/* Support for the generic parts of most COFF variants, for BFD. Copyright 1990, 1991, 1992, 1993, 1994 Free Software Foundation, Inc. Written by Cygnus Support. This file is part of BFD, the Binary File Descriptor library. 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., 675 Mass Ave, Cambridge, MA 02139, USA. */ /* Most of this hacked by Steve Chamberlain, sac@cygnus.com */ /* SECTION coff backends BFD supports a number of different flavours of coff format. The major differences between formats are the sizes and alignments of fields in structures on disk, and the occasional extra field. Coff in all its varieties is implemented with a few common files and a number of implementation specific files. For example, The 88k bcs coff format is implemented in the file @file{coff-m88k.c}. This file @code{#include}s @file{coff/m88k.h} which defines the external structure of the coff format for the 88k, and @file{coff/internal.h} which defines the internal structure. @file{coff-m88k.c} also defines the relocations used by the 88k format @xref{Relocations}. The Intel i960 processor version of coff is implemented in @file{coff-i960.c}. This file has the same structure as @file{coff-m88k.c}, except that it includes @file{coff/i960.h} rather than @file{coff-m88k.h}. SUBSECTION Porting to a new version of coff The recommended method is to select from the existing implementations the version of coff which is most like the one you want to use. For example, we'll say that i386 coff is the one you select, and that your coff flavour is called foo. Copy @file{i386coff.c} to @file{foocoff.c}, copy @file{../include/coff/i386.h} to @file{../include/coff/foo.h}, and add the lines to @file{targets.c} and @file{Makefile.in} so that your new back end is used. Alter the shapes of the structures in @file{../include/coff/foo.h} so that they match what you need. You will probably also have to add @code{#ifdef}s to the code in @file{coff/internal.h} and @file{coffcode.h} if your version of coff is too wild. You can verify that your new BFD backend works quite simply by building @file{objdump} from the @file{binutils} directory, and making sure that its version of what's going on and your host system's idea (assuming it has the pretty standard coff dump utility, usually called @code{att-dump} or just @code{dump}) are the same. Then clean up your code, and send what you've done to Cygnus. Then your stuff will be in the next release, and you won't have to keep integrating it. SUBSECTION How the coff backend works SUBSUBSECTION File layout The Coff backend is split into generic routines that are applicable to any Coff target and routines that are specific to a particular target. The target-specific routines are further split into ones which are basically the same for all Coff targets except that they use the external symbol format or use different values for certain constants. The generic routines are in @file{coffgen.c}. These routines work for any Coff target. They use some hooks into the target specific code; the hooks are in a @code{bfd_coff_backend_data} structure, one of which exists for each target. The essentially similar target-specific routines are in @file{coffcode.h}. This header file includes executable C code. The various Coff targets first include the appropriate Coff header file, make any special defines that are needed, and then include @file{coffcode.h}. Some of the Coff targets then also have additional routines in the target source file itself. For example, @file{coff-i960.c} includes @file{coff/internal.h} and @file{coff/i960.h}. It then defines a few constants, such as @code{I960}, and includes @file{coffcode.h}. Since the i960 has complex relocation types, @file{coff-i960.c} also includes some code to manipulate the i960 relocs. This code is not in @file{coffcode.h} because it would not be used by any other target. SUBSUBSECTION Bit twiddling Each flavour of coff supported in BFD has its own header file describing the external layout of the structures. There is also an internal description of the coff layout, in @file{coff/internal.h}. A major function of the coff backend is swapping the bytes and twiddling the bits to translate the external form of the structures into the normal internal form. This is all performed in the @code{bfd_swap}_@i{thing}_@i{direction} routines. Some elements are different sizes between different versions of coff; it is the duty of the coff version specific include file to override the definitions of various packing routines in @file{coffcode.h}. E.g., the size of line number entry in coff is sometimes 16 bits, and sometimes 32 bits. @code{#define}ing @code{PUT_LNSZ_LNNO} and @code{GET_LNSZ_LNNO} will select the correct one. No doubt, some day someone will find a version of coff which has a varying field size not catered to at the moment. To port BFD, that person will have to add more @code{#defines}. Three of the bit twiddling routines are exported to @code{gdb}; @code{coff_swap_aux_in}, @code{coff_swap_sym_in} and @code{coff_swap_linno_in}. @code{GDB} reads the symbol table on its own, but uses BFD to fix things up. More of the bit twiddlers are exported for @code{gas}; @code{coff_swap_aux_out}, @code{coff_swap_sym_out}, @code{coff_swap_lineno_out}, @code{coff_swap_reloc_out}, @code{coff_swap_filehdr_out}, @code{coff_swap_aouthdr_out}, @code{coff_swap_scnhdr_out}. @code{Gas} currently keeps track of all the symbol table and reloc drudgery itself, thereby saving the internal BFD overhead, but uses BFD to swap things on the way out, making cross ports much safer. Doing so also allows BFD (and thus the linker) to use the same header files as @code{gas}, which makes one avenue to disaster disappear. SUBSUBSECTION Symbol reading The simple canonical form for symbols used by BFD is not rich enough to keep all the information available in a coff symbol table. The back end gets around this problem by keeping the original symbol table around, "behind the scenes". When a symbol table is requested (through a call to @code{bfd_canonicalize_symtab}), a request gets through to @code{coff_get_normalized_symtab}. This reads the symbol table from the coff file and swaps all the structures inside into the internal form. It also fixes up all the pointers