unexpected end of file :-(

unexpected end of file :-(

[Bastian Voigt]

Hello List!

I am working with pthreads-win32 and the microsoft vc++ compiler version 6.
I have a problem I already spent hours on:

I have ported a source file which was perfectly working before, to run 
in an own pthread.
Now the compiler always bitches about "unexpected end of file" (fatal 
error C1004).
On the following MSDN library page a verbose error description can be found:

http://msdn.microsoft.com/library/en-us/vccore/html/C1004.asp

- When I comment out all the pthreads related function calls and the 
#include pthreads statement,
then the file compiles perfectly. And yes, I am pretty sure that those 
lines are syntactically correct.

- When all the used pthread_* functions are defined as macros the file 
also compiles perfectly:
//#include <pthread.h>   
#define pthread_setcancelstate(a, b)
#define PTHREAD_CANCEL_DISABLE 1
#define pthread_cleanup_push(a,b)
#define pthread_exit(a)


So it appears to me that the problem must lie somewhere inside pthread.h ?
BTW my project also includes the IBM SCC libraries
(see http://www-3.ibm.com/security/cryptocards/html/overproduct.shtml).
I have managed to compile and run a little pthreads example program, but 
that was without SCC library import...


Any thoughts on this matter highly appreciated..

Greetings
Bastian Voigt

Re: unexpected end of file :-(

[Ross Johnson]

Hi Bastian,

There haven't been any other reports of this problem. Are 
you using a snapshot version of the pthreads-win32 library, 
or a version out of CVS?

One other thing I notice though - you said that when you 
#define all the pthread functions that you use it works, but 
you have defined
#define pthread_cleanup_push(a,b)
and no
#define pthread_cleanup_pop(a)

Is that an omission? The way these are defined is as a pair 
that forms the start and end of a block. That is, push with 
no pop will create an unbalanced { ... }, and that could be 
the problem.

Regards.
Ross

Bastian Voigt wrote:

> Hello List!
>
> I am working with pthreads-win32 and the microsoft vc++ compiler 
> version 6.
> I have a problem I already spent hours on:
>
> I have ported a source file which was perfectly working before, to run 
> in an own pthread.
> Now the compiler always bitches about "unexpected end of file" (fatal 
> error C1004).
> On the following MSDN library page a verbose error description can be 
> found:
>
> http://msdn.microsoft.com/library/en-us/vccore/html/C1004.asp
>
> - When I comment out all the pthreads related function calls and the 
> #include pthreads statement,
> then the file compiles perfectly. And yes, I am pretty sure that those 
> lines are syntactically correct.
>
> - When all the used pthread_* functions are defined as macros the file 
> also compiles perfectly:
> //#include <pthread.h>   #define pthread_setcancelstate(a, b)
> #define PTHREAD_CANCEL_DISABLE 1
> #define pthread_cleanup_push(a,b)
> #define pthread_exit(a)
>
>
> So it appears to me that the problem must lie somewhere inside 
> pthread.h ?
> BTW my project also includes the IBM SCC libraries
> (see http://www-3.ibm.com/security/cryptocards/html/overproduct.shtml).
> I have managed to compile and run a little pthreads example program, 
> but that was without SCC library import...
>
>
> Any thoughts on this matter highly appreciated..
>
> Greetings
> Bastian Voigt
>

Re: unexpected end of file :-(

[Bastian Voigt]

Ross Johnson wrote:

> #define pthread_cleanup_push(a,b)
> and no
> #define pthread_cleanup_pop(a)
>
> Is that an omission? 

No, it's not :-)

> The way these are defined is as a pair that forms the start and end of 
> a block. That is, push with no pop will create an unbalanced { ... }, 
> and that could be the problem.

Yes, that was the problem. I must admit Iam a pthreads n00b ..Thanks for 
your help!

Regards,
Bastian Voigt