ICU-2038 initial performance test for UnicodeSet

X-SVN-Rev: 10017
This commit is contained in:
Alan Liu 2002-10-10 21:17:07 +00:00
parent 4765328a53
commit 5eb64a5da4
5 changed files with 457 additions and 0 deletions

View File

@ -0,0 +1,65 @@
/*
**********************************************************************
* Copyright (c) 2002, International Business Machines
* Corporation and others. All Rights Reserved.
**********************************************************************
* $Source: /xsrl/Nsvn/icu/icu/source/test/usetperf/Attic/bitset.cpp,v $
**********************************************************************
* 2002-09-20 aliu Created.
*/
#include "unicode/utypes.h"
#include "cmemory.h"
#include "bitset.h"
// TODO: have a separate capacity, so the len can just be set to
// zero in the clearAll() method, and growth can be smarter.
const int32_t SLOP = 8;
const int32_t BYTES_PER_WORD = sizeof(int32_t);
BitSet::BitSet() {
len = SLOP;
data = (int32_t*) uprv_malloc(len * BYTES_PER_WORD);
clearAll();
}
BitSet::~BitSet() {
uprv_free(data);
}
UBool BitSet::get(int32_t bitIndex) const {
uint32_t longIndex = bitIndex >> 5;
int32_t bitInLong = bitIndex & 0x1F;
return (longIndex < len) ? (((data[longIndex] >> bitInLong) & 1) != 0)
: FALSE;
}
void BitSet::set(int32_t bitIndex) {
uint32_t longIndex = bitIndex >> 5;
int32_t bitInLong = bitIndex & 0x1F;
if (longIndex >= len) {
ensureCapacity(longIndex+1);
}
data[longIndex] |= (1 << bitInLong);
}
void BitSet::clearAll() {
for (uint32_t i=0; i<len; ++i) data[i] = 0;
}
void BitSet::ensureCapacity(uint32_t minLen) {
uint32_t newLen = len;
while (newLen < minLen) newLen <<= 1; // grow exponentially
int32_t* newData = (int32_t*) uprv_malloc(newLen * BYTES_PER_WORD);
uprv_memcpy(newData, data, len * BYTES_PER_WORD);
uprv_free(data);
data = newData;
int32_t* p = data + len;
int32_t* limit = data + newLen;
while (p < limit) *p++ = 0;
len = newLen;
}
//eof

View File

@ -0,0 +1,40 @@
/*
**********************************************************************
* Copyright (c) 2002, International Business Machines
* Corporation and others. All Rights Reserved.
**********************************************************************
* $Source: /xsrl/Nsvn/icu/icu/source/test/usetperf/Attic/bitset.h,v $
**********************************************************************
* 2002-09-20 aliu Created.
*/
#ifndef __BITSET_H__
#define __BITSET_H__
#include "unicode/utypes.h"
/**
* A simple, limited clone of the java.util.BitSet.
*/
class BitSet {
uint32_t len;
int32_t* data;
void ensureCapacity(uint32_t minLen);
public:
BitSet();
~BitSet();
UBool get(int32_t bitIndex) const;
void set(int32_t bitIndex);
// Non-java
void clearAll();
// TODO add other methods as needed.
};
#endif

