AppleTrace

AppleTrace is developed for analyzing app's performance on iOS.

>> I have developed a replacement called Messier which is much easier to use. :)

• 中文说明，开发思路及方法
• 搭载MonkeyDev可trace第三方App

Feature

1. User-defined trace section.
2. Trace Objective C methods.

FAQ

Go to Wiki

Clone

git clone https://github.com/everettjf/AppleTrace.git

For stable release , please refer to Releases

Usage

1. Produce trace data.
2. Copy from app's sandbox directory.
3. Merge (all) trace data files into one file trace.json. (There may be more than 1 trace file.)
4. Generate trace.html based on trace.json.

See below for more detail.

1. Produce

Until now , there are 2 ways for generating trace data.

(1) Manual set section.

Call APTBeginSection at the beginning of method ,and APTEndSection at the end of method. For Objective C method (whether instance method or class method), there are APTBegin and APTEnd macro for easy coding.

void anyKindsOfMethod{
    APTBeginSection("process");
    // some code
    APTEndSection("process");
}

- (void)anyObjectiveCMethod{
    APTBegin;
    // some code
    APTEnd;
}

Sample app is sample/ManualSectionDemo.

(2) Dynamic library hooking all objc_msgSend.

Hooking all objc_msgSend methods (based on HookZz). This only support arm64 under debugger ( lldb).

Sample app is sample/TraceAllMsgDemo.

2. Copy

Using any kinds of method, copy <app's sandbox>/Library/appletracedata out of Simulator/RealDevice.

3. Merge

Merge/Preprocess the appletracedata.

python merge.py -d <appletracedata directory>

This will produce trace.json in appletracedata directory.

NOW !!!, you could drop trace.json into Chrome's chrome://tracing. Or if you want to generate a html result, continue to the 4th step.

4. Generate

Run sh get_catapult.sh to get catapult source.

Then generate trace.html using catapult.

python catapult/tracing/bin/trace2html appletracedata/trace.json --output=appletracedata/trace.html
open trace.html

trace.html only support Chrome

SampleData

Open sampledata/trace.html using Chrome.

Thanks

1. HookZz : https://github.com/jmpews/HookZz
2. catapult : https://github.com/catapult-project/catapult

Group

欢迎关注微信订阅号，更多有趣的性能优化点点滴滴。