in the table (represented in the file by offsets from the first symbol in the table) into physical pointers to elements in the new internal table. This involves some work since the meanings of fields change depending upon context: a field that is a pointer to another structure in the symbol table at one moment may be the size in bytes of a structure at the next. Another pass is made over the table. All symbols which mark file names (<> symbols) are modified so that the internal string points to the value in the auxent (the real filename) rather than the normal text associated with the symbol (@code{".file"}). At this time the symbol names are moved around. Coff stores all symbols less than nine characters long physically within the symbol table; longer strings are kept at the end of the file in the string table. This pass moves all strings into memory and replaces them with pointers to the strings. The symbol table is massaged once again, this time to create the canonical table used by the BFD application. Each symbol is inspected in turn, and a decision made (using the @code{sclass} field) about the various flags to set in the @code{asymbol}. @xref{Symbols}. The generated canonical table shares strings with the hidden internal symbol table. Any linenumbers are read from the coff file too, and attached to the symbols which own the functions the linenumbers belong to. SUBSUBSECTION Symbol writing Writing a symbol to a coff file which didn't come from a coff file will lose any debugging information. The @code{asymbol} structure remembers the BFD from which the symbol was taken, and on output the back end makes sure that the same destination target as source target is present. When the symbols have come from a coff file then all the debugging information is preserved. Symbol tables are provided for writing to the back end in a vector of pointers to pointers. This allows applications like the linker to accumulate and output large symbol tables without having to do too much byte copying. This function runs through the provided symbol table and patches each symbol marked as a file place holder (@code{C_FILE}) to point to the next file place holder in the list. It also marks each @code{offset} field in the list with the offset from the first symbol of the current symbol. Another function of this procedure is to turn the canonical value form of BFD into the form used by coff. Internally, BFD expects symbol values to be offsets from a section base; so a symbol physically at 0x120, but in a section starting at 0x100, would have the value 0x20. Coff expects symbols to contain their final value, so symbols have their values changed at this point to reflect their sum with their owning section. This transformation uses the <> field of the @code{asymbol}'s @code{asection} @xref{Sections}. o <> This routine runs though the provided symbol table and uses the offsets generated by the previous pass and the pointers generated when the symbol table was read in to create the structured hierachy required by coff. It changes each pointer to a symbol into the index into the symbol table of the asymbol. o <> This routine runs through the symbol table and patches up the symbols from their internal form into the coff way, calls the bit twiddlers, and writes out the table to the file. */ /* INTERNAL_DEFINITION coff_symbol_type DESCRIPTION The hidden information for an <> is described in a <>: CODE_FRAGMENT . .typedef struct coff_ptr_struct .{ . . {* Remembers the offset from the first symbol in the file for . this symbol. Generated by coff_renumber_symbols. *} .unsigned int offset; . . {* Should the value of this symbol be renumbered. Used for . XCOFF C_BSTAT symbols. Set by coff_slurp_symbol_table. *} .unsigned int fix_value : 1; . . {* Should the tag field of this symbol be renumbered. . Created by coff_pointerize_aux. *} .unsigned int fix_tag : 1; . . {* Should the endidx field of this symbol be renumbered. . Created by coff_pointerize_aux. *} .unsigned int fix_end : 1; . . {* Should the x_csect.x_scnlen field be renumbered. . Created by coff_slurp_symbol_table. *} .unsigned int fix_scnlen : 1; . . {* The container for the symbol structure as read and translated . from the file. *} . .union { . union internal_auxent auxent; . struct internal_syment syment; . } u; .} combined_entry_type; . . .{* Each canonical asymbol really looks like this: *} . .typedef struct coff_symbol_struct .{ . {* The actual symbol which the rest of BFD works with *} .asymbol symbol; . . {* A pointer to the hidden information for this symbol *} .combined_entry_type *native; . . {* A pointer to the linenumber information for this symbol *} .struct lineno_cache_entry *lineno; . . {* Have the line numbers been relocated yet ? *} .boolean done_lineno; .} coff_symbol_type; */ #include "coffswap.h" /* void warning(); */ /* * Return a word with STYP_* (scnhdr.s_flags) flags set to represent the * incoming SEC_* flags. The inverse of this function is styp_to_sec_flags(). * NOTE: If you add to/change this routine, you should mirror the changes * in styp_to_sec_flags(). */ static long sec_to_styp_flags (sec_name, sec_flags) CONST char *sec_name; flagword sec_flags; { long styp_flags = 0; if (!strcmp (sec_name, _TEXT)) { styp_flags = STYP_TEXT; } else if (!strcmp (sec_name, _DATA)) { styp_flags = STYP_DATA; #ifdef TWO_DATA_SECS } else if (!strcmp (sec_name, ".data2")) { styp_flags = STYP_DATA; #endif /* TWO_DATA_SECS */ } else if (!strcmp (sec_name, _BSS)) { styp_flags = STYP_BSS; #ifdef _COMMENT } else if (!strcmp (sec_name, _COMMENT)) { styp_flags = STYP_INFO; #endif /* _COMMENT */ #ifdef _LIB } else if (!strcmp (sec_name, _LIB)) { styp_flags = STYP_LIB; #endif /* _LIB */ #ifdef _LIT } else if (!strcmp (sec_name, _LIT)) { styp_flags = STYP_LIT; #endif /* _LIT */ } else if (!strcmp (sec_name, ".debug")) { #ifdef STYP_DEBUG styp_flags = STYP_DEBUG; #else styp_flags = STYP_INFO; #endif } else if (!strcmp (sec_name, ".stab") || !strncmp (sec_name, ".stabstr", 8)) { styp_flags = STYP_INFO; } /* Try and figure out what it should be */ else if (sec_flags & SEC_CODE) { styp_flags = STYP_TEXT; } else if (sec_flags & SEC_DATA) { styp_flags = STYP_DATA; } else if (sec_flags & SEC_READONLY) { #ifdef STYP_LIT /* 29k readonly