* [rfc] Clear some global variables when reloading object files
@ 2006-08-03 21:46 Daniel Jacobowitz
2006-08-04 21:52 ` Mark Kettenis
0 siblings, 1 reply; 3+ messages in thread
From: Daniel Jacobowitz @ 2006-08-03 21:46 UTC (permalink / raw)
To: gdb-patches, Mark Kettenis
Mark, I encountered a problem today with expression_context_block that
sounded exactly like what you encountered with innermost_block. I
suspect the right thing to do would be to clean up the global variables
entirely, but I agree that it would be a terrible mess to untangle,
and this seems like the next best thing. What do you think?
--
Daniel Jacobowitz
CodeSourcery
2006-08-03 Daniel Jacobowitz <dan@codesourcery.com>
* Makefile.in (objfiles.o, symfile.o): Update.
* objfiles.c: Include "expression.h" and "parser-defs.h".
(free_objfile): Clear global blocks.
* symfile.c: Include "parser-defs.h".
(clear_symtab_users): Clear global blocks.
--- Makefile.in (revision 146239)
+++ Makefile.in (local)
@@ -2441,7 +2441,8 @@ objc-lang.o: objc-lang.c $(defs_h) $(sym
objfiles.o: objfiles.c $(defs_h) $(bfd_h) $(symtab_h) $(symfile_h) \
$(objfiles_h) $(gdb_stabs_h) $(target_h) $(bcache_h) $(mdebugread_h) \
$(gdb_assert_h) $(gdb_stat_h) $(gdb_obstack_h) $(gdb_string_h) \
- $(hashtab_h) $(breakpoint_h) $(block_h) $(dictionary_h) $(source_h)
+ $(hashtab_h) $(breakpoint_h) $(block_h) $(dictionary_h) $(source_h) \
+ $(parser_defs_h) $(expression_h)
observer.o: observer.c $(defs_h) $(observer_h) $(command_h) $(gdbcmd_h) \
$(observer_inc)
obsd-tdep.o: obsd-tdep.c $(defs_h) $(frame_h) $(symtab_h) $(obsd_tdep_h)
@@ -2764,7 +2765,8 @@ symfile.o: symfile.c $(defs_h) $(bfdlink
$(complaints_h) $(demangle_h) $(inferior_h) $(filenames_h) \
$(gdb_stabs_h) $(gdb_obstack_h) $(completer_h) $(bcache_h) \
$(hashtab_h) $(readline_h) $(gdb_assert_h) $(block_h) \
- $(gdb_string_h) $(gdb_stat_h) $(observer_h) $(exec_h)
+ $(gdb_string_h) $(gdb_stat_h) $(observer_h) $(exec_h) \
+ $(parser_defs_h)
symfile-mem.o: symfile-mem.c $(defs_h) $(symtab_h) $(gdbcore_h) \
$(objfiles_h) $(exceptions_h) $(gdbcmd_h) $(target_h) $(value_h) \
$(symfile_h) $(observer_h) $(auxv_h) $(elf_common_h)
--- objfiles.c (revision 146239)
+++ objfiles.c (local)
@@ -34,6 +34,9 @@
#include "target.h"
#include "bcache.h"
#include "mdebugread.h"
+#include "expression.h"
+#include "parser-defs.h"
+
#include "gdb_assert.h"
#include <sys/types.h>
#include "gdb_stat.h"
@@ -441,6 +444,12 @@ free_objfile (struct objfile *objfile)
for example), so we need to call this here. */
clear_pc_function_cache ();
+ /* Clear globals which might have pointed into a removed objfile.
+ FIXME: It's not clear which of these are supposed to persist
+ between expressions and which ought to be reset each time. */
+ expression_context_block = NULL;
+ innermost_block = NULL;
+
/* Check to see if the current_source_symtab belongs to this objfile,
and if so, call clear_current_source_symtab_and_line. */
--- symfile.c (revision 146239)
+++ symfile.c (local)
@@ -51,6 +51,7 @@
#include "block.h"
#include "observer.h"
#include "exec.h"
+#include "parser-defs.h"
#include <sys/types.h>
#include <fcntl.h>
@@ -2537,6 +2538,12 @@ clear_symtab_users (void)
clear_pc_function_cache ();
if (deprecated_target_new_objfile_hook)
deprecated_target_new_objfile_hook (NULL);
+
+ /* Clear globals which might have pointed into a removed objfile.
+ FIXME: It's not clear which of these are supposed to persist
+ between expressions and which ought to be reset each time. */
+ expression_context_block = NULL;
+ innermost_block = NULL;
}
static void
^ permalink raw reply [flat|nested] 3+ messages in thread* Re: [rfc] Clear some global variables when reloading object files
2006-08-03 21:46 [rfc] Clear some global variables when reloading object files Daniel Jacobowitz
@ 2006-08-04 21:52 ` Mark Kettenis
2006-08-08 17:38 ` Daniel Jacobowitz
0 siblings, 1 reply; 3+ messages in thread
From: Mark Kettenis @ 2006-08-04 21:52 UTC (permalink / raw)
To: drow; +Cc: gdb-patches
> Date: Thu, 3 Aug 2006 17:46:42 -0400
> From: Daniel Jacobowitz <drow@false.org>
>
> Mark, I encountered a problem today with expression_context_block that
> sounded exactly like what you encountered with innermost_block. I
> suspect the right thing to do would be to clean up the global variables
> entirely, but I agree that it would be a terrible mess to untangle,
> and this seems like the next best thing. What do you think?
Looks reasonable to me.
> 2006-08-03 Daniel Jacobowitz <dan@codesourcery.com>
>
> * Makefile.in (objfiles.o, symfile.o): Update.
> * objfiles.c: Include "expression.h" and "parser-defs.h".
> (free_objfile): Clear global blocks.
> * symfile.c: Include "parser-defs.h".
> (clear_symtab_users): Clear global blocks.
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [rfc] Clear some global variables when reloading object files
2006-08-04 21:52 ` Mark Kettenis
@ 2006-08-08 17:38 ` Daniel Jacobowitz
0 siblings, 0 replies; 3+ messages in thread
From: Daniel Jacobowitz @ 2006-08-08 17:38 UTC (permalink / raw)
To: Mark Kettenis; +Cc: gdb-patches
On Fri, Aug 04, 2006 at 11:51:23PM +0200, Mark Kettenis wrote:
> > Date: Thu, 3 Aug 2006 17:46:42 -0400
> > From: Daniel Jacobowitz <drow@false.org>
> >
> > Mark, I encountered a problem today with expression_context_block that
> > sounded exactly like what you encountered with innermost_block. I
> > suspect the right thing to do would be to clean up the global variables
> > entirely, but I agree that it would be a terrible mess to untangle,
> > and this seems like the next best thing. What do you think?
>
> Looks reasonable to me.
OK, I checked it in. Let me know if this doesn't fix the crashes you
were seeing?
--
Daniel Jacobowitz
CodeSourcery
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2006-08-08 17:38 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2006-08-03 21:46 [rfc] Clear some global variables when reloading object files Daniel Jacobowitz
2006-08-04 21:52 ` Mark Kettenis
2006-08-08 17:38 ` Daniel Jacobowitz
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox