SQ706A_SCSI_MSCP_Disk_Controller_Oct89.pdf

SQ706A_SCSI_MSCP_Disk_Controller_Oct89.pdf
MODEL SQ706A
DISK CONTROLLER
(HOST ADAPTER)
FOR
SCSI INTERFACE
INSTALLATION AND OPERATION MANUAL
16 October 1989
~II n~~
I
I
DISTRI.BUTED. LOGIC CORPORATION
Street
~565 i~xs~~~gir
Anaheim. California 92806
Telephone: (714) 937·5700
Telex 6836051
TA~LE
OF CONTENTS
PAGE
SECTION 1
DESCRIPTION
. . . . . . . . .. . . . . . . . .
CHARACTERISTICS. • • . • • • • • • • • •
Supports 22-Bit Addressing • • . • • ••
Supports SCSI Common Command Set For
Disk Drives: • • . • • . • • •
SCSI Pass-Through Mode •• ~'. i',... •
Full S,CSI Implementation ••• • y..
Onboard Formatting
Automatic Diagnostic Self Test • • • ••
Onboard Bootstrap PROM . • • • • • • ••
Reassignment of Bad Blocks
16-Entry Command Queue • • • • • • • • •
Single SCSI Command. • • • • • •
Disk Drives Supported.
• • • •
CONTROLLER SPECIFICATIONS
• • •
SECTION 2
INSTALLATION
........
OPERATION • • • • • • •
1-2
1-2
1-21-2 ,
1-2
1-2
1-2
1-3
1-3
1-3
1-7
2-1
2-4
.....
INITIATING COMMUNICATION
Via Controller Serial Port • • • • •
Via Virtual Terminal • • • • • • •
Procedure for Bringing Up··the
Virtual Terminal Via the LSI~11
With Host Adapter Boot Enabled
Procedure for Bringing Up the
Virtual Terminal Via the LSI-11
With Host Adapter Boot Disabled.
Procedure for Bringing Up the
Virtual Terminal Via the
MicroVAX II •
• •••
CONFIGURATION MENU • • • • • • • • •
Standard Diagnostics . . • • • • • •
Host Memory DMA Test • • • • • • .
Display Controller Characteristics •
Set Controller Characteristics •
How to Find Current Address .
For LSI-11 System
For MicroVAX II System .
SCSI Direct Command
Format SCSI Drive .' • • • • • •
SCSI Unit Mapping
ii
1-2
1-2
2-1
HARDWARE BOOTSTRAP JUMPERS
CONTROLLER INSTALLATION
SECTION 3
1-1
3-1
3-1
3-1
3-2
3-2
3-3
3-4
3-5
3-5
3-6
3-8
3-9
3-10
3-10
3-11
3-13
3-17
3-19
TABLE OF CONTENTS
(Continued)
PAGE
SECTION 4
SECTION 5
·....
DIAGNOSTICS
4-1
SETU P AND SELF TEST .
•.•.
FRONT END TEST-ZRCFB3 . . . . . . . . .
DISK EXERCISlrR-ZRCDA 1 . . . . • . /....
MICROVAX II MDM-KDA50-Q .
. . . ,.. •
Fun c t ion a1 Test . . • . • . t. (1 • •
Exerciser
• • . . • . • •
Utility . . . .
4-1
4-2
4-4
4-8
4- 8
4-9
4-9
·........
5-1
ERRORS AND TROUBLESHOOTING
GENERAL . . • . . . • . . .
ERROR LOGS . . • • . • . • • • .. •
SCSI Host Memory Access Error Log
Disk Transfer Error Log • • • • .
SDI Error Log . . . . . • . • • .
Self-Detected Fatal Port/Controller
Errors
• . • •. • • • • • •
5-1
5-2
5-2
5-3
5-4
5-5
LIST OF ILLUSTRATIONS
Figure 1-1
Disk
Figure 2-1
Figure 2-2
Figure 2-3
Controller Configuration • • • •
MicroVAX II Backplane (Typical)
MicroVAX II H9278 Backplane • • •
Syste~,
Simplified . . .
~
1-1
2-2
2-5
2-5
LIST OF TABLES
Table 1-2
Table 1-3
Controller/Q-Bus Interface Lines (Dual
Module)
Controller to Drive - J2
Controller to Terminal - J4
1-4
1-5
1-6
Table 2-1
Jumper
2-3
Table 3-1
IP and
Table 1-1
.......··
·
·
Configuration .
···.·....
SA Hex Addresses
.·..
·
iii
3-12
SECTION 1
DESCRIPTION
This manual describes the installation and operation of Distributed
Logic Corporation (DILOG) Model SQ706A Disk Controller.
This controller is a host adapter which may communicate wi th ei ther an embedded
controller in the ~rive or a bridge controller, which is a controller
that communicates with more than one drive. -The controller is a dualmodule board that interfaces up to seven SCSI disk drives with DEC*
LSI-11/23, -11/23 PLUS, 11173, MICRO/PDP-11 and MicroVAX II computer
systems.
The controller is software compatible with DEC DU drivers
(MSCP) in RT-11, RSX-11M+, RSTS/E, MicroVMS, ULTRIX, UNIX, and MUMPS.
The controller supports both block mode and non-block mode memory.
Figure 1-1 is a simplified diagram of a SCSI disk system.
MICROVAX I I,
MICRO/PDP-11
LSI-11 COMPUTERS
~
Q-BUS
~ SINGLE-ENDED " J2 ~50-PIN CABLE
-"""II1II
.....
rV
0
CONTROL( \4)
CONl'ROL(15)
SCSI
CRIVE
~
...
~
SCSI
CAlVE
1
a:
l;t
w
CATA( Ie)
-I
-I
~
W
~ AOORESS(22)V
()
<
IJ..
a:
w
;-
-"""II1II CONTROLca)
~
......
a:
..-:J
W
-"""II1II
~
POWER
CiROUND
...
a.
::E
0
~
U
w
()
<
IJ..
0
..-a:Z
nI
SCSI
CRIVE
2
a:
0
..W
(J
a:
~
0
w
>
(J)
(J)
~
n/
SCSI
ORlve
3
a:0
w
(J
0
a.
0
a:
~
~
0
a:
(J
'"
~
SCSI
~
CAlVE
~
CRIVE
~
CRIVE
4
~
;~
SCSI
5
t
. ~
TERMINAL
J4
-
Figure 1-1.
SCSI
6
Disk System, Simplified
~~---------~----------------~------------------------- --------------
*DEC LSI-11, MICRO/PDP-11, MicroVAX II, RT-11, RSX, RSTS/E, MICRO/VMS,
MSCP, and DU Driver are registered trademarks of Digital Equipment
Corporation.
UNIVERSAL FORMATTING is a trademark of Distributed Logic Corporation.
1-1
CHARACTERISTICS
Characteristics of the controller are as follows:
o
SUPPORTS 22-BIT ADDRESSING
The controller supports 16-, 1B-, and 22-bit Q-bus addressing.
o
SUPPORTS SCSI COMMON COMMAND
The controller is compatible with SCSI Co~m;n Command Set for
disk drives and connects up to seven SCSI' ul1its (either disks
with embedded SCSI interface or SCSI controller/disk combinations).
The following SCSI commands are, used by the host
adapter:
FORMAT
INQUIRY
MODE SENSE
READ
READ CAPACITY
o
REASSIGN BAD BLOCKS
REQUEST SENSE
START/STOP
TEST UNIT READY
WRITE
SCSI PASS-THROUGH MODE
The controller has.a special mode which permits passing SCSI
commands di rectly from the host through the controller to the
desired unit.
o
FULL SCSI IMPLEMENTATION
The controller offers a full SCSI implementation including disconnect/reconnect and synchronous data transfers.
o
ONBOARD FORMATTING
The onboard formatter uses either the system console or a DLV11compatible connector to provide interactive terminal access.
The
formatter is menu driven and also provides controller and drive
test options.
o
AUTOMATIC DIAGNOSTIC SELF TEST
The controller is suppl ied wi th an onboard automati c sel f-test
diagnostic that is activated each time power is applied.
These
diagnostics are run repeatedly until the controller is brought
online.
o
ONBOARD BOOTSTRAP PROM
The controller contains onboard bootstrap support for RP02,
RL01/02, RM02, RM05, RMBO, RK06/07, RX02, TS-", RSV05, and DU
driver devices.
Onboard jumpers allow selectable bootstrap
addresses, in addition to enabling/disabling the bootstrap.
1-2
o
REASSIGNMENT OF BAD BLOCKS
The controller will handle media flaws by issuing a SCSI Reassign
command.
o
16-ENTRY COMMAND QUEUE
The controller can queue up to 16 commands for up to seven units.
Each unit has a command queue so that the 16 commands may be distributed among the drives in any fashion.
o
SINGLE SCSI COMMAND
The configuration menu allows a single SCSI command to be manually entered and sent to the target.
o
DISK DRIVES SUPPORTED
The controller is compatible with ANSI SCSI spec X3T9.2/85-52 Rev
4B for Direct Access devices.
The coupler is compatible with
disk drives from such manufacturers as:
CONTROL DATA
IOMEGA
MICROPOLIS
PRIAM
TOSHIBA
FUJITSU
KODAK
MINISCRIBE
RODIME
HITACHI
MAXTOR
NORTHERN TELECOM
SIEMENS
For additional drive support, contact the factory.
Table 1-1 lists the Controller/Q-Bus Interface Signals, Table 1-2
lists the Controller/Drive Interface Signals, and Table 1-3 lists the
10-Pin Connector Signals to the terminal.
1-3
Table 1-1.
BUS PIN
I
Controller/Q-Bus Interface Lines (Dual Module)
,
I INPUTI I
IMNEMONIC I OUTPUT I DESCRIPTION
AJ 1, AM 1, BJ 1,
BM', BT', BC2
GND
AN1
BDMA L
Signal Ground and DC return.
0
I
0
I
N/A
Stops program execution. Refresh and OMA Is enabled. Con·
sole operation Is enabled.
AR1
BREF L
N/A
Memory Refresh. Used for Block Mode OMA.
BA1
BDCOK H
I
8Bl
BPOK H
i
eN1
BSACK L
0
SR'
SeVNT L
N/A
BV', AA2, BA2 .
+5
I
+ 5 volt system power.
A02, BD2
+12
N/A
+ 12 volt system power.
AE2
BDOUT L
I/O
Data Out. Valid data from bus master Is on the bus. Inter·
locked with BRPLY.
AF2
BRPLY L
110
Reply from slave to BDOUT or 80!N and durIng IAK.
AH2
BDIN L
I/O
Data Input. Input transfer to master (states master Is ready
tor data). Interlocked with BRPLY.
AJ2
BSYNC L
1/0
Synchronize: becomes active when master places address on
bus; stays active during transfer.
I
I
BWTBT L
I/O
,BIAQ4·7 L
0
AM2
AN2
BIAKll L
BIAK10 L
I
0
AT2
BINIT L
AU2, AV2, BE2, BF2, BDALO L
BH2, BJ2, BK2, BL2, through
8M2, BN2, BP2, BR2, BDAL 15 L
t
BS2, BT2, 8U2, BV2
; .~
I
I
DC power OK. All DC voltages are normal.
I Prlmaiy power OK. vVnen iow activates power fall trap
sequence.
Select Ackn9wledge. Interlocked with BOMGO Indicating
controller Is bus master In a OMA sequence.
,
External Event Interrupt Request.
Write Byte: Indicates output sequence to follow (DATO of
DATOB) or marks byte address time during a DATOB.
' Interru p t Re q uest 4·7.
Serial Interrupt Acknowledge Input and output lines routed
from a Bus, through devices, and back to processor to
establish an Interrupt priority chain.