text/data section */ styp_flags = STYP_LIT; #else styp_flags = STYP_TEXT; #endif /* STYP_LIT */ } else if (sec_flags & SEC_LOAD) { styp_flags = STYP_TEXT; } else if (sec_flags & SEC_ALLOC) { styp_flags = STYP_BSS; } #ifdef STYP_NOLOAD if ((sec_flags & (SEC_NEVER_LOAD | SEC_COFF_SHARED_LIBRARY)) != 0) styp_flags |= STYP_NOLOAD; #endif return (styp_flags); } /* * Return a word with SEC_* flags set to represent the incoming * STYP_* flags (from scnhdr.s_flags). The inverse of this * function is sec_to_styp_flags(). * NOTE: If you add to/change this routine, you should mirror the changes * in sec_to_styp_flags(). */ static flagword styp_to_sec_flags (abfd, hdr) bfd * abfd; PTR hdr; { struct internal_scnhdr *internal_s = (struct internal_scnhdr *) hdr; long styp_flags = internal_s->s_flags; flagword sec_flags = 0; #ifdef STYP_NOLOAD if (styp_flags & STYP_NOLOAD) { sec_flags |= SEC_NEVER_LOAD; } #endif /* STYP_NOLOAD */ /* For 386 COFF, at least, an unloadable text or data section is actually a shared library section. */ if (styp_flags & STYP_TEXT) { if (sec_flags & SEC_NEVER_LOAD) sec_flags |= SEC_CODE | SEC_COFF_SHARED_LIBRARY; else sec_flags |= SEC_CODE | SEC_LOAD | SEC_ALLOC; } else if (styp_flags & STYP_DATA) { if (sec_flags & SEC_NEVER_LOAD) sec_flags |= SEC_DATA | SEC_COFF_SHARED_LIBRARY; else sec_flags |= SEC_DATA | SEC_LOAD | SEC_ALLOC; } else if (styp_flags & STYP_BSS) { #ifdef BSS_NOLOAD_IS_SHARED_LIBRARY if (sec_flags & SEC_NEVER_LOAD) sec_flags |= SEC_ALLOC | SEC_COFF_SHARED_LIBRARY; else #endif sec_flags |= SEC_ALLOC; } else if (styp_flags & STYP_INFO) { /* This should be marked as SEC_DEBUGGING, but that can't be done until we make sure that strip can still work. strip will probably have to preserve the same number of sections to ensure that the section vma matches the section file position. */ } else { sec_flags |= SEC_ALLOC | SEC_LOAD; } #ifdef STYP_LIT /* A29k readonly text/data section type */ if ((styp_flags & STYP_LIT) == STYP_LIT) { sec_flags = (SEC_LOAD | SEC_ALLOC | SEC_READONLY); } #endif /* STYP_LIT */ #ifdef STYP_OTHER_LOAD /* Other loaded sections */ if (styp_flags & STYP_OTHER_LOAD) { sec_flags = (SEC_LOAD | SEC_ALLOC); } #endif /* STYP_SDATA */ return (sec_flags); } #define get_index(symbol) ((long) (symbol)->udata) /* INTERNAL_DEFINITION bfd_coff_backend_data CODE_FRAGMENT Special entry points for gdb to swap in coff symbol table parts: .typedef struct .{ . void (*_bfd_coff_swap_aux_in) PARAMS (( . bfd *abfd, . PTR ext, . int type, . int class, . int indaux, . int numaux, . PTR in)); . . void (*_bfd_coff_swap_sym_in) PARAMS (( . bfd *abfd , . PTR ext, . PTR in)); . . void (*_bfd_coff_swap_lineno_in) PARAMS (( . bfd *abfd, . PTR ext, . PTR in)); . Special entry points for gas to swap out coff parts: . unsigned int (*_bfd_coff_swap_aux_out) PARAMS (( . bfd *abfd, . PTR in, . int type, . int class, . int indaux, . int numaux, . PTR ext)); . . unsigned int (*_bfd_coff_swap_sym_out) PARAMS (( . bfd *abfd, . PTR in, . PTR ext)); . . unsigned int (*_bfd_coff_swap_lineno_out) PARAMS (( . bfd *abfd, . PTR in, . PTR ext)); . . unsigned int (*_bfd_coff_swap_reloc_out) PARAMS (( . bfd *abfd, . PTR src, . PTR dst)); . . unsigned int (*_bfd_coff_swap_filehdr_out) PARAMS (( . bfd *abfd, . PTR in, . PTR out)); . . unsigned int (*_bfd_coff_swap_aouthdr_out) PARAMS (( . bfd *abfd, . PTR in, . PTR out)); . . unsigned int (*_bfd_coff_swap_scnhdr_out) PARAMS (( . bfd *abfd, . PTR in, . PTR out)); . Special entry points for generic COFF routines to call target dependent COFF routines: . unsigned int _bfd_filhsz; . unsigned int _bfd_aoutsz; . unsigned int _bfd_scnhsz; . unsigned int _bfd_symesz; . unsigned int _bfd_auxesz; . unsigned int _bfd_linesz; . boolean _bfd_coff_long_filenames; . void (*_bfd_coff_swap_filehdr_in) PARAMS (( . bfd *abfd, . PTR ext, . PTR in)); . void (*_bfd_coff_swap_aouthdr_in) PARAMS (( . bfd *abfd, . PTR ext, . PTR in)); . void (*_bfd_coff_swap_scnhdr_in) PARAMS (( . bfd *abfd, . PTR ext, . PTR in)); . boolean (*_bfd_coff_bad_format_hook) PARAMS (( . bfd *abfd, . PTR internal_filehdr)); . boolean (*_bfd_coff_set_arch_mach_hook) PARAMS (( . bfd *abfd, . PTR internal_filehdr)); . PTR (*_bfd_coff_mkobject_hook) PARAMS (( . bfd *abfd, . PTR internal_filehdr, . PTR internal_aouthdr)); . flagword (*_bfd_styp_to_sec_flags_hook) PARAMS (( . bfd *abfd, . PTR internal_scnhdr)); . asection *(*_bfd_make_section_hook) PARAMS (( . bfd *abfd, . char *name)); . void (*_bfd_set_alignment_hook) PARAMS (( . bfd *abfd, . asection *sec, . PTR internal_scnhdr)); . boolean (*_bfd_coff_slurp_symbol_table) PARAMS (( . bfd *abfd)); . boolean (*_bfd_coff_symname_in_debug) PARAMS (( . bfd *abfd, . struct internal_syment *sym)); . void (*_bfd_coff_reloc16_extra_cases) PARAMS (( . bfd *abfd, . struct bfd_link_info *link_info, . struct bfd_link_order *link_order, . arelent *reloc, . bfd_byte *data, . unsigned int *src_ptr, . unsigned int *dst_ptr)); . int (*_bfd_coff_reloc16_estimate) PARAMS (( . bfd *abfd, . asection *input_section, . arelent *r, . unsigned int shrink, . struct bfd_link_info *link_info)); . .} bfd_coff_backend_data; . .#define coff_backend_info(abfd) ((bfd_coff_backend_data *) (abfd)->xvec->backend_data) . .#define bfd_coff_swap_aux_in(a,e,t,c,ind,num,i) \ . ((coff_backend_info (a)->_bfd_coff_swap_aux_in) (a,e,t,c,ind,num,i)) . .#define bfd_coff_swap_sym_in(a,e,i) \ . ((coff_backend_info (a)->_bfd_coff_swap_sym_in) (a,e,i)) . .#define bfd_coff_swap_lineno_in(a,e,i) \ . ((coff_backend_info ( a)->_bfd_coff_swap_lineno_in) (a,e,i)) . .#define bfd_coff_swap_reloc_out(abfd, i, o) \ . ((coff_backend_info (abfd)->_bfd_coff_swap_reloc_out) (abfd, i, o)) . .#define bfd_coff_swap_lineno_out(abfd, i, o) \ . ((coff_backend_info (abfd)->_bfd_coff_swap_lineno_out) (abfd, i, o)) . .#define bfd_coff_swap_aux_out(a,i,t,c,ind,num,o) \ . ((coff_backend_info (a)->_bfd_coff_swap_aux_out) (a,i,t,c,ind,num,o)) . .#define bfd_coff_swap_sym_out(abfd, i,o) \ . ((coff_backend_info (abfd)->_bfd_coff_swap_sym_out) (abfd, i, o)) . .#define bfd_coff_swap_scnhdr_out(abfd, i,o) \ . ((coff_backend_info (abfd)->_bfd_coff_swap_scnhdr_out) (abfd, i, o)) . .#define bfd_coff_swap_filehdr_out(abfd, i,o) \ . ((coff_backend_info (abfd)->_bfd_coff_swap_filehdr_out) (abfd, i, o)) . .