GCJ/minGW stacktrace

Fri Jan 24 19:18:00 GMT 2003

Ranjit Mathew writes:
 > Andrew Haley wrote:
 > > I'm still planning to remove the dependence of stack traces on
 > > external programs. It will, however, require the DWARF unwinder. I
 > > intend to write a patch to enable this feature, but keep the old
 > > scheme running.
 > 
 > AFAIK, DW2 EH FDEs don't have symbolic information about the methods,
 > do they? So how will this help? Or am I missing something here?
We have all the symbolic information we need in Java's reflection
data. However, we need to be able to convert a return address into
the start adress of a method, and to do that we need to know where
each method begins and ends. That is the information in the DWARF
unwinder daya.
 > Besides, MinGW uses SJLJ now, which means that methods without a try
 > block would not have unwinding information - how would this scheme
 > work there?
It wouldn't.
Andrew.