[Bug libgcj/62068] New: libjava/prims.cc:807: possible missing call to va_end ?

[Bug libgcj/62068] New: libjava/prims.cc:807: possible missing call to va_end ?

[dcb314 at hotmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62068

            Bug ID: 62068
           Summary: libjava/prims.cc:807: possible missing call to va_end
                    ?
           Product: gcc
           Version: 4.9.2
            Status: UNCONFIRMED
          Severity: minor
          Priority: P3
         Component: libgcj
          Assignee: unassigned at gcc dot gnu.org
          Reporter: dcb314 at hotmail dot com

[trunk/libjava/prims.cc:807]: (error) va_list 'args' was opened but not closed
by va_end().

  for (int i = 0; i < dimensions; ++i)
    {
      jint size = va_arg (args, jint);
      if (size < 0)
    throw new java::lang::NegativeArraySizeException;

Suggest add call to va_end just before throw.

[Bug libgcj/62068] libjava/prims.cc:807: possible missing call to va_end ?

[tromey at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62068

Tom Tromey <tromey at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |tromey at gcc dot gnu.org

--- Comment #1 from Tom Tromey <tromey at gcc dot gnu.org> ---
I thought va_start and va_end could expand to anything,
including code to introduce and close a block, so they
needed to be lexically paired.
This means that the fix here would have to rewrite
the logic in the function a little bit.

[Bug libgcj/62068] libjava/prims.cc:807: possible missing call to va_end ?

[schwab@linux-m68k.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62068

--- Comment #2 from Andreas Schwab <schwab@linux-m68k.org> ---
There is no such requirement for va_start/va_end.  You are probably thinking of
pthread_cleanup_push/pop.

[Bug libgcj/62068] libjava/prims.cc:807: possible missing call to va_end ?

[tromey at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62068

--- Comment #3 from Tom Tromey <tromey at gcc dot gnu.org> ---
It was documented that way on some systems.
Eg: http://www.cs.rit.edu/~hpb/Man/_Man_SunOS_4.1.3_html/html3/varargs.3.html
Perhaps that isn't operative language any more;
I guess I'm mildly curious to know.

[Bug libgcj/62068] libjava/prims.cc:807: possible missing call to va_end ?

[schwab@linux-m68k.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62068

--- Comment #4 from Andreas Schwab <schwab@linux-m68k.org> ---
varargs isn't stdargs.

[Bug libgcj/62068] libjava/prims.cc:807: possible missing call to va_end ?

[tromey at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62068

--- Comment #5 from Tom Tromey <tromey at gcc dot gnu.org> ---
(In reply to Andreas Schwab from comment #4)
> varargs isn't stdargs.

Doh.  Thanks.

[Bug libgcj/62068] libjava/prims.cc:807: possible missing call to va_end ?

[dcb314 at hotmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62068

--- Comment #6 from David Binderman <dcb314 at hotmail dot com> ---
(In reply to Andreas Schwab from comment #2)
> There is no such requirement for va_start/va_end.

I am not sure what you mean here. I checked K & R 2
and va_start and va_end seem to be a matched pair
to me. 

AFAIK, each va_end is 121 with each va_start.

In practice, some systems might be able to handle one 
of them being missing, but the language requirement is
still there.

Unless you know different ?

[Bug libgcj/62068] libjava/prims.cc:807: possible missing call to va_end ?

[schwab@linux-m68k.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62068

--- Comment #7 from Andreas Schwab <schwab@linux-m68k.org> ---
Please reread comment#1.