#define bfd_coff_swap_aouthdr_out(abfd, i,o) \ . ((coff_backend_info (abfd)->_bfd_coff_swap_aouthdr_out) (abfd, i, o)) . .#define bfd_coff_filhsz(abfd) (coff_backend_info (abfd)->_bfd_filhsz) .#define bfd_coff_aoutsz(abfd) (coff_backend_info (abfd)->_bfd_aoutsz) .#define bfd_coff_scnhsz(abfd) (coff_backend_info (abfd)->_bfd_scnhsz) .#define bfd_coff_symesz(abfd) (coff_backend_info (abfd)->_bfd_symesz) .#define bfd_coff_auxesz(abfd) (coff_backend_info (abfd)->_bfd_auxesz) .#define bfd_coff_linesz(abfd) (coff_backend_info (abfd)->_bfd_linesz) .#define bfd_coff_long_filenames(abfd) (coff_backend_info (abfd)->_bfd_coff_long_filenames) .#define bfd_coff_swap_filehdr_in(abfd, i,o) \ . ((coff_backend_info (abfd)->_bfd_coff_swap_filehdr_in) (abfd, i, o)) . .#define bfd_coff_swap_aouthdr_in(abfd, i,o) \ . ((coff_backend_info (abfd)->_bfd_coff_swap_aouthdr_in) (abfd, i, o)) . .#define bfd_coff_swap_scnhdr_in(abfd, i,o) \ . ((coff_backend_info (abfd)->_bfd_coff_swap_scnhdr_in) (abfd, i, o)) . .#define bfd_coff_bad_format_hook(abfd, filehdr) \ . ((coff_backend_info (abfd)->_bfd_coff_bad_format_hook) (abfd, filehdr)) . .#define bfd_coff_set_arch_mach_hook(abfd, filehdr)\ . ((coff_backend_info (abfd)->_bfd_coff_set_arch_mach_hook) (abfd, filehdr)) .#define bfd_coff_mkobject_hook(abfd, filehdr, aouthdr)\ . ((coff_backend_info (abfd)->_bfd_coff_mkobject_hook) (abfd, filehdr, aouthdr)) . .#define bfd_coff_styp_to_sec_flags_hook(abfd, scnhdr)\ . ((coff_backend_info (abfd)->_bfd_styp_to_sec_flags_hook) (abfd, scnhdr)) . .#define bfd_coff_make_section_hook(abfd, name)\ . ((coff_backend_info (abfd)->_bfd_make_section_hook) (abfd, name)) . .#define bfd_coff_set_alignment_hook(abfd, sec, scnhdr)\ . ((coff_backend_info (abfd)->_bfd_set_alignment_hook) (abfd, sec, scnhdr)) . .#define bfd_coff_slurp_symbol_table(abfd)\ . ((coff_backend_info (abfd)->_bfd_coff_slurp_symbol_table) (abfd)) . .#define bfd_coff_symname_in_debug(abfd, sym)\ . ((coff_backend_info (abfd)->_bfd_coff_symname_in_debug) (abfd, sym)) . .#define bfd_coff_reloc16_extra_cases(abfd, link_info, link_order, reloc, data, src_ptr, dst_ptr)\ . ((coff_backend_info (abfd)->_bfd_coff_reloc16_extra_cases)\ . (abfd, link_info, link_order, reloc, data, src_ptr, dst_ptr)) . .#define bfd_coff_reloc16_estimate(abfd, section, reloc, shrink, link_info)\ . ((coff_backend_info (abfd)->_bfd_coff_reloc16_estimate)\ . (abfd, section, reloc, shrink, link_info)) . */ /* See whether the magic number matches. */ static boolean coff_bad_format_hook (abfd, filehdr) bfd * abfd; PTR filehdr; { struct internal_filehdr *internal_f = (struct internal_filehdr *) filehdr; if (BADMAG (*internal_f)) return false; /* if the optional header is NULL or not the correct size then quit; the only difference I can see between m88k dgux headers (MC88DMAGIC) and Intel 960 readwrite headers (I960WRMAGIC) is that the optional header is of a different size. But the mips keeps extra stuff in it's opthdr, so dont check when doing that */ #if defined(M88) || defined(I960) if (internal_f->f_opthdr != 0 && AOUTSZ != internal_f->f_opthdr) return false; #endif return true; } static asection * coff_make_section_hook (abfd, name) bfd * abfd; char *name; { #ifdef TWO_DATA_SECS /* FIXME: This predates the call to bfd_make_section_anyway in make_a_section_from_file, and can probably go away. */ /* On SCO a file created by the Microsoft assembler can have two .data sections. We use .data2 for the second one. */ if (strcmp (name, _DATA) == 0) return bfd_make_section (abfd, ".data2"); #endif return (asection *) NULL; } /* initialize a section structure with information peculiar to this particular implementation of coff */ static boolean coff_new_section_hook (abfd, section) bfd * abfd; asection * section; { section->alignment_power = abfd->xvec->align_power_min; /* Allocate aux records for section symbols, to store size and related info. @@ Shouldn't use constant multiplier here! */ coffsymbol (section->symbol)->native = (combined_entry_type *) bfd_zalloc (abfd, sizeof (combined_entry_type) * 10); #ifdef COFF_SPARC /* This is to allow double-word operations on addresses in data or bss. */ if (strcmp (section->name, ".data") == 0 || strcmp (section->name, ".bss") == 0) section->alignment_power = 3; #endif /* COFF_SPARC */ return true; } #ifdef I960 /* Set the alignment of a BFD section. */ static void coff_set_alignment_hook (abfd, section, scnhdr) bfd * abfd; asection * section; PTR scnhdr; { struct internal_scnhdr *hdr = (struct internal_scnhdr *) scnhdr; unsigned int i; for (i = 0; i < 32; i++) if ((1 << i) >= hdr->s_align) break; section->alignment_power = i; } #else /* ! I960 */ #define coff_set_alignment_hook \ ((void (*) PARAMS ((bfd *, asection *, PTR))) bfd_void) #endif /* ! I960 */ static boolean coff_mkobject (abfd) bfd * abfd; { coff_data_type *coff; abfd->tdata.coff_obj_data = (struct coff_tdata *) bfd_zalloc (abfd, sizeof (coff_data_type)); if (abfd->tdata.coff_obj_data == 0) { bfd_set_error (bfd_error_no_memory); return false; } coff = coff_data (abfd); coff->symbols = (coff_symbol_type *) NULL; coff->conversion_table = (unsigned int *) NULL; coff->raw_syments = (struct coff_ptr_struct *) NULL; coff->raw_linenos = (struct lineno *) NULL; coff->relocbase = 0; /* make_abs_section(abfd);*/ return true; } /* Create the COFF backend specific information. */ static PTR coff_mkobject_hook (abfd, filehdr, aouthdr) bfd * abfd; PTR filehdr; PTR aouthdr; { struct internal_filehdr *internal_f = (struct internal_filehdr *) filehdr; coff_data_type *coff; if (coff_mkobject (abfd) == false) return NULL; coff = coff_data (abfd); coff->sym_filepos = internal_f->f_symptr; coff->flags = internal_f->f_flags; /* These members communicate important constants about the symbol table to GDB's symbol-reading code. These `constants' unfortunately vary among coff implementations... */ coff->local_n_btmask = N_BTMASK; coff->local_n_btshft = N_BTSHFT; coff->local_n_tmask = N_TMASK; coff->local_n_tshift = N_TSHIFT; coff->local_symesz = SYMESZ; coff->local_auxesz = AUXESZ; coff->local_linesz = LINESZ; return (PTR) coff; } /* Determine the machine architecture and type. FIXME: This is target dependent because the magic numbers are defined in the target dependent header files. But there is no