does not support sub-LUNs.
Sub-LUNs are a feature described in the SCSI-1 standard, but the
SCSI-2 specification has eliminated them.

SCSI Commands and Status 3–5

3.2.3 Logical Block Address
ANSI SCSI-2
specification,
Section 6.2.3.

The logical block address on logical units or within a partition
on device volumes must begin with block zero and must be
contiguous up to the last logical block on that logical unit or
within that partition.
A 6-byte command descriptor block contains a 21-bit logical block
address. The 10-byte command descriptor block contains a 32bit logical block address. Logical block addresses in additional
parameter data have their length specified for each occurrence.

3.2.4 Transfer Length
ANSI SCSI-2
specification,
Section 6.2.4.

The transfer length field specifies the amount of data to be
transferred, usually the number of logical blocks. For several
of the commands, the transfer length indicates the requested
number of bytes to be sent as defined in the command description.
For these commands, the transfer length field in the CDB may be
identified by a different name.
Commands that use one byte for transfer length allow a maximum
of 256 blocks of data to be transferred by one command. A
transfer length value of 1 to 255 indicates the number of blocks
that must be transferred. A value of zero indicates 256 blocks.
Commands that use multiple bytes for transfer length allow
a maximum of 65,535 blocks of data to be transferred by one
command. For these commands, a transfer length of zero indicates
that no data transfer must take place. A value of one or greater
indicates the number of blocks that must be transferred.

VMS

VAX based VMS systems use an intermediate buffer for all
incoming and outgoing SCSI data. When the VMS operating
system performs an I/O operation to a SCSI device, it allocates a
portion of this buffer for the duration of the operation. Because
VMS allows multiple concurrent I/O operations on the SCSI bus,
it cannot allocate the full buffer to a single device. The following
list shows the internal buffer size, the maximum data transfer
size, and the recommended data transfer size on the supported
VAX systems:
The internal buffer size is 128 kilobytes on all supported
systems.
The maximum data transfer size is 64 kilobytes on all
supported systems. If a device needs to transfer more than
the maximum transfer size, it must be modified to allow the
data to be returned across multiple SCSI commands. A single
SCSI command cannot transfer more than the maximum size.

3–6 SCSI Commands and Status

For the best overall system throughput, the device should
transfer only 16 kilobytes of data per SCSI command.
Note The restrictions described above that apply to the transfer length
field also apply to the parameter list length and allocation length
fields.

3.2.5 Parameter List Length
ANSI SCSI-2
specification,
Section 6.2.5.

The parameter list length specifies the number of bytes sent
during the Data Out phase. This field typically is used in
command descriptor blocks for parameters that are sent to a
target; for example, it is used with mode parameters, diagnostic
parameters, and log parameters.

3.2.6 Allocation Length
ANSI SCSI-2
specification,
Section 6.2.6.

The allocation length field specifies the maximum number of bytes
that an initiator has allocated for returned data. An allocation
length of zero indicates that no data must be transferred. This
condition must not be considered an error. The target must
terminate the Data In phase when allocation length bytes have
been transferred or when all available data have been transferred
to the initiator, whichever is less.
The allocation length is typically used to return sense data, for
example, mode data, log data, and diagnostic data, to an initiator.

3.2.7 Control Byte
ANSI SCSI-2
specification,
Section 6.2.7.

The control byte is the last byte of every command descriptor
block. The control byte is illustrated in Figure 3–4.
Figure 3–4
Bit

Control Byte
6

Vendor Specific

Reserved

Flag

Link

MLO−003592

Digital does not currently set the link, flag, and vendor-specific
bits in the command descriptor block.

SCSI Commands and Status 3–7

3.3 Status
ANSI SCSI-2
specification,
Section 6.3.

A status byte must be sent from the target to the initiator during
the Status phase at the completion of each command. Figure 3–5
illustrates the status byte.
Figure 3–5
Bit

Status Byte
6

Reserved

Status Byte Code

Reserved
MLO-003609

The contents of the SCSI status byte must conform to rules of the
SCSI-2 specification. Otherwise, there are no restrictions on the
use of this byte. Table 3–2 describes the status byte codes.

3–8 SCSI Commands and Status

Table 3–2

Status Byte Codes

Status Code

Description

Good

The target has successfully completed the
command.

Check Condition

Any error, exception, or abnormal condition
that causes sense data to be set must cause a
Check Condition status. The REQUEST SENSE
command must be issued following a Check
Condition status, to determine the nature of the
condition.

Condition Met

The SEARCH DATA command returns this
status whenever a search condition is satisfied.
The logical block address of the logical block that
satisfies the search may be determined with a
REQUEST SENSE command. This status is also
returned by the PRE-FETCH command when
there is sufficient space in the cache memory for
all of the addressed logical blocks.

Busy

The target is busy. This status is returned
whenever a target is unable to process the
command from an otherwise acceptable initiator
(that is, there are no conflicting reservations).
The recommended initiator recovery action is to
issue the command again at a later time. The
VMS operating system implements the status
byte according to the SCSI-2 specification.

Intermediate

Linked commands are not implemented.

Reservation Conflict

This status is returned whenever a SCSI device
attempts to access a logical unit or an extent
within a logical unit that is reserved with a
conflicting reservation type for another SCSI
device. The recommended initiator recovery
action is to issue the command again at a later
time.

Queue Full

Tagged queuing is not implemented.

VMS

SCSI Commands and Status 3–9

3.4 Command Processing Considerations and Exception
Conditions
ANSI SCSI-2
specification,
Section 6.5.

Sections 3.4.1 through 3.4.4 describe exception conditions and
errors associated with command processing and the sequencing of
commands.

3.4.1 Programmable Operating Definition
ANSI SCSI-2
specification,
Section 6.5.1.

Special applications require that the operating definition of a
logical unit be modified to meet the requirements of a particular
initiator. Digital does not implement this feature. See the ANSI
SCSI-2 specification for additional information.

