=

AST RESEARCH, INC.         TECHNICAL BULLETIN # 1131                4-10-95

=


			INTERPRETATION AND RESOLUTION OF PCI ERROR CODES


PROBLEM

Various PCI error codes may appear during the Power On Self Test (POST) in 
any system with PCI capability after one of the following events: 

1) a flash update of the system BIOS, 
2) installation of a PCI adapter,
3) the addition of an expansion option to a PCI adapter. 

The error codes are not documented and are normally not useful for support 
purposes.  For information on interpretation of PCI error codes, see the 
INTERPRETATION section below.


CAUSE

An error has occurred while configuring a PCI device on the system.  These 
errors are typically related to incorrect settings in the system's EISA or 
ISA configuration; conflicts with other devices installed in the system; or 
incompatibilities between PCI devices when multiple PCI devices have been 
installed. 


SYSTEMS AFFECTED
AST PART NUMBER AND DESCRIPTION

All PCI-based systems.


SOLUTION

To resolve a PCI error, follow the steps below.  PCI error codes are normally 
not useful in troubleshooting or for support purposes.  If an attempt to cure 
an error is unsuccessful, report the problem to the AST Product Support 
(800- 727-1278) for further diagnosis.


NOTE

Before reconfiguring the system, ensure that all system Setup and EISACU or 
ISACU settings have been recorded in the system manual or printed for 
reference.

1. Cold boot the system.

2. Run EISACU or ISACU (depending on your particular system), view the current 
settings, then save and exit.

3. Clear the current CMOS configuration information and reconfigure the system 
using EISACU or ISACU.

4. Remove the PCI boards in reverse order and re-test to determine which board 
is causing the error.

5. Report the error to AST Technical Support (800- 727-1278) for further 
investigation.


INTERPRETATION

PCI error codes are primarily intended for design engineers and hardware 
developers.  They contain specific information regarding the source of the 
error.

PCI error codes may differ if the system has multiple PCI controllers 
installed or PCI to PCI bridges. The following error interpretation is based 
on a single PCI controller (chipset) installed in the system.

The PCI error number must be divided into several parts:

DIGIT     12345678    1 2 3 4 5 6   78
EXAMPLE   83030028 =  8 3 0 3 0 0 | 28

Digit 1:

The first digit ("1" above) indicates the following:
8 = nonfatal error
C = fatal error

Digit 2:

The second digit indicates the PCI subsystem where the error occurred:
1 = Chipset code
2 = ACFG Manager
3 = PCI Manager
4 = PCIC Manager
5 = INT 1Ah Manager
6 = System Resource Manager
7 = ISA Plug and Play Manager

Digit 3:

The third digit indicates the subsystem's error number:
0

Digits 4 and 5:

The forth and fifth digits indicates PCI bus number where the error occurred
0-FF*
*In a system with a single PCI bus, this number will normally be 00.

Digits 7 and 8: 

Digits 7 and 8 are represented in binary:
7   8
16 8 4 2 1   4 2 1  
 0 0 0 0 0   0 0 0

The first 5 binary digits indicate the PCI number where the error occurred.  
This number is set by the manufacturer of the device.

AST has assigned the following device numbers:
	
	0 Neptune PCMC/CDC (CPU board bridge)
	1 NCR SCSI #1
	2 Neptune PCEB (EISA bridge)
	3 NCR SCSI #2 (Manhattan P non-array models only)
	4 PCI slot 1
	5 PCI slot 2
	6 PCI slot 3

The last three digits indicate the function number on the device where the 
error occurred.

Therefore the above example would be interpreted as follows:

EXAMPLE:  PCI error 83030028.
Digit 1:  8 = nonfatal error
Digit 2:  3 = PCI Manager
Digit 3:  PCI Manager error 0.
Digits 4 and 5: PCI bus 0
Digits 7 and 8 (represented in binary): 28 hex = 0 0 1 0 1 0 0 0 binary
	The error occurred in PCI device number 5 .
	The error occurred in device's function 0.