This is the mail archive of the gdb-patches@sourceware.org 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]

[RFA] improved handling of shared lib calls (reverse 2/2)


This change depends on the previous one (epilogue unwinder for i386,
http://sourceware.org/ml/gdb-patches/2009-07/msg00099.html).

With the frame unwinder working properly, we can remove the
special case code in the test for "check for subroutine call"
in handle_inferior_event.

Of course that reveals some hidden problems with shared library
calls, so this patch also addresses those.

Michael

2009-07-03  Michael Snyder  <msnyder@vmware.com>

	* infrun.c (handle_inferior_event): Remove an execution_direction
	check in the "check for subroutine calls" test, and add some 
	additional handling for reverse-stepping through shared libraries.

Index: infrun.c
===================================================================
RCS file: /cvs/src/src/gdb/infrun.c,v
retrieving revision 1.400
diff -u -p -r1.400 infrun.c
--- infrun.c	2 Jul 2009 17:21:06 -0000	1.400
+++ infrun.c	4 Jul 2009 00:27:53 -0000
@@ -3776,9 +3776,8 @@ infrun: not switching back to stepped th
      previous frame must have valid frame IDs.  */
   if (!frame_id_eq (get_stack_frame_id (frame),
 		    ecs->event_thread->step_stack_frame_id)
-      && (frame_id_eq (frame_unwind_caller_id (frame),
-		       ecs->event_thread->step_stack_frame_id)
-	  || execution_direction == EXEC_REVERSE))
+      && frame_id_eq (frame_unwind_caller_id (frame),
+		      ecs->event_thread->step_stack_frame_id))
     {
       CORE_ADDR real_stop_pc;
 
@@ -3806,6 +3805,7 @@ infrun: not switching back to stepped th
       /* Reverse stepping through solib trampolines.  */
 
       if (execution_direction == EXEC_REVERSE
+	  && ecs->event_thread->step_over_calls != STEP_OVER_NONE
 	  && (gdbarch_skip_trampoline_code (gdbarch, frame, stop_pc)
 	      || (ecs->stop_func_start == 0
 		  && in_solib_dynsym_resolve_code (stop_pc))))
@@ -3923,6 +3923,38 @@ infrun: not switching back to stepped th
       return;
     }
 
+  /* Reverse stepping through solib trampolines.  */
+
+  if (execution_direction == EXEC_REVERSE
+      && ecs->event_thread->step_over_calls != STEP_OVER_NONE)
+    {
+      if (gdbarch_skip_trampoline_code (gdbarch, frame, stop_pc)
+	  || (ecs->stop_func_start == 0
+	      && in_solib_dynsym_resolve_code (stop_pc)))
+	{
+	  /* Any solib trampoline code can be handled in reverse
+	     by simply continuing to single-step.  We have already
+	     executed the solib function (backwards), and a few 
+	     steps will take us back through the trampoline to the
+	     caller.  */
+	  keep_going (ecs);
+	  return;
+	}
+      else if (in_solib_dynsym_resolve_code (stop_pc))
+	{
+	  /* Stepped backward into the solib dynsym resolver.
+	     Set a breakpoint at its start and continue, then
+	     one more step will take us out.  */
+	  struct symtab_and_line sr_sal;
+	  init_sal (&sr_sal);
+	  sr_sal.pc = ecs->stop_func_start;
+	  insert_step_resume_breakpoint_at_sal (gdbarch, 
+						sr_sal, null_frame_id);
+	  keep_going (ecs);
+	  return;
+	}
+    }
+
   /* If we're in the return path from a shared library trampoline,
      we want to proceed through the trampoline when stepping.  */
   if (gdbarch_in_solib_return_trampoline (gdbarch,

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