[Bug translator/10454] Raw number statement probes won't work without dwarf info

["mjw at redhat dot com" <sourceware-bugzilla at sourceware dot org>]

------- Additional Comments From mjw at redhat dot com  2009-07-29 11:21 -------
(In reply to comment #3)
> (In reply to comment #1)
> > There is an oddity that I don't understand with this code (when it does work).
> > It looks like if a probe containing a statement_num does have dwarf info then it
> > gets registered through query_func_info (i->entrypc, *i, q); which calls
> > query_statement() as follows:
> 
> I think the problem is in the conditional before that, which says:
> 
>   if (q->has_statement_str)
>     query_statement (i->name, i->decl_file,
>                      lineno, // NB: not q->line !
>                      &(i->die), addr, q);
>   else
>     query_func_info (i->entrypc, *i, q);
> 
> We really want the first case for has_statement_num too.  The else case is for
> function probes, where we want to clamp the address up to the function entry.

I wrote a testcase for this and it seems we do get this right (a statement probe
with address on a process/function with debuginfo gets set exactly on the given
statement). So, I must be reading the code wrong.

I'll attach the testcase. Not checked in yet, since it depends on resolving bug
#10458 first, so we get predictable uaddr() results.

-- 


http://sourceware.org/bugzilla/show_bug.cgi?id=10454

------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.