2002-11-29 Andrew Cagney <ac131313@redhat.com>

[deliverable/binutils-gdb.git] / gdb / frame.h

/* Definitions for dealing with stack frames, for GDB, the GNU debugger.

   Copyright 1986, 1988, 1989, 1990, 1991, 1992, 1993, 1994, 1996,
   1997, 1998, 1999, 2000, 2001, 2002 Free Software Foundation, Inc.

   This file is part of GDB.

   This program is free software; you can redistribute it and/or modify
   it under the terms of the GNU General Public License as published by
   the Free Software Foundation; either version 2 of the License, or
   (at your option) any later version.

   This program is distributed in the hope that it will be useful,
   but WITHOUT ANY WARRANTY; without even the implied warranty of
   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
   GNU General Public License for more details.

   You should have received a copy of the GNU General Public License
   along with this program; if not, write to the Free Software
   Foundation, Inc., 59 Temple Place - Suite 330,
   Boston, MA 02111-1307, USA.  */

#if !defined (FRAME_H)
#define FRAME_H 1

struct symtab_and_line;

/* The frame object.  */

struct frame_info;

/* The frame object's ID.  This provides a per-frame unique identifier
   that can be used to relocate a `struct frame_info' after a target
   resume or a frame cache destruct (assuming the target hasn't
   unwound the stack past that frame - a problem handled elsewhere).  */

struct frame_id
{
  /* The frame's address.  This should be constant through out the
     lifetime of a frame.  */
  /* NOTE: cagney/2002-11-16: The ia64 has two stacks and hence two
     frame bases.  This will need to be expanded to accomodate that.  */
  CORE_ADDR base;
  /* The frame's current PC.  While the PC within the function may
     change, the function that contains the PC does not.  Should this
     instead be the frame's function?  */
  CORE_ADDR pc;
};

/* For every stopped thread, GDB tracks two frames: current and
   selected.  Current frame is the inner most frame of the selected
   thread.  Selected frame is the one being examined by the the GDB
   CLI (selected using `up', `down', ...).  The frames are created
   on-demand (via get_prev_frame()) and then held in a frame cache.  */
/* FIXME: cagney/2002-11-28: Er, there is a lie here.  If you do the
   sequence: `thread 1; up; thread 2; thread 1' you loose thread 1's
   selected frame.  At present GDB only tracks the selected frame of
   the current thread.  But be warned, that might change.  */
/* FIXME: cagney/2002-11-14: At any time, only one thread's selected
   and current frame can be active.  Switching threads causes gdb to
   discard all that cached frame information.  Ulgh!  Instead, current
   and selected frame should be bound to a thread.  */

/* On demand, create the inner most frame using information found in
   the inferior.  If the inner most frame can't be created, throw an
   error.  */
extern struct frame_info *get_current_frame (void);

/* Invalidates the frame cache (this function should have been called
   invalidate_cached_frames).

   FIXME: cagney/2002-11-28: The only difference between
   flush_cached_frames() and reinit_frame_cache() is that the latter
   explicitly sets the selected frame back to the current frame there
   isn't any real difference (except that one delays the selection of
   a new frame).  Code can instead simply rely on get_selected_frame()
   to reinit's the selected frame as needed.  As for invalidating the
   cache, there should be two methods one that reverts the thread's
   selected frame back to current frame (for when the inferior
   resumes) and one that does not (for when the user modifies the
   target invalidating the frame cache).  */
extern void flush_cached_frames (void);
extern void reinit_frame_cache (void);

/* On demand, create the selected frame and then return it.  If the
   selected frame can not be created, this function throws an error.  */
/* FIXME: cagney/2002-11-28: At present, when there is no selected
   frame, this function always returns the current (inner most) frame.
   It should instead, when a thread has previously had its frame
   selected (but not resumed) and the frame cache invalidated, find
   and then return that thread's previously selected frame.  */
extern struct frame_info *get_selected_frame (void);

/* Select a specific frame.  NULL, apparently implies re-select the
   inner most frame.  */
extern void select_frame (struct frame_info *);