3.4.2 Incorrect Initiator Connection
ANSI SCSI-2
specification,
Section 6.5.2.

See the ANSI SCSI-2 specification for information.

ANSI SCSI-2
specification,
Section 6.5.3.

The target’s response to selection of an invalid logical unit is
described in the following paragraphs:

3.4.3 Selection of an Invalid Logical Unit

The target does not support the logical unit (that is, some
targets support the attachment of only one peripheral
device).
In response to an INQUIRY command the target must return
the inquiry data with the peripheral qualifier set.
In response to any other command except REQUEST SENSE
the target must terminate the command with Check Condition
status.
In response to a REQUEST SENSE command the target
must return sense data. The sense key must be set to Illegal
Request and the additional sense code must be set to Invalid
Logical Unit.
The target supports the logical unit, but the logical unit is
not attached to the target.
In response to an INQUIRY command the target must return
the INQUIRY data with the peripheral qualifier set.

3–10 SCSI Commands and Status

In response to any other command except REQUEST SENSE
the target must terminate the command with Check Condition
status.
In response to a REQUEST SENSE command the target
must return sense data. The sense key must be set to Illegal
Request and the additional sense code must be set to Invalid
Logical Unit.
The target supports the logical unit and it is attached, but
the peripheral device is not operational.
In response to an INQUIRY command the target must return
the INQUIRY data with the peripheral qualifier set. The
target’s response to any other command is defined in the
device specification for that device.
The target supports the logical unit but is incapable
of determining if the peripheral is attached or is not
operational when it is not ready.
In response to an INQUIRY command the target must return
the INQUIRY data with the peripheral qualifier set.
In response to a REQUEST SENSE command the target must
return the request sense data with a sense key of No Sense
unless a prior error condition exists. The target’s response to
any other command is defined in the device specification for
that device.

3.4.4 Parameter Rounding
ANSI SCSI-2
specification,
Section 6.5.4.

Certain parameters sent to a target with various commands
contain a range of values. Targets may choose to implement
only selected values from this range. When the target receives
a value that it does not support, it either rejects the command
(Check Condition status with Illegal Request Sense key) or it
rounds the value received to a supported value. The target must
reject unsupported values unless rounding is permitted in the
description of the parameter. See the referenced ANSI SCSI-2
section for additional information.

SCSI Commands and Status 3–11

3.5 Asynchronous Event Notification
ANSI SCSI-2
specification,
Section 6.5.5.

VMS

Asynchronous event notification (AEN) is an optional protocol
used to inform processor devices that an asynchronous event
has occurred. The notification comes in the form of a SEND
command directed to the processor device with a data block
that includes device Request Sense information. Devices that
respond to INQUIRY commands as a processor device type with
asynchronous event notification capability may be notified of
asynchronous events using this protocol. A SCSI device has to be
capable of acting as an initiator in order to perform asynchronous
event notification.
Normally, the VAX CPU running VMS is the only entity that may
act as an initiator on the SCSI bus. However, the user’s device
is also permitted to act as an initiator, as long as it is for the
purpose of selecting and sending a command to the VAX CPU. The
device cannot select or issue a command to any other device on
the bus.
The use of AEN is governed by the permissions fields in the
Asynchronous Event Notification parameters MODE SELECT
page.
AEN is used for four purposes in a SCSI system:
To request that a device promptly inform a processor of an
error condition encountered after command completion
An example is a tape that implements write caching.
Notification of an unable to write condition can be sent to
the processor that initiated the write through the AEN
protocol. An extended contingent allegiance condition may
also establish the same I_T_L nexus used for the SEND
command with an AEN bit of one.
To permit a newly initialized device to inform all the processor
devices on the bus of its availability
To have a device inform processors of other unit attention
conditions
An example is a device that supports removable media. AEN
may be used to report a not ready to ready transition (that is,
the media changed) and also operator-initiated state changes
such as write-protect switch and start/stop switch actuations.
To have a device inform processors of other state changes that
supply sense key 0h information

3–12 SCSI Commands and Status

An example is a tape performing a REWIND command with
the immediate bit set. Notification of beginning of media
reached may be reported through the AEN protocol. In a
similar manner, completion of a CDROM Audio Play operation
started in the immediate mode may be reported through an
AEN notification.
Although other uses of AEN are not prohibited, the AEN protocol
is not intended to be used while an I_T_L nexus exists between
the processor (initiator of the nexus) and the device.
Notification of an asynchronous event is performed using the
SEND command with the AEN bit set to one. The information
identifying the condition being reported must be returned
during the Data In phase of the SEND command. The data sent
beginning at byte 4 of the SEND command AEN data format must
be the same as that which would have been returned to report the
event in response to a REQUEST SENSE command.
An error or unit attention condition successfully reported
using the AEN protocol must not also cause a Check Condition
termination of a subsequent command. Similarly, if a Check
Condition termination of a command from some processor is
caused by an event, an AEN protocol notification must not be
initiated to report the same event to that processor. An event
may be reported with either the Check Condition/Request Sense
protocol or with the AEN protocol, but must not be reported to the
same processor with both.
AEN protocol notification of command related exception conditions
must be sent only to the processor that initiated the operation.
Systems in which devices can become available independent of the
rest of the system may use the AEN protocol to inform processors
when they become ready. In this case the SEND command must
transfer a Request Sense data block with the Unit Attention sense
key. Successful delivery of a unit attention condition notification
through the AEN protocol must clear the unit attention condition
with respect to the receiving processor device.
A device that wishes to perform asynchronous event notification
must conduct a survey to determine which devices respond to the
processor/device type command set and have AEN capability. To
conduct this survey, the device sends an INQUIRY command to
logical unit zero (0) of each device that responds to selection. This
survey must be conducted prior to the first AEN and must be
repeated whenever the device becomes aware of a new device or
the AEN capability state information may have become invalid.