Initialize. Clears devices on 110 bus.
1/0
AR2
AS2
BOMGll V'
BDMG10 L
I
0
AP2
BBS7 L
110
AC1, AD1, BC1, B01, BDAL16 L
BE1, BFl
·BDAL21 L
!
D!roct Mamoii ACCeSS (DMA} i6Qu65t irum controiier; active
low.
BHALT L
AL2 A1 AB1 BP'
I
J
AP1
AK2
t
I
0
Data/address lines, 0·15
DMA Grant Input and Output. Serial OMA priority line from
computer, through" devices and back to computer.
Bank 7 Select. Asserted by bus master when address In upper
4K bank is placed on the bus. Also asserted for Block Mode
DMA.
Extended
1-4
Add~ess
Bits 16·21
j
I
Table 1-2.
Controller to Drive - J2
I PIN NUMBER 1
DESCRIPTION
============1==============1========================================
SIGNAL
DO L
D1 L
D2 L
D3 L
"D4 L
D5 L
D6 L
D7 L
DP L
GROUND
GROUND
GROUND
PWR H
GROUND
GRQUND
ATN L
GROUND
BSY L
ACK L
1
r
I
:
2
4
6
8
10
12
14
16
18
20
22
24
26
28
30
32
34
36
38
RST L
MSG L
SEL L
40
42
44
C/D
46
REO L
48
I/O
50
1
1
t
:
:
I
Data
Data
Data
Data
Da ta
Data
Data
Data
Data
Bus
Bus
Bus
Bus
Bus
Bus
Bus
Bus
Bus
Power OK.
on
0 (LSB)
1
2
3
4
5
6
7 (MSB)
Parity
Green LED on controller is
Attention
I
I
-
I Busy - bus is being used
I
Acknowledge for request data transfer
handshake
Reset
Message being transferred
Select - selects drive or reselects
controller
Control/Data - low indicates control;
high indicates data
Request - request for a request acknowledge data transfer handshake
Input/Output - indicates direction of
~ata movement; low to controller;
high to drive
-~-----~------------~--------------------------------- --------------
All odd pins are ground.
1-5
Table 1-3.
Controller to Terminal -
J~
r===;i;~~~---i--~i~-=I==;i~==~=--------~~Fi~i~I;~-~=-------------====1
'=============1=======1=======
======================================="
UARTCLK_L I
0
:
1
(TTL) UART clock out. Normally inacI
I
I
tive
! !
output if not in THODE.
t-l""!:ln~it-ir.n
~h,f't-~
!""In1-
A
rt'T")
di~pl;y~i;f;~~;ti;~ ;~-TXD*~;~d~;hift
in switch status on RXD*. Open if
not used.
,
SGND
(0)
nr.cdt-iHO
o
o
2
Signal Ground and logic Ground.
3
(RS232) Transmit data in THODE.
..
Active (OJ in Status MODE for ACTIVITY
indication or front panel display
shift data when DTR is inactive (0).
Normally inactive (1).
I
4
(RS232) Data Set Ready if used in
TMODE. Active (1) to indicate a
switch has changed in Status MODE
while DTR is inactive. Open if not
used.
o
5
(RS232) Data Terminal Ready if used
in TMODE. Active in Status MODE to
load input shift register, inactive
to shift data in on RXD* or to enable
ACTIVITY LED.
6
Indexing Key - No Pin for RS232
NOT USED
TMODE*_L
I
7
(TTL) Terminal MODE when grounded.
Enables terminal, diagnostic and formatter functions. Status/Activity
MODE when open. Onboard 1K ohm or
greater pullup.
RXD* L
I
8
(RS232) Receive Data in THODE. Active
(0) in Status MODE for front panel
input shift data when DTR is active.
Normally inactive (1).
SGND
o
9
Signal Gropnd and logic Ground
+5V
o
10
+5 Volt with 1 Amp Fuse F1. Fused +5
Volt supply through JP4 2-3 jumper
Jumper JP4 1-2 to supply +12 Volt for
DLV11-KA RS232 to 20MA adapter.
~~~~-~~--~~-----~~~----~~~-------------------~-~------ ---------------
, -6
I
.
,I
I
j
CONTROLLER SPECIFICATIONS*
MECHANICAL
The controller is completely contained on one dual height module
5.22 inches (13.2 cm) wide by 8.88 inches (22.56 cm) high and
plugs into one dual slot in a Q-bus backplane.
BASE ADDRESS - Factory set at 160334
User selectable address range 160000 - 177774
IP
SA
760334
760336
760340
760342
760344
760346
760354
760356
Consult the factory for additional address requirements
INTERRUPT VECTOR ADDRESS - Host programmable 0-774
NUMBER OF DRIVES SUPPORTED - 8 units on 7 nodes
ADDRESS RANGE - Q-bus memory to 4 Mbytes
DRIVE TRANSFER RATE
Up to 2.0 Mbytes per second Asynchronous.
Up to 3.0 Mbytes per second Synchronous.
PRIORITY LEVEL - Factory set at BR4, user selectable BR5, 6 and 7
DISK DRIVE I/O
ANSI SCSI spec X3T9.2/85-52 Rev
Access devices.
4B
Common Command Set for
~irect
CONNECTOR
A 50-pin ribbon cable type mounted on outer edge of controller
module. Mate is 3M 3452-5000 or equivalent.
POWER - +5 volts at 2.5 amps
ENVIRONMENT
Operating temperature 50 deg. F to 104 deg. F, humidi ty 10% - 90%
non-condensing~ /
SHIPPING WEIGHT - 5 pounds, including documentation
*
Specifications subject to change without notice.
1-7
SECTION 2
INSTALLATION
The padded shipping carton contains the controller board, and if
specified on the sales order, a 50-pin control and data cable to the
first drive, optional data cables to other drives.
Inspect the controller board and its components and the cables for damage.
NOTE
If damage to the board, components on~ t;he board,
or cables is noted, do not install.
Immediately
inform the carrier and DILOG.
HARDWARE BOOTSTRAP JUMPERS
The enabling or disabling of the hardware bootstrap and the selection
of the bootstrap address are coupler parameters that need to be configured before the coupler is installed in the computer backplane.
The coupler contains a bootstrap PROM that can be enabled or disabled
by jumper JP2.
If enabled, the bootstrap address can be changed by
jumper JP1.
NOTES
1.
Installation in MicroVAX II systems requires disabling the bootstrap; remove JP2.
2.
With JP2 removed, the JP1
coupler operation~
3.
If jumper( s) are removed, it is recommended that
they be rotated 90 degrees for storage to prevent
loss, i.e., install with one jumper pin inserted
over only one pin of the jumper location.
jumper has no effect on
Figure 2-1 illustrates the location of the jumpers.
Table 2-1 shows
the jumper position possibilities.
The controller is shipped with
jumpers installed with the bootstrap PROM enabled and bootstrap
address 175000 (octal) selected.
2-1
~~~~~IV~------------~I!~V~~~
I
IL
I
~I
.
W~i
J2
_ _ _ _ _ _ _ _ _ _---oi
i
J4
~M
W
I
!
0
vlP4 JJ111
r:::::J
L1L2 ~
JP10
,---,
I.......I(JJ
o
~JP8
'-.I
o
m
-4
:::0
l>
~
(JJ
C
-4
rn
o
r
o
G1
~
(j
:n
o
n
3
2
0
~'JP3
DJP13
rn
1
JP7
(J)
........
("")
o
:n
1)
n
l
o1)
-<
:n
H
(j')
I
-I
@
~
<D
Ifg
DJP2
I
I
CJJP1
I
----'n'--__--I~
'1,----_ _
Figure 2-1.
Controller
2-2
Configura~ion
Table 2-1.
Jumper Configuration
Jumper J P1
BOOT SELECT
Installed = Boot 775000
Removed
= Boot 773000*
Jumper JP2
BOOT ENABLE
Installed
Removed
= Enable Boot
= Disable Boot*
(must be removed for MicroVAX II systems)
Jumper J P3
FACTORY USE ONLY
Jumper JP4
Installed (etched)
Removed (etch cut)
= +5V for J4 connector
= +12V for J4 connector
Jumper J P7
FACTORY USE ONLY
Jumper JP8, JP10
Installed
= Ilo
Ground connected to Ground (Logic)
= I/O
Ground connected to Shielded Handle
J urn pe r J P11, J P12
Installed
Jumper J P13
FACTORY USE ONLY
Jumper J P14
Installed = I/O Ground connected to Cable Shield.
(Pin 1 of SCSI Interface.)
* When jumper plugs are not installed, they may be retained on the
board by rotating the plug on one pin 90 degrees.
2-3
CONTROLLER INSTALLATION
After the
follows:
jumpers
have
been
positioned
install
the
controller
as
CAUTION
ENSURE ALL POWER IS OFF BEFORE INSTALLING THE CONTROLLER OR CABLES.
DAMAGE TO THE BACKPLANE ASSEMBLY WILL OCCUR IF THE
CONTROLLER IS PLUGGED IN BACKWARDS.
1.
Select the backplane location into which the controller is to be
inserted.
There are several backplane assemblies available from
DEC and other manufacturers.
Figures 2-2 and 2-3 show typical
backplane configurations.
It is important that all option· slots between the processor and
the disk controller be filled to ensure that the daisy-chained
interrupt (BIAK) and DMA (BDMG) signals be complete to the controller slots.
If there must be empty slots between the controller and any option board, the following backplane jumpers must be
installed:
FROM
CO X NS
CO X S2
TO
SIGNAL
CO X M2
CO X R2
BIAK1/LO
BDMG1/LO
2.
Perform this step if the serial port is· to be connected. Connect
the cable from J4 on the controller to the terminal.
Refer to
Table 1-3 for pinouts and descriptions.
3.
Install one end of the cable into the J2 connector on the controller.
ENSURE PIN 1 ON THE CABLE IS MATCHED WITH THE TRIANGLE
ON EACH CONNECTOR AS INDICATED ON FIGURE 2-1.
4.
Ensure the controller is oriented with the components facing row
one, the processor, and gently press both sides until the module
connectors are firmly seated in the backplane.
5.
Connect the other end of the cable to the drive or drives if
daisy-chained,. ~ Ensure the terminator is installed in the last
d r i v e.
;~.
6.
Refer to the disk drive manual for' operating
apply power to the drive(s) and the computer.
2-4
instructions,
and
•
u
lOTI:
eo.,aDeGc. os tbe board .u.c
be
tae1ac toward.
tM PrOC:•• 8Or.
Figure 2-2.
MicroVAX II Backplane (Typical)
DlI!lQ:)RMECT stars (Q/CD)
(J.Lt. cmtEIS Q-IGS stO!S)
noc:r...ssoa
J4D
DCJDc:::J
Figure 2-3.
M1croVAX II H9278 Backplane
2-5
7.
Power-up the system.
Note that there are two green LEDs.
The
leftmost LED is the POWER LED; the rightmost LED is the ACTIVITY
LED.
If the Power LED lights, power to the SCSI bus terminators
has been properly applied; if the Power LED does not light, the
SCSI bus terminators have not received power.
If this should
happen, check the power supply on the system or the fuse on the
controller.
The user should note that the Power LED will turn ON
if either a SCSI device with power applied is connected to the
controller or power is applied to the controller.
This is in accordance with the SCSI specification, which enables bus terminator power to be suppl ied by ei ther the SCSI dev ice or host
adapter.
8.
Upon a good power-up (i.e., the POWER LED lights), the ACTIVITY
LED should turn on for 10 seconds and turn off for 10 seconds.
Each 10-second transition of ON/OFF or OFF/ON of the ACTIVITY LED
indicates a successful pass of the controller self tests.
