Android應用的效能如何測試?JAVA層面可以用TraceView,可是用NDK開發出來的是so,TraceView跟蹤不了怎麼辦?問了Google大神,答案是OProfile!Oprofile 是Linux系統下一個低開銷的系統全域的效能監控工具,利用處理器上所包含的專用的效能監控硬體(若沒有效能監控硬體則使用一個基於計時器的代用品)來收集與效能相關的資料樣品。它獲得關於核心以及系統上的可執行檔的資訊,例如記憶體是何時被引用的;L2緩衝請求的數量;收到的硬體中斷數量等。Oprofile的特點如下:l 無需重新編譯原始碼,如果不進行原始碼及分析,連調試資訊(-g option to gcc)也不是必須的。l 只在核心中插入一個模組。l 可以分析運行於系統之上的所有代碼(禁用中斷的代碼除外)l 系統的額外開銷小,Oprofile會增加1%-8%的系統開銷(取決於採樣頻率)l 相容所有2.2,2.4,2.6核心,可以運行在SMP系統之上l 支援主流CPU架構,包括X86、arm、AVR32、mips、powerpc等Oprofile要想跑在Andorid上,要滿足下面的條件:1.核心要支援2.要將Oprofile移植到Arm平台上下面是移植的全過程:一、Oprofile移植用到的交叉編譯工具如下:arm-2010.09-50-arm-none-linux-gnueabi-i686-pc-linux-gnu.tar.bz2用到的庫如下:popt-1.14.tar.gz binutils-2.21.tar.gzoprofile-0.9.6.tar.gz$ tar xvfz arm-2010.09-50-arm-none-linux-gnueabi-i686-pc-linux-gnu.tar.bz2 -C ~/修改~/.bashrc,添加export PATH=${PATH}:/home/louieli/arm-2010.09/bin$ tar zxvf popt-1.14.tar.gz $ cd popt-1.14$ ac_cv_va_copy=yes ./configure --with-kernel-support --host=arm-none-linux-gnueabi --prefix=/home/louieli/work/popt$ make$ make install$ tar zxvf binutils-2.21.tar.gz$ cd binutils-2.21/$ ./configure --with-kernel-support --host=arm-none-linux-gnueabi --prefix=/home/louieli/work/binutils --enable-shared$ make LDFLAGS="-all-static"可能會出現 cc1: warnings being treated as errors,找到出錯檔案的Makefile檔案,將-Werror去掉$ make install$ tar zxvf oprofile-0.9.6.tar.gz$ cd oprofile-0.9.6/$ ./configure --with-kernel-support --host=arm-none-linux-gnueabi --prefix=/home/louieli/work/oprofile/ --with-extra-libs=/home/louieli/work/popt/lib/ --with-extra-includes=/home/louieli/work/popt/include/ --with-binutils=/home/louieli/work/binutils$ make LDFLAGS="-all-static -L/home/louieli/work/binutils/lib -Xlinker -R -Xlinker /home/louieli/work/binutils/lib -L/home/louieli/work/popt/lib/"$ make install用file 命令查看,我們需要的oprofile檔案都已經變成可以在android上跑的靜態連結檔案了install.sh: Bourne-Again shell script text executableopannotate: ELF 32-bit LSB executable, ARM, version 1 (SYSV), statically linked, for GNU/Linux 2.6.16, not strippedoparchive: ELF 32-bit LSB executable, ARM, version 1 (SYSV), statically linked, for GNU/Linux 2.6.16, not strippedopcontrol: a /system/bin/sh script text executableopgprof: ELF 32-bit LSB executable, ARM, version 1 (SYSV), statically linked, for GNU/Linux 2.6.16, not strippedophelp: ELF 32-bit LSB executable, ARM, version 1 (SYSV), statically linked, for GNU/Linux 2.6.16, not strippedopimport: ELF 32-bit LSB executable, ARM, version 1 (SYSV), statically linked, for GNU/Linux 2.6.16, not strippedopjitconv: ELF 32-bit LSB executable, ARM, version 1 (SYSV), statically linked, for GNU/Linux 2.6.16, not strippedopreport: ELF 32-bit LSB executable, ARM, version 1 (SYSV), statically linked, for GNU/Linux 2.6.16, not strippedoprofiled: ELF 32-bit LSB executable, ARM, version 1 (SYSV), statically linked, for GNU/Linux 2.6.16, not stripped二、編譯linux核心映像a)準備交叉編譯工具鏈android代碼樹中有一個prebuilt項目,包含了我們編譯核心所需的交叉編譯工具。b)設定環境變數$ emacs ~/.bashrc增加如下兩行:export PATH=$PATH:~/android/prebuilt/linux-x86/toolchain/arm-eabi-4.4.0/binexport ARCH=arm儲存後,同步變化:$ source ~/.bashrcc)獲得合適的核心原始碼$ cd ~/android獲得核心原始碼倉庫$ git clone git://android.git.kernel.org/kernel/common.git kernel$ cd kernel$ git branch顯示* android-2.6.27說明你現在在android-2.6.27這個分支上,也是kernel/common.git的預設主分支。顯示所有head分支:$ git branch -a顯示* android-2.6.27remotes/origin/HEAD -> origin/android-2.6.27remotes/origin/android-2.6.25remotes/origin/android-2.6.27remotes/origin/android-2.6.29remotes/origin/android-goldfish-2.6.27remotes/origin/android-goldfish-2.6.29我們選取最新的android-goldfish-2.6.29,其中goldfish是android的模擬器類比的CPU。$ git checkout -b android-goldfish-2.6.29 origin/android-goldfish-2.6.29$ git branch顯示android-2.6.27* android-goldfish-2.6.29我們已經工作在android-goldfish-2.6.29分支上了。d)設定交叉編譯參數開啟kernel目錄下的Makefile檔案,把CROSS_COMPILE指向剛才下載的prebuilt中的arm-eabi編譯器CROSS_COMPILE ?= arm-eabi-把LDFLAGS_BUILD_ID = $(patsubst -Wl$(comma)%,%,/$(call ld-option, -Wl$(comma)–build-id,))這一行注釋掉,並且添加一個空的LDFLAGS_BUILD_ID定義,如下:LDFLAGS_BUILD_ID =e)編譯核心映像$ cd ~/android/kernel$ make goldfish_defconfig$ make menuconfig修改核心配置如下General setup --->[*] Profiling support (EXPERIMENTAL)[ ] Activate markers[*] OProfile system profiling (EXPERIMENTAL) 這是把OProfile直接編進核心,也可以選擇[M] OProfile system profiling (EXPERIMENTAL)會在arch/arm/oprofile檔案夾下產生oprofile.ko,oprofile.ko需要用insmod載入。