File gdb-6.6-scheduler_locking-step-sw-watchpoints2.... of Package gdb

Overview Repositories Revisions Requests Users Attributes Meta

File gdb-6.6-scheduler_locking-step-sw-watchpoints2.patch of Package gdb (Revision 8a55a09e941dbd057e6da99e99f9b032)

Currently displaying revision 8a55a09e941dbd057e6da99e99f9b032 , Show latest

2007-06-25  Jan Kratochvil  <jan.kratochvil@redhat.com>

* inferior.h (enum resume_step): New definition.
	(resume): Change STEP parameter type to ENUM RESUME_STEP.
	* infrun.c (resume): Likewise.  Extend debug printing of the STEP
	parameter.  Lock the scheduler only for intentional stepping.
	(proceed): Replace the variable ONESTEP with tristate RESUME_STEP.
	Set the third RESUME_STEP state according to BPSTAT_SHOULD_STEP.
	(currently_stepping): Change the return type to ENUM RESUME_STEP.
	Return RESUME_STEP_NEEDED if it is just due to BPSTAT_SHOULD_STEP.
	* linux-nat.c (select_singlestep_lwp_callback): Do not focus on
	the software watchpoint events.
	* linux-nat.h (struct lwp_info): Redeclare STEP as ENUM RESUME_STEP.

2007-10-19  Jan Kratochvil  <jan.kratochvil@redhat.com>

* infrun.c (proceed): RESUME_STEP initialized for non-stepping.
	RESUME_STEP set according to STEP only at the end of the function.

2008-02-24  Jan Kratochvil  <jan.kratochvil@redhat.com>

Port to GDB-6.8pre.

Index: gdb-6.8.50.20081209/gdb/inferior.h
===================================================================
--- gdb-6.8.50.20081209.orig/gdb/inferior.h	2008-11-20 01:35:23.000000000 +0100
+++ gdb-6.8.50.20081209/gdb/inferior.h	2008-12-10 01:22:23.000000000 +0100
@@ -168,7 +168,15 @@ extern void reopen_exec_file (void);
 /* The `resume' routine should only be called in special circumstances.
    Normally, use `proceed', which handles a lot of bookkeeping.  */
 
-extern void resume (int, enum target_signal);
+enum resume_step
+  {
+    /* currently_stepping () should return non-zero for non-continue.  */
+    RESUME_STEP_CONTINUE = 0,
+    RESUME_STEP_USER,		/* Stepping is intentional by the user.  */
+    RESUME_STEP_NEEDED		/* Stepping only for software watchpoints.  */
+  };
+
+extern void resume (enum resume_step, enum target_signal);
 
 /* From misc files */
 
Index: gdb-6.8.50.20081209/gdb/infrun.c
===================================================================
--- gdb-6.8.50.20081209.orig/gdb/infrun.c	2008-12-02 20:20:23.000000000 +0100
+++ gdb-6.8.50.20081209/gdb/infrun.c	2008-12-10 01:23:46.000000000 +0100
@@ -73,7 +73,7 @@ static int follow_fork (void);
 static void set_schedlock_func (char *args, int from_tty,
 				struct cmd_list_element *c);
 
-static int currently_stepping (struct thread_info *tp);
+static enum resume_step currently_stepping (struct thread_info *tp);
 
 static int currently_stepping_callback (struct thread_info *tp, void *data);
 
@@ -961,7 +961,7 @@ set_schedlock_func (char *args, int from
    STEP nonzero if we should step (zero to continue instead).
    SIG is the signal to give the inferior (zero for none).  */
 void
-resume (int step, enum target_signal sig)
+resume (enum resume_step step, enum target_signal sig)
 {
   int should_resume = 1;
   struct cleanup *old_cleanups = make_cleanup (resume_cleanups, 0);
@@ -975,10 +975,12 @@ resume (int step, enum target_signal sig
   QUIT;
 
   if (debug_infrun)
-    fprintf_unfiltered (gdb_stdlog,
-                        "infrun: resume (step=%d, signal=%d), "
-			"trap_expected=%d\n",
- 			step, sig, tp->trap_expected);
+    fprintf_unfiltered (gdb_stdlog, "infrun: resume (step=%s, signal=%d), "
+				    "trap_expected=%d\n",
+			(step == RESUME_STEP_CONTINUE ? "RESUME_STEP_CONTINUE"
+			: (step == RESUME_STEP_USER ? "RESUME_STEP_USER"
+			                            : "RESUME_STEP_NEEDED")),
+			sig, tp->trap_expected);
 