The
controller will continue to perform its onboard self tests upon
power-up until brought online by the system (i.e., the ACTIVITY
LED will continue to flash on and off every 10 seconds).
After
it is brought online, the controller will turn on its ACTIVITY
LED only during read and write activity between the controller
and the SCSI device.
9.
If an error is encountered at any point during the self tests, a
5-bit binary error code will be flashed on the ACTIVITY LED, signifying which error has occurred.
(See ERROR CODES for details.)
The most significant bit of the error code is flashed first.
Along flash indicates a binary "1", while a short flash indicates
a binary "0."
This error code will continue to flash until the
board is either powered down or reinitialized.
10.
The system is now ready
cribed in Section 3.
to
operate.
2-6
Format
the
disks
as
des-
SECTION 3
OPERATION
This section covers the operation of the system, including establishing communication with the computer, the controller (host adapter),
and the drives; bootstrapping for MSCP as well as other emulations;
and formatting.
Diagnostics include ZRCFB3, ZRCDA1, and
Section 4 for details.
Error codes
Section 5.
from
the
error
logs
and
MicroVAX/~I;
their
MDM KDA50-Q.
formats
are
listed
See
in
INITIATING COMMUNICATION
VIA CONTROLLER SERIAL PORT
In order to bring up communication via the serial port, a terminal
must be connected to the 10-pin connector (J4) located on the controll e r boa rd.
S e ria 1 communi cat ion s will t h en be gin aut om at i cally
upon power up of the controller or re-boot of the host system.
Serial
communications take place at a rate of 9600 baud; therefore, the terminal must be set to both transmit and receive at this rate.
The
seri al communi ca ti on protocol is 1 start bi t, 1 stop bi t, 8 da ta bi ts
with no parity.
3-1
VIA VIRTUAL TERMINAL
Procedure for Bringing Up
Adapter Boot Enabled
the Virtual
Terminal
Via LSI-11
Wi th Host
In order to bring up communication via the virtual terminal, the
system console is used as the terminal for serial communication.
The system console must be placed in the ODT (Online Dehugging
Technique) mode.
A boot must be initiated by typing on the system console either 77775000G or 7777,3000G, depending on which
boot address is selected on the controller (see Section 2 on
hardware jumpers for details on boot address).
The system console will respond with an n*" as a prompt.
At this point, the
user can type an "FT" which will bring up the Configuration Menu.
NOTE
In the follow i ng exam pI e s, all ch aract ers unde rlined are output by the system; characters not
underlined are input by the user~
EXAMPLE:
If the bootstrap is enabled and the boot address is 175000,
proceed as follows (enter):
~
77775000G
*
(If the bootstrap address is 173000, enter 77773000G.)
3-2
Procedure for Bringing Up
Adapter Boot Disabled
the
Virtual
Terminal
Via
LSI-11
Wi th
Host
If the boot on the controller is disabled, communication via the
virtual terminal can be brought up by typing 77777 (octal) to the
SA address (SA default address = 160336) followed by a carriage
return.
The user must then type 2000G.
The system console will
respond with an rr*n as a prompt.
The user should then type "FT"
to b r i n g up the Co n fig u rat ion Me n u •
In the f 011 ow i n g e x am pI e
underl ined characters are output by the compute.r:
/
NOTE
This procedure will work regardless of whether the
boot address of the controller is enabled or disabl ed.
EXAMPLE:
!
160334/0
~
160336/005400
~
2000G
<CR>
77777
<CR>
Boot for the device as described below.
The controller not only supports standard DEC devices, but also
allows the use of the onboard Configuration Menu.
When DU is
used, the standard DEC emulation is called.
When FT is used, the
onboard Configuration Menu is enabled for use through the system
console.
*
Enter one of the following: DMO, DPQ, DLO, DRO, MSO,
MTO, MUO, DYO, DU, or FT <CR>
NOTE
When making
used.
a
selection,
capital
letters must
be
Definitions are as follows:
DM
DP
DL
DR
MS
MT
MU
DY
DU
FT
= RD06/07 Disk
= RP02/03 Disk
= RL 0 11 0 2 Dis k
Disk
= RM03/05/80
Tape
TS11
=
= Tape
= CTMSCP) Tape
= RX02 Floppy Disk
= DU ernul ati on
onboard Configuration
= Enable
console
3-3
Menu through system
Procedure to Bring Up the Virtual Terminal Via MICRO-VAX II
In order to bring up communication via the virtual terminal, the
system console is used as the terminal for serial communication.
Note that the bootstrap PROM must be disabled. (See Section 2 for
jumper placement.)
Upon powering up the MicroVAX II, the user must set up the
MicroVAX II 1/0 map via the system console, type 3FFF hex to the
SA address (SA default address = 160336) and start executing code
at location 200 hex.
All this can be done by the user as shown
below. After executing this procedure, the Configuration Menu
comes upon the system console. Note that the bootstrap PROM must
be disabled.
EXAMPLE:
»>D/P/L 20088004 800000Q1 <CR> <--Setup MicroVAX II I/O Map
»>D/P/W 20001F40 20 <CR>
<--Setup MicroVAX II I/O Map
»>D/P/W xxxxxxxx 3FFF <CR>
<--Deposit the 3FFF hex in SA address. The values of xxxxxxxx
are hex values of the controller address of the SA register
and are listed in Table 3-1.
'AD 0
D
()~Af
2.22S 200 <CR>
<--Start
10-\.0
executing
code
v
.. " \..0 "" •
(At this point the Configuration Menu should appear.)
NOTE
When a GPX (Graphics Work Station) is used, enter
2223 218 <CR> instead of 200.
3-4
at 200
CONFIGURATION MENU
Upon entering the controller's onboard Configuration Menu,
prompt to appear is:
the first
ARE YOU USING A (P)RINTER OR (C)RT?
Type a flP" if a printer terminal is being used or a ncn if a CRT is
attached, and a carriage return, then the Configuration Menu will
appear as follows:
CONFIGURATION MENU
1
2
3
4
-
STANDARD DIAGNOSTICS
HOST Q-BUS MEMORY DMA TEST
DISPLAY CONTROLLER CHARACTERISTICS
SET CONTROLLER CHARACTERISTICS
5 - SCSI DIRECT COMMAND
6 - FORMAT SCSI DRIVE
7 - SCSI UNIT MAPPING
.ENTER A SELECTION:
Any of the seven selections may be entered at the prompt simply by
typing the number that c~rresponds to the desired selection followed
by a carriage return.
NOTE
A CTRL C (AC) entered during any of the selections
will return to the Configuration Menu.
STANDARD DIAGNOSTICS
Selection 1 of the Configuration Menu enables the controller to run
its onboard diagnostics. Each time a diagnostic test is successfully
passed, a "." will be printed onto the screen. It takes approximately
10 seconds for the controller to make one pass through all the diagnostic tests. The controller will continue to loop on the diagnostic
tests until a CTRL C CAC) is typed on the terminal. (Notice that the
controller does not respond immediately to the AC when in the virtual
terminal mode. It takes several seconds for the controller to respond,
so please be patient.)
Upon recognizing the AC, the controller will return to the Configuration Menu. If an error is encountered during the execution of a diagnostic test, an "E" will be printed onto the sC'reen and the onboard
LED will flash the appropriate error code.
(See documentation on
error code flashing for details.)
The controller will conduct a loop
on error process until either a AC is detected or power on reset is
conducted.
3-5
HOST MEMORY DMA TEST
Selection 2 of the Configuration Menu enables the controller to conduct DMA transfers to and from the host system.
CAUTION
THIS TEST WILL WRITE ALL OF HOST MEMORY -- THEREFORE, ANYTHING RESIDING IN HOST MEMORY AT THE TIME
THIS TEST IS CONDUCTED WILL BE OVERWRI~TEN!I
t est v e r i fie s t hat d a taw r itt e n t 0 and rea d~'" from the h 0 s t i s
val id to ensure that operations between the host and the controller
are functioning properly.
When using this selection, make sure the
host system has finished ini tial izing the mapping registers.
Otherwise, the test will report memory size equal to ~ero.
T his
NOTE
This test will not be
virtual terminal mode.
executed
if
operating
in
Upon· sel ecti ng the HOST MEMORY DMA TEST, the controller will go out
and size the host memory.
The controller will then begin a series of
DMA tests to ensure host-controller DMA communications are functional.
The following will be disabled upon a good pass of the DMA test.
(CTRL-C
ABO~TS
BACK TO MAIN MENU)
SIZING HOST MEMORY ...
MEMORY SIZE = XXXX HEX (IN KBYTES)
TESTING HOST DMA.~ .
... TESTING DMA ON ODD, EVEN ADDRESSES W/EVEN BYTE COUNT
.•• TESTING READ REVERSE
••• TESTING ONE WORD DMA
••• TESTING ONE BYTE DMA (ALSO TESTS BYTE SWAP)
DMA IS OPERATIONAL OVER ENTIRE HOST MEMORY.
The test will continue to run until a CTRL-C is typed on the console.
If the test ever detects an error, an error message will be displayed
on the console. The following is a list of all possible errors.
>
NON-EXISTENT MEMORY ERROR
This error indicates that the controller
write a nonaccessable memory location.
error, the following message is displayed.
attempted to read or
When receiving this
DMA FAILED DUE TO NON-EXISTENT MEMORY
... AT HOST ADDRESS BLOCK LOCATION XXXX HEX
The "xxxx" message above represents the address at which the test
failed. Note that if a memory siie of 0 HEX is calculated at the
begining of the test, the host memory is not accessable to the
controller.
This may be due to the host I/O memory being mapped
incorrectly.
Check the host system memory to make sure that I/O
memory is available to the controller.
3-6
>
DATA MISCOMPARE ERROR
This error indicates that the controller detected a data miscompare error (i.e. the data written and read did not compare).
When receiving this error, the following message is displayed.
DMA FAILED DUE TO DATA MIS-COMPARE
••. AT HOST ADDRESS BLOCK LOCATION XXXX HEX
DATA RECEIVED = XX HEX DATA EXPECTED = XX HEX
>
Q-BUS PARITY ERROR
This error indicates that the controller detected a Q-bus parity
error while performing a DMA transfer.
When receiving this
error, the following message is displayed.
DMA failed due to a QBUS Parity Error
..• AT HOST ADDRESS BLOCK LOCATION XXXX HEX
>
DMAC PARITY ERROR
This error indicates that the controller detected an internal
parity error when reading the controller buffer RAM. When receiving this error, the following message is displayed.
DMA FAILED DUE TO A DMAC PARITY ERROR
Upon receIvIng an error, the test will also print an "E" on the
console and flash an error code on the activity LED. The test will
then attempt to restart the test.
On occasion, the test will be
unable to restart or be aborted via a CTRL-C; therefore a power on
reset must be performed in order to reset the controller.
3-7
DISPLAY CONTROLLER CHARACTERISTICS
Selection 3 of the Configuration Menu allows the user to display on
the terminal the current controller characteristics as determined by
what is currently set in the controller's NOVRAM.
Selection 3
displays the following:
DISPLAY CONTROLLER CHARACTERISTICS
IP/SAADDRESS (IN OCTAL':
DWELL COUNT '( COUNT * BOO NSEC
BURST SIZE (# OF WORDS):
INTERRUPT PRIORITY:
** ** **
= DWELL TIME)!
[XXXXXX]
[XXX]
[XXX]
[XX]
PRESS <CR> TO CONTINUE
** ** **
As seen above, the IP/SA address, the dwell time and burst size for
DMAs, and the controller's interrupt priority to the host are displayed.
The X's in the table above represent the current value of
each controller characteristic.
The address is given in octal value
and all the other values are decimal. After displaying the controller
characteristics, a carriage return, <CR>, will display the Configuration ,Menu again.