View File

@ -0,0 +1,64 @@
/*
**********************************************************************
* Copyright (c) 2002, International Business Machines
* Corporation and others. All Rights Reserved.
**********************************************************************
* $Source: /xsrl/Nsvn/icu/icu/source/test/usetperf/Attic/timer.h,v $
**********************************************************************
* 2002-09-20 aliu Created.
*/
#ifndef __PERFTIMER_H__
#define __PERFTIMER_H__
#include "unicode/utypes.h"
// Derived from Ram's perftime.h
//----------------------------------------------------------------------
// Win32
#if defined(WIN32) || defined(_WIN32) || defined(WIN64) || defined(_WIN64)
#include <windows.h>
class Timer {
LARGE_INTEGER tstart, tend;
public:
Timer() {}
inline void start() {
QueryPerformanceCounter(&tstart);
}
inline double stop() {
QueryPerformanceCounter(&tend);
LARGE_INTEGER freq;
int result = QueryPerformanceFrequency(&freq);
return ((double)(tend.QuadPart - tstart.QuadPart))/((double)freq.QuadPart);
}
};
//----------------------------------------------------------------------
// UNIX
#else
#include <sys/time.h>
class Timer {
struct timeval tstart, tend;
struct timezone tz;
public:
Timer() {}
inline void start() {
gettimeofday(&tstart, &tz);
}
inline double stop() {
gettimeofday(&tend, &tz);
double t1, t2;
t1 = (double)tstart.tv_sec + (double)tstart.tv_usec*1e-6;
t2 = (double)tend.tv_sec + (double)tend.tv_usec*1e-6;
return t2-t1;
}
};
#endif
#endif