particular need for this. If the magic numbers were moved to a separate file, this function would be target independent and would also be much more successful at linking together COFF files for different architectures. */ static boolean coff_set_arch_mach_hook (abfd, filehdr) bfd *abfd; PTR filehdr; { long machine; enum bfd_architecture arch; struct internal_filehdr *internal_f = (struct internal_filehdr *) filehdr; machine = 0; switch (internal_f->f_magic) { #ifdef I386MAGIC case I386MAGIC: case I386PTXMAGIC: case I386AIXMAGIC: /* Danbury PS/2 AIX C Compiler */ case LYNXCOFFMAGIC: /* shadows the m68k Lynx number below, sigh */ arch = bfd_arch_i386; machine = 0; break; #endif #ifdef A29K_MAGIC_BIG case A29K_MAGIC_BIG: case A29K_MAGIC_LITTLE: arch = bfd_arch_a29k; machine = 0; break; #endif #ifdef MC68MAGIC case MC68MAGIC: case M68MAGIC: #ifdef MC68KBCSMAGIC case MC68KBCSMAGIC: #endif #ifdef APOLLOM68KMAGIC case APOLLOM68KMAGIC: #endif #ifdef LYNXCOFFMAGIC case LYNXCOFFMAGIC: #endif arch = bfd_arch_m68k; machine = 68020; break; #endif #ifdef MC88MAGIC case MC88MAGIC: case MC88DMAGIC: case MC88OMAGIC: arch = bfd_arch_m88k; machine = 88100; break; #endif #ifdef Z8KMAGIC case Z8KMAGIC: arch = bfd_arch_z8k; switch (internal_f->f_flags & F_MACHMASK) { case F_Z8001: machine = bfd_mach_z8001; break; case F_Z8002: machine = bfd_mach_z8002; break; default: return false; } break; #endif #ifdef I960 #ifdef I960ROMAGIC case I960ROMAGIC: case I960RWMAGIC: arch = bfd_arch_i960; switch (F_I960TYPE & internal_f->f_flags) { default: case F_I960CORE: machine = bfd_mach_i960_core; break; case F_I960KB: machine = bfd_mach_i960_kb_sb; break; case F_I960MC: machine = bfd_mach_i960_mc; break; case F_I960XA: machine = bfd_mach_i960_xa; break; case F_I960CA: machine = bfd_mach_i960_ca; break; case F_I960KA: machine = bfd_mach_i960_ka_sa; break; } break; #endif #endif #ifdef U802ROMAGIC case U802ROMAGIC: case U802WRMAGIC: case U802TOCMAGIC: arch = bfd_arch_rs6000; machine = 6000; break; #endif #ifdef WE32KMAGIC case WE32KMAGIC: arch = bfd_arch_we32k; machine = 0; break; #endif #ifdef H8300MAGIC case H8300MAGIC: arch = bfd_arch_h8300; machine = bfd_mach_h8300; /* !! FIXME this probably isn't the right place for this */ abfd->flags |= BFD_IS_RELAXABLE; break; #endif #ifdef H8300HMAGIC case H8300HMAGIC: arch = bfd_arch_h8300; machine = bfd_mach_h8300h; /* !! FIXME this probably isn't the right place for this */ abfd->flags |= BFD_IS_RELAXABLE; break; #endif #ifdef SH_ARCH_MAGIC case SH_ARCH_MAGIC: arch = bfd_arch_sh; machine = 0; break; #endif #ifdef H8500MAGIC case H8500MAGIC: arch = bfd_arch_h8500; machine = 0; break; #endif #ifdef SPARCMAGIC case SPARCMAGIC: #ifdef LYNXCOFFMAGIC case LYNXCOFFMAGIC: #endif arch = bfd_arch_sparc; machine = 0; break; #endif default: /* Unreadable input file type */ arch = bfd_arch_obscure; break; } bfd_default_set_arch_mach (abfd, arch, machine); return true; } #ifdef SYMNAME_IN_DEBUG static boolean symname_in_debug_hook (abfd, sym) bfd * abfd; struct internal_syment *sym; { return SYMNAME_IN_DEBUG (sym) ? true : false; } #else #define symname_in_debug_hook \ (boolean (*) PARAMS ((bfd *, struct internal_syment *))) bfd_false #endif /* SUBSUBSECTION Writing relocations To write relocations, the back end steps though the canonical relocation table and create an @code{internal_reloc}. The symbol index to use is removed from the @code{offset} field in the symbol table supplied. The address comes directly from the sum of the section base address and the relocation offset; the type is dug directly from the howto field. Then the @code{internal_reloc} is swapped into the shape of an @code{external_reloc} and written out to disk. */ static boolean coff_write_relocs (abfd) bfd * abfd; { asection *s; for (s = abfd->sections; s != (asection *) NULL; s = s->next) { unsigned int i; struct external_reloc dst; arelent **p = s->orelocation; if (bfd_seek (abfd, s->rel_filepos, SEEK_SET) != 0) return false; for (i = 0; i < s->reloc_count; i++) { struct internal_reloc n; arelent *q = p[i]; memset ((PTR) & n, 0, sizeof (n)); n.r_vaddr = q->address + s->vma; #ifdef R_IHCONST /* The 29k const/consth reloc pair is a real kludge. The consth part doesn't have a symbol; it has an offset. So rebuilt that here. */ if (q->howto->type == R_IHCONST) n.r_symndx = q->addend; else #endif if (q->sym_ptr_ptr) { if (q->sym_ptr_ptr == bfd_abs_section.symbol_ptr_ptr) /* This is a relocation relative to the absolute symbol. */ n.r_symndx = -1; else { n.r_symndx = get_index ((*(q->sym_ptr_ptr))); /* Take notice if the symbol reloc points to a symbol we don't have in our symbol table. What should we do for this?? */ if (n.r_symndx > obj_conv_table_size (abfd)) abort (); } } #ifdef SWAP_OUT_RELOC_OFFSET n.r_offset = q->addend; #endif #ifdef SELECT_RELOC /* Work out reloc type from what is required */ SELECT_RELOC (n, q->howto); #else n.r_type = q->howto->type; #endif coff_swap_reloc_out (abfd, &n, &dst); if (bfd_write ((PTR) & dst, 1, RELSZ, abfd) != RELSZ) return false; } } return true; } /* Set flags and magic number of a coff file from architecture and machine type. Result is true if we can represent the arch&type, false if not. */ static boolean coff_set_flags (abfd, magicp, flagsp) bfd * abfd; unsigned *magicp; unsigned short *flagsp; { switch (bfd_get_arch (abfd)) { #ifdef Z8KMAGIC case bfd_arch_z8k: *magicp = Z8KMAGIC; switch (bfd_get_mach (abfd)) { case bfd_mach_z8001: *flagsp = F_Z8001; break; case bfd_mach_z8002: *flagsp = F_Z8002; break; default: return false; } return true; #endif #ifdef I960ROMAGIC case bfd_arch_i960: { unsigned flags; *magicp = I960ROMAGIC; /* ((bfd_get_file_flags(abfd) & WP_TEXT) ? I960ROMAGIC : I960RWMAGIC); FIXME??? */ switch (bfd_get_mach (abfd)) { case bfd_mach_i960_core: flags = F_I960CORE; break; case bfd_mach_i960_kb_sb: flags = F_I960KB; break; case bfd_mach_i960_mc: flags = F_I960MC; break; case bfd_mach_i960_xa: flags = F_I960XA; break; case bfd_mach_i960_ca: flags = F_I960CA; break; case bfd_mach_i960_ka_sa: flags = F_I960KA; break; default: return false; } *flagsp = flags; return true; } break; #endif #ifdef I386MAGIC case bfd_arch_i386: *magicp = I386MAGIC; #ifdef LYNXOS /* Just overwrite the usual value if we're doing Lynx. */ *magicp = LYNXCOFFMAGIC; #endif