/* Given a FRAME, return the next (more inner, younger) or previous
   (more outer, older) frame.  */
extern struct frame_info *get_prev_frame (struct frame_info *);
extern struct frame_info *get_next_frame (struct frame_info *);

/* Given a frame's ID, relocate the frame.  Returns NULL if the frame
   is not found.  */
extern struct frame_info *frame_find_by_id (struct frame_id id);

/* Base attributes of a frame: */

/* The frame's `resume' address.  Where the program will resume in
   this frame.  */
extern CORE_ADDR get_frame_pc (struct frame_info *);

/* Closely related to the resume address, various symbol table
   attributes that are determined by the PC.  Note that for a normal
   frame, the PC refers to the resume address after the return, and
   not the call instruction.  In such a case, the address is adjusted
   so that it (approximatly) identifies the call site (and not return
   site).

   NOTE: cagney/2002-11-28: The frame cache could be used to cache the
   computed value.  Working on the assumption that the bottle-neck is
   in the single step code, and that code causes the frame cache to be
   constantly flushed, caching things in a frame is probably of little
   benefit.  As they say `show us the numbers'.

   NOTE: cagney/2002-11-28: Plenty more where this one came from:
   find_frame_block(), find_frame_partial_function(),
   find_frame_symtab(), find_frame_function().  Each will need to be
   carefully considered to determine if the real intent was for it to
   apply to the PC or the adjusted PC.  */
extern void find_frame_sal (struct frame_info *frame,
			    struct symtab_and_line *sal);

/* Return the frame address from FI.  Except in the machine-dependent
   *FRAME* macros, a frame address has no defined meaning other than
   as a magic cookie which identifies a frame over calls to the
   inferior (um, SEE NOTE BELOW).  The only known exception is
   inferior.h (PC_IN_CALL_DUMMY) [ON_STACK]; see comments there.  You
   cannot assume that a frame address contains enough information to
   reconstruct the frame; if you want more than just to identify the
   frame (e.g. be able to fetch variables relative to that frame),
   then save the whole struct frame_info (and the next struct
   frame_info, since the latter is used for fetching variables on some
   machines) (um, again SEE NOTE BELOW).

   NOTE: cagney/2002-11-18: Actually, the frame address isn't
   sufficient for identifying a frame, and the counter examples are
   wrong!

   Code that needs to (re)identify a frame must use get_frame_id() and
   frame_find_by_id() (and in the future, a frame_compare() function
   instead of INNER_THAN()).  Two reasons: an architecture (e.g.,
   ia64) can have more than one frame address (due to multiple stack
   pointers) (frame ID is going to be expanded to accomodate this);
   successive frameless function calls can only be differientated by
   comparing both the frame's base and the frame's enclosing function
   (frame_find_by_id() is going to be modified to perform this test). 

   The generic dummy frame version of PC_IN_CALL_DUMMY() is able to
   identify a dummy frame using only the PC value.  So the frame
   address is not needed.  In fact, most PC_IN_CALL_DUMMY() calls now
   pass zero as the frame/sp values as the caller knows that those
   values won't be used.  Once all architectures are using generic
   dummy frames, PC_IN_CALL_DUMMY() can drop the sp/frame parameters.
   When it comes to finding a dummy frame, the next frame's frame ID
   (with out duing an unwind) can be used (ok, could if it wasn't for
   the need to change the way the PPC defined frame base in a strange
   way).

   Modern architectures should be using something like dwarf2's
   location expression to describe where a variable lives.  Such
   expressions specify their own debug info centric frame address.
   Consequently, a generic frame address is pretty meaningless.  */

extern CORE_ADDR get_frame_base (struct frame_info *);

/* Return the per-frame unique identifer.  Can be used to relocate a
   frame after a frame cache flush (and other similar operations).  */
extern void get_frame_id (struct frame_info *fi, struct frame_id *id);

/* The frame's level: 0 for innermost, 1 for its caller, ...; or -1
   for an invalid frame).  */
extern int frame_relative_level (struct frame_info *fi);

