Re: symtab/1534: gdb current segfaults when debugee a.out changes

The following reply was made to PR symtab/1534; it has been noted by GNATS.

From: David Carlton <carlton@xxxxxxxxxx>
To: Daniel Jacobowitz <drow@xxxxxxxxxx>
Cc: rrh@xxxxxxxx, gdb-gnats@xxxxxxxxxxxxxxxxxx
Subject: Re: symtab/1534: gdb current segfaults when debugee a.out changes
Date: Fri, 23 Jan 2004 09:13:12 -0800

 On 23 Jan 2004 14:58:00 -0000, Daniel Jacobowitz <drow@xxxxxxxxxx> said:
  
 > I'm pretty sure this is yours.
 
 Given that I wrote several of the functions in that backtrace, I'm
 inclined to agree. :-)
 
 > The symtab is on the objfile's obstacks, which are wiped when it is
 > reloaded, so you probably need to wipe the symtab pointer too.  See
 > reread_symbols.
 
 Thanks for the diagnosis.  Robert, can you see if this patch helps?
 
 Index: symfile.c
 ===================================================================
 RCS file: /cvs/src/src/gdb/symfile.c,v
 retrieving revision 1.116
 diff -u -p -r1.116 symfile.c
 --- symfile.c  19 Jan 2004 19:56:02 -0000      1.116
 +++ symfile.c  23 Jan 2004 17:09:52 -0000
 @@ -1918,6 +1918,7 @@ reread_symbols (void)
              objfile->symtabs = NULL;
              objfile->psymtabs = NULL;
              objfile->free_psymtabs = NULL;
 +            objfile->cp_namespace_symtab = NULL;
              objfile->msymbols = NULL;
              objfile->sym_private = NULL;
              objfile->minimal_symbol_count = 0;