SCSI Commands and Status 3–13

After conducting the survey, the device verifies that each processor
device with AEN capability is ready to receive asynchronous
event notifications by issuing a TEST UNIT READY command.
If a Check Condition status is returned, a REQUEST SENSE
command is issued. This procedure clears any pending Unit
Attention conditions. A processor device that reports AEN
capability and responds to a TEST UNIT READY command with
Good status must be ready to accept a SEND command with an
AEN bit of one.
Note A SCSI device that uses AEN at initialization time must provide a
means to defeat these notifications. You can do this with a switch
or jumper wire. Devices that implement saved parameters may
alternatively save the AEN permissions either on an individual
SCSI ID or systemwide option. In any case, a device must conduct
a survey with INQUIRY commands to be sure that the devices on
the SCSI bus are appropriate destinations for SEND commands
with an AEN bit of one. (The devices on the bus or the SCSI ID
assignments may have changed.)

3.6 Contingent Allegiance Condition
ANSI SCSI-2
specification,
Section 6.6.

The Contingent Allegiance condition exists following the return
of a Check Condition or Command Terminated status and
may optionally exist following an unexpected disconnect. The
Contingent Allegiance condition is preserved for the I_T_x nexus
until it is cleared. The Contingent Allegiance condition is cleared
upon the generation of a hard Reset condition, by the receipt of an
ABORT message, by a BUS DEVICE RESET message, or by any
subsequent command for the I_T_x nexus, except a REQUEST
SENSE command. The Contingent Allegiance condition is cleared
following the return of the sense data for the I_T_x nexus. While
the Contingent Allegiance condition exists, the target must
preserve the sense data for the initiator.

3.7 Extended Contingent Allegiance Condition
VMS

The VMS operating system does not support the extended
Contingent Allegiance condition. See the ANSI SCSI-2
specification, Section 6.7, for information.

3.8 Unit Attention Condition
ANSI SCSI-2
specification,
Section 6.9

A Unit Attention condition begins for each initiator whenever
the target is reset by a BUS DEVICE RESET message, a hard
Reset condition, or by a power-on reset. See the ANSI SCSI-2
specification for information.

3–14 SCSI Commands and Status

4
Characteristics of Most Device Types
This chapter describes the general characteristics expected of
most SCSI devices. It provides the following information:
A model for all device types, Section 4.1. See also the ANSI
SCSI-2 specification, Section 7.1.
Commands for all device types, Section 4.2. See also the ANSI
SCSI-2 specification, Section 7.2.
For the parameters applicable for all device types, see the ANSI
SCSI-2 specification, Section 7.3.

4.1 Model for Most Device Types
ANSI SCSI-2
specification,
Section 7.1.

The model described in Section 4.1.1 and Section 4.1.2 provides
information for most device types; it does not define any new
requirements or alter any requirements defined in the ANSI
SCSI-2 specification.

4.1.1 SCSI Addresses
ANSI SCSI-2
specification,
Section 7.1.1.

There are two levels of addresses within the SCSI architecture:
the SCSI device address and the logical unit number or target
routine number.
SCSI Device Address
SCSI devices occupy (that is, respond to) one address on the SCSI
bus. Generally the SCSI device provides switches or jumpers to
select one of the eight available addresses (0–7). The address
is used during bus arbitration and selection or reselection of
SCSI devices. Each device on the SCSI bus is assigned a unique
address.

Characteristics of Most Device Types 4–1

Normally, the SCSI device address is set when the system is
configured and it remains static thereafter. Some systems and
devices provide vendor-specific means to alter this address.
Logical Units
Each target has one or more logical units, beginning with logical
unit zero. Each target can have a maximum of eight logical units.
These logical units are usually mapped directly to peripheral
devices. The logical units may be a portion of a peripheral device
or may comprise multiple peripheral devices.
An initiator can determine whether a target implements a
logical unit by issuing an INQUIRY command and examining the
returned peripheral qualifier and peripheral device type.
The concept of a logical unit is not defined for an initiator. (A
SCSI device may implement both the initiator role and the target
role. In such cases, logical units are defined only for the target
role.)
Target Routines
An optional feature of the SCSI architecture permits each target
to have one or more target routines, beginning with target routine
zero. Implement this routine using the ANSI SCSI-2 specification.
There are a maximum of eight target routines. These target
routines are processes that execute directly on the target; they
are not associated with a particular logical unit or peripheral
device. Target routines are addressed using the LUNTAR bit of
the IDENTIFY message.
Target routines are principally intended to return information
about the target. The only valid commands are INQUIRY and
REQUEST SENSE.

4.1.2 Commands Implemented by SCSI Targets
ANSI SCSI-2
specification,
Section 7.1.2.

This section describes five commands that SCSI targets
implement: INQUIRY, RECEIVE DIAGNOSTIC RESULTS,
REQUEST SENSE, SEND DIAGNOSTIC, and TEST UNIT
READY. These commands are used to configure the system, to test
targets, and to return important information concerning errors
and exception conditions.

4–2 Characteristics of Most Device Types

VMS

The VMS operating system allows commands of up to 128 bytes in
length. There are no restrictions on the size or type of command
that may be sent, as long as the commands conform to the SCSI-2
specification.
INQUIRY Command

ANSI SCSI-2
specification,
section 7.2.5

A system may use the INQUIRY command to determine the
configuration of the SCSI bus. Target devices respond with
information that includes their type and standard level and
may include the vendor’s identification, model number, and
other useful information. A SCSI device may take longer to get
certain portions of this information, especially if it retrieves the
information from the medium.
RECEIVE DIAGNOSTIC RESULTS Command

ANSI SCSI-2
specification,
Section 7.2.13.

