[commit] frame_chain_invalid() -> legacy_frame_chain_invalid()

[Andrew Cagney <ac131313 at redhat dot com>]

Hello,

This is a strong hint that the frame_chain_valid() function, that was 
once so critical to gdb, is almost no more.

committed,
Andrew

2003-04-06  Andrew Cagney  <cagney at redhat dot com>

	* frame.h (legacy_frame_chain_valid): Rename frame_chain_valid.
	Update comment.
	* frame.c (legacy_saved_regs_this_id): Update.
	(legacy_get_prev_frame): Update.
	* xstormy16-tdep.c: Update comment.
	* sparc-tdep.c (sparc_frame_chain): Update comment.
	* blockframe.c (legacy_frame_chain_valid): Update.
 
Index: blockframe.c
===================================================================
RCS file: /cvs/src/src/gdb/blockframe.c,v
retrieving revision 1.66
diff -u -r1.66 blockframe.c
--- blockframe.c	5 Apr 2003 18:54:38 -0000	1.66
+++ blockframe.c	6 Apr 2003 18:34:12 -0000
@@ -554,12 +554,12 @@
 	  && (pc) <= (CALL_DUMMY_ADDRESS () + DECR_PC_AFTER_BREAK));
 }
 
-/* Function: frame_chain_valid 
-   Returns true for a user frame or a call_function_by_hand dummy frame,
-   and false for the CRT0 start-up frame.  Purpose is to terminate backtrace.  */
+/* Returns true for a user frame or a call_function_by_hand dummy
+   frame, and false for the CRT0 start-up frame.  Purpose is to
+   terminate backtrace.  */
 
 int
-frame_chain_valid (CORE_ADDR fp, struct frame_info *fi)
+legacy_frame_chain_valid (CORE_ADDR fp, struct frame_info *fi)
 {
   /* Don't prune CALL_DUMMY frames.  */
   if (DEPRECATED_USE_GENERIC_DUMMY_FRAMES
Index: frame.c
===================================================================
RCS file: /cvs/src/src/gdb/frame.c,v
retrieving revision 1.100
diff -u -r1.100 frame.c
--- frame.c	6 Apr 2003 01:13:58 -0000	1.100
+++ frame.c	6 Apr 2003 18:34:13 -0000
@@ -821,7 +821,7 @@
       gdb_assert (DEPRECATED_FRAME_CHAIN_P ());
       base = DEPRECATED_FRAME_CHAIN (next_frame);
 
-      if (!frame_chain_valid (base, next_frame))
+      if (!legacy_frame_chain_valid (base, next_frame))
 	return;
     }
   if (base == 0)
@@ -1246,7 +1246,7 @@
       gdb_assert (DEPRECATED_FRAME_CHAIN_P ());
       address = DEPRECATED_FRAME_CHAIN (this_frame);
 
-      if (!frame_chain_valid (address, this_frame))
+      if (!legacy_frame_chain_valid (address, this_frame))
 	return 0;
     }
   if (address == 0)
Index: frame.h
===================================================================
RCS file: /cvs/src/src/gdb/frame.h,v
retrieving revision 1.84
diff -u -r1.84 frame.h
--- frame.h	6 Apr 2003 01:13:59 -0000	1.84
+++ frame.h	6 Apr 2003 18:34:13 -0000
@@ -455,10 +455,14 @@
 extern void *frame_obstack_zalloc (unsigned long size);
 #define FRAME_OBSTACK_ZALLOC(TYPE) ((TYPE *) frame_obstack_zalloc (sizeof (TYPE)))
 
-/* If DEPRECATED_FRAME_CHAIN_VALID returns zero it means that the
-   given frame is the outermost one and has no caller.  */
+/* If legacy_frame_chain_valid() returns zero it means that the given
+   frame is the outermost one and has no caller.
 
-extern int frame_chain_valid (CORE_ADDR, struct frame_info *);
+   This method has been superseeded by the per-architecture
+   frame_unwind_pc() (returns 0 to indicate an invalid return address)
+   and per-frame this_id() (returns a NULL frame ID to indicate an
+   invalid frame).  */
+extern int legacy_frame_chain_valid (CORE_ADDR, struct frame_info *);
 
 extern void generic_save_dummy_frame_tos (CORE_ADDR sp);
 
Index: sparc-tdep.c
===================================================================
RCS file: /cvs/src/src/gdb/sparc-tdep.c,v
retrieving revision 1.86
diff -u -r1.86 sparc-tdep.c
--- sparc-tdep.c	6 Apr 2003 01:13:59 -0000	1.86
+++ sparc-tdep.c	6 Apr 2003 18:34:14 -0000
@@ -437,16 +437,17 @@
      about the chain value.  If it really is zero, we detect it later
      in sparc_init_prev_frame.
      
-     Note:  kevinb/2003-02-18:  The constant 1 used to be returned
-     here, but, after some recent changes to frame_chain_valid(),
-     this value is no longer suitable for causing frame_chain_valid()
-     to "not worry about the chain value."  The constant ~0 (i.e,
-     0xfff...) causes the failing test in frame_chain_valid() to
-     succeed thus preserving the "not worry" property.  I had considered
-     using something like ``get_frame_base (frame) + 1''.  However, I think
-     a constant value is better, because when debugging this problem,
-     I knew that something funny was going on as soon as I saw the
-     constant 1 being used as the frame chain elsewhere in GDB.  */
+     Note: kevinb/2003-02-18: The constant 1 used to be returned here,
+     but, after some recent changes to legacy_frame_chain_valid(),
+     this value is no longer suitable for causing
+     legacy_frame_chain_valid() to "not worry about the chain value."
+     The constant ~0 (i.e, 0xfff...) causes the failing test in
+     legacy_frame_chain_valid() to succeed thus preserving the "not
+     worry" property.  I had considered using something like
+     ``get_frame_base (frame) + 1''.  However, I think a constant
+     value is better, because when debugging this problem, I knew that
+     something funny was going on as soon as I saw the constant 1
+     being used as the frame chain elsewhere in GDB.  */
 
   return ~ (CORE_ADDR) 0;
 }
Index: xstormy16-tdep.c
===================================================================
RCS file: /cvs/src/src/gdb/xstormy16-tdep.c,v
retrieving revision 1.44
diff -u -r1.44 xstormy16-tdep.c
--- xstormy16-tdep.c	1 Apr 2003 17:17:29 -0000	1.44
+++ xstormy16-tdep.c	6 Apr 2003 18:34:15 -0000
@@ -748,9 +748,8 @@
 
 /* Function: xstormy16_frame_saved_pc
    Returns the return address for the selected frame. 
-   Called by frame_info, frame_chain_valid, and sometimes by
-   get_prev_frame.
-*/
+   Called by frame_info, legacy_frame_chain_valid, and sometimes by
+   get_prev_frame.  */
 
 static CORE_ADDR
 xstormy16_frame_saved_pc (struct frame_info *fi)