Below is a list of the default values of the controller configuration
characteristics:
DEFAULT VALUE
IP/SA ADDRESS (IN OCTAL):
DWELL COUNT (COUNT * BOO NSEC
BURST SIZE (# OF WORDS):
INTERRUPT PRIORITY:
= DWELL TIME):
[16033 4 ]
[02]
COB]
[04]
3-B
SET CONTROLLER CHARACTERISTICS
Selection 4 of the ,Configuration Menu allows the user to set the
controller characteristics (i.e., set up the NOVRAM).
Selection 4
displays the following:
SET CONTROLLER CHARACTERISTICS
[ ] = CURRENT
CONFIGURATION, <CR>
= DEFAULTS
IP/SA ADDRESS:
DWELL COUNT:
(COUNT *800 NSEC = DWELL TIME)
BURST SIZE (# OF WORDS):
INTERRUPT PRIORITY:
TO CURRENT SETTING
[XXxxxxJ
[XXXL ~)
{USER RESPONSE}
{USER RESPONSE}
[XXX]
[XX]
{USER RESPONSE}
{USER RESPONSE}
SAVE NEW CONFIGURATION (Y/N)?
In order to set the controller characteristics, the user is prompted
for all the information. The user is first prompted for the IP/SA
address.
The current setting is displayed alpng with the "IP/SA
Address" message. A prompt then appears and waits for user response.
The ~ser response in the above table is indicated by the message
"{user response}." The user now has the option of changing the current IP/SA address. He can do this simply by typing a new address (in
octal) at the prompt followed by a carriage return. If the user does
not wish to change the address, he need simply type a carriage return
and the address will remain unchanged. In either case, the user will
then be prompted for the dwell time.
He will again have the same
option as before; he can either change the value or leave it at its
current setting. He will then be prompted for the next piece of information, and this process will continue until all the information on
the controller characteristics has been prompted for. The controller
will then prompt the user as to whether or not the new configuration
is to be saved in the controller's NOVRAM. If the user chooses not to
save the new configuration, the controller will simply display a mes'sage saying that the new configuration was not saved. A <CR> at this
time will simply take the user back to the Configuration Menu. If the
user chooses to save the new configuration, the controller will respond with the following:
NEW CONFIGURATION SAVED IN NOVRAM •..
REBOOT SYSTEM TO CONFIGURE CONTROLLER HARDWARE!
The host system only needs to be rebooted if the IP/SA address was
changed; otherwise, the user need only type a <CR> to go back to the
Configuration Menu.
During the prompting of the controller characteristics, the controller
will respond with an "Invalid Response" message if the user response
is considered invalid.
The controller expects all user responses to
be in decimal except for the IP/SA address which it expects to see in
octal (for appropriate octal IP/SA addresses see the options listed
below) .
3-9
Below are lists of valid options for each of the controller configuration characteristics.
IP/SA ADDRESS (IN OCTAL):
SUGGESTED ADDRESSES:
[160000 - 177774J
[i72150,i60334,i60340,16034 4 ,
160354,160360 160374J
j
DWELL COUNT:
(COUNT * 800 NSEC = DWELL TIME)
BURST SIZE:
[1-63J
INTERRUPT PRIORITY:
[4-7J
How To Find Current Address
In case the user ever changes the controller slave address (IP/SA
address) and forgets what value he set it to, the following steps
should be taken:
FOR LSI=11 SYSTEM:
1.
Enable the boot address on the controller and make sure
no other controller is using the same boot address (see
details on hardware jumpers).
CAUTION
MAKE SURE BOARD IS POWERED
CHANGING JUMPERS!
DOWN BEFORE
2.
Power up the board, and get into ODT mode. Conduct a
boot by typing on the system console either 77775000G
or 77773000G, depending on what boot address the user
has enabled. (See section on INITIATING COMMUNICATION
for details on boot procedure~)
3.
Wait for the
4.
Halt the host system
location o.
Location
address.
"*"
prompt.
How to examine address
processor and look at address
zero should contain the IP/SA
a
using aDT:
(allows the user to examine location 0)
3-10
FOR MICRO-VAX II SYSTEM:
i.
Conduct a power on reset on the board (i.e., power down
the board and power it 'back up).
2.
Check all the possible SA addresses that can be selected on the controller via the system console. (See Table
3-1 for listing of suggested IP and SA addresses, refer
to hex address only.)
How to examipe addresses on MicroVAX II:
".
222E/P/W xxxxxxxx <CR>
<---- Allow,s user to exam i ne
loca~ion xxxxxxxx, a hex
address from Table 3-1.
If required to check the range 160000Q-17774Q, a
suggested method is to start examining at location
1600000 as shown above (corresponding hex address for
MicroVAX II is 20000000 calculated as shown below Table
3-1. After examining the first address, the user can
continue checking subsequent addresses by simply typing
the following:
222 E <CR>
3.
<---- allows the user to examine subsequent
hex addresses.
Once the user finds the address he suspects is the controller slave address, he should deposit a 0 in the
corresponding IP address. If the controller responds
with a OBOOH in the SA address, then the correct slave
address has been found. Deposit a 0 in the IP address
as shown below:
How to deposit a 0 on MicroVAX II:
222D/P/W xxxxxxxx 0 <CR> <---- Allows user to deposit a
o at location xxxxxxxx,
a hex address from Table
3 -1 •
4.
If the slave address does not respond with OBOOH, then
go back to step 2 of this procedure.
3-11
Table 3-1.
IP and SA Hex Addresses
*
r---------------------------------------------~~---------------~-----l
.
IP REGISTER I SA REGISTER I
SA REGISTER
I
IP REGISTER
OCT AL ADDRESS I OCTAL ADDRESS I
HEX ADDRESS
I
HEX ADDRESS
I
I
I FOR MICROVAX II I FOR MICROVAX II I
~~~=============,====~===~~~====l=================:=================~I
I,
*
I
172150
172152
2000146A
20001468
160334
160336
200000DE
200000DC
160340
160342
200000E2
200000EO
160344
160346
200000E6
200000E4
160354
160356
200000EE
200000EC
Suggested Addresses only; to be entered
MicroVAX II.
II
in place of xxxxxxxx for
For addresses other than the above (a~dress range 1600000-177740)
perform the following to calculate hex address for MicroVAX II:
A.
Convert the 13 least significant
octal to hexadecimal.
B.
Add 200000000.
bits of
the address
in
For example, if the octal address is 177740, the hex value is obtained
as follows:
A.
177774
1FFC.
octal with
B.
1FFC + 20000000
13 LS bits
= 20001FFC.
3-12
=
17774, converted
to hex =
SCSI DIRECT COMMAND
Selection 5 of the ~onfiguration Menu allows the user to input and
execute SCSI commands.
At this point, all user responses must be
input in hex. All of the controller's responses will be displayed in
hex.
Upon entering the SCSI DIRECT COMMAND mode the user is prompted for
the SCSI device node number, logical unit number, and number of bytes
in the command descriptor .block (CDB).
The user is then asked to
input the SCSI command descriptor block CCDB) one' b~te at a time. The
user is prompted as follows:
ENTER IF OF COMMAND BYTES:
ENTER SCSI COMMAND DESCRIPTOR BLOCK (CDB):
[E.G., FOR A FORMAT COMMAND, LUN = 00]
BYTE
BYTE
BYTE
BYTE
BYTE
BYTE
flO
=
=
=
fl3 =
#4 =
#5 =
#1
#2
{USER RESPONSE}
04
00
00
00
00
00
DATA TRANSFERS LIMITED TO 1 SECTOR [512 BYTES]
ENTER
ENTER
ENTER
ENTER
ENTER
ENTER
COMMAND
COMMAND
COMMAND
COMMAND
COMMAND
COMMAND
BYTE
BYTE
BYTE
BYTE
BYTE
BYTE
#0
fl1
=
=
#2 =
#3 =
114 =
fl5 =
{USER
{USER
{USER
{USER
{USER
{USER
RESPONSE}
RESPONSE}
RESPONSE}
RESPONSE}
RESPONSE}
RESPONSE}
In order to input the command descriptor block, the user need simply
type, in hex, each command byte when prompted for. An example of a
FORMAT command is given as part of the prompt message to demonstrate
how the user is to enter the command properly. The prompting of the
command bytes will continue until the specified number of command
bytes have been received.
Note that all user responses above are
indicated by "{user response}."
If the user ever inputs an invalid
response, the controller will respond with an tt** Invalid Setting"
message.
The user will then again be prompted for the appropriate
information. Also, the user must note that all data transfer commands
are limited to 512 bytes. An attempt to transfer more than 512 bytes
of data will result in an error during the execution of the command.
3-13
Upon recelvlng the command descriptor block, the controller will display the c;ommand bytes entered.
The user is then prompted as to
whether or not the command bytes displayed are correct.
If the command bytes are correct, the user need simply type nyu for yes and the
controller will continue on wi t h i ts next prompt.
If the command
bytes are incorrect, the user must then type an "N" for no.
At this
point the controller will respond with a second prompt asking if the
user wishes to go back to the Configuration Menu.
If the user responds wi th a "Y" for yes: the controller .. will take him back to the
Configuration Menu, otherwise an "N" for no will take the user back to
reenter a SCSI command descriptor block (COB).
Once the user has input a command correctly, the controller will
prompt him as to whether or not he wishes to send data out during the
ex e cut ion 0 f the com man d •
Th e use r nee don 1 y sen d d a tao u t i f the
current command to be executed sends data to the target; otherwise,
the user should respond to the prompt with an "N" for no.
If data
does need to be sent out to the SCSI device, the user should type a
flY" for yes, at which point the controller will prompt him for the
number of data bytes to be entered.
The user will then be asked to
input the data bytes one byte at a time.
NOTE
The maximum number of data
tered is 512 bytes.
ENTER # OF DATA BYTES:
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
BYTE
BYTE
BYTE
BYTE
BYTE
BYTE
BYTE
BYTE
BYTE
BYTE
BYTE
#0000 = {USER
= {USER
= {USER
= {USER
= {USER
= {USER
= {USER
= {USER
= {USER
= {USER
= {USER
110001
fFOO02
110003
110004
110005
110006
110007
110008
IFOO09
/FOOOA
bytes
that
can be
en-
{USER RESPONSE}
RESPONSE}
RESPONSE}
RESPONSE}
RESPONSE}
RESPONSE}
RESPONSE}
RESPONSE}
RESPONSE}
RESPONSE}
RESPONSE}
RESPONSE}
In order to inputu~ tQe data bytes, the user need simply type, in hex,
each data byte when prompted for.
The prompting of the data bytes
will continue until --the specified number of data bytes have been received.
Once all of the data has been input, the controller will display the data bytes.
It will then prompt the user as to whether or
not the data ,displayed is correct.
If the data is correct, the user
need simply type a "Y" for yes and the controller will continue on to
the next prompt.
If the data is incorrect, the user must type an "N"
for no, at which point the controller will prompt the user to see if
he wishes to go back to the Configuration Menu.
If the user types a
nyn for yes, the controller will go back to the Configuration Menu.
If the user types an "N" for no, then the controller will prompt the
user for new data.
3-14
Upon inputting correct data or specifying that no data need be sent
out, the user will be prompted as to whether or not the command
entered is to be executed. The user is prompted as follows:
CAUTION
COMMANDS WILL BE EXECUTED AS ENTERED.
DATA ON THE DRIVE MAY BE OVERWRITTEN.
CURRENT
EXECUTE COMMAND (YIN)?
~
>
The caution indicates that the SCSI device will reG~ive the command as
entered by the user; therefore, the user should" ehsure that the command is correct before executing.
