For Magnetic Tape Devices
Code Feature X'80' 7 Track X'40' Dual Density X'20' Translate X'10' Data Conversion
For Direct Access Storage Devices
Feature
Rotational Position Sensing (RPS) Extended Sense Bytes (24 bytes)
Top Half of 2314 Used as 2311
Bottom Half of 2314 Used as 2311
35MB Data Module (mounted) 70MB Data Module (mounted)
Code X'80' X'40' X'20' X'10' X'08' X'04' X'02' X'01' Reserve/Release are valid CCW operation codes 3330V Virtual MSS volume
For special devices
Code Feature X'20' Type II channel adapter for 370X X'10' Type I channel adapter for 370X Figure 69 (Part 5 of 5). CP Device Classes, Types, Models, and Features
Introduction to Debugging 521
Identifying and Locating a Pageable Module
If a program check PSW or SVC PSW points to an address beyond the end of the CP resident nucleus, the failing module is a page able module. The CP system load
map identifies the end of the resident nucleus. .
Go to the address indicated in the PSW. Backtrack to the beginning of that page
frame. The first eight bytes of that page frame (the page frame containing the
address pointed to by the PSW) contains the name of the first pageable module
loaded into the page. If multiple modules exist within the same page frame, identi­
fy the module using the load map and failing address displacement within the page
frame. In most cases, register 12 points directly to the name.
To locate a pageable module whose address is shown in the load map, use the sys­
tem VMBLOK segment and page tables. For example, if the address in the load
map is 55000, use the segment and page tables to locate the module at segment 5,
page 5.
VMDUMP Records: Format and Content
When a user issues the VMDUMP command, CP dumps virtual storage of the
user's virtual machine. CP stores this dump on the reader spool file of a virtual
machine that the user specified as an operand on the VMDUMP command. CP writes the storage dump to the spool file as a series of logical records. Each
spool file record and each logical dump record is 4096-bytes long. However,
because each spool file record contains a header, one logical dump record does not
fit into one spool file record. For this reason, CP splits a logical dump record into
two parts. CP writes one part to one spool file record and the other part to an
adjacent spool file record. The size of each part varies depending upon the amount
of space remaining in the spool file record that CP is currently using. Thus, each
logical dump record spans two spool file records. --Fig 'fI1' unknown --shows
the format of spool file records, the format of logical dump records, and how log­
ical dump records span spool file records.
The first spool file record contains a spool page buffer linkage block (SPLINK) fol­
lowed by a TAG area followed by dump information. All other spool file records
contain only a SPLINK followed by dump information.
A SPLINK, which contains data needed to locate information in the associated
spool file record, has the following format:
hexadecimal
offset
o
4
8
C
length
4 bytes
4 bytes
4 bytes
4 bytes
content
the DASD location (DCHR) of the next page
buffer
the DASD location (DCHR) of the previous
page buffer
binary zeros
the number of data records in the buffer
The TAG area contains either binary zeros or user supplied data. If a virtual
machine program or the user has issued the TAG command, the TAG area contains
the information provided via this command. Otherwise it contains binary zeros.
522 VM/SP System Programmer's Guide
Previous Page Next Page