Natively compiled SWT segfaults under Windows
Olivier Parisy
ml.olivier.parisy@free.fr
Tue Aug 15 09:47:00 GMT 2006
Ranjit Mathew a écrit :
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>> Olivier Parisy wrote:
>>> I was also able to compile and link basic SWT code, but it dies silently
>> at run time.
>>>> Executing it under GDB, I got:
>> Program received signal SIGSEGV, Segmentation fault.
>> fallback_backtrace(_Unwind_Reason_Code (*)(_Unwind_Context*, void*),
>> _Jv_UnwindState*) (
>>>> The current stack-unwinding code for GCJ/Win32 assumes that
> all code is compiled to use frame pointers to link together
> call frames on the stack. I suspect that the SWT code has
> not been compiled thus (i.e. using -fno-omit-frame-pointer).
>Should I only compile java code (and the SWT jar) with this flag, or
should the dll be rebuilt with those frame pointers, too?
> Instead of unceremoniously SIGSEGV-ing, perhaps we can
> find the base of the current thread's stack and limit
> the values we examine to that address and the current
> top-of-stack.
>Considering that the current behavior is for the application to die
silently excepted when run in a debugger, that would definitely be an
improvement.
Best regards,
Olivier Parisy.
More information about the Java
mailing list