Also, in case the user issues a
write command, he should be aware that any data on the SCSI device at
the location of the wri te command will be overwri tten.
If the user
chooses not to execute the SCSI command, he should type an "Nfl at
which point the controller will take him back to the Configuration
Menu. If the user chooses to execute the command, he need simply type
a "Y. If
At this pOint the controller will display the following message:
EXECUTING SCSI COMMAND
Upon completion of the command a number of different messages may appear. The different messages and their meanings are as defined below:
>
SCSI COMMAND COMPLETED WITH NO ERRORS
This message indicates that the command was successful and no
errors occurred during its execution.
If this message is displayed after executing a read command, the controller will automatically display the read data from the SCSI device.
>
CHECK CONDITION ON SCSI COMMAND
This message indicates that a check condition occurred during the
execution of the command.
The controller will automatically request sense data from the SCSI device and display the sense data
following this message.
>
SCSI BUS PARITY ERROR
This message indicates that a SCSI bus parity error was detected
during the execution of the command and subsequent retries of the
command also produced parity errors.
>
SCSI BUS PARITY ERROR
RETRY WAS SUCCESSFUL
This message indicates that a SCSI bus parity error was detected
during the execution of the command, but during a subsequent retry the command executed with no errors.
3-15
>
QBUS PARITY ERROR
This message indicates that a QBUS parity error occurred during
an access of QBUS memory.
>
CONTROLLER PARITY ERROR DURING ACCESS OF DRAM
This message
during
>
~n
indicates that a controller parity
of the controller's DynDmic RAM.
error occurred
~~~~ss
SCSI BUS RESET OCCURRED DURING COMMAND EXECUTION
This message indicates that a SCSI bus reset occurred during the
execution of the command; therefore, the command was not campI eted.
>
SCSI DEVICE NOT RESPONDING
This message indicates that the SCSI device is not responding to
the command issued.
>
TRANSFER TRUNCATED TO SINGLE SECTOR {512 BYTES}
This message indicates that a multi-sector data transfer command
was attempted.
The command was therefore truncated to a single
sector command.
If this message is displayed after a read command, only a single sector of data will be displayed.
>
COMMAND TIMEOUT
This message indicates the requested command took over an hour
without any activity on the SCSI bus.
Upon the completion of the
user with the following:
command,
the controller will
prompt
the
ISSUE ANOTHER SCSI COMMAND (YIN)?
At this point the user can either issue another SCSI command simply by
typing a "Y" for yes, or he can go back to the Configuration Menu by
typing an "N" for no.
3-16
FORMAT SCSI DRIVE
Selection 6 of the Configuration Menu allows the user to format a SCSI
drive.
Upon entering FORMAT SCSI DRIVE mode, the user is prompted for
all the SCSI drive information needed in order to format the drive.
The prompting is as follows:
FORMAT SCSI DRIVE
NODE
H OF DRIVE TO FORMAT'(0-7):
{USER, RESPONSE}
LOGICAL UNIT NUMBER (LUN) OF DRIVE TO FORMAT (Cf-3')': {USER RESPONSE}
DRIVE FORMAT INTERLEAVE (0-7):
{USER RESPONSE}
FORMAT USING PRIMARY DEFECT LIST ONLY (YIN)?
The user is first prompted for the SCSI node of the drive to be formatted (node selection varies between 0 and 7).
The user need simply
respond by typing a value between 0 and 7 followed by a <CR).
(Note
that all user responses above are indicated by "{user response}".)
The ,user is next prompted for the Logical Unit Number (LUN) of the
drive to be formatted (LUN selection ranges from 0 through 3).
The
user need simply respond.by typing a value between 0 and 3 followed by
a <CR).
The user is next prompted for the drive format interleave. Interleaving is a technique of assigning successive addresses to sectors
which are physically separated on the disk, in order to reduce access
time.
A 3 to 1 interleave requires three rotations of the disk to
transfer one track.
The range for interleaving on the host adapter is
from 0 to 7, where an interleave of 0 allows the SCSI drive to set its
own defaul t interleave and 1 through 7 allows a 1 to 1 through 7 to 1
interleave, respectively.
To set the interleave, the 'user need simply
type a value between 0 and 7 followed by a <CR).
The user is prompted if only the primary defect list will be used as
the flaw map when formatting.
Responding with a nyn (Yes) will return
the disk to the condition as originally shipped.
No flaws found and
reassigned since the original defect list will be added.
Responding
with an "N" (No) will include flaws found and reassigned in the growth
list.
Depending on the response, one of the following messages will
be displayed:
PRIMARY DEFECT LIST HAS BEEN SELECTED
OR
PRIMARY AND GROWTH DEFECT LISTS HAVE BEEN SELECTED
3-17
After inputting all the requested
respond with the following message:
information,
the
controller
will
CAUTION
ALL DATA WILL BE LOST TO THE SELECTED DRIVE.
CONTINUE WITH FORMAT (YIN)?
The caution serves to warn the user that the data on the drive will be
lost once the format command has taken place.
Therefore, if the user
does not wish to destroy- the data on the disk, he can simply reply
with an "N" at the prompt followed by a <CR) and the controller will
take him back to the Configuration Menu.
If the user wishes to continue with the format, he then need simply type a "Y" at the prompt
followed by a <CR).
Once the formatting has begun, the controller will print the following
message:
FORMATTING SELECTED DRIVE
This message will remain until the drive has completed the format command.
At this point, if the SCSI drive formats with no errors, the controller will print a "Completed" message.
Otherwise, if any errors occur,
one of the SCSI DIRECT COMMAND error messages will be displayed.
(See
SCSI DIRECT COMMAND error message documentation for details.)
3-18
SCSI UNIT MAPPING
Selection 7 of the Configuration Menu allows the user to select the
SCSI unit map of the host adapter. The user can map up to 8 different
host units, -selec~ing each individual node and logical unit number
(LUN) for each SCSI device.
The user can also select the node at
which the host adapter is to reside.
The user should note that at
this point, the controller expects inputs in decimal.
NOTE
Logical mapping of units must be done for all possible devices. If it has not been mapped, a device cannot be .used by the system. ~ Tf another
unit must be added, the map must be/ qhanged offline and then the sy stem rebooted.
~
,',.(
Upon entering the SCSI UNIT MAPPING mode, the current SC~I unit map
table is displayed, as de~ermined by the contents of the NOVRAM. The
display is as shown below:
Host Adapter is at Node XX
SCSI UNIT MAPPING
------------------~
SETUP
HOST UNIT
IF
01
XX
02
03
04
05
06
07
08
XX
XX
XX
XX
XX
XX
XX
NODE / LOGICAL UNIT II (LUN) BYTE SWAP
XX
XX
XX
XX
XX
XX
XX
XX
XX
XX
XX
XX
XX
xx
vv
xx
AA
XX
XX
XX
XX
XX
XX
XX
XX
The HOST UNIT column defines the MSr.P lo~ical address assigned to the
device node and logical unit II (LUN). For example, Host TTnit 00 would
be recognized by VMS as DUAO. DUBO, etc., de-pending on the adapter
address.' Addi tional logical dev ices connected to the adapter on the
SCSI bus require a successive Host Logical Unit number, i.e., DUA1,
DUA2 (DUB1, DUB2) etc. be assigned. The Node column ,defines the SCSI
device (intelligent peripheral or controller) node. address.
The
LOGICAL UNIT ,
(LUN) column defines the SCSI address fro the
peripheral attached ·to the mapped device NODE.
Note that in the
majority of cases, the LUN II will always be zero; the vast majority of
peripherals connected to a node contain a built in controller.
The
master controller/peripheral with attached slave peripherals without
controllers is rare.
t
The constraining factor is that there can never be more than eight (8)
nodes on an SCSI bus (including the host adapter). The maximum number
of peripheral device nodes and LUN's is defined by the MSCP driver.
not the adapter, for the particular device (typically four for MSCP
disk drives).
As seen above, the SCSI unit map displays the HOST (logical) lJNTT, the
peripheral device NODE / LOGICAL UNIT # (LUN), and BYTE SWAP
configuration for 8 different setups available on the Host Adapter.
In addition it specifies the Node that the Host Adapter occupies. The
xx's in the table above represent the current configuration numbers
for each setup. The default SCSI unit set at the factory is shown on
the next page, followed by an example of the SCSI unit map for adding
additional peripheral devices to the adapter. shown below.
3-19
SCSI UNIT MAPPING
------------------HOST UNIT
SETUP Ii
Host Adapter is at Node 07
NODE / LOGICAL UNIT t! (LUN)
00
00
01
00
02
No current setup
03
04
05
06
07
08
No
No
No
No
setup
setup
setup
setup
- _.. ,.No ""'1.....,,,..,...,,,,,,,4- h:)C'l"utJ
No current setup
current
current
current
current
"' .... 4 4
<;;;41'"'
Example SCSI UNIT MAP for a three, intelligent peripheral system:
SCSI UNIT MAPPING
SETUP IF
01
02
03
04
05
06
07
08
Host Adapter is at Node 07
HOST UNIT
00
02
02
No current
No current
No current
No current
No current
NODE / LOGICAL UNIT t! (LUN)
00
00
00
01
00
02
setup
setup
setup
setup
~ot-l'n
...,
..... ..,\At-'
3-20
The default setting of the host adapter does not setup the last seven
available SCSI units.
As seen above, when a SCSI unit has not been
setup, the "No current setup" message will appear.
Following the display of the SCSI unit map,· a menu of
options for SCSI UNIT MAPPING mode is displayed.
o-
1
2
3
4
-
the available
RETURN TO CONFIGURATION MENU
CHANGE UNIT MAPPING
CHANGE HO~T ADApTER NODE
SAVE UNIT MAPPING IN NOV RAM
RESTORE DEFAULT UNIT MAPPING
ENT~R
A SELECTION:
In order to choose an option, the user need simply type the
the option he wishes to oerform followed by a <CR).
number of
OPTiON 0 will simply return the user to the Configuration Menu.
OPTION 1 will allow the user to change any of the current setups on
the SCSI unit map.
Upon entering this option, the user will be
prompted for the setup # he wishes to change.
The user need simply
type the number of the setup he wishes to modify at the prompt followed by a <CR). At this point, if the setup the user wishes to modify
currently occupies a valid setup (i.e., the "No current setup" is Nor
displayed), he will be prompted as to whether or not he wishes to
remove the setup. If the user chooses to remove the setup by typing a
"Y" at the orompt, the SCSI unit map will be redisplayed and the setup
removed will display .. the "No current setup" messa~e. If the user
chooses not to remove the setup by typing an "N" at"the prompt, he
will then be prompted for the new host unit, node, and lo~ical unit
number he wishes to set.
If the setup the user wishes to modify has
no valid setup (i.e., the "No current setup" message is displayed), he
will simply be prompted for the new setup consisting of the host unit,
node, and logical unit number.
In any case, at the completion of the
prompting, the SCSI unit map will be redisplayed with the modifications made by the user.
OPTION 2
adapter.
will allow the user to change the
3-21
node occupied by the host
OPTION 3 will allOW the user to save the currently displayed SCSI unit
map in the NOVRAM. This will allow the oarticular map saved to remain
as part of the confi~uration of the host adapter even when the board
has no pow~r applied.
The map will be preserved for the life of the
NOVRAM or until another map is saved in the NOVRAM. In order to save
the map in the NOVRAM of the host adapter, the following specification
must be met by the SCSI unit map.
The following specifications exclude those setups which are not valid.
1.
All setups must occupy unique host unit numbers.
2.
All setups must occupy
logical unit number.
3.