/* Return the frame's type.  Some are real, some are signal
   trampolines, and some are completly artificial (dummy).  */

enum frame_type
{
  /* A true stack frame, created by the target program during normal
     execution.  */
  NORMAL_FRAME,
  /* A fake frame, created by GDB when performing an inferior function
     call.  */
  DUMMY_FRAME,
  /* In a signal handler, various OSs handle this in various ways.
     The main thing is that the frame may be far from normal.  */
  SIGTRAMP_FRAME
};
extern enum frame_type get_frame_type (struct frame_info *);

/* FIXME: cagney/2002-11-10: Some targets want to directly mark a
   frame as being of a specific type.  This shouldn't be necessary.
   PC_IN_SIGTRAMP() indicates a SIGTRAMP_FRAME and PC_IN_CALL_DUMMY()
   indicates a DUMMY_FRAME.  I suspect the real problem here is that
   get_prev_frame() only sets initialized after INIT_EXTRA_FRAME_INFO
   as been called.  Consequently, some targets found that the frame's
   type was wrong and tried to fix it.  The correct fix is to modify
   get_prev_frame() so that it initializes the frame's type before
   calling any other functions.  */
extern void deprecated_set_frame_type (struct frame_info *,
				       enum frame_type type);

/* Unwind the stack frame so that the value of REGNUM, in the previous
   (up, older) frame is returned.  If VALUEP is NULL, don't
   fetch/compute the value.  Instead just return the location of the
   value.  */
extern void frame_register_unwind (struct frame_info *frame, int regnum,
				   int *optimizedp, enum lval_type *lvalp,
				   CORE_ADDR *addrp, int *realnump,
				   void *valuep);

/* More convenient interface to frame_register_unwind().  */
/* NOTE: cagney/2002-09-13: Return void as one day these functions may
   be changed to return an indication that the read succeeded.  */

extern void frame_unwind_signed_register (struct frame_info *frame,
					  int regnum, LONGEST *val);

extern void frame_unwind_unsigned_register (struct frame_info *frame,
					    int regnum, ULONGEST *val);

/* Get the value of the register that belongs to this FRAME.  This
   function is a wrapper to the call sequence ``frame_unwind_register
   (get_next_frame (FRAME))''.  As per frame_register_unwind(), if
   VALUEP is NULL, the registers value is not fetched/computed.  */

extern void frame_register (struct frame_info *frame, int regnum,
			    int *optimizedp, enum lval_type *lvalp,
			    CORE_ADDR *addrp, int *realnump,
			    void *valuep);

/* More convenient interface to frame_register().  */
/* NOTE: cagney/2002-09-13: Return void as one day these functions may
   be changed to return an indication that the read succeeded.  */

extern void frame_read_signed_register (struct frame_info *frame,
					int regnum, LONGEST *val);

extern void frame_read_unsigned_register (struct frame_info *frame,
					  int regnum, ULONGEST *val);

/* Map between a frame register number and its name.  A frame register
   space is a superset of the cooked register space --- it also
   includes builtin registers.  */

extern int frame_map_name_to_regnum (const char *name, int strlen);
extern const char *frame_map_regnum_to_name (int regnum);

/* Unwind the PC.  Strictly speaking return the resume address of the
   calling frame.  For GDB, `pc' is the resume address and not a
   specific register.  */

extern CORE_ADDR frame_pc_unwind (struct frame_info *frame);

\f
/* Return the location (and possibly value) of REGNUM for the previous
   (older, up) frame.  All parameters except VALUEP can be assumed to
   be non NULL.  When VALUEP is NULL, just the location of the
   register should be returned.

   UNWIND_CACHE is provided as mechanism for implementing a per-frame
   local cache.  It's initial value being NULL.  Memory for that cache
   should be allocated using frame_obstack_alloc().

