33
33
created during processing should be written. If
34
34
not specified the current directory is used. The
35
35
work files created during processing are called
36
- "cmpout.bin" and "expout .txt".
36
+ "cmpout.bin", "expout.txt" and "md5sum .txt".
37
37
38
38
OPTIONS
39
39
40
40
-t Translate ASCII/EBCDIC (default = no)
41
41
-a Algorithm to be used (default = 0)
42
- -r Number of random tests (default = 100)
42
+ -r Number of random tests (default = 4)
43
+ -r Number of speed test repeats (default = 4)
43
44
-n No hard-coded test cases (default = all)
44
45
-bn Use only buffer sizes set 'n' (default = all)
45
46
-on Use only buffer offsets set 'n' (default = all)
46
47
-speed Speed test; -r = repeats (see NOTES below)
48
+ -z Zero Padding [enabled:requested] (see NOTES below)
49
+ -w Zero Padding Alignment (default = 8 bit)
47
50
48
51
EXAMPLES
49
52
50
- cmpsctst .\files .\dicts .\work > EXTREMELY-long-test.log
51
- cmpsctst .\files\small .\dicts .\work -r 0 > non-random-test.log
52
- cmpsctst .\files\small .\dicts .\work -n -r 16 > randoms-test.log
53
- cmpsctst .\files\large .\dicts .\work -speed -r 1000 > speed-test.log
53
+ cmpsctst .\files .\dicts .\work -z > EXTREMELY-long-test.log
54
+ cmpsctst .\files\small .\dicts .\work -r 0 -z -w 3 > VERY-long- non-random-test.log
55
+ cmpsctst .\files\small .\dicts .\work -n -r 3 -z 0:1 > short- randoms-test.log
56
+ cmpsctst .\files\large .\dicts .\work -speed -r 100 > speed-test.log
54
57
55
58
NOTES
56
59
94
97
95
98
Because a default test using the many built-in (hard-coded)
96
99
buffer size and offset values can cause cmpsctst to run for
97
- an *VERY* long time since there are so many of them (24336,
100
+ a *VERY* long time since there are so many of them (33124
98
101
times #of test files, times #of test dictionaries = total
99
102
number of tests, each of which does a full compression and
100
103
expansion) the -n (no-non-random) option allows you to skip
101
104
these tests to allow cmpsctst to finish much more quickly.
102
105
103
106
You can also use the -bn and -on options to choose a smaller
104
107
subset of buffer size and/or offset values too. There are 3
105
- buffer size sets with 2, 6 and 4 values in each of their sets
106
- respectively, and 2 sets of offset values with 9 and 6 values
107
- in each of their sets, respectively.
108
+ buffer size sets with 3, 6 and 4 values in each of their
109
+ respective sets, and 2 sets of offset values with 9 and 5
110
+ values in each of their respective sets. Specifying -b3 -o2
111
+ for example, will cause 400 total tests to be performed for
112
+ each file/dictionary pair. Thus for a set of 2 files and 3
113
+ dictionaries, a grand total of 2400 tests will be performed.
108
114
109
115
The -r value defines either the number of random buffer size
110
116
and offset tests to perform or else the number of repeats to
126
132
specific buffer size and offset values, you need to call the
127
133
CMPSCTST.EXE tool yourself (i.e. don't use cmpsctst.rexx).
128
134
135
+ The '-z' (Zero Padding) option controls CMPSC-Enhancement
136
+ Facility. Specify the option as two 0/1 values separated by
137
+ a single colon. The first 0/1 defines whether the facility
138
+ should be simulated as being enabled or not. The second 0/1
139
+ controls whether the Zero Padding option (GR0 bit 46) should
140
+ be set (requested) or not in the compression/expansion call.
141
+ If the -z option is not specified the default is 0:0. If the
142
+ option is specified but without any arguments then it's 1:1.
143
+
144
+ The '-w' (Zero Padding Alignment) option controls adjustment
145
+ of the model-dependent storage boundary used by zero padding.
146
+ The value should be specified as a power of 2 number of bits
147
+ ranging from 1 to 12 (i.e. zero pad to 2-4096 byte boundary).
148
+
129
149
All dictionaries must be in RAW BINARY format and MUST use
130
150
the following file naming convention:
131
151
166
186
167
187
VERSION
168
188
169
- 2.1 (May 22, 2012)
189
+ 2.4 (July 2012)
170
190
171
191
-------------------------------------------------------------------------------
172
192
173
- CMPSC Instruction Testing Tool, version 2.1 .0
193
+ CMPSC Instruction Testing Tool, version 2.4 .0
174
194
Copyright (C) 2012 Software Development Laboratories
175
195
176
196
Options:
@@ -188,7 +208,10 @@ Options:
188
208
189
209
-0 Format-0 (default)
190
210
-1 Format-1
191
- -s Symbol Size (1-5 or 9-13; default = 2)
211
+ -s Symbol Size (1-5 or 9-13; default = 2)
212
+
213
+ -z Zero Padding (enabled:requested; see NOTES)
214
+ -w Zero Padding Alignment (default = 8 bit)
192
215
193
216
-t Translate (ASCII <-> EBCDIC as needed)
194
217
-v Verbose [filename]
@@ -204,7 +227,7 @@ Returns:
204
227
Examples:
205
228
206
229
CMPSCTST -c -i 8192:*:foo.txt -o *:4095:foo.cmpsc -r 1000 \
207
- -t -d cdict.bin -x edict.bin -1 -s 10 -v rpt.log
230
+ -t -d cdict.bin -x edict.bin -1 -s 10 -v rpt.log -z 0:1
208
231
209
232
CMPSCTST -e -i foo.cmpsc -o foo.txt -t -x edict.bin -s 10 -q
210
233
@@ -227,6 +250,21 @@ Notes:
227
250
If you want your buffers to always be page aligned then you need to
228
251
specify 0 for the offset. Dictionaries will always be page aligned.
229
252
253
+ The '-z' (Zero Padding) option controls CMPSC-Enhancement Facility.
254
+ Specify the option as two 0/1 values separated by a single colon.
255
+ If the -z option is not specified the default is 0:0. If the option
256
+ is specified but without any arguments then the default is 1:1.
257
+ The first 0/1 defines whether the facility should be enabled or not.
258
+ The second 0/1 controls whether the GR0 zero padding option bit 46
259
+ should be set or not (i.e. whether the zero padding option should be
260
+ requested or not). The two values together allow testing the proper
261
+ handling of the facility since zero padding may only be attempted
262
+ when both the Facility bit and the GR0 option bit 46 are both '1'.
263
+
264
+ The '-w' (Zero Padding Alignment) option allows adjusting the model-
265
+ dependent integral storage boundary for zero padding. The value is
266
+ specified as a power of 2 number of bits ranging from 1 to 12.
267
+
230
268
Use the '-t' (Translate) option when testing using ASCII test files
231
269
since most dictionaries are intended for EBCDIC data. If specified,
232
270
the test data is first internally translated from ASCII to EBCDIC
0 commit comments