The RECEIVE DIAGNOSTIC RESULTS command requests that
analytical data be sent to the initiator after completion of a SEND
DIAGNOSTIC command. If the target supports the optional page
format, the page code field in the SEND DIAGNOSTIC command
specifies the format of the returned data.
To ensure that the diagnostic command information is not
destroyed by a command sent from another initiator, the
SEND DIAGNOSTIC command must either be linked to the
RECEIVE DIAGNOSTIC RESULTS command or the logical
unit must be reserved.
Although diagnostic software is generally device-specific, this
command and the SEND DIAGNOSTIC command provide
a means to isolate the operating system software from the
device-specific diagnostic software. Hence the operating
system can remain device-independent. These commands
allow diagnostic software to be transferred more easily to
other operating systems.
REQUEST SENSE Command

ANSI SCSI-2
specification,
section 7.2.14

Whenever a Contingent Allegiance condition is established (see
Section 3.6), the initiator that received the error must issue a
REQUEST SENSE command to receive the sense data describing
what caused the Contingent Allegiance condition. If the initiator
issues some other command, the sense data is lost.

Characteristics of Most Device Types 4–3

SEND DIAGNOSTIC Command
ANSI SCSI-2
specification,
section 7.2.15

The SEND DIAGNOSTIC command provides a means to request
the target to perform a self-test. While the test is target-specific,
the means of requesting the test is standardized and the response
is Good status if all is well or Check Condition status if the test
fails.
The SEND DIAGNOSTIC command also provides other
powerful features when used in conjunction with the RECEIVE
DIAGNOSTIC RESULTS command, but this capability is optional.
All SCSI devices should implement the Common Command Set
commands and messages as described in Chapter 2 and Chapter 3
of this guide.
All fields in the INQUIRY command packet must be properly
filled.
Device-dependent diagnostics should be implemented as follows:
Developers should implement two levels of the SCSI SEND
DIAGNOSTIC commands.
–

Device SCSI controller test.
The SlfTST bit should be set in the SEND DIAGNOSTIC
command packet. The preferred execution time is less
than 1 second.

–

Specific media access tests.
The SlfTst and UnitOffline bits should be set in the SEND
DIAGNOSTIC command packet. The preferred execution
time is less than 60 seconds.

Developers should implement the RECEIVE DIAGNOSTIC
commands with information that permits field replaceable
unit (FRU) isolation (95 percent fault detection, 90 percent
isolation to FRU) within the device.
Note Vendor-specific commands (that is, commands not found in Digital
documentation or the ANSI SCSI-2 specification) cannot be tested.
TEST UNIT READY Command
ANSI SCSI-2
specification,
section 7.2.16

The TEST UNIT READY command allows an initiator to poll a
logical unit until it is ready without the need to allocate space for
returned data. It is especially useful for checking the cartridge
status of logical units with removable media. Targets are expected
to respond promptly to indicate the current status of the device
(that is, a target must avoid lengthy disconnections in an attempt
to respond with Good status).

4–4 Characteristics of Most Device Types

4.2 Commands for Most Device Types
ANSI SCSI-2
specification,
Section 7.2.

Table 4–1 lists the operation codes for commands that apply to
most device types.
Table 4–1

Commands for Most Device Types

Command Name

Operation
Code

Type

ANSI SCSI-2
Specification
Section

CHANGE DEFINITION

40h

7.2.1

COMPARE

39h

7.2.2

COPY

18h

7.2.3

COPY AND VERIFY

3Ah

7.2.4

INQUIRY

12h

7.2.5

MODE SELECT(6)

15h

7.2.8

MODE SENSE(6)

1Ah

7.2.10

READ BUFFER

3Ch

7.2.12

READ CAPACITY

25h

8.2.7

RECEIVE DIAGNOSTIC
RESULTS

1Ch

7.2.13

REQUEST SENSE

03h

7.2.14

SEND DIAGNOSTIC

1Dh

7.2.15

START/STOP UNIT

18h

8.2.17

TEST UNIT READY

00h

7.2.16

WRITE BUFFER

3Bh

7.2.17

Key
M—Mandatory command implementation. All commands must be implemented
according to the ANSI SCSI-2 specification.
O—Optional command implementation. All commands should be implemented
according to the ANSI SCSI-2 specification.
Z—Device-type specific command implementation.

Characteristics of Most Device Types 4–5

5
Power-On Protocols and Initialization
Procedures
ANSI SCSI-2
specification,
Appendix F.

This chapter describes Digital’s requirements for system
initialization of SCSI devices. These requirements must be
implemented in addition to the requirements listed in the ANSI
SCSI-2 specification.
The topics are organized as follows:
Dynamic Reconfiguration of Devices, Section 5.1
External Boxes, Section 5.2
Device Behavior Following Power-On, Section 5.3
Device Behavior Following Bus Reset, Section 5.4

5.1 Dynamic Reconfiguration of Devices
Devices on the SCSI bus may not be added to the bus, removed
from the bus, or re-cabled while the operating system is running.
Caution Failure to meet this requirement may cause loss of user data or
system failure.

5.2 External Boxes
Devices that reside outside the main system box should remain
powered on at all times while the system is in operation.

Power-On Protocols and Initialization Procedures 5–1

Some powered-off SCSI devices fail to present high impedance
to the SCSI bus, leaving the bus useless.
A powered-off device can reduce the terminator power on the
bus to unacceptable levels, causing corruption of user data or
system failure.
The device can spike various SCSI bus signals during poweron or power-off, leading to corruption of user data or system
failure.
Note Many users disregard warnings and cycle the power on external
boxes while the system is running, because it often seems to work.
Digital recommends that devices be designed to minimize this
possibility and that external devices remain powered on at all
times during system operation. Powering off devices can lead to
data loss and hardware failure.

