New file: rs6000-core.c.

[binutils.git] / bfd / doc / bfdinfo

Info file bfdinfo, produced by Makeinfo, -*- Text -*- from input file
bfd.texinfo.

   This file documents the BFD library.

   Copyright (C) 1991 Free Software Foundation, Inc.

   Permission is granted to make and distribute verbatim copies of
this manual provided the copyright notice and this permission notice
are preserved on all copies.

   Permission is granted to copy and distribute modified versions of
this manual under the conditions for verbatim copying, subject to the
terms of the GNU General Public License, which includes the provision
that the entire resulting derived work is distributed under the terms
of a permission notice identical to this one.

   Permission is granted to copy and distribute translations of this
manual into another language, under the above conditions for modified
versions.

\1f
File: bfdinfo,  Node: Top,  Next: Overview,  Prev: (dir),  Up: (dir)

   This file documents the binary file descriptor library libbfd.

* Menu:

* Overview::                    Overview of BFD
* History::                     History of BFD
* Backends::                    Backends
* Porting::                     Porting
* Future::                      Future
* Index::                       Index

BFD body:
* Memory usage::
* Sections::
* Symbols::
* Archives::
* Formats::
* Relocations::
* Core Files::
* Targets::
* Architecturs::
* Opening and Closing::
* Internal::
* File Caching::

BFD backends:
* a.out backends::
* coff backends::

\1f
File: bfdinfo,  Node: Overview,  Next: History,  Prev: Top,  Up: Top

Introduction
************

   Simply put, BFD is a package which allows applications to use the
same routines to operate on object files whatever the object file
format.  A different object file format can be supported simply by
creating a new BFD back end and adding it to the library.

   BFD is split into two parts; the front end and the many back ends.

   * memory, and various canonical data structures. The front end also
     decides which back end to use, and when to call back end routines.

   * end provides a set of calls which the BFD front end can use to
     maintain its canonical form. The back ends also may keep around
     information for their own use, for greater efficiency.

\1f
File: bfdinfo,  Node: History,  Next: How It Works,  Prev: Overview,  Up: Top

History
=======

   One spur behind BFD was the desire, on the part of the GNU 960 team
at Intel Oregon, for interoperability of applications on their COFF and
b.out file formats.  Cygnus was providing GNU support for the team, and
Cygnus was contracted to provide the required functionality.

   The name came from a conversation David Wallace was having with
Richard Stallman about the library: RMS said that it would be quite
hard--David said "BFD".  Stallman was right, but the name stuck.

   At the same time, Ready Systems wanted much the same thing, but for
different object file formats: IEEE-695, Oasys, Srecords, a.out and 68k
coff.

   BFD was first implemented by Steve Chamberlain ([email protected]),
John Gilmore ([email protected]),  K. Richard Pixley ([email protected]) and
David Wallace ([email protected]) at Cygnus Support in Palo Alto,
California.

\1f
File: bfdinfo,  Node: How It Works,  Next: History,  Prev: Porting,  Up: Top

How It Works
============

   To use the library, include `bfd.h' and link with `libbfd.a'.        

   BFD provides a common interface to the parts of an object file for
a calling application.

   When an application sucessfully opens a target file (object,
archive or whatever) a pointer to an internal structure is returned.
This pointer points to a structure called `bfd', described in
`include/bfd.h'.  Our convention is to call this pointer a BFD, and
instances of it within code `abfd'.  All operations on the target
object file are applied as methods to the BFD.  The mapping is defined
within `bfd.h' in a set of macros, all beginning `bfd'_.

   For example, this sequence would do what you would probably expect:
return the number of sections in an object file attached to a BFD
`abfd'.


     #include "bfd.h"
     
     unsigned int number_of_sections(abfd)
     bfd *abfd;
     {
       return bfd_count_sections(abfd);
     }

   lisp

   The abstraction used within BFD is that an object file has a header,
a number of sections containing raw data, a set of relocations, and
some symbol information. Also, BFDs opened for archives have the
additional attribute of an index and contain subordinate BFDs. This
approach is fine for a.out and coff, but loses efficiency when applied
to formats such as S-records and IEEE-695.

What BFD Version 1 Can Do
=========================

   As different information from the the object files is required, BFD
reads from different sections of the file and processes them.  For
example a very common operation for the linker is processing symbol
tables.  Each BFD back end provides a routine for converting between
the object file's representation of symbols and an internal canonical
format. When the linker asks for the symbol table of an object file,
it calls through the memory pointer to the relevant BFD back end
routine which reads and converts the table into a canonical form.  The
linker then operates upon the canonical form. When the link is
finished and the linker writes the output file's symbol table, another
BFD back end routine is called which takes the newly created symbol
table and converts it into the chosen output format.