View File

@ -0,0 +1,124 @@
/*
**********************************************************************
* Copyright (c) 2002, International Business Machines
* Corporation and others. All Rights Reserved.
**********************************************************************
* $Source: /xsrl/Nsvn/icu/icu/source/test/usetperf/Attic/usetperf.cpp,v $
**********************************************************************
* 2002-09-20 aliu Created.
*/
#include <stdio.h>
#include "unicode/utypes.h"
#include "unicode/uniset.h"
#include "unicode/uchar.h"
#include "unicode/usetiter.h"
#include "bitset.h"
#include "timer.h"
#define LENGTH(a) (sizeof(a)/sizeof(a[0]))
int main(int argc, const char *argv[]) {
Timer timer;
BitSet bs;
UnicodeSet us;
int32_t i, j, n, temp;
UChar32 cp;
double t;
int32_t PROPS[] = {
// category iterations for add, contains, iterator
U_TITLECASE_LETTER, 100, 100, 20000000,
U_UNASSIGNED, 30, 100, 20000000,
};
for (j=0; j<LENGTH(PROPS); j+=4) {
UCharCategory prop = (UCharCategory) PROPS[j];
printf("\nGetting characters for character category %d\n", prop);
bs.clearAll();
int32_t total = 0;
for (cp=0; cp<0x110000; ++cp) {
if (u_charType(cp) == prop) {
bs.set((int32_t) cp);
++total;
}
}
printf("Total characters: %d\n", total);
// add()
n = PROPS[j+1];
printf("Testing add() x %d...", n);
timer.start();
for (i=0; i<n; ++i) {
us.clear();
for (cp=0; cp<0x110000; ++cp) {
if (bs.get((int32_t) cp)) {
us.add(cp);
}
}
}
t = timer.stop();
printf("result: %f sec => %f ms/loop\n", t, t*1e3/n);
// contains()
n = PROPS[j+2];
printf("Testing contains() x %d...", n);
temp = 0;
timer.start();
for (i=0; i<n; ++i) {
us.clear();
for (cp=0; cp<0x110000; ++cp) {
if (us.contains(cp)) {
temp += cp;
}
}
}
t = timer.stop();
printf("result: %f sec => %f ms/loop\n", t, t*1e3/n);
// iterator
n = PROPS[j+3];
printf("Testing iterator x %d...", n);
temp = 0;
timer.start();
for (i=0; i<n; ++i) {
UnicodeSetIterator uit(us);
while (uit.next()) {
temp += uit.getCodepoint();
}
}
t = timer.stop();
printf("result: %f sec => %f ns/loop\n", t, t*1e9/n);
}
char* PAT[] = {
