Patch to fix PR9861

[David Daney]

TJ Laurenzo wrote:
>> > I would expect similar behavior from GDB (except that in GDB, the
>> > lack of demangling would really complicate things more than the
>> > same lack in binutils).  So, my opinion is that GDB is the main
>> > problem here, since this patch effectively makes Java debugging
>> > very difficult with an unpatched GDB.
>>
>>It does.  However, it might be possible to persuade people to ship
>>updated bnutils on free operating systems.  Unfree systems are going
>>to be much more problematic, though.
> 
> I'm not sure we are talking about a hard requirement to update
> binutils.  The old version works.  There are a few instances that
> could cause confusion but which could be solved with a FAQ entry that
> says to upgrade binutils to some version to get more specific error
> messages (ie.  Unresolved symbols in the output from ld would not
> pretty-print but would show raw mangled form).
> 
> As for GDB, I think it does become non-negotiable to upgrade that in
> order to get anything but the most rudimentary debugging support.  I
> have to ask, though, how many people who are using a bleeding-edge
> version of gcj will be unwilling to update to a new version of GDB?

I am not a gdb version expert, but I seem to recall reading somewhere 
that the gdb that ships with FC[34] has some RedHat local patches to 
allow it to work better with their patched kernel/glibc.  I may be being 
a worry wart, but I am concerned that since I use FC3 that I will not be 
able to use gdb's CVS HEAD.

Not that that should be a reason for not applying the patch.  If we need 
to change the mangling to be able to implement according to the JLS then 
sooner may be better than later.

David Daney