\1f
File: bfdinfo,  Node: BFD information loss,  Next: Mechanism,  Prev: BFD outline,  Up: BFD

Information Loss
----------------

   *Some information is lost due to the nature of the file format.*
The output targets supported by BFD do not provide identical
facilities, and information which may be described in one form has
nowhere to go in another format. One example of this is alignment
information in `b.out'. There is nowhere in an `a.out' format file to
store alignment information on the contained data, so when a file is
linked from `b.out' and an `a.out' image is produced, alignment
information will not propagate to the output file. (The linker will
still use the alignment information internally, so the link is
performed correctly).

   Another example is COFF section names. COFF files may contain an
unlimited number of sections, each one with a textual section name. If
the target of the link is a format which does not have many sections
(eg `a.out') or has sections without names (eg the Oasys format) the
link cannot be done simply. You can circumvent this problem by
describing the desired input-to-output section mapping with the linker
command language.

   *Information can be lost during canonicalization.* The BFD internal
canonical form of the external formats is not exhaustive; there are
structures in input formats for which there is no direct
representation internally.  This means that the BFD back ends cannot
maintain all possible data richness through the transformation between
external to internal and back to external formats.

   This limitation is only a problem when an application reads one
format and writes another.  Each BFD back end is responsible for
maintaining as much data as possible, and the internal BFD canonical
form has structures which are opaque to the BFD core, and exported
only to the back ends. When a file is read in one format, the
canonical form is generated for BFD and the application. At the same
time, the back end saves away any information which may otherwise be
lost. If the data is then written back in the same format, the back
end routine will be able to use the canonical form provided by the BFD
core as well as the information it prepared earlier.  Since there is a
great deal of commonality between back ends, this mechanism is very
useful. There is no information lost for this reason when linking or
copying big endian COFF to little endian COFF, or `a.out' to `b.out'. 
When a mixture of formats is linked, the information is only lost from
the files whose format differs from the destination.

\1f
File: bfdinfo,  Node: Mechanism,  Prev: BFD information loss,  Up: BFD

Mechanism
---------

   The greatest potential for loss of information is when there is
least overlap between the information provided by the source format,
that stored by the canonical format, and the information needed by the
destination format. A brief description of the canonical form may help
you appreciate what kinds of data you can count on preserving across
conversions.

*files*
     Information on target machine architecture, particular
     implementation and format type are stored on a per-file basis.
     Other information includes a demand pageable bit and a write
     protected bit.  Note that information like Unix magic numbers is
     not stored here--only the magic numbers' meaning, so a `ZMAGIC'
     file would have both the demand pageable bit and the write
     protected text bit set.  The byte order of the target is stored
     on a per-file basis, so that big- and little-endian object files
     may be linked with one another.

*sections*
     Each section in the input file contains the name of the section,
     the original address in the object file, various flags, size and
     alignment information and pointers into other BFD data structures.

*symbols*
     Each symbol contains a pointer to the object file which originally
     defined it, its name, its value, and various flag bits.  When a
     BFD back end reads in a symbol table, the back end relocates all
     symbols to make them relative to the base of the section where
     they were defined.  This ensures that each symbol points to its
     containing section.  Each symbol also has a varying amount of
     hidden data to contain private data for the BFD back end.  Since
     the symbol points to the original file, the private data format
     for that symbol is accessible.  `gld' can operate on a collection
     of symbols of wildly different formats without problems.

     Normal global and simple local symbols are maintained on output,
     so an output file (no matter its format) will retain symbols
     pointing to functions and to global, static, and common
     variables.  Some symbol information is not worth retaining; in
     `a.out' type information is stored in the symbol table as long
     symbol names.  This information would be useless to most COFF
     debuggers; the linker has command line switches to allow users to
     throw it away.

     There is one word of type information within the symbol, so if the
     format supports symbol type information within symbols (for
     example COFF, IEEE, Oasys) and the type is simple enough to fit
     within one word (nearly everything but aggregates) the
     information will be preserved.

*relocation level*
     Each canonical BFD relocation record contains a pointer to the
     symbol to relocate to, the offset of the data to relocate, the
     section the data is in and a pointer to a relocation type
     descriptor. Relocation is performed effectively by message
     passing through the relocation type descriptor and symbol
     pointer. It allows relocations to be performed on output data
     using a relocation method only available in one of the input
     formats. For instance, Oasys provides a byte relocation format. 
     A relocation record requesting this relocation type would point
     indirectly to a routine to perform this, so the relocation may be
     performed on a byte being written to a COFF file, even though 68k
     COFF has no such relocation type.

