NAME

autoconf − diagnostics from the DYNIX autoconfiguration code

DESCRIPTION

When DYNIX bootstraps, it probes the machine it is running on and locates controllers, drives, and other devices, printing out what it finds on the console.  This process is driven by a system configuration table which is processed by config(8) and compiled into each kernel.

Devices in SB8000 backplane slots are noted (processor boards, memory controllers, SCED controllers and MULTIBUS adaptors).  Devices which are deconfigured, or which failed powerup diagnostics, are noted also. 

Devices plugged into the SB8000 backplane are located by a very deterministic procedure using each board’s SLIC to determine board types. 

MULTIBUS (or SCSI bus) devices are located by probing to see if their control-status registers (or target adapters and units) respond.  If not, they are reported as not found. 

A generic system may be built which allows the user to pick its root device at boot time.  If such a system is booted with the RB_ASKNAME option of (see reboot(2)), then the name of the root device is read from the console terminal at boot time (as part of the boot string), and any available device may be used.

SEE ALSO

intro(4), config(8)

DIAGNOSTICS

In general, DYNIX reports the SLIC id number of each of the boards it finds in the SB8000 bus:
 
%d processors; slic %d %d ... The processors listed in this message are available for use by DYNIX. 

Not using processors: slic %d %d ...  The processors listed in this message are either deconfigured, or failed powerup diagnostics. 

%d MULTIBUS Adapters; slic %d %d ...  The MULTIBUS adapters listed in this message were found in the system. 

No MULTIBUS Adapters No MULTIBUS adapters were found in the system. 

Not using MULTIBUS Adapters: slic %d %d ...  The listed MULTIBUS adapters were not used by DYNIX because they were deconfigured or failed powerup diagnostics. 

Not using memory boards: slic %d %d...  The listed memory boards were not used; they were either deconfigured or failed powerup diagnostics. 

Discontiguous memory [0x%x,0x%x) unused.  The present version of DYNIX does not support non-contiguous arrays of memory.  The memory described above will not be used by DYNIX. 

%d SCSI/Ether controllers; slic %d %d ...  The listed SCSI/Ether controllers were found in the system. 

Not using SCSI/Ether Controllers: slic %d %d ...  The listed SCSI/Ether controllers were deconfigured or failed powerup diagnostics. 

Pseudo devices: The kind and number of pseudo devices (those for which there is no corresponding physical device) are listed. 

%d pty Pseudo-terminals used by networking and some utilities.  %d inet Internet networking was configured into this DYNIX kernel.  %d loop The Internet loopback device was configured into the DYNIX kernel.  %d ether Internet interfaces for Ethernet are configured into this DYNIX kernel.  %d shm Prototype shared memory is configured into this DYNIX kernel. 

MBAd DMA window at 0x%x.  The physical address of the quarter-megabyte MULTIBUS adapter I/O window in the megabyte MULTIBUS address space. 

%s%d found at MBAd%d csr 0x%x, bin %d vec %d MB level %d, %d maps at %d This message reports the existence of a MULTIBUS controller found on a MULTIBUS Adapter.  Some fields in the above message may be missing for some devices.  For example,
"zt0 at MBAd0 csr 0x200, bin 5 vec 0 MB level 7, 34 maps at 0"
says that a Xylogics tape controller was found on MULTIBUS adapter 0, at I/O address 0x200. It interrupts DYNIX on bin 5 with interrupt vector 0. On the MULTIBUS, it interrupts on MULTIBUS level 7. It needed 34 MULTIBUS adapter map registers, maps 0 through 33.

%s%d at MBAd%d csr 0x%x: MB level %d overlaps previous use.  The specified MULTIBUS controller has a MULTIBUS interrupt level that conflicts with another controller on the same MULTIBUS adapter. 

%s%d at MBAd%d csr 0x%x: can’t fit %d map registers.  The specified MULTIBUS controller requested more map registers than were available on the specified MULTIBUS adapter. 

%s%d not found.  The specified controller (e.g. "zt0") was not found in the MULTIBUS. 

%s%d found at SEC%d target adapter %d unit %d bin %d vec %d input q %d output q %d This message describes a device found on a SCSI/Ether controller.  Some fields may be missing for some devices.  As an example,
“sd0 found at SEC0 target adapter 6 unit 0 bin 5 vec 0 inputq 50 output q 50”
describes a SCSI disk found on the SCSI bus at target adapter 6, logical unit 0. It will interrupt DYNIX on interrupt bin 5 with vector 0. The length of the device input queue (for commands to the SCSI/Ether controller) and output queue (for acknowledgements from the SCSI/Ether controller) are 50 entries. As another example,
“se0 found at SEC0 bin 5 vec 3 inputq 200 outputq 200”
and
“se1 found at SEC0 bin 5 vec 4 inputq 10 outputq 10”
are the lines describing the Ethernet interface on the SCSI/Ether controller. The first line describes the Ethernet input device; the second the Ethernet output device.

%s%d not found.  The listed SCSI/Ether device was not found in the system. 

Too many vectors in bin %d.  An internal consistency check failed; there were too many interrupts allocated to the interrupt bin listed in the message. 

WARNING: Swap partition[%d], dev=(%d,%d) does not exist.  The swap partition listed in the message exists on a disk that was not found in the system. 

illegal config: %s device can’t be probed for wildcards.  A device with no probe routine was wildcarded in the configuration.  With no probe routine, there is no way to attempt to find the specified device. 

SEC_probe: invalid config: non-scsi device has unit wildcard SCSI/Ether devices that are not on the SCSI bus should not have unit numbers wildcarded in the configuration file. 

%s%d: ioconf.c error, already initialized: target %d unit %d An error exists in the ioconf.c configuration file.  Somehow, DYNIX attempted to initialize the same device on the SCSI/Ether controller twice. 

can’t initialize SCSI/Ether controller %d: status %d The SCSI/Ether controller listed in the message could not be initialized.  This may be a sign of a hardware problem in the SCSI/Ether controller. 

MBAd%d: can’t map c-list (%d maps needed, %d available).  There were not enough maps available to map terminal c-lists in the specified MULTIBUS adapter. 

c-list @ MBAd%d: %d maps @ %d.  This message describes the number and location of MULTIBUS adapter maps allocated for terminal character lists (c-lists). 

MBAd%d: %d maps unused.  Some MULTIBUS adapter map registers were unused. 

4BSD