-
Notifications
You must be signed in to change notification settings - Fork 84
/
Copy pathunixbench.py
144 lines (124 loc) · 6.41 KB
/
unixbench.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
import os
import re
from autotest.client import test, utils
from autotest.client.shared import error
class unixbench(test.test):
version = 2
def initialize(self):
self.job.require_gcc()
self.err = None
# http://www.tux.org/pub/tux/niemi/unixbench/unixbench-4.1.0.tgz
def setup(self, tarball='unixbench-4.1.0.tar.bz2'):
tarball = utils.unmap_url(self.bindir, tarball, self.tmpdir)
utils.extract_tarball_to_dir(tarball, self.srcdir)
os.chdir(self.srcdir)
utils.system('patch -p1 < %s/unixbench.patch' % self.bindir)
utils.system('patch -p1 < %s/Makefile.patch' % self.bindir)
utils.make()
utils.system('rm pgms/select')
def run_once(self, args='', stepsecs=0):
vars = ('TMPDIR=\"%s\" RESULTDIR=\"%s\"' %
(self.tmpdir, self.resultsdir))
if stepsecs:
# change time per subtest from unixbench's defaults of
# 10 secs for small tests, 30 secs for bigger tests
vars += ' systime=%i looper=%i seconds=%i'\
' dhrytime=%i arithtime=%i' \
% ((stepsecs,) * 5)
os.chdir(self.srcdir)
try:
utils.system(vars + ' ./Run ' + args)
finally:
times_path = os.path.join(self.resultsdir, 'times')
# The 'times' file can be needlessly huge as it contains warnings
# and error messages printed out by small benchmarks that are
# run in a loop. It can easily compress 100x in such cases.
if os.path.exists(times_path):
utils.system("gzip -9 '%s'" % (times_path,), ignore_status=True)
report_path = os.path.join(self.resultsdir, 'report')
self.report_data = open(report_path).readlines()[9:]
def cleanup(self):
# check err string and possible throw
if self.err is not None:
raise error.TestError(self.err)
def check_for_error(self, words):
l = len(words)
if l >= 3 and words[-3:l] == ['no', 'measured', 'results']:
# found a problem so record it in err string
key = '_'.join(words[:-3])
if self.err is None:
self.err = key
else:
self.err = self.err + " " + key
return True
else:
return False
def postprocess_iteration(self):
keyval = {}
for line in self.report_data:
if not line.strip():
break
words = line.split()
# look for problems first
if self.check_for_error(words):
continue
# we should make sure that there are at least
# 6 guys before we start accessing the array
if len(words) >= 6:
key = '_'.join(words[:-6])
key = re.sub('\W', '', key)
value = words[-6]
keyval[key] = value
for line in self.report_data:
if 'FINAL SCORE' in line:
keyval['score'] = line.split()[-1]
break
self.write_perf_keyval(keyval)
""" Here is a sample report file:
BYTE UNIX Benchmarks (Version 4.1.0)
System -- Linux adrianbg 2.6.18.5 #1 SMP Thu J Start Benchmark Run: Tue Sep 1
9 interactive users.
21:03:50 up 5 days, 7:38, 9 users, load average: 0.71, 0.40, 0.25
lrwxrwxrwx 1 root root 4 Aug 15 09:53 /bin/sh -> bash
/bin/sh: symbolic link to `bash'
/dev/sda6 192149596 91964372 90424536 51% /home
Dhrystone 2 using register variables 7918001.7 lps (10.0 secs, 10 samples)
System Call Overhead 1427272.7 lps (10.0 secs, 10 samples)
Process Creation 11508.6 lps (30.0 secs, 3 samples)
Execl Throughput 4159.7 lps (29.7 secs, 3 samples)
File Read 1024 bufsize 2000 maxblocks 1708109.0 KBps (30.0 secs, 3 samples)
File Write 1024 bufsize 2000 maxblocks 788024.0 KBps (30.0 secs, 3 samples)
File Copy 1024 bufsize 2000 maxblocks 452986.0 KBps (30.0 secs, 3 samples)
File Read 256 bufsize 500 maxblocks 508752.0 KBps (30.0 secs, 3 samples)
File Write 256 bufsize 500 maxblocks 214772.0 KBps (30.0 secs, 3 samples)
File Copy 256 bufsize 500 maxblocks 143989.0 KBps (30.0 secs, 3 samples)
File Read 4096 bufsize 8000 maxblocks 2626923.0 KBps (30.0 secs, 3 samples)
File Write 4096 bufsize 8000 maxblocks 1175070.0 KBps (30.0 secs, 3 samples)
File Copy 4096 bufsize 8000 maxblocks 793041.0 KBps (30.0 secs, 3 samples)
Shell Scripts (1 concurrent) 4417.4 lpm (60.0 secs, 3 samples)
Shell Scripts (8 concurrent) 1109.0 lpm (60.0 secs, 3 samples)
Shell Scripts (16 concurrent) 578.3 lpm (60.0 secs, 3 samples)
Arithmetic Test (type = short) 1843690.0 lps (10.0 secs, 3 samples)
Arithmetic Test (type = int) 1873615.8 lps (10.0 secs, 3 samples)
Arithmetic Test (type = long) 1888345.9 lps (10.0 secs, 3 samples)
Arithmetic Test (type = float) 616260.3 lps (10.0 secs, 3 samples)
Arithmetic Test (type = double) 615942.1 lps (10.0 secs, 3 samples)
Arithoh 18864899.5 lps (10.0 secs, 3 samples)
Dc: sqrt(2) to 99 decimal places 161726.0 lpm (30.0 secs, 3 samples)
Recursion Test--Tower of Hanoi 89229.3 lps (20.0 secs, 3 samples)
INDEX VALUES
TEST BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 7918001.7 678.5
Double-Precision Whetstone 55.0 1948.2 354.2
Execl Throughput 43.0 4159.7 967.4
File Copy 1024 bufsize 2000 maxblocks 3960.0 452986.0 1143.9
File Copy 256 bufsize 500 maxblocks 1655.0 143989.0 870.0
File Copy 4096 bufsize 8000 maxblocks 5800.0 793041.0 1367.3
Pipe Throughput 12440.0 1048491.9 842.8
Pipe-based Context Switching 4000.0 300778.3 751.9
Process Creation 126.0 11508.6 913.4
Shell Scripts (8 concurrent) 6.0 1109.0 1848.3
System Call Overhead 15000.0 1427272.7 951.5
=========
FINAL SCORE 902.1
"""