This is the mail archive of the gdb@sources.redhat.com mailing list for the GDB project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [Fwd: gdb "call" very slow on large C++ program; 2 possible patches]


Daniel Jacobowitz writes:
 > On Fri, Dec 07, 2001 at 11:21:32AM -0500, Elena Zannoni wrote:
 > > Yes, something like that happened. I am thinking to send a patch to
 > > restore some sanity to this function. But...
 > > 
 > > ...actually, I think that if search_symbols on FUNCTION_NAMESPACES is
 > > equivalent to make_symbol_overload_list(), we could just get rid of
 > > make_symbol_overload_list() *and* overload_list_add_symbol().
 > > 
 > > It would be nice to clean this up.
 > > 
 > > Daniel, does search_symbols produces a correct overload list? At first
 > > sight it would look so.
 > 
 > Well, I'm not immediately familiar with this code.  At first thought,
 > there might be some complexity because search_symbols takes a regexp,
 > but that's easily resolvable.  More importantly,
 > make_symbol_overload_list respects scope.  So no, the two are not
 > equivalent.
 > 

I was looking at the fact that make_symbol_overload_list is used only
for c++ non-member functions, and that search_symbols limits the
search to functions.

 > I found at least one more bug in overload_list_add_symbol (extraneous
 > realloc calls) while looking this over, though.  I also understand
 > where those loops came from, more or less; see make_completion_list. 
 > This function (and its comments) need some housecleaning.
 > 

Yes, there were calls (which were wrong) to overload_list_add_symbol,
in each loop. Sometimes ago, the #if 0 was lost, and so were the calls
in the loops.

Elena



  /* Comment and #if 0 from Rajiv Mirani <mirani@cup.hp.com>.
     However, leaving #if 0's around is uncool.  We need to figure out
     what this is really trying to do, decide whether we want that,
     and either fix it or delete it.  --- Jim Blandy, Mar 1999 */

  /* ??? RM: What in hell is this? overload_list_add_symbol expects a symbol,
   * not a partial_symbol or a minimal_symbol. And it looks at the type field
   * of the symbol, and we don't know the type of minimal and partial symbols
   */
#if 0
  /* Look through the partial symtabs for all symbols which begin
     by matching OLOAD_NAME.  Add each one that you find to the list.  */

  ALL_PSYMTABS (objfile, ps)
    {
      /* If the psymtab's been read in we'll get it when we search
         through the blockvector.  */
      if (ps->readin) continue;

      for (psym = objfile->global_psymbols.list + ps->globals_offset;
           psym < (objfile->global_psymbols.list + ps->globals_offset
                   + ps->n_global_syms);
           psym++)
        {
          /* If interrupted, then quit. */
          QUIT;
          overload_list_add_symbol (*psym, oload_name);
        }

      for (psym = objfile->static_psymbols.list + ps->statics_offset;
           psym < (objfile->static_psymbols.list + ps->statics_offset
                   + ps->n_static_syms);
           psym++)
        {
          QUIT;
          overload_list_add_symbol (*psym, oload_name);
        }
    }

  /* At this point scan through the misc symbol vectors and add each
     symbol you find to the list.  Eventually we want to ignore
     anything that isn't a text symbol (everything else will be
     handled by the psymtab code above).  */

  ALL_MSYMBOLS (objfile, msymbol)
    {
      QUIT;
      overload_list_add_symbol (msymbol, oload_name);
    }
#endif

  /* Search upwards from currently selected frame (so that we can
     complete on local vars.  */


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]