*line numbers*
     Object formats can contain, for debugging purposes, some form of
     mapping between symbols, source line numbers, and addresses in
     the output file.  These addresses have to be relocated along with
     the symbol information.  Each symbol with an associated list of
     line number records points to the first record of the list.  The
     head of a line number list consists of a pointer to the symbol,
     which allows divination of the address of the function whose line
     number is being described. The rest of the list is made up of
     pairs: offsets into the section and line numbers. Any format
     which can simply derive this information can pass it successfully
     between formats (COFF, IEEE and Oasys).

\1f
File: bfdinfo,  Node: BFD front end,  Next: BFD back end,  Prev: Mechanism,  Up: Top

BFD front end
*************

typedef bfd
===========

   Pointers to bfd structs are the cornerstone of any application using
`libbfd'. References though the BFD and to data in the BFD give the
entire BFD functionality.

   Here is the BFD struct itself.  This contains the major data about
the file, and contains pointers to the rest of the data.

     struct _bfd 
     {

   The filename the application opened the BFD with.

       CONST char *filename;

   A pointer to the target jump table.

       struct bfd_target *xvec;

   To avoid dragging too many header files into every file that
includes `bfd.h', IOSTREAM has been declared as a "char *", and MTIME
as a "long".  Their correct types, to which they are cast when used,
are "FILE *" and "time_t".

   The iostream is the result of an fopen on the filename.

       char *iostream;

   Is the file being cached *Note File Caching::.

       boolean cacheable;

   Marks whether there was a default target specified when the BFD was
opened. This is used to select what matching algorithm to use to chose
the back end.

       boolean target_defaulted;

   The caching routines use these to maintain a least-recently-used
list of BFDs (*note File Caching::.).

       struct _bfd *lru_prev, *lru_next;

   When a file is closed by the caching routines, BFD retains state
information on the file here:

       file_ptr where;

   and here:

       boolean opened_once;

       boolean mtime_set;

   File modified time

       long mtime;

   Reserved for an unimplemented file locking extension.

     int ifd;

   The format which belongs to the BFD.

       bfd_format format;

   The direction the BFD was opened with

       enum bfd_direction {no_direction = 0,
                            read_direction = 1,
                            write_direction = 2,
                            both_direction = 3} direction;

   Format_specific flags

       flagword flags;

   Currently my_archive is tested before adding origin to anything. I
believe that this can become always an add of origin, with origin set
to 0 for non archive files.

       file_ptr origin;

   Remember when output has begun, to stop strange things happening.

       boolean output_has_begun;

   Pointer to linked list of sections

       struct sec  *sections;

   The number of sections

       unsigned int section_count;

   Stuff only useful for object files: The start address.

       bfd_vma start_address;

   Used for input and output

       unsigned int symcount;

   Symbol table for output BFD

       struct symbol_cache_entry  **outsymbols;

   Architecture of object machine, eg m68k

       enum bfd_architecture obj_arch;

   Particular machine within arch, e.g. 68010

       unsigned long obj_machine;

   Stuff only useful for archives:

       PTR arelt_data;              
       struct _bfd *my_archive;     
       struct _bfd *next;           
       struct _bfd *archive_head;   
       boolean has_armap;

   Used by the back end to hold private data.

       PTR tdata;

   Used by the application to hold private data

       PTR usrdata;

   Where all the allocated stuff under this BFD goes (*note Memory
Usage::.).

       struct obstack memory;
     };

`bfd_set_start_address'
.......................

   Marks the entry point of an output BFD. Returns `true' on success,
`false' otherwise.

     boolean bfd_set_start_address(bfd *, bfd_vma);

`bfd_get_mtime'
...............

   Return cached file modification time (e.g. as read from archive
header for archive members, or from file system if we have been called
before); else determine modify time, cache it, and return it.

     long bfd_get_mtime(bfd *);

`stuff'
.......