"['A-Za-z\\u00C0-\\u00C5\\u00C7-\\u00CF\\u00D1-\\u00D6\\u00D9-\\u00DD\\u00E0-\\u00E5\\u00E7-\\u00EF\\u00F1-\\u00F6\\u00F9-\\u00FD\\u00FF-\\u010F\\u0112-\\u0125\\u0128-\\u0130\\u0134-\\u0137\\u0139-\\u013E\\u0143-\\u0148\\u014C-\\u0151\\u0154-\\u0165\\u0168-\\u017E\\u01A0-\\u01A1\\u01AF-\\u01B0\\u01CD-\\u01DC\\u01DE-\\u01E1\\u01E6-\\u01ED\\u01F0\\u01F4-\\u01F5\\u01F8-\\u01FB\\u0200-\\u021B\\u021E-\\u021F\\u0226-\\u0233\\u1E00-\\u1E99\\u1EA0-\\u1EF9\\u212A-\\u212B]",
"['.0-9A-Za-z~\\u00C0-\\u00C5\\u00C7-\\u00CF\\u00D1-\\u00D6\\u00D9-\\u00DD\\u00E0-\\u00E5\\u00E7-\\u00EF\\u00F1-\\u00F6\\u00F9-\\u00FD\\u00FF-\\u010F\\u0112-\\u0125\\u0128-\\u0130\\u0134-\\u0137\\u0139-\\u013E\\u0143-\\u0148\\u014C-\\u0151\\u0154-\\u0165\\u0168-\\u017E\\u01A0-\\u01A1\\u01AF-\\u01B0\\u01CD-\\u01DC\\u01DE-\\u01E3\\u01E6-\\u01ED\\u01F0\\u01F4-\\u01F5\\u01F8-\\u021B\\u021E-\\u021F\\u0226-\\u0233\\u0301\\u0303-\\u0304\\u0306-\\u0307\\u0310\\u0314-\\u0315\\u0323\\u0325\\u0331\\u0341\\u0344\\u0385-\\u0386\\u0388-\\u038A\\u038C\\u038E-\\u0390\\u03AC-\\u03B0\\u03CC-\\u03CE\\u03D3\\u0403\\u040C\\u040E\\u0419\\u0439\\u0453\\u045C\\u045E\\u04C1-\\u04C2\\u04D0-\\u04D1\\u04D6-\\u04D7\\u04E2-\\u04E3\\u04EE-\\u04EF\\u1E00-\\u1E99\\u1EA0-\\u1EF9\\u1F01\\u1F03-\\u1F05\\u1F07\\u1F09\\u1F0B-\\u1F0D\\u1F0F\\u1F11\\u1F13-\\u1F15\\u1F19\\u1F1B-\\u1F1D\\u1F21\\u1F23-\\u1F25\\u1F27\\u1F29\\u1F2B-\\u1F2D\\u1F2F\\u1F31\\u1F33-\\u1F35\\u1F37\\u1F39\\u1F3B-\\u1F3D\\u1F3F\\u1F41\\u1F43-\\u1F45\\u1F49\\u1F4B-\\u1F4D\\u1F51\\u1F53-\\u1F55\\u1F57\\u1F59\\u1F5B\\u1F5D\\u1F5F\\u1F61\\u1F63-\\u1F65\\u1F67\\u1F69\\u1F6B-\\u1F6D\\u1F6F\\u1F71\\u1F73\\u1F75\\u1F77\\u1F79\\u1F7B\\u1F7D\\u1F81\\u1F83-\\u1F85\\u1F87\\u1F89\\u1F8B-\\u1F8D\\u1F8F\\u1F91\\u1F93-\\u1F95\\u1F97\\u1F99\\u1F9B-\\u1F9D\\u1F9F\\u1FA1\\u1FA3-\\u1FA5\\u1FA7\\u1FA9\\u1FAB-\\u1FAD\\u1FAF-\\u1FB1\\u1FB4\\u1FB8-\\u1FB9\\u1FBB\\u1FC4\\u1FC9\\u1FCB\\u1FCE\\u1FD0-\\u1FD1\\u1FD3\\u1FD8-\\u1FD9\\u1FDB\\u1FDE\\u1FE0-\\u1FE1\\u1FE3\\u1FE5\\u1FE8-\\u1FE9\\u1FEB-\\u1FEC\\u1FEE\\u1FF4\\u1FF9\\u1FFB\\u212A-\\u212B\\uE04D\\uE064]",
"[\\u0901-\\u0903\\u0905-\\u0939\\u093C-\\u094D\\u0950-\\u0954\\u0958-\\u096F]",
};
UErrorCode ec = U_ZERO_ERROR;
n = 2000;
for (j=0; j<LENGTH(PAT); ++j) {
printf("\nApplying pattern %s x %d...", PAT[j], n);
UnicodeSet set;
UnicodeString pat(PAT[j], "");
timer.start();
for (i=0; i<n; i++) {
set.applyPattern(pat, ec);
}
t = timer.stop();
printf("result: %f sec => %f us/loop\n", t, t*1e6/n);
}
return 0;
}