5.3 Device Behavior Following Power-On
VMS

The device must meet the following requirements after it is
powered on.
Within 5 seconds of power-on, the device must be able to
respond to a SCSI bus selection so that the VMS operating
system knows that there is an active device at this SCSI ID.
However, the device is permitted to return any of the following
errors until it is ready for normal operation:
–

Busy status.

–

Check Condition status, followed by Not Ready sense key
in response to the next REQUEST SENSE command.

–

Check Condition status, followed by Unit Attention
sense key in response to the next REQUEST SENSE
command. (Note that Unit Attention may be returned only
once per power-up/bus reset, as described in the SCSI-2
specification.)

Within 15 seconds of power-on, the device must be able to
successfully complete an INQUIRY command so that the
device can be identified.
The VMS operating system does not specify a timeout value
by which devices must respond to normal commands following
power-on, because this value is determined by the user-written
class driver.

5–2 Power-On Protocols and Initialization Procedures

5.4 Device Behavior Following Bus Reset
VMS

A SCSI bus reset may cause some devices such as tape drives to
lose user data. Therefore, only the CPU can set the SCSI RST
signal, and then only as a last resort. The device is prohibited
from setting the SCSI RST signal. The device must behave as
follows after a SCSI bus reset, except when the reset occurs as the
result of power-on.
The device must be able to respond to a SCSI bus selection
within 5 seconds of the bus reset. However, the device is
permitted to return one of the following once it begins to
accept bus selections:
–

Busy status.

–

Check Condition status followed by Not Ready sense key
in response to the next REQUEST SENSE command.

–

CHECK CONDITION status, followed by Unit Attention
sense key in response to the next REQUEST SENSE
command. (Note that Unit Attention may be returned only
once per power-up/bus reset, as described in the SCSI-2
specification.)
The device may not return any other status values or sense
keys unless it is ready for normal operation.
The device must be able to respond to normal commands
within 10 seconds of the bus reset.
The device must be able to handle multiple bus resets in
succession, with no ill effects. The device must recover within the
time limits specified above, following the last bus reset.

Power-On Protocols and Initialization Procedures 5–3

A
Associated Documents
The manuals associated with Digital’s implementation of the
Small Computer System Interface (SCSI) are listed below.
American National Standard for Information Systems—
Small Computer System Interface-2 (SCSI-2), X3T9.2/89-042
specification
The SCSI-2 specification (Revision 10b) is a draft of a proposed
standard. Copies of the draft document may be purchased
from Global Engineering Documents, 2805 McGaw, Irvine, CA
92714, (800) 854-7179 or (714) 261-1455. Refer to document
X3.131-198X.
American National Standard for Information Systems—Small
Computer System Interface (SCSI), X3.131-1986 specification
The document can be ordered from the American National
Standards Institute, Inc., 1430 Broadway, New York, NY,
10018. This document is known as the SCSI-1 standard.
Small Computer System Interface: An Overview, Order No.
EK–SCSIS–OV
Provides an overview of Digital’s implementation of the SCSI-2
requirements. (Available on compact disc.)
This document can be ordered from Digital Equipment
Corporation, P. O. Box CS2008, Nashua, NH 03061. For an
internal order, write Publishing and Circulation Services
(P&CS), NRO3-1/W3, Digital Equipment Corporation,
Northboro, MA 01532.
VMS Version 5.3 Small Computer System Interface (SCSI)
Device Support Manual, Order No. AA–PAJ2A–TE

Associated Documents A–1

Describes the mechanisms the VMS operating system provides
that allow a SCSI device that is not supplied by Digital to be
attached to certain VAXstation and MicroVAX systems.
This document can be ordered from Digital Equipment
Corporation, P. O. Box CS2008, Nashua, NH 03061. For an
internal order, write the Software Supply Business (SSB),
Digital Equipment Corporation, Westminster, MA 01473.
VMS Device Support Manual
Describes the components of a VMS device driver and the
basic rules to which device drivers not supplied by Digital
must adhere.
VMS I/O User’s Reference Manual
Contains information necessary to issue I/O requests from an
application program to I/O device drivers supplied as part of
the VMS operating system.
VMS System Dump Analyzer Utility Manual
Provides assistance in investigating system failures.
VMS Delta/XDelta Utility Manual
Provides information on debugging VMS device driver code.
VAX MACRO and Instruction Set Reference Manual
Describes the VAX MACRO assembly language. VMS device
drivers are written in VAX MACRO.
VMS System Services Reference Manual
Describes the high-level language interface to the I/O
subsystem of the VMS operating system.
IEZ11 Software Installation Guide,
Order No. AA–PA9DA–TE
IEZ11 User’s Guide, Order No. AA–NU45A–TE
IEZ11 Hardware Installation Guide,
Order No. AA–PA9EA–TE
For additional information, refer to:
NCR Corporation. SCSI: Understanding Small Computers.
Englewood Cliffs, N.J.: Prentice-Hall, 1990.

A–2 Associated Documents

B
ANSI SCSI-2 Specification Cross-Reference
Table B–1 is a cross-reference to the draft ANSI SCSI-2
specification. Use the table to find the Digital information in
this manual associated with a numbered SCSI-2 specification
section.
Table B–1

Cross-Reference to ANSI SCSI-2 Specification

ANSI SCSI-2
Specification

Subject

Where to Look
In This Guide

Section 4.1

Physical Description

Section 1.1

Secton 4.2

Cable Requirements

Section 1.2

Section 4.2.1

Single-Ended Cable

Section 1.2

Section 4.3

Connector Requirements

Section 1.3

Section 4.4

Electrical Description

Section 1.4

Section 4.4.1

Single-Ended Signals

Section 1.4.1

Section 4.4.3

Terminator Power

Section 1.4.2

Section 4.4.4

Reserved Lines

Section 1.4.4

Section 4.5

SCSI Bus

Section 1.5

Section 4.6

SCSI Bus Signals

Table 1–4

Section 4.6.1

Signal Values

Section 1.6.1

Section 4.6.1

Signal Sources

Section 1.6.2

Section 4.7

SCSI Bus Timing

Table 1–5

Section 5.1

SCSI Bus Phases

Section 2.1
(continued on next page)

ANSI SCSI-2 Specification Cross-Reference B–1

Table B–1 (Cont.)