     #define bfd_sizeof_headers(abfd, reloc) \
          BFD_SEND (abfd, _bfd_sizeof_headers, (abfd, reloc))
     
     #define bfd_find_nearest_line(abfd, section, symbols, offset, filename_ptr, func, line_ptr) \
          BFD_SEND (abfd, _bfd_find_nearest_line,  (abfd, section, symbols, offset, filename_ptr, func, line_ptr))
     
     #define bfd_debug_info_start(abfd) \
             BFD_SEND (abfd, _bfd_debug_info_start, (abfd))
     
     #define bfd_debug_info_end(abfd) \
             BFD_SEND (abfd, _bfd_debug_info_end, (abfd))
     
     #define bfd_debug_info_accumulate(abfd, section) \
             BFD_SEND (abfd, _bfd_debug_info_accumulate, (abfd, section))
     
     #define bfd_stat_arch_elt(abfd, stat) \
             BFD_SEND (abfd, _bfd_stat_arch_elt,(abfd, stat))
     
     #define bfd_coff_swap_aux_in(a,e,t,c,i) \
             BFD_SEND (a, _bfd_coff_swap_aux_in, (a,e,t,c,i))
     
     #define bfd_coff_swap_sym_in(a,e,i) \
             BFD_SEND (a, _bfd_coff_swap_sym_in, (a,e,i))
     
     #define bfd_coff_swap_lineno_in(a,e,i) \
             BFD_SEND ( a, _bfd_coff_swap_lineno_in, (a,e,i))

   lisp

\1f
File: bfdinfo,  Node: Memory Usage,  Next: Sections,  Prev: bfd,  Up: Top

Memory Usage
============

   BFD keeps all its internal structures in obstacks. There is one
obstack per open BFD file, into which the current state is stored.
When a BFD is closed, the obstack is deleted, and so everything which
has been allocated by libbfd for the closing file will be thrown away.

   BFD will not free anything created by an application, but pointers
into `bfd' structures will be invalidated on a `bfd_close'; for
example, after a `bfd_close' the vector passed to
`bfd_canonicalize_symtab' will still be around, since it has been
allocated by the application, but the data that it pointed to will be
lost.

   The general rule is not to close a BFD until all operations
dependent upon data from the BFD have been completed, or all the data
from within the file has been copied. To help with the management of
memory, there is a function (`bfd_alloc_size') which returns the
number of bytes in obstacks associated with the supplied BFD. This
could be used to select the greediest open BFD, close it to reclaim
the memory, perform some operation and reopen the BFD again, to get a
fresh copy of the data structures.

\1f
File: bfdinfo,  Node: Sections,  Next: Symbols,  Prev: Memory Usage,  Up: Top

Sections
========

   Sections are supported in BFD in `section.c'.

   The raw data contained within a BFD is maintained through the
section abstraction.  A single BFD may have any number of sections,
and keeps hold of them by pointing to the first, each one points to
the next in the list.

* Menu:

* Section Input::
* Section Output::
* typedef asection::
* section prototypes::

\1f
File: bfdinfo,  Node: Section Input,  Next: Section Output,  Up: Sections

Section Input
-------------

   When a BFD is opened for reading, the section structures are created
and attached to the BFD.

   Each section has a name which describes the section in the outside
world - for example, `a.out' would contain at least three sections,
called `.text', `.data' and `.bss'.

   Sometimes a BFD will contain more than the 'natural' number of
sections. A back end may attach other sections containing constructor
data, or an application may add a section (using bfd_make_section) to
the sections attached to an already open BFD. For example, the linker
creates a supernumary section `COMMON' for each input file's BFD to
hold information about common storage.

   The raw data is not necessarily read in at the same time as the
section descriptor is created. Some targets may leave the data in
place until a `bfd_get_section_contents' call is made. Other back ends
may read in all the data at once - For example; an S-record file has
to be read once to determine the size of the data. An IEEE-695 file
doesn't contain raw data in sections, but data and relocation
expressions intermixed, so the data area has to be parsed to get out
the data and relocations.

\1f
File: bfdinfo,  Node: Section Output,  Next: typedef asection,  Prev: Section Input,  Up: Sections

Section Output
--------------

   To write a new object style BFD, the various sections to be written
have to be created. They are attached to the BFD in the same way as
input sections, data is written to the sections using
`bfd_set_section_contents'.

   The linker uses the fields `output_section' and `output_offset' to
create an output file.

   The data to be written comes from input sections attached to the
output sections.  The output section structure can be considered a
filter for the input section, the output section determines the vma of
the output data and the name, but the input section determines the
offset into the output section of the data to be written.

   Eg to create a section "O", starting at 0x100, 0x123 long,
containing two subsections, "A" at offset 0x0 (ie at vma 0x100) and
"B" at offset 0x20 (ie at vma 0x120) the structures would look like:



        section name          "A"
          output_offset   0x00
          size            0x20
          output_section ----------->  section name    "O"
                                  |    vma             0x100
        section name          "B" |    size            0x123
          output_offset   0x20    |
          size            0x103   |
          output_section  --------|

   lisp

\1f
File: bfdinfo,  Node: typedef asection,  Next: section prototypes,  Prev: Section Output,  Up: Sections

typedef asection
----------------

   The shape of a section struct:

     typedef struct sec {

   The name of the section, the name isn't a copy, the pointer is the
same as that passed to bfd_make_section.