   Register window architectures (eg SPARC) should note that REGNUM
   identifies the register for the previous frame.  For instance, a
   request for the value of "o1" for the previous frame would be found
   in the register "i1" in this FRAME.  */

typedef void (frame_register_unwind_ftype) (struct frame_info *frame,
					    void **unwind_cache,
					    int regnum,
					    int *optimized,
					    enum lval_type *lvalp,
					    CORE_ADDR *addrp,
					    int *realnump,
					    void *valuep);

/* Same as for registers above, but return the address at which the
   calling frame would resume.  */

typedef CORE_ADDR (frame_pc_unwind_ftype) (struct frame_info *frame,
					   void **unwind_cache);

/* Describe the saved registers of a frame.  */

#if defined (EXTRA_FRAME_INFO) || defined (FRAME_FIND_SAVED_REGS)
/* XXXX - deprecated */
struct frame_saved_regs
  {
    /* For each register R (except the SP), regs[R] is the address at
       which it was saved on entry to the frame, or zero if it was not
       saved on entry to this frame.  This includes special registers
       such as pc and fp saved in special ways in the stack frame.

       regs[SP_REGNUM] is different.  It holds the actual SP, not the
       address at which it was saved.  */

    CORE_ADDR regs[NUM_REGS];
  };
#endif

/* We keep a cache of stack frames, each of which is a "struct
   frame_info".  The innermost one gets allocated (in
   wait_for_inferior) each time the inferior stops; current_frame
   points to it.  Additional frames get allocated (in
   get_prev_frame) as needed, and are chained through the next
   and prev fields.  Any time that the frame cache becomes invalid
   (most notably when we execute something, but also if we change how
   we interpret the frames (e.g. "set heuristic-fence-post" in
   mips-tdep.c, or anything which reads new symbols)), we should call
   reinit_frame_cache.  */

struct frame_info
  {
    /* Nominal address of the frame described.  See comments at
       get_frame_base() about what this means outside the *FRAME*
       macros; in the *FRAME* macros, it can mean whatever makes most
       sense for this machine.  */
    CORE_ADDR frame;

    /* Address at which execution is occurring in this frame.
       For the innermost frame, it's the current pc.
       For other frames, it is a pc saved in the next frame.  */
    CORE_ADDR pc;

    /* Level of this frame.  The inner-most (youngest) frame is at
       level 0.  As you move towards the outer-most (oldest) frame,
       the level increases.  This is a cached value.  It could just as
       easily be computed by counting back from the selected frame to
       the inner most frame.  */
    /* NOTE: cagney/2002-04-05: Perhaphs a level of ``-1'' should be
       reserved to indicate a bogus frame - one that has been created
       just to keep GDB happy (GDB always needs a frame).  For the
       moment leave this as speculation.  */
    int level;

    /* The frame's type.  */
    enum frame_type type;

    /* For each register, address of where it was saved on entry to
       the frame, or zero if it was not saved on entry to this frame.
       This includes special registers such as pc and fp saved in
       special ways in the stack frame.  The SP_REGNUM is even more
       special, the address here is the sp for the previous frame, not
       the address where the sp was saved.  */
    /* Allocated by frame_saved_regs_zalloc () which is called /
       initialized by FRAME_INIT_SAVED_REGS(). */
    CORE_ADDR *saved_regs;	/*NUM_REGS + NUM_PSEUDO_REGS*/

#ifdef EXTRA_FRAME_INFO
    /* XXXX - deprecated */
    /* Anything extra for this structure that may have been defined
       in the machine dependent files. */
      EXTRA_FRAME_INFO
#endif

    /* Anything extra for this structure that may have been defined
       in the machine dependent files. */
    /* Allocated by frame_obstack_alloc () which is called /
       initialized by INIT_EXTRA_FRAME_INFO */
    struct frame_extra_info *extra_info;

    /* If dwarf2 unwind frame informations is used, this structure holds all
       related unwind data.  */
    struct context *context;

    /* Unwind cache shared between the unwind functions - they had
       better all agree as to the contents.  */
    void *unwind_cache;

    /* See description above.  The previous frame's registers.  */
    frame_register_unwind_ftype *register_unwind;

    /* See description above.  The previous frame's resume address.
       Save the previous PC in a local cache.  */
    frame_pc_unwind_ftype *pc_unwind;
    int pc_unwind_cache_p;
    CORE_ADDR pc_unwind_cache;

    /* Pointers to the next (down, inner, younger) and previous (up,
       outer, older) frame_info's in the frame cache.  */
    struct frame_info *next; /* down, inner, younger */
    int prev_p;
    struct frame_info *prev; /* up, outer, older */
  };