Cross-Reference to ANSI SCSI-2 Specification

ANSI SCSI-2
Specification

Subject

Where to Look
In This Guide

Section 5.1.1

Bus Free Phase

Section 2.1.1

Section 5.1.2

Arbitration Phase

Section 2.1.2

Section 5.1.3

Selection Phase

Section 2.1.3

Section 5.1.4

Reselection Phase

Section 2.1.4

Section 5.1.5

Information Transfer Phases

Section 2.1.5

Section 5.1.6

Command Phase

Section 2.1.6

Section 5.1.7

Data Phase

Section 2.1.7

Section 5.1.8

Status Phase

Section 2.1.8

Section 5.1.9

Message Phase

Section 2.1.9

Section 5.1.10

Signal Restrictions Between Phases

Section 2.1.10

Section 5.2

SCSI Bus Conditions

Section 2.2

Section 5.2.1

Attention Condition

Section 2.2.1

Section 5.2.2

Reset Condition

Section 2.2.2

Section 5.3

SCSI Bus Phase Sequences

Section 2.3

Section 5.4

SCSI Pointers

Section 2.4

Section 5.5

SCSI Message System

Section 2.5

Section 5.6

Messages

Section 2.6

Section 5.6.1

ABORT Message

Section 2.6.1

Section 5.6.5

COMMAND COMPLETE Message

Section 2.6.2

Section 5.6.6

DISCONNECT Message

Section 2.6.3

Section 5.6.19

RESTORE POINTERS Message

Section 2.6.4

Section 5.6.20

SAVE DATA POINTER Message

Section 2.6.5

Section 6.1

Command Implementation
Requirements

Section 3.1

Section 6.1.1

Reserved

Section 3.1.1

Section 6.1.2

Operation Code Types

Section 3.1.2

Section 6.2

Command Descriptor Block

Section 3.2

Section 6.2.1

Operation Code

Section 3.2.1

Section 6.2.2

Logical Unit Number

Section 3.2.2

Section 6.2.3

Logical Block Address

Section 3.2.3
(continued on next page)

B–2 ANSI SCSI-2 Specification Cross-Reference

Table B–1 (Cont.)

Cross-Reference to ANSI SCSI-2 Specification

ANSI SCSI-2
Specification

Subject

Where to Look
In This Guide

Section 6.2.4

Transfer Length

Section 3.2.4

Section 6.2.5

Parameter List Length

Section 3.2.5

Section 6.2.6

Allocation Length

Section 3.2.6

Section 6.2.7

Control Byte

Section 3.2.7

Section 6.3

Status

Section 3.3

Secton 6.5

Command Processing Considerations
and Exception Conditions

Section 3.4

Section 6.5.1

Programmable Operating Definition

Section 3.4.1

Section 6.5.2

Incorrect Initiator Connection

Section 3.4.2

Section 6.5.3

Selection of an Invalid Logical Unit

Section 3.4.3

Section 6.5.4

Parameter Rounding

Section 3.4.4

Section 6.5.5

Asynchronous Event Notification

Section 3.5

Section 6.6

Contingent Allegiance Condition

Section 3.6

Section 6.7

Extended Contingent Allegiance
Condition

Section 3.7

Section 6.9

Unit Attention Condition

Section 3.8

Section 7.1

Model for All Device Types

Section 4.1

Section 7.1.1

SCSI Addresses

Section 4.1.1

Section 7.1.2

Commands Implemented by All
SCSI Devices

Section 4.1.2

Section 7.2

Commands for All Device Types

Section 4.2

Appendix F

Power-On Protocols and Initialization Chapter 5
Procedures

ANSI SCSI-2 Specification Cross-Reference B–3

Glossary
ANSI
American National Standards Institute.
arbitrate
To select one device from several devices that are seeking access to the SCSI bus
concurrently.
assertion
The process of driving a signal low, to the true state (0 V–0.5 V).
asynchronous event notification (AEN)
A SCSI protocol that allows a SCSI device (usually a target) to inform the processor
(usually the initiator) that an event has occurred asynchronously with respect to the
processor’s current stream of execution.
AWG
American Wire Gauge.
bus
A flat or twisted-pair cable composed of individual identical circuits. A computer’s
SCSI port and SCSI peripheral devices can be installed anywhere along the cable.
bus free
No SCSI device is actively using the bus; therefore, the bus is available for use.

Glossary–1

byte
A binary character string made up of eight bits.
central processing unit (CPU)
The part of the computer system that controls the interpretation and execution of
instructions.
CISC
Complex Instruction Set Computer.
class driver
A component of the SCSI class/port architecture that acts as an interface between
the user and the SCSI port. In the VMS implementation, the SCSI class driver
translates I/O functions as specified in a user’s $QIO request to a SCSI command
targeted to a device on the SCSI bus.
Common Command Set (CCS)
A subset of eighteen commands in the SCSI command set. CCS is a proposal before
the ANSI X3T9.2 subcommittee.
command descriptor block (CDB)
The structure used to communicate requests from an initiator to a target.
connect
The function that occurs when an initiator selects a target to start an operation. A
connection can only occur between an initiator and a target.
contact
The electrically conductive portion of a connector associated with a single conductor
in a cable.
controller
A computer module that interprets signals between the host and a peripheral
device.
CPU
See central processing unit.