         CONST char *name;

   The next section in the list belonging to the BFD, or NULL.

         struct sec *next;

   The field flags contains attributes of the section. Some of these
flags are read in from the object file, and some are synthesized from
other information.

     flagword flags;

     #define SEC_NO_FLAGS   0x000

   Tells the OS to allocate space for this section when loaded.  This
would clear for a section containing debug information only.

     #define SEC_ALLOC      0x001

   Tells the OS to load the section from the file when loading.  This
would be clear for a .bss section

     #define SEC_LOAD       0x002

   The section contains data still to be relocated, so there will be
some relocation information too.

     #define SEC_RELOC      0x004

   Obsolete

     #define SEC_BALIGN     0x008

   A signal to the OS that the section contains read only data.

     #define SEC_READONLY   0x010

   The section contains code only.

     #define SEC_CODE       0x020

   The section contains data only.

     #define SEC_DATA        0x040

   The section will reside in ROM.

     #define SEC_ROM        0x080

   The section contains constructor information. This section type is
used by the linker to create lists of constructors and destructors
used by `g++'. When a back end sees a symbol which should be used in a
constructor list, it creates a new section for the type of name (eg
`__CTOR_LIST__'), attaches the symbol to it and builds a relocation.
To build the lists of constructors, all the linker has to to is
catenate all the sections called `__CTOR_LIST__' and relocte the data
contained within - exactly the operations it would peform on standard
data.

     #define SEC_CONSTRUCTOR 0x100

   The section is a constuctor, and should be placed at the end of the
..

     #define SEC_CONSTRUCTOR_TEXT 0x1100

     #define SEC_CONSTRUCTOR_DATA 0x2100

     #define SEC_CONSTRUCTOR_BSS  0x3100

   The section has contents - a bss section could be `SEC_ALLOC' |
`SEC_HAS_CONTENTS', a debug section could be `SEC_HAS_CONTENTS'

     #define SEC_HAS_CONTENTS 0x200

   An instruction to the linker not to output sections containing this
flag even if they have information which would normally be written.

     #define SEC_NEVER_LOAD 0x400

   The base address of the section in the address space of the target.

        bfd_vma vma;

   The size of the section in bytes of the loaded section. This
contains a value even if the section has no contents (eg, the size of
`.bss').

        bfd_size_type size;

   If this section is going to be output, then this value is the
offset into the output section of the first byte in the input section.
Eg, if this was going to start at the 100th byte in the output
section, this value would be 100.

        bfd_vma output_offset;

   The output section through which to map on output.

        struct sec *output_section;

   The alignment requirement of the section, as an exponent - eg 3
aligns to 2^3 (or 8)

        unsigned int alignment_power;

   If an input section, a pointer to a vector of relocation records for
the data in this section.

        struct reloc_cache_entry *relocation;

   If an output section, a pointer to a vector of pointers to
relocation records for the data in this section.

        struct reloc_cache_entry **orelocation;

   The number of relocation records in one of the above

        unsigned reloc_count;

   Which section is it 0..nth

        int index;

   Information below is back end specific - and not always used or
updated

   File position of section data

        file_ptr filepos;

   File position of relocation info

        file_ptr rel_filepos;

   File position of line data

        file_ptr line_filepos;

   Pointer to data for applications

        PTR userdata;

        struct lang_output_section *otheruserdata;

   Attached line number information

        alent *lineno;

   Number of line number records

        unsigned int lineno_count;

   When a section is being output, this value changes as more
linenumbers are written out

        file_ptr moving_line_filepos;

   what the section number is in the target world

        unsigned int target_index;

        PTR used_by_bfd;

   If this is a constructor section then here is a list of the
relocations created to relocate items within it.

        struct relent_chain *constructor_chain;

   The BFD which owns the section.

        bfd *owner;

     } asection ;

\1f
File: bfdinfo,  Node: section prototypes,  Next: Section,  Prev: typedef section,  Up: Sections

section prototypes
------------------

`bfd_get_section_by_name'
.........................

   Runs through the provided ABFD and returns the `asection' who's
name matches that provided, otherwise NULL. *Note Sections::, for more
information.

     asection * bfd_get_section_by_name(bfd *abfd, CONST char *name);

`bfd_make_section'
..................

