1 04/05/84 cumulative_page_trace, cpt
2
3 Syntax: cpt command_line -control_args
4
5
6 Function: accumulates page trace data so that the total set of pages
7 used during the invocation of a command or subsystem can be determined.
8 The command accumulates data from one invocation of itself to the next.
9 The output is in tabular format showing all pages that have been
10 referenced by your process. You can obtain the same trace as that
11 produced by page_trace.
12
13
14 Arguments:
15 command_line
16 is a character string to be interpreted by the command processor as
17 a command line. If this character string contains blanks, enclose
18 it in quotes. All procedures invoked as a result of processing this
19 command line are metered by cumulative_page_trace.
20
21
22 Control arguments:
23 -count, -ct
24 prints the accumulated results, giving the number of each page and
25 the number of faults for each page. Do not use -count with -print
26 or -total see "Notes" below.
27 -flush
28 clears primary memory before each invocation of the command line and
29 after each interrupt. This helps you determine the number of page
30 faults but increases the cost.
31 -interrupt N, -int N
32 interrupts execution every N virtual CPU milliseconds for page fault
33 sampling. The default is 500 CPU milliseconds.
34 -long, -lg
35 produces output in long format, giving full pathnames.
36
37
38 -loop N
39 calls the command to be metered N times.
40 -print, -pr
41 prints the accumulated results, giving the number of each page
42 referenced. Do not use -print with -count or -total see "Notes."
43 -print_linkage_faults
44 prints all accumulated linkage faults and calls the hcs_$make_ptr
45 entry point.
46 -reset, -rs
47 resets the table of accumulated data. If the table is not reset,
48 data from the current use of cumulative_page_trace is added to that
49 obtained earlier in the process.
50 -short, -sh
51 formats output for a line length of 80.
52
53
54 -sleep N
55 waits for N seconds after each call to the command being metered.
56 -temp_dir path, -td path
57 creates temporary segments, used for flushing main memory, in the
58 directory identified by path. The default is to create them in the
59 process directory. Use -temp_dir with -flush.
60 -timers
61 includes all faults between signal and restart.
62
63
64 -total, -tt
65 prints the total number of page and segment faults and the number of
66 pages referenced for each segment. Do not use -total with -count or
67 -print see "Notes."
68 -trace_linkage_faults
69 accumulates linkage, page, and segment fault information.
70 -trace path
71 writes the trace on the segment "path" using an I/O switch named
72 "cpt.out"; cumulative_page_trace attaches and detaches this switch.
73
74
75 Notes: This command operates by sampling and reading the system trace
76 array after invocation of a command and at repeated intervals.
77
78 At least one of three generic operations must be requested. They may
79 all be combined and, if so, are performed in the following order:
80 resetting the table of accumulated data, calling the command to be
81 metered, applying the specified control arguments, and printing the
82 results in the specified format.
83
84 If 500 milliseconds is too long, messages indicate that some page
85 faults may have been missed; choose then a smaller value, but the cost
86 of a smaller value is high and may cause additional side effects. If
87 the command or subsystem to be metered includes the taking of CPUT
88 interrupts, then supply -timers. This control argument includes some
89 of the page faults of the metering mechanism as well.
90
91
92 Only one of -count, -print, or -total can be given. Each of these
93 control arguments produces printed output in a different format. If
94 more than one format is desired, the command must be invoked once for
95 each format.
96
97 For -flush to operate correctly, the directory used for temporary
98 segments must have sufficient quota for as many pages as there are in
99 main memory.