Add a script that generates a string table of 8-bits

1 files changed, 203 insertions, 0 deletions

diff --git a/tests/benchmarks/corelib/tools/qstring/generatelist_char.pl b/tests/benchmarks/corelib/tools/qstring/generatelist_char.pl
new file mode 100755
index 0000000..1c7003a
--- /dev/null
+++ b/tests/benchmarks/corelib/tools/qstring/generatelist_char.pl
@@ -0,0 +1,203 @@
+#!/usr/bin/perl
+# -*- mode: utf-8; tabs: nil -*-
+## Copyright (C) 2011 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.
+##
+## $QT_BEGIN_LICENSE:LGPL$
+## No Commercial Usage
+## This file contains pre-release code and may not be distributed.
+## You may use this file in accordance with the terms and conditions
+## contained in the Technology Preview License Agreement accompanying
+## this package.
+##
+## GNU Lesser General Public License Usage
+## Alternatively, 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.
+##
+## If you have questions regarding the use of this file, please contact
+## Nokia at qt-info@nokia.com.
+##
+##
+##
+##
+##
+##
+##
+##
+## $QT_END_LICENSE$
+#
+# 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 printCharArray($$$) {
+    $str = $_[0];
+    $align = $_[1] & 0x3f;
+    $offset = $_[2];
+
+    $headpadding = $align & 0xf;
+    $tailpadding = 16 - (($len + $headpadding) & 0xf);
+    $multiplecachelines = ($align + $len) > 0x40;
+
+    if ($multiplecachelines) {
+        # if this string crosses into a new cacheline, then
+        # replicate the result
+        $headpadding |= ($offset & ~0x3f);
+        $headpadding += 0x40
+            if ($headpadding < $offset);
+        $headpadding -= $offset;
+        ++$cachelinecrosses;
+    }
+
+    if ($headpadding > 0) {
+        print "    \"";
+        for $i (1..$headpadding) {
+            printf "\\%o", 256-$i;
+        }
+        print "\"\n";
+    }
+
+    print "        \"";
+    for ($i = 0; $i < $len; $i++) {
+        $c = substr($str, $i, 1);
+        if (ord($c) >= 0x20 && ord($c) <= 0x7f) {
+            print $c;
+        } else {
+            printf "\\%o\"\"", ord($c);
+        }
+    }
+
+    if ($tailpadding > 0) {
+        print "\"\n    \"";
+        for $i (1..$tailpadding) {
+            printf "\\%o", 256-$i;
+        }
+    }
+    print "\" // ", $offset + $headpadding + $len + $tailpadding;
+    print "+" if $multiplecachelines;
+    print "\n";
+
+    return ($offset + $headpadding, $offset + $headpadding + $len + $tailpadding);
+}
+
+print "// This is a generated file - DO NOT EDIT\n\n";
+
+print "#include \"data.h\"\n\n";
+
+$varname = shift @ARGV;
+print "const char " . $varname . "Data[] __attribute__((aligned(64))) = {\n";
+$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;
+    read IN, $b, $len;
+
+    <IN>;                       # Eat the newline
+
+    if ($len == 0) {
+        $data[$count]->{offset1} = $offset;
+        $data[$count]->{offset2} = $data[$count]->{offset1};
+        ++$data[$count]->{offset2} if ($sameptr eq "DIFF");
+    } else {
+        print "    // #$count\n";
+        ($data[$count]->{offset1}, $offset) =
+            printCharArray($a, $data[$count]->{align1}, $offset);
+        die if ($offset & 0xf) != 0;
+
+        if ($sameptr eq "DIFF") {
+            ($data[$count]->{offset2}, $offset) =
+                printCharArray($b, $data[$count]->{align2}, $offset);
+        } else {
+            $data[$count]->{offset2} = $data[$count]->{offset1};
+        }
+        print "\n";
+    }
+    ++$count;
+
+    $totalsize += $len;
+    $maxlen = $len if $len > $maxlen;
+}
+print "};\n";
+close IN;
+
+print "const struct StringCollection " . $varname . "[] = {\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} & 0xf) != ($data[$i]->{align1} & 0xf));
+    die if (($data[$i]->{offset2} & 0xf) != ($data[$i]->{align2} & 0xf));
+}
+print "};\n";
+
+print "const int " . $varname . "Count = $count;\n";
+print "const int " . $varname . "MaxLen = $maxlen;\n";
+printf "// average comparison length: %.4f\n", ($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;