summaryrefslogtreecommitdiffstats
path: root/tests/benchmarks/corelib/tools/qstring/generatelist.pl
blob: 1f10b638b22443eeeabaf00d5ef75dd7721cce99 (plain)
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
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
#!/usr/bin/perl
## Copyright (C) 2012 Nokia Corporation and/or its subsidiary(-ies).
## All rights reserved.
## Contact: Nokia Corporation (qt-info@nokia.com)
##
## This file is part of the QtCore module of the Qt Toolkit.
##
## $QT_BEGIN_LICENSE:LGPL$
## GNU Lesser General Public License Usage
## This file may be used under the terms of the GNU Lesser General Public
## License version 2.1 as published by the Free Software Foundation and
## appearing in the file LICENSE.LGPL included in the packaging of this
## file. Please review the following information to ensure the GNU Lesser
## General Public License version 2.1 requirements will be met:
## http://www.gnu.org/licenses/old-licenses/lgpl-2.1.html.
##
## In addition, as a special exception, Nokia gives you certain additional
## rights. These rights are described in the Nokia Qt LGPL Exception
## version 1.1, included in the file LGPL_EXCEPTION.txt in this package.
##
## GNU General Public License Usage
## Alternatively, this file may be used under the terms of the GNU General
## Public License version 3.0 as published by the Free Software Foundation
## and appearing in the file LICENSE.GPL included in the packaging of this
## file. Please review the following information to ensure the GNU General
## Public License version 3.0 requirements will be met:
## http://www.gnu.org/copyleft/gpl.html.
##
## Other Usage
## Alternatively, this file may be used in accordance with the terms and
## conditions contained in a signed written agreement between you and Nokia.
##
##
##
##
##
## $QT_END_LICENSE$
#
# Parses a file (passed as argument) that contains a dump of pairs of
# strings and generates C source code including said data.
#
# The format of the file is:
#   LEN = <len> <keyword> <align1> <align2>\n<data1><data2>\n
# where:
#   LEN		the literal string "LEN"
#   <len>	the length of the data, in 16-bit words
#   <keyword>	the literal string "SAME" or "DIFF"
#   <align1>    the alignment or pointer value of the first data
#   <align2>    the alignment or pointer value of the second data
#   <data1>     the first data
#   <data2>     the second data
#   \n          newline
#
# The code to write this data would be:
#   fprintf(out, "LEN = %d %s %d %d\n", len,
#          (p1 == p2) ? "SAME" : "DIFF",
#          uint(quintptr(p1)) & 0xfff, uint(quintptr(p2)) & 0xfff);
#   fwrite(p1, 2, len, out);
#   fwrite(p2, 2, len, out);
#   fwrite("\n", 1, 1, out);

sub printUshortArray($$$) {
    $str = $_[0];
    $align = $_[1] & 0x1f;
    $offset = $_[2];

    die if ($align & 1) != 0;
    $align /= 2;

    $len = (length $str) / 2;
    $headpadding = $align & 0x7;
    $tailpadding = 8 - (($len + $headpadding) & 0x7);
    $multiplecachelines = ($align + $len) > 0x20;

    if ($multiplecachelines) {
	# if this string crosses into a new cacheline, then
	# replicate the result
	$headpadding |= ($offset & ~0x1f);
	$headpadding += 0x20
	    if ($headpadding < $offset);
	$headpadding -= $offset;
	++$cachelinecrosses;
    }
    for $i (1..$headpadding) {
	print 65536-$i,",";
    }
    print "\n        " if ($headpadding > 0);
    print "    " if ($headpadding == 0);

    for ($i = 0; $i < $len * 2; $i += 2) {
	print " ", ord(substr($str, $i, 1)) +
	    ord(substr($str, $i + 1, 1)) * 256,
	    ",";
    }
    print "\n    " if ($tailpadding > 0);

    for $i (1..$tailpadding) {
	print 65536-$i, ",";
    }
    print " // ", $offset + $headpadding + $len + $tailpadding;
    print "+" if $multiplecachelines;

    return ($offset + $headpadding, $offset + $headpadding + $len + $tailpadding);
}

print "// This is a generated file - DO NOT EDIT\n\n";

print "#include \"data.h\"\n\n";

print "const ushort stringCollectionData[] __attribute__((aligned(64))) = {\n";
$count = 0;
$offset = 0;
$totalsize = 0;
$maxlen = 0;
$cachelinecrosses = 0;

open IN, "<" . $ARGV[0];
while (1) {
    $line = readline(*IN);
    last unless defined($line);
    $line =~ /LEN = (\d+) (\w+) (\d+) (\d+)/;
    $len = $1;
    $data[$count]->{len} = $len;
    $sameptr = $2;
    $data[$count]->{align1} = $3 - 0;
    $data[$count]->{align2} = $4 - 0;

    # statistics
    $alignhistogram{$3 & 0xf}++;
    $alignhistogram{$4 & 0xf}++;
    $samealignments{$3 & 0xf}++ if ($3 & 0xf) == ($4 & 0xf);

    read IN, $a, $len * 2;
    read IN, $b, $len * 2;

    <IN>;			# Eat the newline

    if ($len == 0) {
	$data[$count]->{offset1} = $offset;
	$data[$count]->{offset2} = $data[$count]->{offset1};
	++$data[$count]->{offset2} if ($sameptr eq "DIFF");
    } else {
	print "    // #$count\n";
	print "    ";
	($data[$count]->{offset1}, $offset) =
	    printUshortArray($a, $data[$count]->{align1}, $offset);
	print "\n    ";
	die if ($offset & 0x7) != 0;

	if ($sameptr eq "DIFF") {
	    ($data[$count]->{offset2}, $offset) =
		printUshortArray($b, $data[$count]->{align2}, $offset);
	    print "\n\n";
	} else {
	    $data[$count]->{offset2} = $data[$count]->{offset1};
	    print "\n\n";
	}
    }
    ++$count;

    $totalsize += $len;
    $maxlen = $len if $len > $maxlen;
}
print "};\n";
close IN;

print "const struct StringCollection stringCollection[] = {\n";
for $i (0..$count-1) {
    print "    {",
        $data[$i]->{len}, ", ",
        $data[$i]->{offset1}, ", ",
        $data[$i]->{offset2}, ", ",
        $data[$i]->{align1}, ", ",
        $data[$i]->{align2},
        "},     // #$i\n";
    next if $data[$i]->{len} == 0;
    die if (($data[$i]->{offset1} & 0x7) != ($data[$i]->{align1} & 0xf)/2);
    die if (($data[$i]->{offset2} & 0x7) != ($data[$i]->{align2} & 0xf)/2);
}
print "};\n";

print "const int stringCollectionCount = $count;\n";
print "const int stringCollectionMaxLen = $maxlen;\n";
printf "// average comparison length: %.4f\n", ($totalsize * 1.0 / $count);
printf "// cache-line crosses: %d (%.1f%%)\n",
    $cachelinecrosses, ($cachelinecrosses * 100.0 / $count / 2);

print "// alignment histogram:\n";
for $key (sort { $a <=> $b } keys(%alignhistogram)) {
    $value = $alignhistogram{$key};
    $samealigned = $samealignments{$key};
    printf "//   0xXXX%x = %d (%.1f%%) strings, %d (%.1f%%) of which same-aligned\n",
	$key, $value, $value * 100.0 / ($count*2),
        $samealigned, $samealigned * 100.0 / $value;
    $samealignedtotal += $samealigned;
}
printf "//   total  = %d (100%) strings, %d (%.1f%%) of which same-aligned\n",
    $count * 2, $samealignedtotal, $samealignedtotal * 100 / $count / 2;