return true; break; #endif #ifdef MC68MAGIC case bfd_arch_m68k: #ifdef APOLLOM68KMAGIC *magicp = APOLLO_COFF_VERSION_NUMBER; #else *magicp = MC68MAGIC; #endif #ifdef LYNXOS /* Just overwrite the usual value if we're doing Lynx. */ *magicp = LYNXCOFFMAGIC; #endif return true; break; #endif #ifdef MC88MAGIC case bfd_arch_m88k: *magicp = MC88OMAGIC; return true; break; #endif #ifdef H8300MAGIC case bfd_arch_h8300: switch (bfd_get_mach (abfd)) { case bfd_mach_h8300: *magicp = H8300MAGIC; return true; case bfd_mach_h8300h: *magicp = H8300HMAGIC; return true; } break; #endif #ifdef SH_ARCH_MAGIC case bfd_arch_sh: *magicp = SH_ARCH_MAGIC; return true; break; #endif #ifdef SPARCMAGIC case bfd_arch_sparc: *magicp = SPARCMAGIC; #ifdef LYNXOS /* Just overwrite the usual value if we're doing Lynx. */ *magicp = LYNXCOFFMAGIC; #endif return true; break; #endif #ifdef H8500MAGIC case bfd_arch_h8500: *magicp = H8500MAGIC; return true; break; #endif #ifdef A29K_MAGIC_BIG case bfd_arch_a29k: if (abfd->xvec->byteorder_big_p) *magicp = A29K_MAGIC_BIG; else *magicp = A29K_MAGIC_LITTLE; return true; break; #endif #ifdef WE32KMAGIC case bfd_arch_we32k: *magicp = WE32KMAGIC; return true; break; #endif #ifdef U802TOCMAGIC case bfd_arch_rs6000: case bfd_arch_powerpc: *magicp = U802TOCMAGIC; return true; break; #endif default: /* Unknown architecture */ /* return false; -- fall through to "return false" below, to avoid "statement never reached" errors on the one below. */ break; } return false; } static boolean coff_set_arch_mach (abfd, arch, machine) bfd * abfd; enum bfd_architecture arch; unsigned long machine; { unsigned dummy1; unsigned short dummy2; if (! bfd_default_set_arch_mach (abfd, arch, machine)) return false; if (arch != bfd_arch_unknown && coff_set_flags (abfd, &dummy1, &dummy2) != true) return false; /* We can't represent this type */ return true; /* We're easy ... */ } /* Calculate the file position for each section. */ static void coff_compute_section_file_positions (abfd) bfd * abfd; { asection *current; asection *previous = (asection *) NULL; file_ptr sofar = FILHSZ; #ifndef I960 file_ptr old_sofar; #endif if (bfd_get_start_address (abfd)) { /* A start address may have been added to the original file. In this case it will need an optional header to record it. */ abfd->flags |= EXEC_P; } if (abfd->flags & EXEC_P) sofar += AOUTSZ; sofar += abfd->section_count * SCNHSZ; for (current = abfd->sections; current != (asection *) NULL; current = current->next) { /* Only deal with sections which have contents */ if (!(current->flags & SEC_HAS_CONTENTS)) continue; /* Align the sections in the file to the same boundary on which they are aligned in virtual memory. I960 doesn't do this (FIXME) so we can stay in sync with Intel. 960 doesn't yet page from files... */ #ifndef I960 { /* make sure this section is aligned on the right boundary - by padding the previous section up if necessary */ old_sofar = sofar; sofar = BFD_ALIGN (sofar, 1 << current->alignment_power); if (previous != (asection *) NULL) { previous->_raw_size += sofar - old_sofar; } } #endif /* FIXME, in demand paged files, the low order bits of the file offset must match the low order bits of the virtual address. "Low order" is apparently implementation defined. Add code here to round sofar up to match the virtual address. */ current->filepos = sofar; sofar += current->_raw_size; #ifndef I960 /* make sure that this section is of the right size too */ old_sofar = sofar; sofar = BFD_ALIGN (sofar, 1 << current->alignment_power); current->_raw_size += sofar - old_sofar; #endif #ifdef _LIB /* Force .lib sections to start at zero. The vma is then incremented in coff_set_section_contents. This is right for SVR3.2. */ if (strcmp (current->name, _LIB) == 0) bfd_set_section_vma (abfd, current, 0); #endif previous = current; } obj_relocbase (abfd) = sofar; } #ifndef RS6000COFF_C /* If .file, .text, .data, .bss symbols are missing, add them. */ /* @@ Should we only be adding missing symbols, or overriding the aux values for existing section symbols? */ static boolean coff_add_missing_symbols (abfd) bfd *abfd; { unsigned int nsyms = bfd_get_symcount (abfd); asymbol **sympp = abfd->outsymbols; asymbol **sympp2; unsigned int i; int need_text = 1, need_data = 1, need_bss = 1, need_file = 1; for (i = 0; i < nsyms; i++) { coff_symbol_type *csym = coff_symbol_from (abfd, sympp[i]); CONST char *name; if (csym) { /* only do this if there is a coff representation of the input symbol */ if (csym->native && csym->native->u.syment.n_sclass == C_FILE) { need_file = 0; continue; } name = csym->symbol.name; if (!name) continue; if (!strcmp (name, _TEXT)) need_text = 0; #ifdef APOLLO_M68 else if (!strcmp (name, ".wtext")) need_text = 0; #endif else if (!strcmp (name, _DATA)) need_data = 0; else if (!strcmp (name, _BSS)) need_bss = 0; } } /* Now i == bfd_get_symcount (abfd). */ /* @@ For now, don't deal with .file symbol. */ need_file = 0; if (!need_text && !need_data && !need_bss && !need_file) return true; nsyms += need_text + need_data + need_bss + need_file; sympp2 = (asymbol **) bfd_alloc_by_size_t (abfd, nsyms * sizeof (asymbol *)); if (!sympp2) { bfd_set_error (bfd_error_no_memory); return false; } memcpy (sympp2, sympp, i * sizeof (asymbol *)); if (need_file) { /* @@ Generate fake .file symbol, in sympp2[i], and increment i. */ abort (); } if (need_text) sympp2[i++] = coff_section_symbol (abfd, _TEXT); if (need_data) sympp2[i++] = coff_section_symbol (abfd, _DATA); if (need_bss) sympp2[i++] = coff_section_symbol (abfd, _BSS); BFD_ASSERT (i == nsyms); bfd_set_symtab (abfd, sympp2, nsyms); return true; } #endif /* ! defined (RS6000COFF_C) */ /* SUPPRESS 558 */ /* SUPPRESS 529 */ static boolean coff_write_object_contents (abfd) bfd * abfd; { asection *current; unsigned int count; boolean hasrelocs = false; boolean haslinno = false; file_ptr reloc_base; file_ptr lineno_base; file_ptr sym_base; file_ptr scn_base; file_ptr data_base; unsigned long reloc_size = 0; unsigned long lnno_size = 0; asection *text_sec = NULL; asection *data_sec = NULL; asection *bss_sec = NULL; struct internal_filehdr internal_f; struct internal_aouthdr internal_a; bfd_set_error (bfd_error_system_call); /* Number the output sections, starting from one on the