/* Values for the source flag to be used in print_frame_info_base(). */
enum print_what
  { 
    /* Print only the source line, like in stepi. */
    SRC_LINE = -1, 
    /* Print only the location, i.e. level, address (sometimes)
       function, args, file, line, line num. */
    LOCATION,
    /* Print both of the above. */
    SRC_AND_LOC, 
    /* Print location only, but always include the address. */
    LOC_AND_ADDRESS 
  };

/* Allocate additional space for appendices to a struct frame_info.
   NOTE: Much of GDB's code works on the assumption that the allocated
   saved_regs[] array is the size specified below.  If you try to make
   that array smaller, GDB will happily walk off its end. */

#ifdef SIZEOF_FRAME_SAVED_REGS
#error "SIZEOF_FRAME_SAVED_REGS can not be re-defined"
#endif
#define SIZEOF_FRAME_SAVED_REGS \
        (sizeof (CORE_ADDR) * (NUM_REGS+NUM_PSEUDO_REGS))

extern void *frame_obstack_alloc (unsigned long size);
extern void frame_saved_regs_zalloc (struct frame_info *);

/* Define a default FRAME_CHAIN_VALID, in the form that is suitable for most
   targets.  If FRAME_CHAIN_VALID returns zero it means that the given frame
   is the outermost one and has no caller.

   XXXX - both default and alternate frame_chain_valid functions are
   deprecated.  New code should use dummy frames and one of the
   generic functions. */

extern int file_frame_chain_valid (CORE_ADDR, struct frame_info *);
extern int func_frame_chain_valid (CORE_ADDR, struct frame_info *);
extern int nonnull_frame_chain_valid (CORE_ADDR, struct frame_info *);
extern int generic_file_frame_chain_valid (CORE_ADDR, struct frame_info *);
extern int generic_func_frame_chain_valid (CORE_ADDR, struct frame_info *);
extern void generic_save_dummy_frame_tos (CORE_ADDR sp);



#ifdef FRAME_FIND_SAVED_REGS
/* XXX - deprecated */
#define FRAME_INIT_SAVED_REGS(FI) get_frame_saved_regs (FI, NULL)
extern void get_frame_saved_regs (struct frame_info *,
				  struct frame_saved_regs *);
#endif

extern struct block *get_frame_block (struct frame_info *,
                                      CORE_ADDR *addr_in_block);

/* Return the `struct block' that belongs to the selected thread's
   selected frame.  If the inferior has no state, return NULL.

   NOTE: cagney/2002-11-29:

   No state?  Does the inferior have any execution state (a core file
   does, an executable does not).  At present the code tests
   `target_has_stack' but I'm left wondering if it should test
   `target_has_registers' or, even, a merged target_has_state.

   Should it look at the most recently specified SAL?  If the target
   has no state, should this function try to extract a block from the
   most recently selected SAL?  That way `list foo' would give it some
   sort of reference point.  Then again, perhaphs that would confuse
   things.

   Calls to this function can be broken down into two categories: Code
   that uses the selected block as an additional, but optional, data
   point; Code that uses the selected block as a prop, when it should
   have the relevant frame/block/pc explicitly passed in.

   The latter can be eliminated by correctly parameterizing the code,
   the former though is more interesting.  Per the "address" command,
   it occures in the CLI code and makes it possible for commands to
   work, even when the inferior has no state.  */

extern struct block *get_selected_block (CORE_ADDR *addr_in_block);

extern struct symbol *get_frame_function (struct frame_info *);

extern CORE_ADDR frame_address_in_block (struct frame_info *);

extern CORE_ADDR get_pc_function_start (CORE_ADDR);

extern struct block *block_for_pc (CORE_ADDR);

extern struct block *block_for_pc_sect (CORE_ADDR, asection *);

extern int frameless_look_for_prologue (struct frame_info *);

extern void print_frame_args (struct symbol *, struct frame_info *,
			      int, struct ui_file *);

extern struct frame_info *find_relative_frame (struct frame_info *, int *);

extern void show_and_print_stack_frame (struct frame_info *fi, int level,
					int source);

extern void print_stack_frame (struct frame_info *, int, int);

extern void print_only_stack_frame (struct frame_info *, int, int);

extern void show_stack_frame (struct frame_info *);

extern void print_frame_info (struct frame_info *, int, int, int);

extern void show_frame_info (struct frame_info *, int, int, int);

extern struct frame_info *block_innermost_frame (struct block *);

