This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: [PATCH] add 'rs6000_in_function_epilogue_p()' (Revised)
On Fri, Dec 02, 2005 at 01:19:28PM -0800, Jim Blandy wrote:
> > > Whatever we do immediately, the underlying cause here is that
> > > unwinding isn't working at every instruction. If it were, then the
> > > call to frame_find_by_id would find the frame containing the watched
> > > variable, and watchpoint_check would correctly conclude that the
> > > variable was still in scope.
> > >
> > > Paul, does your test program have Dwarf CFI for the functions in question?
> > >
> > I have attached a couple of dumps (and a typescript show how they where made), the executable, and the source.
>
> Okay, thanks. I don't have a ppc64 toolchain handy; could you post
> the output from objdump -d testwatch? The readelf dump that I wanted
> was 'readelf -wF' or '-wf'; but since you provided the executable,
> that's fine.
We don't even use CFI for PowerPC, still, because of the debacle with
incorrect register numbers recently fixed in GCC.
--
Daniel Jacobowitz
CodeSourcery, LLC