first section with a name which doesn't start with a *. @@ The code doesn't make this check. Is it supposed to be done, or isn't it?? */ count = 1; for (current = abfd->sections; current != (asection *) NULL; current = current->next) { current->target_index = count; count++; } if (abfd->output_has_begun == false) { coff_compute_section_file_positions (abfd); } if (abfd->sections != (asection *) NULL) { scn_base = abfd->sections->filepos; } else { scn_base = 0; } if (bfd_seek (abfd, scn_base, SEEK_SET) != 0) return false; reloc_base = obj_relocbase (abfd); /* Make a pass through the symbol table to count line number entries and put them into the correct asections */ lnno_size = coff_count_linenumbers (abfd) * LINESZ; data_base = scn_base; /* Work out the size of the reloc and linno areas */ for (current = abfd->sections; current != NULL; current = current->next) { /* We give section headers to +ve indexes */ if (current->target_index > 0) { reloc_size += current->reloc_count * RELSZ; data_base += SCNHSZ; } } lineno_base = reloc_base + reloc_size; sym_base = lineno_base + lnno_size; /* Indicate in each section->line_filepos its actual file address */ for (current = abfd->sections; current != NULL; current = current->next) { if (current->target_index > 0) { if (current->lineno_count) { current->line_filepos = lineno_base; current->moving_line_filepos = lineno_base; lineno_base += current->lineno_count * LINESZ; } else { current->line_filepos = 0; } if (current->reloc_count) { current->rel_filepos = reloc_base; reloc_base += current->reloc_count * RELSZ; } else { current->rel_filepos = 0; } } } /* Write section headers to the file. */ internal_f.f_nscns = 0; if (bfd_seek (abfd, (file_ptr) ((abfd->flags & EXEC_P) ? (FILHSZ + AOUTSZ) : FILHSZ), SEEK_SET) != 0) return false; { #if 0 unsigned int pad = abfd->flags & D_PAGED ? data_base : 0; #endif unsigned int pad = 0; for (current = abfd->sections; current != NULL; current = current->next) { struct internal_scnhdr section; if (current->target_index > 0) { internal_f.f_nscns++; strncpy (&(section.s_name[0]), current->name, 8); #ifdef _LIB /* Always set s_vaddr of .lib to 0. This is right for SVR3.2 Ian Taylor . */ if (strcmp (current->name, _LIB) == 0) section.s_vaddr = 0; else #endif section.s_vaddr = current->lma + pad; section.s_paddr = current->lma + pad; section.s_size = current->_raw_size - pad; /* If this section has no size or is unloadable then the scnptr will be 0 too */ if (current->_raw_size - pad == 0 || (current->flags & (SEC_LOAD | SEC_HAS_CONTENTS)) == 0) { section.s_scnptr = 0; } else { section.s_scnptr = current->filepos; } section.s_relptr = current->rel_filepos; section.s_lnnoptr = current->line_filepos; section.s_nreloc = current->reloc_count; section.s_nlnno = current->lineno_count; if (current->reloc_count != 0) hasrelocs = true; if (current->lineno_count != 0) haslinno = true; section.s_flags = sec_to_styp_flags (current->name, current->flags); if (!strcmp (current->name, _TEXT)) { text_sec = current; } else if (!strcmp (current->name, _DATA)) { data_sec = current; #ifdef TWO_DATA_SECS } else if (!strcmp (current->name, ".data2")) { data_sec = current; #endif /* TWO_DATA_SECS */ } else if (!strcmp (current->name, _BSS)) { bss_sec = current; } #ifdef I960 section.s_align = (current->alignment_power ? 1 << current->alignment_power : 0); #endif { SCNHDR buff; coff_swap_scnhdr_out (abfd, §ion, &buff); if (bfd_write ((PTR) (&buff), 1, SCNHSZ, abfd) != SCNHSZ) return false; } pad = 0; } } } /* OK, now set up the filehdr... */ /* Don't include the internal abs section in the section count */ /* We will NOT put a fucking timestamp in the header here. Every time you put it back, I will come in and take it out again. I'm sorry. This field does not belong here. We fill it with a 0 so it compares the same but is not a reasonable time. -- gnu@cygnus.com */ internal_f.f_timdat = 0; if (bfd_get_symcount (abfd) != 0) internal_f.f_symptr = sym_base; else internal_f.f_symptr = 0; internal_f.f_flags = 0; if (abfd->flags & EXEC_P) internal_f.f_opthdr = AOUTSZ; else internal_f.f_opthdr = 0; if (!hasrelocs) internal_f.f_flags |= F_RELFLG; if (!haslinno) internal_f.f_flags |= F_LNNO; if (0 == bfd_get_symcount (abfd)) internal_f.f_flags |= F_LSYMS; if (abfd->flags & EXEC_P) internal_f.f_flags |= F_EXEC; if (!abfd->xvec->byteorder_big_p) internal_f.f_flags |= F_AR32WR; else internal_f.f_flags |= F_AR32W; /* FIXME, should do something about the other byte orders and architectures. */ memset (&internal_a, 0, sizeof internal_a); /* Set up architecture-dependent stuff */ { unsigned int magic = 0; unsigned short flags = 0; coff_set_flags (abfd, &magic, &flags); internal_f.f_magic = magic; internal_f.f_flags |= flags; /* ...and the "opt"hdr... */ #ifdef A29K #ifdef ULTRA3 /* NYU's machine */ /* FIXME: This is a bogus check. I really want to see if there * is a .shbss or a .shdata section, if so then set the magic * number to indicate a shared data executable. */ if (internal_f.f_nscns >= 7) internal_a.magic = SHMAGIC; /* Shared magic */ else #endif /* ULTRA3 */ internal_a.magic = NMAGIC;/* Assume separate i/d */ #define __A_MAGIC_SET__ #endif /* A29K */ #ifdef I960 internal_a.magic = (magic == I960ROMAGIC ? NMAGIC : OMAGIC); #define __A_MAGIC_SET__ #endif /* I960 */ #if M88 #define __A_MAGIC_SET__ internal_a.magic = PAGEMAGICBCS; #endif /* M88 */ #if APOLLO_M68 #define __A_MAGIC_SET__ internal_a.magic = APOLLO_COFF_VERSION_NUMBER; #endif #if M68 || WE32K #define __A_MAGIC_SET__ /* Never was anything here for the 68k */ #endif /* M68 || WE32K */ #if I386 #define __A_MAGIC_SET__ internal_a.magic = ZMAGIC; #endif /* I386 */ #if RS6000COFF_C #define __A_MAGIC_SET__ internal_a.magic = (abfd->flags & D_PAGED) ? RS6K_AOUTHDR_ZMAGIC : (abfd->flags & WP_TEXT) ? RS6K_AOUTHDR_NMAGIC : RS6K_AOUTHDR_OMAGIC; #endif #ifndef __A_MAGIC_SET__ #include "Your aouthdr magic number is not being set!" #else #undef __A_MAGIC_SET__ #endif } /* Now should write relocs, strings, syms */ obj_sym_filepos (abfd) = sym_base; if (bfd_get_symcount (abfd) != 0) { #ifndef RS6000COFF_C if (!coff_add_missing_symbols (abfd)) return false; #endif if (!coff_renumber_symbols (abfd)) return false; coff_mangle_symbols (abfd); if (! coff_write_symbols (abfd)) return