No setup can occupy the same node as the host adapter.
a unique
combination of
node and
Unless the above specifications are me~, t.he host adapter will not
allow the SCSI unit map to be saved in the NOVRAM. If the ~CSI map is
not allowed to be saved, the following'message will be displayed:
SCSI UNIT MAP NOT SAvED IN NuVRAM DUE TU OVERLAPP1NG
OF HOST UNIT NUMBERS OR NODE/LOGICAL UNIt NUMBERS.
If the above specifications are met, the controller will then display
a message saying that the map was saved in the NOVRAM. A (eR) at this
point will redisplay the SCSI unit map.
OPTION 4 will allow the user to restore the ~efault SCSI 11nit map as
set by the factory. Upon entering this option, the user is prompted
as to whether or not he wishes to save the default mapping in th~
NOVRAM.
If he chooses to save the mapping in the NOVRAM, he need
simply type a "Y" at the prompt followed by a <CR> at ,which point the
default map will be saved and redisplayed. If the user chooses not to
save the map in the NOVRAM, he need simply type an "N" at the prompt,
at which point the default map will be displayed but not saved.
3-22
SECTION 4
DIAGNOSTICS
Two DEC RC25 diagnostics may be used to test the controller. They are
ZRCFB3, Front End Test, and ZRCDA1, Disk Exerciser.
MicroVAX II, MDM
KDA50-Q, diagnostics are also included.
SETUP AND SELF TEST
Install the controller as descri bed in Secti on 2.
Appl Y power to the
system, and verify that both green LEDs light.
Install the XXDP+
diagnostic floppy in the floppy drive and boot the system.
When booting is completed, the XXDP+ sign-on will appear:
XXDP-SM SMALL MONITOR VERSION 2
BOOT FROM DYO
28KW MEMORY
·UNIBUS SYSTEM
RESTART ADDR: 152010.
THIS IS XXDP-SM TYPE "H" OR "H/L" FOR HELP
(NOTE:
28KW
= 28
Kilowords)
4-1
FRONT END TEST - ZRCFB3
The controller will only support tests 1-8 which must be selected by
the user.
These tests will bring the controller through initialization several times and do extensi ve checks on the DMA capa bil i ty.
Once the prompt ". n has appeared, type the following command I ine to
start ZRCFB3 diagnostic:
• R Z RCFB3
The system will echo the filename to let the user know that the file
is being loaded •
• R ZRCFB3
ZRCFB3.BIN
the diagnostic has been loaded,
will appear on the user's console.
When
the diagnostic
startup message
DRSSM-FO
CZRCF-A-O
RC25 FRONT END/HOST DIAGNOSTIC
UNIT IS AZTEC RC25 PLATTER
RSTRT ADR 145676
DR)
The diagnostic can be started by typing the following command line:
DR)START/TEST:1-8<CR)
The above command line instructs the diagnostic supervisor to start
the test but initiate only tests 1 through 8.
The supervisor will
then prompt the user for hardware or software changes.
CHANG E HW (L)
?
The diagnostic must be informed of the hardware parameters of the system under test. Enter the following information.
CHANGE HW (L)
? Y<CR)
Enter the number of controllers that are being tested.
I! UNITS (D)
?,':-1~·CR>
",
The diagnostic will" then prompt the user to enter the following information for th~ number of units that have been selected. The following
is an example:
UNIT 0
IP ADDRESS (0) 172150 ? <CR)
VECTOR (0) 154 ? <CR)
BR LEVEL (0) 5 ? <CR)
PLATTER ADDRESS[ES] (D) ? O<CR)
4-2
The platter address is the unit number of the disk drive under test.
Since the controller does not support the tests which require a disk,
this question is not. appropriate but must be answered to start the
diagnostic.
Once the hardware questions are answered, the supervisor
will prompt for software changes.
CHAN GE SW (L)
?
The software question can be answered NO because the controller does
not support the tests which require a disk drive.
CHANGE SW (L)
? N<CR>
The diagnostic will print each test as it runs and will
user of any errors that occur.
TESTING UNIT II:
TEST
TEST
TEST
TEST
TEST
'TEST
TEST
TEST
1
2
3
4
5
6
7
8
o
IP_REGISTER:172150
inform the
PLATTER II: 0
REGISTER EXISTENCE TEST
STEP 1 READ/WRITE POWERUP DIAGNOSTICS
DIAGNOSTIC WRAP TEST
VECTOR AND BR LEVEL TEST
STEP 1-3 READ/WRITE DIAGNOSTIC
PURGE POLL TEST
SMALL RING TEST
LARGE RING TEST
When the diagnostic has completed all the tests, the end
message will be printed and the diagnostic will be restarted.
DZ RCF
EOP
1
o TOTAL ERRORS
DR>EXIT<CR>
4-3
of
pass
DISK EXERCISER - ZRCDA1
The controller is also compatib e with the multi-drive exerc ser,
ZRCDA1.
This diagnostic will br ng the controller online and ssue
random record numbers to the select-ed unites).
This diagnostic also
supports multiple controllers as well as multiple units on a single
controller~
Once the XXDP prompt" " is displayed,
lowing command line:
run ZRCDA1 by typing the fol-
.R ZRCDA1<CR)
The system will echo the filename to inform the user that the program
is being loaded:
.R ZRCDA1
ZRCDA1.BIN
Once the diagnosti~ is loaded, the diagnostics startup message will be
displayed on the user's console:
DRSSM-FO
CZRCD-A-O
RC25 DISK EXERCISER
UNIT IS SINGLE RC25 PLATTER
RSTRT ADR 145676
DR)
Patch as follows:
ADDRESS
IS
SHOULD BE
PATCH 1
26070
26072
'6237
50
12737
1 43326
PATCH 2
30644
1003
1367
PATCH 3
30704
1003
1367
PATCH 4
37522
1416
240
Start the test
the diagnostic supervisor prompt nDR)" appears.
after~
DR)START<CR)·
The supervisor will then prompt toe user to change hardware or software default parameters:
CH ANG E HW (L)
?
The diagnostic must be informed of the hardware,parameters of the system under test.
Enter the following command line to change hardware
parameters:
CHANGE HW (L)
? Y<CR>
4-4
En t e r the
tested:
nurn be r
II UNITS CD)
of
co nt roll e r s
0
r
d r i v est hat
are
cur r e n t 1 y be i n g
? 1<CR>
If 2 is entered the next prompt will appear twice so that the second
controller or second drive may be selected.
The diagnostic will prompt the user to enter the following· information
for the number of units that have been selected:
UNIT 0
IP ADDRESS (0) 172150 ? <CR>
VECTOR (0) 154 ? <CR>
BR LEVEL CD) 5 ? <CR>
PLATTER ADDRESS (UNIT PLUG) CD) 0 ? <CR>
ALLOW WRITES TO CUSTOMER DATA AREA ON THIS PLATTER (L) ?
** WARNING - CUSTOMER DATA AREA MAY BE OVERWRITTEN I ...
CONFIRM (L) ? Y<CR>
Y<CR>
Th e pI a t t era d d res sis the un i t nurn be r 0 f the dis k d r i v e un de r t est.
The 'customer data area is the host data area of the disk drive and is
used to test the controller.
Backup any data in the host partition,
if necessary, before continuing with the diagnostic.
After the hardware questions are answered,
the user for any software changes:
CHANG E SW (L)
the supervisor will prompt
?
The user can take the defaul t software val ues because the dri ves are
larger than an RC25.
If the tests are being. run on a contracted unit,
some of the software val ues may have to be changed to prevent errors
from occurring.
If the selected unit is fully formatted,
typing NO to the software query:
CH AN GE SW CL )
To change
query:
the
the
default values
by
? N<CR>
software
CHAN GE SW (L)
use
defaul t
val ues,
answer
YES
to
the
software
? Y<CR>
The user will theu" be prompted by the supervisor to input the following information:
ERROR LIMIT (0 FOR NO LIMIT) CD)
32? <CR>
Enter the error limit that must be reached before a unit is deselected
by the diagnostic. The default value of 32 is used.
TRANSFER LIMIT IN MEGABYTES (0 FOR NO LIMIT)
4-5
CD)
2? <CR>
Enter the number of bytes to be transferred between the controller and
the diagnostic.
Effectively, this selects the time required to reach
an END OF PASS.
SUPPRESS PRINTING ERROR LOG MESSAGES (L)
Y? <CR)
The default value should be used unless multiple errors occur and more
information is required to resolve the problem.
RUN DM EXERCISER INStEAD OF MULTI-DRIVE SUBTES!. (L) N ? <CR)
The default must always be taken because the controller does not sup:
port Diagnostic Mode (DM) of operation.
RANDOM SEEK MODE (L)
Y? <CR)
The best possible test is to seek randomly across the entire disk surface. Therefore, the default value should be used.
STARTING TRACK (D)
O? <CR)
The starting track number is to be entered.
(Because the controller
does .not emulate the RC25, the questions regarding starting and ending
track numbers do not apply. If the unit under test is fully formatted
and is more than 40 meg~bytes, the diagnostic will not overflow the
cylinder address.)
ENDING TRACK CD)
1641? <CR)
Enter the endi ng track number.
If the uni t
unit (40 megabytes), the default can be used.
is 1 arger than an RC25
READ-COMPARES PERFORMED AT THE CONTROLLER (L) N ? <CR)
The default value is used to require the controller to compare the
data read with host memory.
THE REMAINING QUESTIONS APPLY ONLY TO UNPROTECTED PLATTERS.
The user can use the default values for the remaining questions.
WRITE ONLY (L) N ? <CR>
The disk drive under test
always take the default.
is
never
a write
only
disk;
therefore,
WRITE-COMPARES PERFORMED AT THE CONTROLLER (L) N ? <CR)
The controller will perform write checks if the default is taken.
CHECK ALL WRITES AT HOST BY READING (L) Y ? <CR>
4-6
The diagnostic will NOT issue read commands to check the data just
written if the default is taken.
Otherwise, the diagnostic software
will perform the wri~e check function.
USER DEFINED DATA PATTERN (L) N ? <CR)
The data pattern used in the diagnostic is
prompt testing, always use the default value.
worst
case.
To
ensure
SELECT PREDEFINED DATA PATTERN (0 FOR SEQUENTIAL SELECTION)
(D) 0 ? <CR>
Always use the default value.
The test
answered.
will
begin
after
the
hardware
and
software
questions
are
INIT SUB TEST START
ABOUT TO VERIFY VECTOR 154(0) FOR DEVICE 172150(0) ••• COMPLETED
The diagnostic will run until the transfer limit is reached.
After
the limit has been reached, the diagnostic will print status information about the unit under test and display the END OF PASS message:
CZRCD EOP
1
o TOTAL ERRS
4-7
MICROVAX II - MDM KDA50-Q
The controller is compatible with the MicroVAX II diagnostics as des=
cribed below.
Parts of the Functional Test, the Exerciser, and the
Utility Test are listed as either PASS or FAIL.
Where FAIL is indicated, the error advisory is described.
NOTE
If MicroVAX Diagnostic Monitor (MDM) is to be run,
unit mapping (NOVRAM) must have all devices connected.
FUNCTIONAL TEST
Test 1 - PASS - Controller Wrap Mode Test
This test tests the ability of the controller to read
register contents from the host and then turn around
the same value back to the host.
in the SA
and write
Test 2 - FAIL - Controller Interrupt Test
This test tests the controller's ability to interrupt at the correct IPL and vector.
It also verifies that the controller can be
successfully initialized by the host.
The error advisory
terrupted above BR
BR level is NOVRAM
pass if the NOVRAM
ler.
for this test failure was, "The controller inlevel 4," which is true since the controller's
selectable and was set to 5.
The test will
is set to interrupt priority on the control-
Test 3 - FAIL - Controller RAM Test
This test writes data patterns to the controller RAM and then
reads the data patterns back to the host where they are verified.
