Re: [patch] java run-time stack trace

[Andrew Haley <aph at redhat dot com>]

Bryce McKinlay writes:
 > Per Bothner wrote:
 > 
 > > This patch accomplishes that.  It does have the undesirable side-effect that
 > > C or C++ functions are are written Java-style (i.e no * or ::):
 > >
 > >    at 0x40186645: _Jv_FindClass(_Jv_Utf8Const, java.lang.ClassLoader) (/home/bothner/GNU/linux/lib/libgcj.so.1)
 > >
 > > but I think that is reasonable enough for a Java stack printer.
 > >
 > > Any objections to checking this in (now, before the branch)?
 > 
 > - drop the PC address display, because it takes up precious space
 > and isn't really interesting for 99% of debugging tasks

I've had *very* emphatic demands for the PC display in embedded
systems for post mortem dumps.  It's not like you can always attach a
gdb to the process.  Embedded systems are a very important part of
users of gcj.

It's clear that people have very different requirements for backtrace:
any change should be configurable, and IMO runtime configurable.

 > - print line number information for frames from shared libs. In
 > order for this to work we'd need to either open up multiple
 > sessions of addr2line (one for each shared lib);

Yes, that would be an improvement.

 > or just call into libbfd directly.

This is a GPL issue: if it had been possible legally to link with
libbfd we would have done that already instead of all this messing
about invoking an external program.

Andrew.