   /* Some targets (e.g. Solaris x86) have a kernel bug when stepping
      over an instruction that causes a page fault without triggering
@@ -1127,9 +1129,10 @@ a command like `return' or `jump' to con
 	     individually.  */
 	  resume_ptid = inferior_ptid;
 	}
-      else if ((scheduler_mode == schedlock_on)
+      else if (scheduler_mode == schedlock_on
 	       || (scheduler_mode == schedlock_step
-		   && (step || singlestep_breakpoints_inserted_p)))
+		   && (step == RESUME_STEP_USER
+		       || singlestep_breakpoints_inserted_p)))
 	{
 	  /* User-settable 'scheduler' mode requires solo thread resume. */
 	  resume_ptid = inferior_ptid;
@@ -1302,7 +1305,7 @@ proceed (CORE_ADDR addr, enum target_sig
   struct gdbarch *gdbarch = get_regcache_arch (regcache);
   struct thread_info *tp;
   CORE_ADDR pc = regcache_read_pc (regcache);
-  int oneproc = 0;
+  enum resume_step resume_step = RESUME_STEP_CONTINUE;
 
   if (step > 0)
     step_start_function = find_pc_function (pc);
@@ -1322,13 +1325,13 @@ proceed (CORE_ADDR addr, enum target_sig
 	   actually be executing the breakpoint insn anyway.
 	   We'll be (un-)executing the previous instruction.  */
 
-	oneproc = 1;
+	resume_step = RESUME_STEP_USER;
       else if (gdbarch_single_step_through_delay_p (gdbarch)
 	       && gdbarch_single_step_through_delay (gdbarch,
 						     get_current_frame ()))
 	/* We stepped onto an instruction that needs to be stepped
 	   again before re-inserting the breakpoint, do so.  */
-	oneproc = 1;
+	resume_step = RESUME_STEP_USER;
     }
   else
     {
@@ -1359,13 +1362,13 @@ proceed (CORE_ADDR addr, enum target_sig
 	 is required it returns TRUE and sets the current thread to
 	 the old thread. */
       if (prepare_to_proceed (step))
-	oneproc = 1;
+	resume_step = RESUME_STEP_USER;
     }
 
   /* prepare_to_proceed may change the current thread.  */
   tp = inferior_thread ();
 
-  if (oneproc)
+  if (resume_step == RESUME_STEP_USER)
     {
       tp->trap_expected = 1;
       /* If displaced stepping is enabled, we can step over the
@@ -1451,8 +1454,13 @@ proceed (CORE_ADDR addr, enum target_sig
   /* Reset to normal state.  */
   init_infwait_state ();
 
+  if (step)
+    resume_step = RESUME_STEP_USER;
+  if (resume_step == RESUME_STEP_CONTINUE && bpstat_should_step ())
+    resume_step = RESUME_STEP_NEEDED;
+
   /* Resume inferior.  */
-  resume (oneproc || step || bpstat_should_step (), tp->stop_signal);
+  resume (resume_step, tp->stop_signal);
 
   /* Wait for it to stop (if not standalone)
      and in any case decode why it stopped, and act accordingly.  */
@@ -3690,10 +3698,16 @@ currently_stepping_callback (struct thre
   return tp != data && currently_stepping_thread (tp);
 }
 
-static int
+static enum resume_step
 currently_stepping (struct thread_info *tp)
 {
-  return currently_stepping_thread (tp) || bpstat_should_step ();
+  if (currently_stepping_thread (tp))
+    return RESUME_STEP_USER;
+
+  if (bpstat_should_step ())
+    return RESUME_STEP_NEEDED;
+
+  return RESUME_STEP_CONTINUE;
 }
 
 /* Inferior has stepped into a subroutine call with source code that
Index: gdb-6.8.50.20081209/gdb/linux-nat.c
===================================================================
--- gdb-6.8.50.20081209.orig/gdb/linux-nat.c	2008-12-02 08:57:36.000000000 +0100
+++ gdb-6.8.50.20081209/gdb/linux-nat.c	2008-12-10 01:22:23.000000000 +0100
@@ -2343,7 +2343,10 @@ count_events_callback (struct lwp_info *
 static int
 select_singlestep_lwp_callback (struct lwp_info *lp, void *data)
 {
-  if (lp->step && lp->status != 0)
+  /* We do not focus on software watchpoints as we would not catch
+     STEPPING_PAST_SINGLESTEP_BREAKPOINT breakpoints in some other thread
+     as they would remain pending due to `Push back breakpoint for %s'.  */
+  if (lp->step == RESUME_STEP_USER && lp->status != 0)
     return 1;
   else
     return 0;
Index: gdb-6.8.50.20081209/gdb/linux-nat.h
===================================================================
--- gdb-6.8.50.20081209.orig/gdb/linux-nat.h	2008-07-27 23:12:40.000000000 +0200
+++ gdb-6.8.50.20081209/gdb/linux-nat.h	2008-12-10 01:22:23.000000000 +0100
@@ -55,8 +55,8 @@ struct lwp_info
   /* If non-zero, a pending wait status.  */
   int status;
 
-  /* Non-zero if we were stepping this LWP.  */
-  int step;
+  /* The kind of stepping of this LWP.  */
+  enum resume_step step;
 
   /* Non-zero si_signo if this LWP stopped with a trap.  si_addr may
      be the address of a hardware watchpoint.  */

Places

File gdb-6.6-scheduler_locking-step-sw-watchpoints2.patch of Package gdb (Revision 8a55a09e941dbd057e6da99e99f9b032)

Places