Re: Forwarding and the GNU runtime one more time (was ANN: GNUstep 2.6.0 for openSUSE 11.4)

[David Chisnall]

On 22 Jun 2011, at 11:23, Nicola Pero wrote:

> I don't think that should be a problem with the GNU runtime. :-)
> 
> The forwarding IMP you get is (should be) a normal function that you can call
> without special consequences on the stacktrace.  It's exactly the same
> function that gets called when forwarding is done during a normal method
> invocation. ;-)


The forwarding IMP you get is generated by libffi.  It does not export any 
DWARF data, so it will also uglify the back trace.  It will usually show up as 
three addresses of the form ?? in a back trace.

Note that there is a mechanism in recent versions of GDB for providing DWARF 
data for JIT'd functions, so if anyone cared sufficiently then it would be 
possible to fix this.

David

--
This email complies with ISO 3103