$make f)測試產生的核心映像$ emulator -avd myavd -kernel ~/android/kernel/arch/arm/boot/zImage三、Oprofile在android模擬器中的使用1.先看一下opcontrol的參數# opcontrolopcontrol: usage: -l/--list-events list event types and unit masks -?/--help this message -v/--version show version --init loads the oprofile module and oprofilefs --setup give setup arguments (may be omitted) --status show configuration --start-daemon start daemon without starting profiling -s/--start start data collection -d/--dump flush the collected profiling data -t/--stop stop data collection -h/--shutdown stop data collection and kill daemon -V/--verbose[=all,sfile,arcs,samples,module,misc,ext] be verbose in the daemon log --reset clears out data from current session --save=name save data from current session to session_name --deinit unload the oprofile module and oprofilefs -e/--event=eventspec Choose an event. May be specified multiple times. Of the form "default" or "name:count:unitmask:kernel:user", where : name: event name, e.g. CPU_CLK_UNHALTED or RTC_INTERRUPTS count: reset counter value e.g. 100000 unitmask: hardware unit mask e.g. 0x0f kernel: whether to profile kernel: 0 or 1 user: whether to profile userspace: 0 or 1 -p/--separate=type,[types] Separate profiles as follows : none: no profile separation library: separate shared library profiles per-application kernel: same as library, plus kernel profiles thread: per-thread/process profiles cpu: per CPU profiles all: all of the above -c/--callgraph=#depth enable callgraph sample collection with a maximum depth. Use 0 to disable callgraph profiling. --session-dir=dir place sample database in dir instead of default location (/var/lib/oprofile) -i/--image=name[,names] list of binaries to profile (default is "all") --vmlinux=file vmlinux kernel image --no-vmlinux no kernel image (vmlinux) available --kernel-range=start,end kernel range vma address in hexadecimal --buffer-size=num kernel buffer size in sample units --buffer-watershed kernel buffer watershed in sample units (2.6 only= --cpu-buffer-size=num per-cpu buffer size in units (2.6 only) --note-table-size kernel notes buffer size in notes units (2.4 only) --xen Xen image (for Xen only) --active-domains=<list> List of domains in profiling session (for Xen only) (list contains domain ids separated by commas)2.使用方法將我們之前編譯好的oprofile和busybox裝入模擬器執行oprofile目錄中的install.sh 將oprofile裝入模擬器adb push busybox /data/busybox$adb shell //進入模擬器shell#chmod 777 /data/busybox# /data/busybox --install /data/busybox#export PATH=/data/busybox:$PATH:/data/oprofile# mount -o remount rw /# mount -o rw,remount -t yaffs2 /dev/mtdblock3 /system# touch /etc/mtab# echo nodev /dev/oprofile oprofilefs rw 0 0>/etc/mtab# mkdir /dev/oprofile# mount -t oprofilefs nodev /dev/oprofile //這一句很重要,沒有這一句會出現下面的錯誤# opcontrol --init cat: can't open '/dev/oprofile/cpu_type': No such file or directoryUnable to open cpu_type file for readingMake sure you have done opcontrol --initcpu_type 'unset' is not validyou should upgrade oprofile or force the use of timer mode# opcontrol --init //初始化,只需運行一次# opcontrol --setup --callgraph=2 --session-dir=/data/first --no-vmlinuxUsing 2.6+ OProfile kernel interface.Using log file /data/first/samples/oprofiled.logDaemon started.Profiler running.# opcontrol --statusDaemon running: pid 637Separate options: nonevmlinux file: noneImage filter: noneCall-graph depth: 2# opcontrol --start //啟動profilerUsing 2.6+ OProfile kernel interface.Using log file /var/lib/oprofile/samples/oprofiled.logDaemon started.Profiler running.# /data/test/test //運行我們的程式 ( 我的測試程式通過這條指令編譯arm-none-linux-gnueabi-gcc -g -o test test.c -static -fno-omit-frame-pointer)in cin ain bin ain cin bin ain a# opcontrol --dump //收集採樣資料# opcontrol --stop //停止profilerStopping profiling.#opreport --session-dir=/data/first -l /data/test/test //查看報告CPU: CPU with timer interrupt, speed 0 MHz (estimated)Profiling through timer interruptsamples % symbol name11291 79.9589 a1129 7.9952 b853 6.0406 main848 6.0052 c現在我們就可以根據oprofile的輸出對我們的程式進行最佳化了。如果有哪位同學也想試一把的話,一定要用linux。這種移植環境很重要,我之前就在測試機(win7+cygwin)上浪費了很多時間。這裡有打包好的工具,大家可以下載。其中kernel-qemu就是我們之前編譯好的核心,替換掉Android SDK中的kernel-qemu就行了。祝各位好運!