Exception handler question

[Neal Gafter]

This is necessary to guarantee that the block is not exited until the
monitor has been exited, even in the presence of asynchronous
exceptions.

On Mon, Nov 17, 2008 at 1:48 PM, Roman Kennke <roman.kennke at aicas.com> wrote:
> Hi,
>
> I have a question regarding exception handlers. If I have code like
> this:
>
> synchronized (o)
>     {
>       x();
>     }
>
> produces bytecode like this:
>
> 0:   aload_1
>     1:   dup
>     2:   astore_2
>     3:   monitorenter
> #   4:   aload_0
> #   5:   invokevirtual   #2; //Method x:()V
> #   8:   aload_2
> #   9:   monitorexit
>     10:  goto    18
> *   13:  astore_3
> *   14:  aload_2
> *   15:  monitorexit
>     16:  aload_3
>     17:  athrow
>     18:  return
>   Exception table:
>    from   to  target type
> #    4    10    13   any
> *   13    16    13   any
>
> The code marked with # has an exception handler, which is marked with *.
> This is clearly ok, it is made so that the monitors are correctly exited
> when an exception is thrown in the sync block.
>
> My question is, why does the exception handler * point to itself?
> Wouldn't that catapult the VM into a loop, if monitorexit fails? A
> scenario that comes to mind is when x() is a JNI function that does a
> monitorexit. I'd like to understand why this code is generated like
> this.
>
> /Roman
> --
> Dipl.-Inform. (FH) Roman Kennke, Software Engineer, http://kennke.org
> aicas Allerton Interworks Computer Automated Systems GmbH
> Haid-und-Neu-Straße 18 * D-76131 Karlsruhe * Germany
> http://www.aicas.com   * Tel: +49-721-663 968-48
> USt-Id: DE216375633, Handelsregister HRB 109481, AG Karlsruhe
> Geschäftsführer: Dr. James J. Hunt
>
>