Glossary–2

device
The general name for any unit connected to the system that is capable of receiving,
storing, or transmitting data. See also controller.
device driver
A set of routines and tables that the system uses to process an I/O request for a
particular device type.
disconnect
The action that occurs when a SCSI device releases control of the SCSI bus, allowing
it to go to the Bus Free phase.
DMA
Direct Memory Access.
field
A group of one or more contiguous bits. Fields containing only one bit are usually
referred to as the xx bit instead of the xx field.
EMC
Electromagnetic compatibility.
ESD
Electrostatic discharge.
host adapter
A device that connects a host system to the SCSI bus. The device usually performs
the lower layers of the SCSI protocol and normally operates as an initiator. This
function may be integrated into the host system.
initial connection
A connection that exists from the assertion of the BSY signal in a Selection phase
until the next Bus Free phase occurs.
initiator
A SCSI device (usually a host system) that requests an operation to be performed by
another SCSI device (a target).

Glossary–3

invalid
Illegal (reserved) or unsupported. Refers to an illegal or unsupported bit, field, or
code value.
I/O process
A process consisting of one initial connection and zero or more reconnections, all
pertaining to a single command. An active I/O process is presently in execution.
The current I/O process is the one that is presently connected on the SCSI bus. I/O
connections pertain to a nexus in which one or more command descriptor blocks
are usually transferred. An I/O process begins with the establishment of a nexus,
and normally ends with the Bus Free phase following the successful transfer of a
COMMAND COMPLETE, a RELEASE RECOVERY, or an ABORT message. An
I/O process also ends when a hard Reset condition occurs, an unexpected Bus Free
phase occurs, or when the Bus Free phase occurs following a BUS DEVICE RESET
message.
I_T nexus
A nexus that exists between an initiator and a target.
I_T_x nexus
A nexus that is either an I_T_L or I_T_R nexus.
interrupt priority level (IPL)
The level at which a software or hardware interrupt is generated on a VAX system.
There are 32 interrupt priority levels: IPL 0 is the lowest, 31 is the highest. The
levels arbitrate contention for processor service. For example, a device cannot
interrupt a processor if the processor is currently executing at an IPL greater than
the IPL of the device’s interrupt request.
logical unit
A physical or virtual peripheral device that is addressable through a target.
logical unit number (LUN)
An encoded 3-bit identifier for the logical unit.
LSB
Least significant byte.

Glossary–4

LUN
See logical unit number.
MSB
Most significant byte.
negation
The process of driving a signal high, to the false state (2.5 V–5.5 V). The cable
terminators bias the signal by placing the driver in a high impedance condition.
nexus
The point at which an initial connection is established and that ends with the
completion of the I/O process. A nexus may specify a single logical unit or target
routine by the successful transfer of an IDENTIFY message.
OEM
Original Equipment Manufacturer.
one
A true signal value.
open system
A vendor-neutral computing environment that is compliant with international
standards, permits system and network interoperability or software applications
portability, including consistency of data and human access, and satisfies one or
more of a business’s functional requirements.
page
Parameter structures used by several commands. Pages are identified with a value
known as a page code.
parameter
A structure that contains one or more fields.

Glossary–5

peripheral device
A physical device that can be attached to a SCSI device, which in turn connects
to the SCSI bus. It provides the CPU with additional memory storage or
communications capability. Examples of peripheral devices are: magnetic disks,
printers, optical disks, and magnetic tapes.
port
See SCSI port
port driver
A component of the VMS SCSI class/port architecture that transmits and receives
SCSI commands and data. It knows the details of transmitting data from the local
processor’s SCSI port hardware across the SCSI bus.
priority
The ranking of devices on the bus during arbitration.
processor
A functional section of hardware in a computer that changes instructions into a
form the computer understands, and executes these instructions.
reconnect
The act of reviving a nexus to continue an I/O process. A target reconnects to an
initiator by using the Reselection and Message In phases after winning arbitration.
An initiator reconnects to a target by using the Selection and Message Out phases
after winning arbitration.
reserved
The term used for bits, fields, and code values that are set aside for future
standardization.
RISC
Reduced Instruction Set Computer.
saved pointers
A tracking process in which one pointer is for the command, one pointer is for the
data, and one pointer is for the status.

Glossary–6

SCSI
An interface designed for connecting peripheral devices to computer systems. SCSI
is defined by an American National Standards Institute (ANSI) standard; in this
guide, SCSI refers to either SCSI-1 or SCSI-2.
SCSI-1
ANSI Small Computer System Interface (X3.131–1986).
SCSI-2
Proposed ANSI Small Computer System Interface-2 (X3.131–198x).
SCSI device
A host computer adapter, a peripheral controller, or intelligent peripheral that can
be attached to the SCSI bus.
SCSI device address
The octal representation of the unique address (7–0) assigned to a SCSI device. This
address is normally assigned and set in the SCSI device during system installation.
SCSI device ID
The bit-significant representation of the SCSI address referring to one of the signal
lines DB(7–0).
VMS systems: Typically, both the MicroVAX 3100 and VAXstation 3100 systems
are assigned device ID 6 and assert at DB(6); the VAXstation 3520/3540 CPU is
assigned device ID 7 and asserts at DB(7).
ULTRIX systems: The VAXstation 3100, VAXstation 3520/3540, DECstation 2100
and 3100 systems, as well as the DECsystem 3100, are assigned device ID 6, and
the ULTRIX kernel asserts at DB(7). ULTRIX customers can reconfigure the SCSI
ID to their needs.
SCSI port
Software: the channel that controls communications to and from a specific SCSI bus
in the system. Hardware: The name of the logical socket at the back of the system
unit to which a SCSI device is connected.
signal release
The act of allowing the cable terminators to bias the signal to the false state by
placing the driver in the high impedance condition.