   This function creates a new empty section called NAME and attaches
it to the end of the chain of sections for the BFD supplied. An
attempt to create a section with a name which is already in use,
returns the old section by that name instead.

   Possible errors are:

`invalid_operation'
     If output has already started for this BFD.

`no_memory'
     If obstack alloc fails.

     asection * bfd_make_section(bfd *, CONST char *name);

`bfd_set_section_flags'
.......................

   Attempts to set the attributes of the section named in the BFD
supplied to the value. Returns true on success, false on error. 
Possible error returns are:

`invalid operation'
     The section cannot have one or more of the attributes requested.
     For example, a .bss section in `a.out' may not have the
     `SEC_HAS_CONTENTS' field set.

     boolean bfd_set_section_flags(bfd *, asection *, flagword);

`bfd_map_over_sections'
.......................

   Calls the provided function FUNC for each section attached to the
BFD ABFD, passing OBJ as an argument. The function will be called as
if by

       func(abfd, the_section, obj);

     void bfd_map_over_sections(bfd *abfd, void (*func)(), PTR obj);

   This is the prefered method for iterating over sections, an
alternative would be to use a loop:

        section *p;
        for (p = abfd->sections; p != NULL; p = p->next)
           func(abfd, p, ...)

`bfd_set_section_size'
......................

   Sets SECTION to the size VAL. If the operation is ok, then `true'
is returned, else `false'.

   Possible error returns:

`invalid_operation'
     Writing has started to the BFD, so setting the size is invalid

     boolean bfd_set_section_size(bfd *, asection *, bfd_size_type val);

`bfd_set_section_contents'
..........................

   Sets the contents of the section SECTION in BFD ABFD to the data
starting in memory at DATA. The data is written to the output section
starting at offset OFFSET for COUNT bytes.

   Normally `true' is returned, else `false'. Possible error returns
are:

`no_contents'
     The output section does not have the `SEC_HAS_CONTENTS'
     attribute, so nothing can be written to it.

`and some more too'
   This routine is front end to the back end function
`_bfd_set_section_contents'.

     boolean bfd_set_section_contents(bfd *abfd,        
              asection *section,
              PTR data,
              file_ptr offset,
              bfd_size_type count);

`bfd_get_section_contents'
..........................

   This function reads data from SECTION in BFD ABFD into memory
starting at LOCATION. The data is read at an offset of OFFSET from the
start of the input section, and is read for COUNT bytes.

   If the contents of a constuctor with the `SEC_CONSTUCTOR' flag set
are requested, then the LOCATION is filled with zeroes.

   If no errors occur, `true' is returned, else `false'.  Possible
errors are:

`unknown yet'
     boolean bfd_get_section_contents(bfd *abfd, asection *section, PTR location,
              file_ptr offset, bfd_size_type count);

\1f
File: bfdinfo,  Node: Symbols,  Next: Archives,  Prev: Sections,  Up: To

Symbols
=======

   BFD trys to maintain as much symbol information as it can when it
moves information from file to file. BFD passes information to
applications though the `asymbol' structure. When the application
requests the symbol table, BFD reads the table in the native form and
translates parts of it into the internal format. To maintain more than
the infomation passed to applications some targets keep some
information 'behind the sceans', in a structure only the particular
back end knows about. For example, the coff back end keeps the
original symbol table structure as well as the canonical structure
when a BFD is read in. On output, the coff back end can reconstruct
the output symbol table so that no information is lost, even
information unique to coff which BFD doesn't know or understand. If a
coff symbol table was read, but was written through an a.out back end,
all the coff specific information would be lost. (.. until BFD 2 :).

   The symbol table of a BFD is not necessarily read in until a
canonicalize request is made. Then the BFD back end fills in a table
provided by the application with pointers to the canonical information.

   To output symbols, the application provides BFD with a table of
pointers to pointers to `asymbol's. This allows applications like the
linker to output a symbol as read, since the 'behind the sceens'
information will be still available.