/* NOTE: cagney/2002-09-13: There is no need for this function.
   Instead either of frame_unwind_signed_register() or
   frame_unwind_unsigned_register() can be used.  */
extern CORE_ADDR deprecated_read_register_dummy (CORE_ADDR pc,
						 CORE_ADDR fp, int);
extern void generic_push_dummy_frame (void);
extern void generic_pop_current_frame (void (*)(struct frame_info *));
extern void generic_pop_dummy_frame (void);

extern int generic_pc_in_call_dummy (CORE_ADDR pc,
				     CORE_ADDR sp, CORE_ADDR fp);

/* NOTE: cagney/2002-06-26: Targets should no longer use this
   function.  Instead, the contents of a dummy frames registers can be
   obtained by applying: frame_register_unwind to the dummy frame; or
   get_saved_register to the next outer frame.  */

extern char *deprecated_generic_find_dummy_frame (CORE_ADDR pc, CORE_ADDR fp);

extern void generic_fix_call_dummy (char *dummy, CORE_ADDR pc, CORE_ADDR fun,
				    int nargs, struct value **args,
				    struct type *type, int gcc_p);

/* The function generic_get_saved_register() has been made obsolete.
   GET_SAVED_REGISTER now defaults to the recursive equivalent -
   generic_unwind_get_saved_register() - so there is no need to even
   set GET_SAVED_REGISTER.  Architectures that need to override the
   register unwind mechanism should modify frame->unwind().  */
extern void deprecated_generic_get_saved_register (char *, int *, CORE_ADDR *,
						   struct frame_info *, int,
						   enum lval_type *);

extern void generic_save_call_dummy_addr (CORE_ADDR lo, CORE_ADDR hi);

extern void get_saved_register (char *raw_buffer, int *optimized,
				CORE_ADDR * addrp,
				struct frame_info *frame,
				int regnum, enum lval_type *lval);

extern int frame_register_read (struct frame_info *frame, int regnum,
				void *buf);

/* From stack.c.  */
extern void args_info (char *, int);

extern void locals_info (char *, int);

extern void (*selected_frame_level_changed_hook) (int);

extern void return_command (char *, int);


/* NOTE: cagney/2002-11-27:

   You might think that the below global can simply be replaced by a
   call to either get_selected_frame() or select_frame().

   Unfortunatly, it isn't that easy.

   The relevant code needs to be audited to determine if it is
   possible (or pratical) to instead pass the applicable frame in as a
   parameter.  For instance, DEPRECATED_DO_REGISTERS_INFO() relied on
   the deprecated_selected_frame global, while its replacement,
   PRINT_REGISTERS_INFO(), is parameterized with the selected frame.
   The only real exceptions occure at the edge (in the CLI code) where
   user commands need to pick up the selected frame before proceeding.

   This is important.  GDB is trying to stamp out the hack:

   saved_frame = deprecated_selected_frame;
   deprecated_selected_frame = ...;
   hack_using_global_selected_frame ();
   deprecated_selected_frame = saved_frame;

   Take care!  */

extern struct frame_info *deprecated_selected_frame;


/* NOTE: cagney/2002-11-28:

   These functions are used to explicitly create and set the inner
   most (current) frame vis:

   set_current_frame (create_new_frame (read_fp(), stop_pc)));

   Such code should be removed.  Instead that task can be left to
   get_current_frame() which will update things on-demand.

   The only vague exception is found in "infcmd.c" (and a few
   architectures specific files) as part of the code implementing the
   command ``(gdb) frame FRAME PC''.  There, the frame should be
   created/selected in a single shot.  */

extern void set_current_frame (struct frame_info *);
extern struct frame_info *create_new_frame (CORE_ADDR, CORE_ADDR);

#endif /* !defined (FRAME_H)  */