The error advisory for this test failure was, "The controller SA
register contains error code: 100016 <octal>."
The controller
does not allow direct access to its Data Buffer RAM.
T est 4 - PAS S' - Fu n c t ion Test (V e r i f y' Mod e )
This test enables the controller and disk drives.
An MSCP read
and access command are issued to test the read and posi tioner
circuitry of the drives.
Test 5 - PASS - Functional Test (Service
Mod~)
This test operates the same as Test 4 except in Service Mode.
4-8
EXERCISER
Test 1 - PASS - Exerciser (Verify Mode)
This test performs random MSCP read and access operations on all
of the drives selected for test.
Data integrity is verified.
The exerciser will run for 3 minutes.
Test 2 - PASS - Exerciser (Service Mode)
This test operates the same as Test 1 except in Service Mode.
UTILITY
Test 1 - PASS - MSCP Exerciser
This test is run the same as Test 2 of the Exerciser Section if
run in the Read-Only Mode.
This test (when run in the Read/Write Mode) will perform random
MSCP read, access, wri te, and erase commands.
Error log and
attention messages are enabled during this test.
The exerciser
will run for 3 minutes.
Test 2 - FAIL - DUP Functional Test
This test loads a DUP program into the controller for
fault isolation of controller and disk drives.
improved
The error advisory for this test failure was, "The controller SA
register contains error code:
100016 <octal>. II
The controller
does not allow direct access to its Data Buffer RAM nor is it architecturally designed to run "downloaded" programs.
Test 3 - FAIL - DUP RA Series Disk Formatter
This test loads a DUP program into the controller for disk for,matting.
The error advis9ry for this test failure was, "The controller SA
register contains error code:
100016' <octal>."
The controller
does not allow direct access to its Data Buffer RAM nor is it architecturally designed to run "downloaded" programs.
4-9
SECTION 5
ERRORS AND TROUBLESHOOTING
GENERAL
The SCSI cable that connects the
terminated at both ends.
,Since
nated, this must be one end of
attached to the cable, the drive
o nl y d r i v e wit h term ina t ion.
host adapter to the drives must be
the host adapter is already termithe cable.
Jf 'several drives are
at the end ofJ::rr~r cable must be the
' ,
If you are experiencing problems, Option 5 of the Configuration Menu
allows a single SCSI cDmmand to be given directly to the drive.
This
is documented in Section 3, "SCSI Direct Command."
For example, a SCSI Test Unit Ready command is as follows:
6 bytes - 0,0,0,0,0,0
Enter and execute this command.
If it cannot be executed, the system
may return SCSI Request Sense command information.
Consul t the drive
manual to decode this information.
NOTE:
After a bus reset, the SCSI
Test Uni t Ready command will usually return an error indicating that
the bus was reset.
Execute the command again to clear the error.
Another example is a SCSI Read command.
This is as follows:
6 bytes - 8,0,0,0,1,0
°
This will read Logical Block
and display it on the screen.
If not
successful, the SCSI Request Sense command information may appear on
the screen.
Consult the drive manual to decode this information.
The SCSI Request Sense command information returned f'rom a SCSI command that failed, is ·error status returned from the drive.
This
allows drive vendors to give general SCSI status and vendor unique
i n for mat ion.
Ge n era 1 S CSIs tat us i s s urn mar i zed i nth e f 0 u r bit s 0 f
word three of the returned information.
This is as follows (called
the Sense Key):
°1 -- NORECOVERED
SENSE.
2 -
3 4
S
6
7
-
No specific information to return.
ERROR. Last command completed successfully
but with some drive recovery action.
NOT READY.
Drive not ready.
MEDIUM ERROR. Media error.
HARDWARE ERROR.
ILLEGAL REQUEST •.
UNIT ATTENTION.
Drive reset or media changed.
DATA PROTECT.
Consult the drive manual for additional Request Sense information.
5-1
ERROR LOGS
Host Memory Access, Disk Transfer, and SDI error logs and their formats are listed below;
are fatal errors as posted in the
Also
SA register.
ERRORS - SCSI HOST ADAPTER
---~--~--~--------~------------~---------------------
HOST MEMORY ACCESS ERROR LOG
o
8 7
31
16 15
+---~~----~-----------~---+------------+------------+
COMMAND REFERENCE NUMBER
I··
I .,
0
~--------~~---------~--~--+-------------------------+
I
SEQUENCE NUMBER
I
U-NIT NUMBER
I
4
I
8
+-------------------------+------------+------------+
L,
EVENT- CODE..
I
FLAGS
i
FORMAT
+-----------------~~-~----+------------+------------+
,
I
I
CONTROLLER ID
+
12
+
16
+----------------~..:---j..---+------------+------------+
I
UNDEFINED,
UNDEFINED lADAPTER REV
1
20
+--~----~-----------------+--~~~~==~~~~+------------+
HOST MEMORY ADDRESS
24
~--~---------------~-------------------+------------+
CONTROLLER DEPENDENT INFORMATION
+--------------------------------------+------------+
DMA ADDRESS
lOBUS STATUS I 2e
+ - -"".- __ ,.. _____ +_ .. ____ ....... __ ___ ..., __ "__ .. ___ .. ""'" + __ ... _." ....
+
~
I
BUFFER LO I
~
BYTE COUNT
1IiiiIilf.., _
~
... _ _
! MODIFIERS
32
+---~--~--~--+----~-------+------------+------------+
UNDEF:INED I
BURST
I
DWELL
I
BUFFER HI I 36
+------------+------------+------------+------------+
Fl ags
=
Format
=
Event Code =
a
(sequence number reset)
1 (host memory access error)
29H (odd transfer address in DU command)
49H (odd byte count in DU command)
~9H (nonexistent memory error)
5-2
DISK TRANSFER ERROR LOG
16 1 5
31
o
8 7
+-------------------------+--~---------+------------+
I
I
COMMAND REFERENCE NUMBER
+-----------------------~-+-------------------------+
SEQUENC~E NUMBER
UNIt· NUMBER
+-------------------------+------------+------------+
EVENT CODE
,FL·AGS ",- I
FO RMAT' '
>
~
0
4
8
+_ .. - - - ~ .... - -- -- .... - ---'-'- - - - ......... -+- ......... ---.-- ---:'~+-f~------"'-""-+
I
:
CONTROLLER ID
+
12
+
16
+-------------------------+-----------~+------------+
UNDEFINED
CDB SIZE
I ADAPTER REV: 20
+----------------------~--~---~--------+------------+
B YT ES 0 TO 5 OF TH E CU RRENT scsl: COB: 24
- +-------------------------+
+
I
UNIT ID
I
28
+-------------------------+--------'-----+------------+
I
BYTES 6 TO 9 OF THE CU RRENT SCSI CDB
I 32
+---------------------------------------------------+
SCSI INQUIRY COMMAND (BYTES 16 TO 19)
36
40
LOGICAL BLOCK ·NUMBER
+-----------------------------------~----~~---------+
CONTROLLER DEPENDENT INFORMATION
(Taken from the req uest sense command executed after -error)
+------------+------------+------------+------------+
I
:
FRU FAILED
.I
I
ADDITIONAL
I· SENSE KEY
I
SENSE KEY
I
I·
144
BYTE 15
BYTE 14
I BYTE 12
I. ~YTE 2
I ,',
+- .... __ ................ - -+- ---- ... -. -- ----+- ---- -. ... --- .. -+- - ... -~------- -_ ......... +\0, ~"
FIELD POj[NTE~
VENDOR UNIQUE
48
B YT E l7~
B YTE- 1 6
BYTE 19
BYTE ·18
+- - - - - - - - - - - - +- - - - - - - - - ,- - "!",':-"- - - - - I
VENDOR UNIQUE
I
I BYTE 23
I BYTE 22
I BYTE 21
BYTE 20
: . 52
I
+------------
+-.. . -. -:
+
+------------+------------+------------+------------+
VENDOR UNIQUE '
I
BYTE 27
BYTE 26
BYTE 25
BYTE 24
1'56
+------------+------------+------------+------------+
Flags
Format
Event Code
= 0 (sequence numbe'r reset,)
= 2 (disk transfer error)
= 48H (LBN too large)
68H
E8H
108H
148H
(inconsistent internal control
(SCSI bus reset)
(recovered (soft) error)
(reselection timeout error)
5-3
structure)
..... ..-.<.....
...
.~.
-mo
..
~~--~~--~~--~-----~--~--------~---~------~~~~~-------
.'
:;', , ; " ' " ,:
S.DI ERROR LOG
.____ ,,' '
---~~------------------------------------------------
+-------------------------+---------------------~-~~tI
~r:;_nfTJ;'Mrt'
~JTTMCt:"D
r
UNIT NUHBER
..,J.,. .~~
4
... ~~'';'.:'~- :.::~.:~:.:.'.:~.:~:~~ -;-~.:~~- .;.~----- - - --+- ------ ---~.~,!, ~ "..- .~
'.:1 ~ ': FLAGS
:
F"O RMA't:· ': h'~, B -:
"\+- -<';.i.-:"- - - - - - +- - - - - - .; - -- :'1-2;.!,
::i
.r -:.:--:':. : . -;:- :-: ~':,: ,;-'-~.,.--:- --..,-:
. EVEN! .CODE
.
H'
.~ .• },~ : ..~:~~ '~~ ':~. '[email protected] Ei'R':I D'
...~ .;+-.•. , .... _ .,'
I
..... "
,1")(16 '
. +..;~~ ___ :.._~-.:.:.:.:~--:~---:.·~~~+-.;,.---t;...------+-----------~'~;'~.~ .~. J
UND~FIN E:Q ~'
"I ··CBD S IZ E : ADAPTER REV'-.il egO \
... ,!f-"...
+- -.:.---.:. BYTES
--- --""!\-~~-~,~,~,--;;;';+,"-! -.:;,.~~~-- ----+- ---- ----: -:~=:*._., ~. "•..
0 'TO 5 OF' TH E CU RRENT SCS I CDB
,{ ('¢4
+-------------------~---~-~I
UNIT
ID
. I
...,
"I"'''
"."
..
!.f-..... ---.-'~~--: -.- ---,- ------:-;.. . - - - ...... .,...- .. __ ._--
I
9..
-----+-- - .... --------,-.. t,~
,:,
r·o ,.I.. . ... :
TO .9 .,OF .:rH E ·CU R~RENT SCSI CDB
k -;3:2
+ - -':' ~ - ..: ~~ .... - .. - -~~ - ... -.,,-.,; .. :-~ .. ..:": ~ ..;,~l.: ....... - _ .. - ...... - ........... _ .. _ .. - ... :t'.. ... __ ........... ~
I;·
" 'SCSI INQ'UIR'Y: COMM-AND ·(·BYTES 16 TO 19)
_'_'Tt~,;,i6
+ ~.- ~ - - - .....«~
-~~~
..::..~
~
+.
-'
:
: , i . , " . 1 LOGicAi.:.B'LOCK NUMBER
M-'I;~40
~
BYTES_
. . . . . . . . ". . . . . .,. . . ,.., . . . . . . . . . .
. . . . . .;... . . ::..
............ -
.. -
-
_
...... -
... -
............
,+~~-~-----~-~::--------------~~==------------------=~--~-') ,.
. ,,~_'.,' _ ..
CONTROLLER DE PENDENT INFO RNATION
(Taken from the r..e,q ue stsense command executed after error}
+--_ ... _- .... - . . --+-----;-...... - ... ~--.:.+-~.-~----'----+ .. -----------+ ..... ~ .... ,' --,
I FRQ ~AILED I ADDITIONAL: SENSE K~Y' f~~~
I ~ENSE KEY : ' , L~44·~...
