Trace tables

Collated ECB trace table

The collated ECB trace table provides a snapshot of system activity by scanning every ECB when a system error occurs. Each ECB in the z/TPF system maintains a separate macro trace table. The most recent trace entries are extracted and displayed. The ECB address, the I-stream in which the macro was issued, the time stamp, macro name, parameters, the state of the system (problem or supervisor), and the key (0 or 1) that the system was in when the macro was issued are extracted. This provides an abbreviated real-time trace of all system activity. The ZSTRC command controls system trace options.

The DOF utility prints the trace table when a dump is issued. The entries are displayed in reverse chronological order, beginning with the most recent entry. The collated macro trace display is included once and contains all I-stream activity. For an example of the collated ECB trace table, see Collated ECB trace table.

Collated input/output (I/O) trace table

The collated input/output (I/O) trace table shows the I/O activity at the time that a system error occurs. The table contains information for each I-stream that was active at the time of the system error. For an example of the collated I/O trace table, see Collated input/output (I/O) trace table.

ECB heap trace table

The ECB heap trace table provides information about the latest ECB heap storage requests and releases for an ECB. Specifically, the ECB heap trace table contains information about CALOC, MALOC, RALOC, and FREEC macro requests, and calloc, calloc64, malloc, malloc64, realloc, realloc64 and free C function requests. Use the ZSTRC command to turn on or turn off ECB heap trace.

ECB heap trace does not track mymalloc calls. All mymalloc API calls appear in the C function trace because a mymalloc macro calls a corresponding mymalloc C function.

The ECB heap trace information is included in the ECB virtual memory (EVM) portion of the dump. For an example, see ECB virtual memory (EVM). You also can display the information with the ZDECB command.

ECB level socket trace table

The ECB level socket trace table provides information about all socket APIs that were issued by a particular ECB. Use the ZSTRC command to turn on or turn off the ECB level socket trace. For an example, see ECB level socket trace.

Systems Network Architecture (SNA) input/output (I/O) trace table

The SNA I/O trace table contains trace information for SNA I/O interrupts that occur during Network Control Program (NCP) exchange identifier (XID) exchanges, adjacent link station (ALS) XID exchanges, and channel-to-channel (CTC) XID exchanges. Information about significant steps in XID7 processing for CTC devices is also logged in this table. The trace table also contains detailed trace information for NCP, ALS, and CTC asynchronous interrupts. Normal data transfer operations and attention-only interrupts that occur during NCP and ALS data transfer operations are not traced. For more information about the SNA I⁄O trace table, see SNA I/O trace facility.

The SNA I/O trace table is included in system error dumps that have the control program keyword (ICP) specified in the dump override table. The DOF utility prints the SNA I/O trace table when these dumps are issued and labels it with the XID dump tag. For information about how to interpret the contents of the SNA I/O trace table, see the IOTBL data macro.

Path information unit (PIU) input/output (I/O) trace table

The path information unit (PIU) trace table contains a copy of the data transferred between the z/TPF system and remote resources. This trace table is included in system error dumps that have the SNA keyword (ISNA) specified in the dump override table.

When system error dumps are issued, the DOF utility prints the PIU trace table and labels it with the PIU dump tag.

For more information about the PIU trace table, see Using the Path Information Unit (PIU) trace facility. For more information about how to interpret the contents of the PIU trace table, see the IPTBL data macro.

Communication control program trace routine

The communication control program (CCP) trace routine provides a sequential record of CCP macros, start I/O and halt I/O activity, and multiplexer interrupts that occur on one line, or on one or more groups of lines. The CCP trace routine is started, modified, and ended by commands. These messages and the collected CCP trace data are stored in a main storage block. Optionally, you can use this block in main storage only as a wraparound log or, when full, to write its contents to tape. If it is a main storage log only, it is displayed unformatted whenever a dump occurs. If written to the RTA, the DOF utility produces a formatted list of all items traced, including the commands used for the CCP trace activity. On any dump that occurs, the wraparound CCP trace main storage block is displayed with a record of the last 40 activities traced. For more information about CCP trace operations, see Communications control program trace.