Glossary–7

SII
System Interface Interconnect, a SCSI controller chip.
status
One byte of information sent from a target to an initiator upon completion of each
command.
System Interface Interconnect
A gate array developed by Digital that manages the SCSI bus and the following
activities: SCSI bus selection, bus to memory transfers, and DISCONNECT and
RECONNECT commands.
target
A SCSI device that performs an operation requested by an initiator.
unexpected disconnect
A disconnection that occurs as a result of a protocol error.
vendor-specific
A term that refers to bits, fields, or code values that are not defined in this guide,
and that may be used in a different way in vendor implementations.
zero
A false signal value.

Glossary–8

Index

Allocation length
description of, 3–7
Asynchronous event notification
description of, 3–12
device availability notification, 3–12
error condition, 3–12
how it works, 3–13
state changes, 3–12
Unit Attention condition, 3–12

Bus phases (Cont.)
Data Out, 2–7
Information Transfer, 2–5
Message, 2–2, 2–7
Reselection, 2–2, 2–4
Selection, 2–2, 2–3
sequences, 2–11
signal restrictions between, 2–7
signal sequence diagram, 2–12
Status, 2–2, 2–7
transition sequences, 2–10

Bus
communication on, 1–19, 1–20
initiator and target, 1–21
SCSI, description of, 1–19
signals, 1–22
timing values, 1–25, 1–26
Bus conditions
asynchronous, 2–8
Attention, 2–8
Reset, 2–8
Bus phases
Arbitration, 2–2, 2–3
Bus Free, 2–2, 2–3
Command, 2–2, 2–6
Data, 2–2
Data In, 2–7

Cable
50-conductor A cable, 1–2
50-conductor flat, 1–3
Digital recommendation, 1–3
impedance, 1–2
length of, 1–2
ordering information, 1–4, 1–5
requirements, 1–2
25-signal twisted-pair, 1–3
CDB
command descriptor block, 3–3
Command
exception conditions, 3–10
for most device types, 4–5
implementation requirements, 3–2
implemented by SCSI targets, 4–2

Index–1

Command (Cont.)
INQUIRY, 4–3
operation code, 3–5
operation code types, 3–2
parameter rounding, 3–11
processing considerations, 3–10
RECEIVE DIAGNOSTIC RESULTS,
4–3
REQUEST SENSE, 4–3
reserved bits, field, bytes, values, 3–2
selection of invalid logical unit, 3–10
SEND DIAGNOSTIC, 4–4
TEST UNIT READY, 4–4
Command descriptor block (CDB)
definition of, 3–3
for 10-byte commands, 3–4
for 6-byte commands, 3–3
Connector
nonshielded, 1–5
68-pin, pin assignments, 1–15
68-pin, systems used on, 1–12
68-pin assembly, 1–12
68-pin configuration, 1–13
50-pin nonshielded cable, 1–6
50-pin nonshielded device, 1–6
50-pin nonshielded pin assignments,
1–10
50-pin shielded cable, 1–8
50-pin shielded device, 1–8
50-pin shielded pin assignments,
1–11
68-pin strain relief, 1–14
requirements, 1–5
reserved lines, 1–19
shielded, 1–5
Contingent Allegiance condition
description of, 3–14
Control byte
description of, 3–7
Cross-reference
to ANSI SCSI-2 specification, B–1

D
Development considerations, ix, xi

Index–2

Device address, 4–1
Device type
characteristics, 4–1
Diagnostics
implementation, 4–4
Documentation
where to order, A–1

E
Electrical description
reserved lines, 1–19
Electrical specifications, 1–16
Error condition
AEN, 3–12
source of, 5–2
unexpected disconnect, 2–3
Extended Contingent Allegiance
condition
unsupported, 3–14

H
Host adapter
systems with, 2–6

I
Incorrect initiator connection, 3–10
Information transfer
asynchronous, 2–6
bus phases, 2–5
synchronous, 2–6
Initialization procedures, 5–1

L
Logical block address
description of, 3–6
Logical unit
characteristics of, 4–2
invalid, 3–10
programmable operating definition,
3–10
target does not support, 3–10
target supports, 3–10

Logical unit (Cont.)
target supports attached, 3–11
target supports unattached, 3–11
Logical unit number
description of, 3–5

Power-on protocols (Cont.)
dynamic reconfiguration, 5–1
external boxes, 5–1

SCSI
addresses, 4–1
bus signals, 1–22
cable requirements, 1–2
configuring a device, 1–20
daisy-chained devices, 1–2
Digital’s implementation, viii
Digital’s systems with, 2–6
electrical specifications, 1–16
messages, list of, 2–21
sample configurations, 1–21
SCSI-2 specification, x, xi
SCSI bus
timing values, 1–26
SCSI ID
assignments, 1–20
Signal
SCSI bus, 1–22
single-ended, 1–16
sources, 1–23, 1–24
termination, 1–18
values, 1–23
Standards
conformance to, xi
Status byte
code description, 3–9
description of, 3–8

Message
ABORT, 2–22
codes, mnemonics, 2–18
COMMAND COMPLETE, 2–22
DISCONNECT, 2–23
extended, codes, 2–19
extended, format, 2–19
format, 2–16
implemented by Digital, 2–21
implemented by VMS, 2–19
RESTORE POINTERS, 2–14, 2–23
SAVE DATA POINTER, 2–14, 2–23
SCSI message system, 2–16
table of codes, 2–17

O
Opcodes
types of, 3–2
Ordering information
cables, 1–4, 1–5
documentation, A–1
terminators, 1–16

P
Parameter list length
description of, 3–7
Physical specifications, 1–2
Pin assignments
See Connector
Pointers
saved, 2–14
saved, diagram of, 2–14
Power-on protocols
bus reset, 5–3
device behavior, 5–2

T
Target routines
characteristics of, 4–2
Terminator
ordering information, 1–16
TERMPWR pin, 1–17
Transfer length
description of, 3–6

Index–3

U
Unit Attention condition
description of, 3–14

V
VMS
development considerations, ix

Index–4