B YT E 1 5
. rB YT E J 4
I BYT E 1 2
I
BYT E 2
I ,,'
+,;.;.~.:.;..~~;...~~ .- - - - ... +- ... --- ..~--...----+--
:
VENDOR UNIQUE
1 'BYTE 19'
B'YTE 18
r
'C.·'
I
I
. . ~\_ .. ------+----~~~~--""'+.", . . . , . , ...~
,~g";!T E
FIELD POINTER
17
B YT E 1 6
I
1.\:
L ,~,"8-..,-~.
+------------+---------~--+------------+------------+
I ~..
~,,'. , .. , ' '~', VENDO-R UN IQ UE L , ~~'
_, ~, _.
I BYTE 23
. BYTE 22
I BYTE 21
BYTE 20
I 52',~
, .
~.
~..
•
, .
•
ll'··
.+------------+-----------~+--~~--------+------------+
'VENDOR' UNIQ'UE
'11I
::.
i
I
\.:;"
BYTE
27 ...
. .,
+- ---- -,.~~~
~ J~"
(
.,
j
"
l'
'.,'
BYTE 26 .
,LI
.
BYTE
"7
25
BYTE
24
'
.
,~
..,
";'," ,
,~
: 56
--.+-.,,----------. -+;....:.--........ -- _ ...... - .. +- ---- - - _ ... -- ..... +
.-'..
.....
.'
..
t
",
F1 a g s
'~Q "(s e'q u e n c e ,n.um·b ~ r res e t )
For mat,..
..
3 (SD I .:e r r'o r )- f
'
Eve n t Cb de"
' 2 B (c om mil nd-' tim e 0 u t )
;4B (buffer overrun)
'6BR (tar,get not· responding to selection)~,<'.
CBH (parity error - retry successful)
EBH (drive error)
10BH (parity error - retry unsuccessful)
23H (removable media not mounted)
6H (write protected)
5H (media format error)
- ......... , .........
_..
... _ _ ......
--.... -..-+
~
.. ~ ..... v ....
,0'''''-
"'I
----~~-~~----------------~-----~~------------------------------~~-----
SELF-DETECTED FATAL PORT/CONTR.OLLER ERRORS>
Fatal Error Code Definitions - Error will be posted in th~ SA Register
bit~ 7 to O.
Errors 1 to 31 decimal will bef~ash~d.with the activity
LED' .or'(L2)
MSB
first,
1 = long and 0'=
,'flash.
_ .• '
, ••
--$.nort
, ~',.i"
\."
,
~
,
.r
..~"
J
..
'
: -D~~- --HEX ---OCT -- i ----DE SC Ri PT I.e; -:.:-: -: -- ~-:.~ -- ------~~--.,,. ---------
---l
===~'==:: =, =========== : ==========,= ======== ========"= =='====== =============:2
~.;,
od 1"
'01
01:;
[.QBUS Command Pa'cket' Re"aq"f:r
.,
............+.;.~.:- _ .. .., ......... ~ ............... : .. - ... - ........................ - ................ ~ ............ -- ..... _ ... _ ............... ~ .......... ..,,.. ...... ,.. .. -
~ '0 0~2 '
02
02
I; [ Q
BUS: Com'm'"a:f.{(f
Pac ke t Wr i t e ]
_________
________ ... _____ .. __
_ ... _________________ t ___
,::, OO~ ~'03
03
;~'~o4 ,r::- 04
04
;
___ ... t.. ..~:.;..:;.~ _____ ... _ ... __
; 50S 1:
05
~
~~~_~~--
...
-~~-
... - - -
[Buffer, RAM par~ty errorJ
t ... :,
. .:..
-~-[B~";;:RAM:d;~;~;,~;,6;j~:-,----~~'------------~_ ........... ~ ~ ~! ..... ;;; ..".~':!' ... :;;;. ~':_":; ... ;,.~~ ......... .; .......... ~ ....................... __ ........ _ ..... ___ ..
........... o:t.:;;:,:.: _ ......... _ ......... _ ...... _
--~~~~~
~
05
;
... ~~~~~~~---~~
[E
PIi 0 M
Che c k sam] ";
~-~--~~~~-~~~--~~-~~
_.
;~ ,"'
'
... ~----- ... ~ ... ---~---~~-~--~-~---
~d-06,,·
06
06
I ; [QBUS COimmandi-R,eS"pons""e'Ring Read] ,
--- -:::.~- -- - - - - - - - - - I - -- - -~~~,.. - ~';".'- ..;-~ - ~--~ - --- - -'- - --:---.::-, -,-'r -- - - - - ---';do107
07 I ; [QBUS" C.ornmarrdl-R'e·~po--:ns!= "Ring Wri te]!
, ., ,.
I....
t'
i-'
" , ..! '
- _ ..... .:1'"_ ... -. ................................. I ............... _ .... --1- ........ ~ .. - .... ...- -- - : - .. _ ti. ...................................... -
C(£o~
08
10
-- ..........7--·;";====== ... - - - - -
~b91•;.. _,
•
09
11
I ; [QBUS,I.n-t'e-r~r'tlpt l '"(NOr USED"
i
~ ..................... -
.... _ .... ..
.'
l--------,~,--~·;;.-..l-·:..:.;..----------------~~-----------
II
RESERVED -
-.. 010 . . . .OA
. ---------I----..
12
RESERVED
--~--
I
---~------~
.
.. ',..
~-~-~-~~~-~---------------
. •'
.
":' ;
. . -----
-----~-------------l------~~----~-~~--~-~-~-:---~--~----~--~-------Of"f~i
OB
13
I · [QBUS' DMAl('NOTU~SEDJ ~,'
i
e
_____ ~~.~ ___________ I _~ _____ ~ ___ ~ _;...:____ ~l_..:. _. _________________________ _
1 4 ::_______
; [CPUi·¥'ailure]
__ 012'
J:.;, _J ____OC,_________
..:.::-:.. _________":_ '___________._-..,. -:-"___
o1 3; 0 D
1 5 I ; [ Wa t ch do g <T i me r"L -, ,'; -: ;, . ! \ ~ ,
~
,
I-
-:.
.,
••, ' .
•
.......
"'-
__________ _
:~.::-,,,,
11!'
I'
-----~-------------I-----------------~-~~-~--r--~~~--~ -~--r---------
o1 4:
0E
16
I R E SE RYE D .
015 ';" 'OF
17
I ; Int-'errupt Wr'i te (ijOT USED)
__ "';iJ~ _L____________
'.~
~
!
'.
1_ - - - - -:.. .-_ - ---- ... - - - - - - ' '; -~ -- - ':.,':.::::": : : :.. .: :. ~ : - - - - - - - - - - -
-~~- ... ----------~----, ... ----~~~~~-~--------~- ... ~-~----~-~~---~----------~
01~ -~
10
20
; MAINTENANCE READ/WRITE Ihval id Regi~'n;
.;.....
Identifier
------!-----------11
21
'['r5U\.Dr~yerJ
-------~~---~---~---~---------~----------------; MAINTENANCE WRITE Loa~~b N6n~Co~dable;
Controller [DU Dr~yerJ,
.... - _ ....... - -. ..... -. -~.-I'-'" - ....... ,..,....;..._.- ... -: - ............ - .......... -- ........... - ................... ....
018
019
12
13
22
23
24
; Controller RAM Error: ('hon:;"parit'y)"
................................
~ -"-.~..~'"
.
-.- .......................................
-- .......... -...... _........... _. . ...
"
..
-
; INIT S'~qu.:ence Er'F9-r JDU Driver Initialization]
-----------~~-~~~~~~~--~~----------------------~
RESERVED
i, '
~
5-·5
----------~~-~--------~-~~~----~----------------------------~~-------'sELF-DETECTED FATAL PO RT / CONTROLLERE RRORS (co nt. )
----------------------------------------------------------_._--------____ .. ___ . . _ _ . .
..
. . ---.. . .
. . ..
"" ... -
,
,
',..
.'", ~ ".
"
~-----.'
.
--~-----.~--'-c-~-
'
,
.
.,..~-----
~
~~~"!!"'!!!o' . . . ~-~""'!!!'&~
...
'!!'!!!'~~
.
'
"~
. .- - .....-~~~---~
--15---- ?5_~.- -;-CDU -Ii ~ i~; ~ -1 ~ it ;:;i ;:~; t;:; ~ -Ii;:; g~;; t1; "',f;i]' ~~; j 1
~ '-'-----------'---·-----------'--.-,
"
DEC'"
'OCT,"~
HE:X
D,E'SCRIPTION" .;:
I
"
------~------------ -----~----~-~--~---------------------~--.~------
I • - 021 ~-
.
f.,'';''.~ 022
-.,...~
.
..
...
..
O,i3
,
,.6'
-: --- -
:-~O~4-.... - -- ~
~-
-,--~-,---'---------------'--""r.-'---''"''--,----1,
26
-.~ - ~ ....-
'j
E'EPROM ':Checksum "Failure
..... '~.~ 1---"':""" ... -" ...,":~'-
---'~.~ .... -
1
--..;, _"!ilia - -.-.- - _ .... - - - - - - - - - -
17" ' ,27 ,,:fr'.., ,;S;BIC
. Regist'er
Error~'
.
",,".
,
.
,.,~- ... -......-.~~~:~.~...;.; .... ~
i
I
,
--'8 -::--:-30~- [-~-~~tc~,;;~; ~i i;; :-E;;~; ----------------- -~~ -,. ---]
:-- --:- ... - --,-'- - .. - ....... l - .. - -. ... -- ..
~.--
-- - ............ - - .. - - -"-- .. - - .......... - ............... - ... ~.
~ ... ~.~~
.. :-"'_.... I
",' ~~~~~~-~~~~':--~-~~-if';,-:~~'~::-':::~:--------~-------------~-----·~---~'~-11
A
r r r . -,
. . . ~-o 27----'B- ~- -33.--!-~ -QBus -p~~;; -F;ii~;; --------------------------, --0 28--,- ~ ; c----3-4 -- i-~ -N ~~;;1; t; ~ t- M;; ~;; -D;t;~ t; d-d~ ~ i ~ ~ -H~;:t i ---;--1
:.
026
,
QBUS DMAE
32
1
'.:
0
I
.
I
"
,"
.
'
Controller DMA's'
. I ,
,
'
"""
i
I
,-~O'29----1D:;--3:5-~ i -~-p.;;it;-E;;~~-D;t;dt;d-d~;i~~-H~;tic~;t;~ii·;;-l
"
'
'1
DMAts
t
OMAt s
'
030
I
.. -,1£...... -~---:-~--;;..~-------- ...,--'--,~--'..;;--------:-~----------------,
36 : 1,QBIC Data Compare error durlng
.'.'~.;,.;.,,~-,;.~-'
HostfControlle~
'
I
-------------------f-------..
-------------~------------------~----~~-~
031,
1F
3 7 : ; port b parity error detected
--032----20----;;o--1-~-~~;t-~-~;;it;-;;;~;-d;t;~t;d-----------"--"-~1
-
-
-
"'"!' ... - . - , - . - . - , -
-~
-
-
033 . . 21 "
-
-
41
-
-
- -..: - -,- - - - -.- - - - - - - - - - - - - - - - - - - - - - - - - - - - - -'- - - -1
!I - ,-. - -RESERVED
.
' . ,
j
., --034----22----;;2--!-~-~~;~-;-;;;it;-;;;~;-d;t;~t;d-------,-------~--l
035
23
43
I
NOT USED
I
,I~---------~--~~----~'t,--~---------------------------------~-~---------~
.. __ ..:
______________________________________ • __ ,
IL _____________
255.,
FF )77
!
~o
I
~_
I
~~_~
~-~--~
5-6
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement