java.lang.ClassCastException: java.lang.Integer cannot be cast to java.lang.String

[moooonk]

Hello @jbachorik
I use btrace-1.3.11 do some check in my project, and found this Exception.
I found my AOP funtion can not attach on some kind of Class with @location(value=Kind.SYNC_ENTRY).
I don't sure my problem has something to do with this Exception,but I think this Exception is bug.

btrace DEBUG: java.lang.ClassCastException: java.lang.Integer cannot be cast to java.lang.String
java.lang.ClassCastException: java.lang.Integer cannot be cast to java.lang.String
at com.sun.btrace.runtime.InstrumentingMethodVisitor.fromSlotType(InstrumentingMethodVisitor.java:1464)
at com.sun.btrace.runtime.InstrumentingMethodVisitor.visitVarInsn(InstrumentingMethodVisitor.java:683)
at com.sun.btrace.org.objectweb.asm.MethodVisitor.visitVarInsn(Unknown Source)
at com.sun.btrace.util.templates.TemplateExpanderVisitor.visitVarInsn(TemplateExpanderVisitor.java:79)
at com.sun.btrace.org.objectweb.asm.MethodVisitor.visitVarInsn(Unknown Source)
at com.sun.btrace.org.objectweb.asm.MethodVisitor.visitVarInsn(Unknown Source)
at com.sun.btrace.org.objectweb.asm.MethodVisitor.visitVarInsn(Unknown Source)
at com.sun.btrace.org.objectweb.asm.ClassReader.a(Unknown Source)
at com.sun.btrace.org.objectweb.asm.ClassReader.b(Unknown Source)
at com.sun.btrace.org.objectweb.asm.ClassReader.accept(Unknown Source)
at com.sun.btrace.runtime.BTraceClassReader.accept(BTraceClassReader.java:174)
at com.sun.btrace.org.objectweb.asm.ClassReader.accept(Unknown Source)
at com.sun.btrace.runtime.BTraceClassReader.accept(BTraceClassReader.java:183)
at com.sun.btrace.runtime.InstrumentUtils.accept(InstrumentUtils.java:177)
at com.sun.btrace.runtime.InstrumentUtils.accept(InstrumentUtils.java:171)
at com.sun.btrace.runtime.BTraceClassWriter.instrument(BTraceClassWriter.java:82)
at com.sun.btrace.runtime.BTraceTransformer.transform(BTraceTransformer.java:184)
at sun.instrument.TransformerManager.transform(TransformerManager.java:188)
at sun.instrument.InstrumentationImpl.transform(InstrumentationImpl.java:428)
at sun.instrument.InstrumentationImpl.retransformClasses0(Native Method)
at sun.instrument.InstrumentationImpl.retransformClasses(InstrumentationImpl.java:144)
at com.sun.btrace.agent.Client.retransformLoaded(Client.java:478)
at com.sun.btrace.agent.Main$3.run(Main.java:692)
at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
at java.util.concurrent.FutureTask.run(FutureTask.java:266)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
at java.lang.Thread.run(Thread.java:745)

[jbachorik]

Could you share the method source causing this problem?
Thanks!

[moooonk]

Source here.
And with my test,this source can run in btrace-1.3.8.3 no Exception.

package com.sun.btrace.samples;

import com.sun.btrace.annotations.;
import static com.sun.btrace.BTraceUtils.;

@btrace(unsafe=true) public class AllMethods {
@OnMethod(
clazz="com.bt.pjg.game.backpack.BackpackService",
method="/.*/",
location=@location(value=Kind.SYNC_ENTRY)
)
public static void m() {
print("entered SYNC");
}

@OnMethod(
    clazz="com.bt.pjg.game.backpack.BackpackService",
    method="/.*/",
	location=@Location(value=Kind.ENTRY)
)
public static void m2() {
    print("entered run");
}

}

[jbachorik]

I guess BackpackService is a closed source. I will try to figure out the problem from the available data. This is all due to stackframe maps in bytcode - it is very complex area without extensive documentation unfortunately :( Thanks for patience!

[moooonk]

BackpackService is My company project source.I'm sorry.It's not convenient for me to share the detail source.
But I can share the source structure.I hope it will help you。

public class BackpackService {
private static final BackpackService instance = new BackpackService();

private BackpackManger manager;

private BackpackService() {
	
}

public static BackpackService instance() {
	return instance;
}

public void init() {
	
}

public CacheMap<ItemBase> getItemMap(long userId) {
	
}

public List<ItemBase> getItemList(long userId) {
	
}

public ItemBase getItemById(long userId, int itemId) {
	
}

public int getItemCount(long userId, int itemId) {
	
}

public ItemBase addItem(User user, int itemId, int count, GetApproach getApproach) {

}

public List<ItemBase> addItem(User user, Collection<M2QEntry> entryList, GetApproach getApproach) {

}

private void throwInvalidParamsException(M2QEntry entry) {
	
}

public boolean decrItem(User user, int itemId, int count, DecReason decReason) {
	
}

public boolean decrItem(User user, Collection<M2QEntry> entryList, DecReason decReason) {
	
	
}

public void replaceItem(User user, ItemBase itemBase) {
	
}

private void replaceItem(User user, Collection<ItemBase> itemBaseList) {
	
}

private void pushChange(User user, Collection<ItemBase> itemBaseList) {
	
}

public List<UserItem> changeToUserItem(Collection<ItemBase> itemBaseList) {
	synchronized (this) {
		synchronized (itemBaseList) {
			
			}			
			return userItemList;				
		}
	}
}

public short useTool(User user, List<M2QEntry> m2qList, long targetId) {
	
}

public boolean checkEnough(User user, int itemId, int count) {
	
}

public boolean checkEnough(User user, Collection<M2QEntry> entryList) {
	
}

}

[jbachorik]

Can you, please, try with #341 fix applied? I discovered another corner case in stackframe map generation and fixed that - might fix also your issue, with some luck.

[jbachorik]

Ping. Any chance to test this fix to see whether the problem goes away?

[moooonk]

Unfortunately,I donwload new btrace-bin-1.3.11 release and test.But this bug still here.

[jbachorik]

Could you try a custom built BTrace containing commit #527d86f265a0b534791309b42303879aa44da7ae ?
I added more debug info there to figure out what is throwing the stack tracing logic off.

[moooonk]

log here:

---

btrace WARNING: Unknown slot type: -2

---

---

btrace WARNING: Unknown slot type: -2

---

---

---

[jbachorik]

Any chance of sharing the compiled version of that class? I can sign whatever non-disclosure agreement if necessary and you can send it directly to my email.
Without that, I am afraid, I won't be able to solve this problem - there seems to be some peculiarity in the stackframe maps in that class and it is throwing off the algorithm used to recompute them in instrumented methods.

[jbachorik]

Hi, I addressed this issue in develop branch. Can you, please, verify?