* Menu:

* Reading Symbols::
* Writing Symbols::
* typedef asymbol::
* symbol handling functions::

\1f
File: bfdinfo,  Node: Reading Symbols,  Next: Writing Symbols,  Prev: Symbols,  Up: Symbols

Reading Symbols
---------------

   There are two stages to reading a symbol table from a BFD;
allocating storage, and the actual reading process. This is an excerpt
from an appliction which reads the symbol table:


       unsigned int storage_needed;
       asymbol **symbol_table;
       unsigned int number_of_symbols;
       unsigned int i;
     
       storage_needed = get_symtab_upper_bound (abfd);
     
       if (storage_needed == 0) {
          return ;
       }
       symbol_table = (asymbol **) malloc (storage_needed);
         ...
       number_of_symbols = 
          bfd_canonicalize_symtab (abfd, symbol_table); 
     
       for (i = 0; i < number_of_symbols; i++) {
          process_symbol (symbol_table[i]);
       }

   lisp

   All storage for the symbols themselves is in an obstack connected to
the BFD, and is freed when the BFD is closed.

\1f
File: bfdinfo,  Node: Writing Symbols,  Next: typedef asymbol,  Prev: Reading Symbols,  Up: Symbols

Writing Symbols
---------------

   Writing of a symbol table is automatic when a BFD open for writing
is closed. The application attaches a vector of pointers to pointers
to symbols to the BFD being written, and fills in the symbol count.
The close and cleanup code reads through the table provided and
performs all the necessary operations. The outputing code must always
be provided with an 'owned' symbol; one which has come from another
BFD, or one which has been created using `bfd_make_empty_symbol'.

   An example showing the creation of a symbol table with only one
element:


     #include "bfd.h"
     main() 
     {
       bfd *abfd;
       asymbol *ptrs[2];
       asymbol *new;
     
       abfd = bfd_openw("foo","a.out-sunos-big");
       bfd_set_format(abfd, bfd_object);
       new = bfd_make_empty_symbol(abfd);
       new->name = "dummy_symbol";
       new->section = (asection *)0;
       new->flags = BSF_ABSOLUTE | BSF_GLOBAL;
       new->value = 0x12345;
     
       ptrs[0] = new;
       ptrs[1] = (asymbol *)0;
     
       bfd_set_symtab(abfd, ptrs, 1);
       bfd_close(abfd);
     }
     
     ./makesym 
     nm foo
     00012345 A dummy_symbol

   lisp

   Many formats cannot represent arbitary symbol information; for
instance the `a.out' object format does not allow an arbitary number
of sections. A symbol pointing to a section which is not one of
`.text', `.data' or `.bss' cannot be described.

\1f
File: bfdinfo,  Node: typedef asymbol,  Next: symbol handling functions,  Prev: Writing Symbols,  Up: Symbols

typedef asymbol
---------------

   An `asymbol' has the form:

     typedef struct symbol_cache_entry 
     {

   A pointer to the BFD which owns the symbol. This information is
necessary so that a back end can work out what additional (invisible to
the application writer) information is carried with the symbol.

       struct _bfd *the_bfd;

   The text of the symbol. The name is left alone, and not copied - the
application may not alter it.

        CONST char *name;

   The value of the symbol.

        symvalue value;

   Attributes of a symbol:

     #define BSF_NO_FLAGS    0x00

   The symbol has local scope; `static' in `C'. The value is the
offset into the section of the data.

     #define BSF_LOCAL  0x01

   The symbol has global scope; initialized data in `C'. The value is
the offset into the section of the data.

     #define BSF_GLOBAL 0x02

   Obsolete

     #define BSF_IMPORT 0x04

   The symbol has global scope, and is exported. The value is the
offset into the section of the data.

     #define BSF_EXPORT 0x08

   The symbol is undefined. `extern' in `C'. The value has no meaning.

     #define BSF_UNDEFINED      0x10    

   The symbol is common, initialized to zero; default in `C'. The
value is the size of the object in bytes.

     #define BSF_FORT_COMM      0x20    

   A normal `C' symbol would be one of: `BSF_LOCAL', `BSF_FORT_COMM', 
`BSF_UNDEFINED' or `BSF_EXPORT|BSD_GLOBAL'

   The symbol is a debugging record. The value has an arbitary meaning.

     #define BSF_DEBUGGING      0x40

   The symbol has no section attached, any value is the actual value
and is not a relative offset to a section.

     #define BSF_ABSOLUTE       0x80

   Used by the linker

     #define BSF_KEEP        0x10000
     #define BSF_KEEP_G      0x80000

   Unused

     #define BSF_WEAK        0x100000
     #define BSF_CTOR        0x200000 
     #define BSF_FAKE        0x400000

   The symbol used to be a common symbol, but now it is allocated.

     #define BSF_OLD_COMMON  0x800000

   The default value for common data.

     #define BFD_FORT_COMM_DEFAULT_VALUE 0

   In some files the type of a symbol sometimes alters its location in
an output file - ie in coff a `ISFCN' symbol which is also `C_EXT'
symbol appears where it was declared and not at the end of a section. 
This bit is set by the target BFD part to convey this information.