View File

@ -0,0 +1,164 @@
# Microsoft Developer Studio Project File - Name="usetperf" - Package Owner=<4>
# Microsoft Developer Studio Generated Build File, Format Version 6.00
# ** DO NOT EDIT **
# TARGTYPE "Win32 (x86) Console Application" 0x0103
CFG=usetperf - Win32 Debug
!MESSAGE This is not a valid makefile. To build this project using NMAKE,
!MESSAGE use the Export Makefile command and run
!MESSAGE
!MESSAGE NMAKE /f "usetperf.mak".
!MESSAGE
!MESSAGE You can specify a configuration when running NMAKE
!MESSAGE by defining the macro CFG on the command line. For example:
!MESSAGE
!MESSAGE NMAKE /f "usetperf.mak" CFG="usetperf - Win32 Debug"
!MESSAGE
!MESSAGE Possible choices for configuration are:
!MESSAGE
!MESSAGE "usetperf - Win32 Release" (based on "Win32 (x86) Console Application")
!MESSAGE "usetperf - Win32 Debug" (based on "Win32 (x86) Console Application")
!MESSAGE "usetperf - Win64 Release" (based on "Win32 (x86) Console Application")
!MESSAGE "usetperf - Win64 Debug" (based on "Win32 (x86) Console Application")
!MESSAGE
# Begin Project
# PROP AllowPerConfigDependencies 0
# PROP Scc_ProjName ""
# PROP Scc_LocalPath ""
CPP=cl.exe
RSC=rc.exe
!IF "$(CFG)" == "usetperf - Win32 Release"
# PROP BASE Use_MFC 0
# PROP BASE Use_Debug_Libraries 0
# PROP BASE Output_Dir "Release"
# PROP BASE Intermediate_Dir "Release"
# PROP BASE Target_Dir ""
# PROP Use_MFC 0
# PROP Use_Debug_Libraries 0
# PROP Output_Dir "Release"
# PROP Intermediate_Dir "Release"
# PROP Ignore_Export_Lib 0
# PROP Target_Dir ""
MTL=midl.exe
# ADD BASE CPP /nologo /W3 /GX /O2 /D "WIN32" /D "NDEBUG" /D "_CONSOLE" /D "_MBCS" /FD /c
# ADD CPP /nologo /G6 /MD /W3 /GX /O2 /I "..\..\..\include" /D "WIN32" /D "NDEBUG" /D "_CONSOLE" /D "_MBCS" /FD /c
# ADD BASE RSC /l 0x409 /d "NDEBUG"
# ADD RSC /l 0x409 /d "NDEBUG"
BSC32=bscmake.exe
# ADD BASE BSC32 /nologo
# ADD BSC32 /nologo
LINK32=link.exe
# ADD BASE LINK32 kernel32.lib user32.lib gdi32.lib winspool.lib comdlg32.lib advapi32.lib shell32.lib ole32.lib oleaut32.lib uuid.lib odbc32.lib odbccp32.lib kernel32.lib user32.lib gdi32.lib winspool.lib comdlg32.lib advapi32.lib shell32.lib ole32.lib oleaut32.lib uuid.lib odbc32.lib odbccp32.lib /nologo /subsystem:console /machine:I386
# ADD LINK32 icuuc.lib kernel32.lib user32.lib gdi32.lib winmm.lib winspool.lib comdlg32.lib advapi32.lib shell32.lib ole32.lib oleaut32.lib uuid.lib odbc32.lib odbccp32.lib /nologo /subsystem:console /machine:I386 /libpath:"..\..\..\lib"
!ELSEIF "$(CFG)" == "usetperf - Win32 Debug"
# PROP BASE Use_MFC 0
# PROP BASE Use_Debug_Libraries 1
# PROP BASE Output_Dir "Debug"
# PROP BASE Intermediate_Dir "Debug"
# PROP BASE Target_Dir ""
# PROP Use_MFC 0
# PROP Use_Debug_Libraries 1
# PROP Output_Dir "Debug"
# PROP Intermediate_Dir "Debug"
# PROP Ignore_Export_Lib 0
# PROP Target_Dir ""
MTL=midl.exe
# ADD BASE CPP /nologo /W3 /Gm /GX /ZI /Od /D "WIN32" /D "_DEBUG" /D "_CONSOLE" /D "_MBCS" /FD /GZ /c
# ADD CPP /nologo /G6 /MDd /W3 /Gm /GX /ZI /Od /I "..\..\..\include" /D "WIN32" /D "_DEBUG" /D "_CONSOLE" /D "_MBCS" /FD /GZ /c
# ADD BASE RSC /l 0x409 /d "_DEBUG"
# ADD RSC /l 0x409 /d "_DEBUG"
BSC32=bscmake.exe
# ADD BASE BSC32 /nologo
# ADD BSC32 /nologo
LINK32=link.exe
# ADD BASE LINK32 kernel32.lib user32.lib gdi32.lib winspool.lib comdlg32.lib advapi32.lib shell32.lib ole32.lib oleaut32.lib uuid.lib odbc32.lib odbccp32.lib kernel32.lib user32.lib gdi32.lib winspool.lib comdlg32.lib advapi32.lib shell32.lib ole32.lib oleaut32.lib uuid.lib odbc32.lib odbccp32.lib /nologo /subsystem:console /debug /machine:I386 /pdbtype:sept
# ADD LINK32 icuucd.lib winmm.lib kernel32.lib user32.lib gdi32.lib winspool.lib comdlg32.lib advapi32.lib shell32.lib ole32.lib oleaut32.lib uuid.lib odbc32.lib odbccp32.lib /nologo /subsystem:console /debug /machine:I386 /pdbtype:sept /libpath:"..\..\..\lib"
!ELSEIF "$(CFG)" == "usetperf - Win64 Release"
