为什么我的堆中有两个空的ThreadAbortException
本文关键字:两个 ThreadAbortException 我的 为什么 | 更新日期: 2023-09-27 18:23:49
我正在研究WinDBG的内存转储,我想知道为什么堆中有两个System.Threading.ThreadAbortExceptions
,它们都是空的。
对于其他三个发现的异常,我理解它们为什么存在,以及它们是默认创建的:
-
System.ExecutionEngineException
-
System.StackOverflowException
-
System.OutOfMemoryException
WinDBG输出
0:000>!dumpeap-类型System.Threading.ThreadException
Address MT Size
010210fc 79330ef8 72
01021144 79330ef8 72
total 2 objects
Statistics:
MT Count TotalSize Class Name
79330ef8 2 144 System.Threading.ThreadAbortException
Total 2 objects
0:000>!pe 010210fc
Exception object: 010210fc
Exception type: System.Threading.ThreadAbortException
Message: <none>
InnerException: <none>
StackTrace (generated):<none>
StackTraceString: <none>
HResult: 80131530
0:000>!pe 01021144
Exception object: 01021144
Exception type: System.Threading.ThreadAbortException
Message: <none>
InnerException: <none>
StackTrace (generated): <none>
StackTraceString: <none>
HResult: 80131530
所以我的问题是:
- 这两个也是默认创建的吗?如果是,为什么会有两个
- 如果没有,为什么它们是空的
内存转储来自Windows服务。
使用转储中的线程信息进行更新
0:000>!线程
ThreadCount: 14
UnstartedThread: 0
BackgroundThread: 8
PendingThread: 0
DeadThread: 4
Hosted Runtime: no
PreEmptive GC Alloc Lock
ID OSID ThreadOBJ State GC Context Domain Count APT Exception
0 1 11d4 0015c538 a020 Enabled 00000000:00000000 00163aa0 0 MTA
2 2 71c 0016f6a0 b220 Enabled 00000000:00000000 00163aa0 0 MTA (Finalizer)
3 4 1914 0019ac48 180b220 Enabled 6a205b0c:6a207910 00163aa0 0 MTA (Threadpool Worker)
5 6 1bd4 001b1580 200b020 Enabled 00000000:00000000 00163aa0 0 MTA
6 7 16a4 001bd260 200b220 Enabled 6a1dc7b8:6a1dd910 00163aa0 0 MTA
7 8 870 001c4a58 200b220 Enabled 6a1da740:6a1db910 00163aa0 0 MTA
8 9 2204 001cf798 b220 Enabled 00000000:00000000 00163aa0 0 MTA
9 d 4d8 0021cb98 80a220 Enabled 00000000:00000000 00163aa0 0 MTA (Threadpool Completion Port)
10 e 1b70 002227c0 200b220 Enabled 6a27d820:6a27d910 00163aa0 0 MTA
11 89 2224 68a3fbd0 880b220 Enabled 00000000:00000000 00163aa0 0 MTA (Threadpool Completion Port)
XXXX 11 0 2336e658 8801820 Enabled 00000000:00000000 00163aa0 0 Ukn (Threadpool Completion Port)
XXXX 46 0 16d17270 8801820 Enabled 00000000:00000000 00163aa0 0 Ukn (Threadpool Completion Port)
XXXX 3a 0 16ca7a70 8801820 Enabled 00000000:00000000 00163aa0 0 Ukn (Threadpool Completion Port)
XXXX 3b 0 10e64250 8801820 Enabled 00000000:00000000 00163aa0 0 Ukn (Threadpool Completion Port)
它们是由CLR预先分配的。我将让这些来自CLR SSCLI20版本的片段来讲述这个故事:
从clr/src/vm/clrex.cpp,CLRException::GetThrowable():
// If creating a normal ThreadAbortException fails, due to OOM or StackOverflow,
// use a pre-created one.
// We do not won't to change a ThreadAbortException into OOM or StackOverflow, because
// it will cause recursive call when escalation policy is on:
// Creating ThreadAbortException fails, we throw OOM. Escalation leads to ThreadAbort.
// The cycle repeats.
throwable = GetPreallocatedThreadAbortException();
相同文件,CLRException::GetPreallocatedRudeThreadAbortException()方法:
// When we are hosted, we pre-create this exception.
// This function should be called only if the exception has been created.
_ASSERTE(g_pPreallocatedRudeThreadAbortException);
return ObjectFromHandle(g_pPreallocatedRudeThreadAbortException);
"循环重复"可能需要一些解释:OutOfMemoryException->Thread.Art()->new ThreadAbortException()->OutOfMemory Exception->Thread.art()。等等。
出于完全相同的原因,同一源文件还具有GetPreallocatedOutOfMemoryException()、GetPrealloatedStackOverflowException()和GetPreallocationExecutionEngineException()。