Re: [Boost-users] Visual C++ 8 - Example Crashes with Access Violation

From: Ovanes Markarian (om_boost_at_[hidden])
Date: 2006-04-27 11:25:11

This is really a strange behavior. I can your sample app in VC 2003 without any crashes...

Try to comment the output to std::cout after your timer has elapsed... Maybe there is a problem
with writing to streams and cout somehow does not block and main exists while cout is written... ?

On Thu, April 27, 2006 2:42, Rush Manbert said:
> Brown Gabe wrote:
>> Hi Everyone,
>>
>>   Hopefully this isn't some noob question, but here goes...  I'm trying to
>> run boost 1.3.1 in Visual C++ 8 VC8 and I keep getting an access violation
>> at the termination of my thread.  I've tried looking to see what the problem
>> is but I can't seem to determine the problem.  I'm using the plain example
>> thread.cpp
>>
> <snip>
>
>> I get the expected out from the application. Upon tracing my code, the
>> exception only occurs when the worker thread exits.
>>
>> Exception occured in threadex.c on line 348.
>> Unhandled exception at 0x7c9012b4 in test.exe: 0xC0000005: Access violation
>> reading location 0xcccccccc.
>>
>> "    __try {
>>             _endthreadex (
>>                 ( (unsigned (__CLR_OR_STD_CALL *)(void
>> *))(((_ptiddata)ptd)->_initaddr) )
>>                 ( ((_ptiddata)ptd)->_initarg ) ) ;"
>>
>> Any thoughts on what might cause an error like this???
>>
>> --
>
> Others may have more intelligent, or more focused, suggestions, but here
> are two things to check:
>
> 1) 0xcccccccc looks very much like the contents of an uninitialized
> variable when running under the debug heap management code. My guess is
> that you are using an uninitialized pointer value whose contents are
> 0xcccccccc.
>
> See http://www.codeguru.com/cpp/w-p/win32/tutorials/article.php/c9535__1
> the follow the "continued" link to the second page. The fill values are
> listed there.
>
> 2) I assume you're using VS 2005? The Debug | Exceptions... dialog lets
> you set it to break when an exception is thrown, rather than fielding an
> unhandled exception. I don't know anything about debugging threads, though.
>
> Best of luck,
> Rush
>