# PROP BASE Use_MFC 0
# PROP BASE Use_Debug_Libraries 0
# PROP BASE Output_Dir "Release"
# PROP BASE Intermediate_Dir "Release"
# PROP BASE Target_Dir ""
# PROP Use_MFC 0
# PROP Use_Debug_Libraries 0
# PROP Output_Dir "Release"
# PROP Intermediate_Dir "Release"
# PROP Ignore_Export_Lib 0
# PROP Target_Dir ""
MTL=midl.exe
# ADD BASE CPP /nologo /W3 /GX /O2 /D "WIN64" /D "NDEBUG" /D "_CONSOLE" /D "_MBCS" /FD /c
# ADD CPP /nologo /MD /W3 /GX /Zi /O2 /Op /I "..\..\..\include" /D "WIN64" /D "NDEBUG" /D "_CONSOLE" /D "_MBCS" /D "_IA64_" /D "WIN32" /D "_AFX_NO_DAO_SUPPORT" /FD /QIA64_fmaopt /Wp64 /Zm600 /c
# ADD BASE RSC /l 0x409 /d "NDEBUG"
# ADD RSC /l 0x409 /d "NDEBUG"
BSC32=bscmake.exe
# ADD BASE BSC32 /nologo
# ADD BSC32 /nologo
LINK32=link.exe
# ADD BASE LINK32 kernel32.lib user32.lib gdi32.lib winspool.lib comdlg32.lib advapi32.lib shell32.lib ole32.lib oleaut32.lib uuid.lib odbc32.lib odbccp32.lib kernel32.lib user32.lib gdi32.lib winspool.lib comdlg32.lib advapi32.lib shell32.lib ole32.lib oleaut32.lib uuid.lib odbc32.lib odbccp32.lib /nologo /subsystem:console /machine:IX86 /machine:IA64
# ADD LINK32 icuuc.lib kernel32.lib user32.lib gdi32.lib winmm.lib winspool.lib comdlg32.lib advapi32.lib shell32.lib ole32.lib oleaut32.lib uuid.lib odbc32.lib odbccp32.lib /nologo /subsystem:console /machine:IX86 /libpath:"..\..\..\lib" /machine:IA64
!ELSEIF "$(CFG)" == "usetperf - Win64 Debug"
# PROP BASE Use_MFC 0
# PROP BASE Use_Debug_Libraries 1
# PROP BASE Output_Dir "Debug"
# PROP BASE Intermediate_Dir "Debug"
# PROP BASE Target_Dir ""
# PROP Use_MFC 0
# PROP Use_Debug_Libraries 1
# PROP Output_Dir "Debug"
# PROP Intermediate_Dir "Debug"
# PROP Ignore_Export_Lib 0
# PROP Target_Dir ""
MTL=midl.exe
# ADD BASE CPP /nologo /W3 /Gm /GX /ZI /Od /D "WIN64" /D "_DEBUG" /D "_CONSOLE" /D "_MBCS" /FD /GZ /c
# ADD CPP /nologo /MDd /W3 /Gm /GX /Zi /Od /Op /I "..\..\..\include" /D "WIN64" /D "_DEBUG" /D "_CONSOLE" /D "_MBCS" /D "_IA64_" /D "WIN32" /D "_AFX_NO_DAO_SUPPORT" /FD /GZ /QIA64_fmaopt /Wp64 /Zm600 /c
# ADD BASE RSC /l 0x409 /d "_DEBUG"
# ADD RSC /l 0x409 /d "_DEBUG"
BSC32=bscmake.exe
# ADD BASE BSC32 /nologo
# ADD BSC32 /nologo
LINK32=link.exe
# ADD BASE LINK32 kernel32.lib user32.lib gdi32.lib winspool.lib comdlg32.lib advapi32.lib shell32.lib ole32.lib oleaut32.lib uuid.lib odbc32.lib odbccp32.lib kernel32.lib user32.lib gdi32.lib winspool.lib comdlg32.lib advapi32.lib shell32.lib ole32.lib oleaut32.lib uuid.lib odbc32.lib odbccp32.lib /nologo /subsystem:console /debug /machine:IX86 /pdbtype:sept /machine:IA64
# ADD LINK32 icuucd.lib winmm.lib kernel32.lib user32.lib gdi32.lib winspool.lib comdlg32.lib advapi32.lib shell32.lib ole32.lib oleaut32.lib uuid.lib odbc32.lib odbccp32.lib /nologo /subsystem:console /incremental:no /debug /machine:IX86 /pdbtype:sept /libpath:"..\..\..\lib" /machine:IA64
!ENDIF
# Begin Target
# Name "usetperf - Win32 Release"
# Name "usetperf - Win32 Debug"
# Name "usetperf - Win64 Release"
# Name "usetperf - Win64 Debug"
# Begin Group "Source Files"
# PROP Default_Filter "cpp;c;cxx;rc;def;r;odl;idl;hpj;bat"
# Begin Source File
SOURCE=.\bitset.cpp
# ADD CPP /I "..\..\common"
# End Source File
# Begin Source File
SOURCE=.\usetperf.cpp
# ADD CPP /I "..\..\common"
# End Source File
# End Group
# Begin Group "Header Files"
# PROP Default_Filter "h;hpp;hxx;hm;inl"
# End Group
# Begin Group "Resource Files"
# PROP Default_Filter "ico;cur;bmp;dlg;rc2;rct;bin;rgs;gif;jpg;jpeg;jpe"
# End Group
# End Target
# End Project