     #define BSF_NOT_AT_END    0x40000

   Signal that the symbol is the label of constructor section.

     #define BSF_CONSTRUCTOR   0x1000000

   Signal that the symbol is a warning symbol. If the symbol is a
warning symbol, then the value field (I know this is tacky) will point
to the asymbol which when referenced will cause the warning.

     #define BSF_WARNING       0x2000000

   Signal that the symbol is indirect. The value of the symbol is a
pointer to an undefined asymbol which contains the name to use instead.

     #define BSF_INDIRECT     0x4000000

       flagword flags;

   A pointer to the section to which this symbol is relative, or 0 if
the symbol is absolute or undefined. Note that it is not sufficient to
set this location to 0 to mark a symbol as absolute - the flag
`BSF_ABSOLUTE' must be set also.

       struct sec *section;

   Back end special data. This is being phased out in favour of making
this a union.

       PTR udata;       
     } asymbol;

\1f
File: bfdinfo,  Node: symbol handling functions,  Next: Symbols,  Prev: typedef asymbol,  Up: Symbols

Symbol Handling Functions
-------------------------

`get_symtab_upper_bound'
........................

   Returns the number of bytes required in a vector of pointers to
`asymbols' for all the symbols in the supplied BFD, including a
terminal NULL pointer. If there are no symbols in the BFD, then 0 is
returned.


     #define get_symtab_upper_bound(abfd) \
          BFD_SEND (abfd, _get_symtab_upper_bound, (abfd))

   lisp

`bfd_canonicalize_symtab'
.........................

   Supplied a BFD and a pointer to an uninitialized vector of pointers. 
This reads in the symbols from the BFD, and fills in the table with
pointers to the symbols, and a trailing NULL. The routine returns the
actual number of symbol pointers not including the NULL.


     #define bfd_canonicalize_symtab(abfd, location) \
          BFD_SEND (abfd, _bfd_canonicalize_symtab,\
                       (abfd, location))

   lisp

`bfd_set_symtab'
................

   Provided a table of pointers to to symbols and a count, writes to
the output BFD the symbols when closed.

     boolean bfd_set_symtab(bfd *, asymbol **, unsigned int );

`bfd_print_symbol_vandf'
........................

   Prints the value and flags of the symbol supplied to the stream
file.

     void bfd_print_symbol_vandf(PTR file, asymbol *symbol);

`bfd_make_empty_symbol'
.......................

   This function creates a new `asymbol' structure for the BFD, and
returns a pointer to it.

   This routine is necessary, since each back end has private
information surrounding the `asymbol'. Building your own `asymbol' and
pointing to it will not create the private information, and will cause
problems later on.


     #define bfd_make_empty_symbol(abfd) \
          BFD_SEND (abfd, _bfd_make_empty_symbol, (abfd))

   lisp

\1f
File: bfdinfo,  Node: Archives,  Next: Formats,  Prev: Symbols,  Up: Top

Archives
========

   Gumby, you promised to write this bit...

   Archives are supported in BFD in `archive.c'.

   An archive is represented internally just like another BFD, with a
pointer to a chain of contained BFDs. Archives can be created by
opening BFDs, linking them together and attaching them as children to
another BFD and then closing the parent BFD.

`bfd_get_next_mapent'
.....................

   What this does

     symindex bfd_get_next_mapent(bfd *, symindex, carsym **);

`bfd_set_archive_head'
......................

   Used whilst processing archives. Sets the head of the chain of BFDs
contained in an archive to NEW_HEAD. (see chapter on archives)

     boolean bfd_set_archive_head(bfd *output, bfd *new_head);

`bfd_get_elt_at_index'
......................

   Return the sub bfd contained within the archive at archive index n.

     bfd * bfd_get_elt_at_index(bfd *, int);

`bfd_openr_next_archived_file'
..............................

   Initially provided a BFD containing an archive and NULL, opens a BFD
on the first contained element and returns that. Subsequent calls to
bfd_openr_next_archived_file should pass the archive and the previous
return value to return a created BFD to the next contained element. 
NULL is returned when there are no more.

     bfd* bfd_openr_next_archived_file(bfd *archive, bfd *previous);

\1f
File: bfdinfo,  Node: Formats,  Next: Relocations,  Prev: Archives,  Up: Top

File Formats
============

   A format is a BFD concept of high level file contents. The formats
supported by BFD are:

`bfd_object'
     The BFD may contain data, symbols, relocations and debug info.

`bfd_archive'
     The