false; if (! coff_write_linenumbers (abfd)) return false; if (! coff_write_relocs (abfd)) return false; } if (text_sec) { internal_a.tsize = bfd_get_section_size_before_reloc (text_sec); internal_a.text_start = internal_a.tsize ? text_sec->vma : 0; } if (data_sec) { internal_a.dsize = bfd_get_section_size_before_reloc (data_sec); internal_a.data_start = internal_a.dsize ? data_sec->vma : 0; } if (bss_sec) { internal_a.bsize = bfd_get_section_size_before_reloc (bss_sec); } internal_a.entry = bfd_get_start_address (abfd); internal_f.f_nsyms = bfd_get_symcount (abfd); /* now write them */ if (bfd_seek (abfd, (file_ptr) 0, SEEK_SET) != 0) return false; { FILHDR buff; coff_swap_filehdr_out (abfd, (PTR) & internal_f, (PTR) & buff); if (bfd_write ((PTR) & buff, 1, FILHSZ, abfd) != FILHSZ) return false; } if (abfd->flags & EXEC_P) { AOUTHDR buff; coff_swap_aouthdr_out (abfd, (PTR) & internal_a, (PTR) & buff); if (bfd_write ((PTR) & buff, 1, AOUTSZ, abfd) != AOUTSZ) return false; } return true; } static boolean coff_set_section_contents (abfd, section, location, offset, count) bfd * abfd; sec_ptr section; PTR location; file_ptr offset; bfd_size_type count; { if (abfd->output_has_begun == false) /* set by bfd.c handler */ coff_compute_section_file_positions (abfd); #ifdef _LIB /* If this is a .lib section, bump the vma address so that it winds up being the number of .lib sections output. This is right for SVR3.2. Shared libraries should probably get more generic support. Ian Taylor . */ if (strcmp (section->name, _LIB) == 0) ++section->lma; #endif /* Don't write out bss sections - one way to do this is to see if the filepos has not been set. */ if (section->filepos == 0) return true; if (bfd_seek (abfd, (file_ptr) (section->filepos + offset), SEEK_SET) != 0) return false; if (count != 0) { return (bfd_write (location, 1, count, abfd) == count) ? true : false; } return true; } #if 0 static boolean coff_close_and_cleanup (abfd) bfd *abfd; { if (!bfd_read_p (abfd)) switch (abfd->format) { case bfd_archive: if (!_bfd_write_archive_contents (abfd)) return false; break; case bfd_object: if (!coff_write_object_contents (abfd)) return false; break; default: bfd_set_error (bfd_error_invalid_operation); return false; } /* We depend on bfd_close to free all the memory on the obstack. */ /* FIXME if bfd_release is not using obstacks! */ return true; } #endif static PTR buy_and_read (abfd, where, seek_direction, size) bfd *abfd; file_ptr where; int seek_direction; size_t size; { PTR area = (PTR) bfd_alloc (abfd, size); if (!area) { bfd_set_error (bfd_error_no_memory); return (NULL); } if (bfd_seek (abfd, where, seek_direction) != 0 || bfd_read (area, 1, size, abfd) != size) return (NULL); return (area); } /* buy_and_read() */ /* SUBSUBSECTION Reading linenumbers Creating the linenumber table is done by reading in the entire coff linenumber table, and creating another table for internal use. A coff linenumber table is structured so that each function is marked as having a line number of 0. Each line within the function is an offset from the first line in the function. The base of the line number information for the table is stored in the symbol associated with the function. The information is copied from the external to the internal table, and each symbol which marks a function is marked by pointing its... How does this work ? */ static boolean coff_slurp_line_table (abfd, asect) bfd *abfd; asection *asect; { LINENO *native_lineno; alent *lineno_cache; BFD_ASSERT (asect->lineno == (alent *) NULL); native_lineno = (LINENO *) buy_and_read (abfd, asect->line_filepos, SEEK_SET, (size_t) (LINESZ * asect->lineno_count)); lineno_cache = (alent *) bfd_alloc (abfd, (size_t) ((asect->lineno_count + 1) * sizeof (alent))); if (lineno_cache == NULL) { bfd_set_error (bfd_error_no_memory); return false; } else { unsigned int counter = 0; alent *cache_ptr = lineno_cache; LINENO *src = native_lineno; while (counter < asect->lineno_count) { struct internal_lineno dst; coff_swap_lineno_in (abfd, src, &dst); cache_ptr->line_number = dst.l_lnno; if (cache_ptr->line_number == 0) { coff_symbol_type *sym = (coff_symbol_type *) (dst.l_addr.l_symndx + obj_raw_syments (abfd))->u.syment._n._n_n._n_zeroes; cache_ptr->u.sym = (asymbol *) sym; sym->lineno = cache_ptr; } else { cache_ptr->u.offset = dst.l_addr.l_paddr - bfd_section_vma (abfd, asect); } /* If no linenumber expect a symbol index */ cache_ptr++; src++; counter++; } cache_ptr->line_number = 0; } asect->lineno = lineno_cache; /* FIXME, free native_lineno here, or use alloca or something. */ return true; } static boolean coff_slurp_symbol_table (abfd) bfd * abfd; { combined_entry_type *native_symbols; coff_symbol_type *cached_area; unsigned int *table_ptr; unsigned int number_of_symbols = 0; if (obj_symbols (abfd)) return true; if (bfd_seek (abfd, obj_sym_filepos (abfd), SEEK_SET) != 0) return false; /* Read in the symbol table */ if ((native_symbols = coff_get_normalized_symtab (abfd)) == NULL) { return (false); } /* on error */ /* Allocate enough room for all the symbols in cached form */ cached_area = (coff_symbol_type *) bfd_alloc (abfd, (size_t) (bfd_get_symcount (abfd) * sizeof (coff_symbol_type))); if (cached_area == NULL) { bfd_set_error (bfd_error_no_memory); return false; } /* on error */ table_ptr = (unsigned int *) bfd_alloc (abfd, (size_t) (bfd_get_symcount (abfd) * sizeof (unsigned int))); if (table_ptr == NULL) { bfd_set_error (bfd_error_no_memory); return false; } else { coff_symbol_type *dst = cached_area; unsigned int last_native_index = bfd_get_symcount (abfd); unsigned int this_index = 0; while (this_index < last_native_index) { combined_entry_type *src = native_symbols + this_index; table_ptr[this_index] = number_of_symbols; dst->symbol.the_bfd = abfd; dst->symbol.name = (char *) (src->u.syment._n._n_n._n_offset); /* We use the native name field to point to the cached field. */ src->u.syment._n._n_n._n_zeroes = (long) dst; dst->symbol.section = coff_section_from_bfd_index (abfd, src->u.syment.n_scnum); dst->symbol.flags = 0; dst->done_lineno = false; switch (src->u.syment.n_sclass) { #ifdef I960 case C_LEAFEXT: #if 0 dst->symbol.value = src->u.syment.n_value - dst->symbol.section->vma; dst->symbol.flags = BSF_EXPORT | BSF_GLOBAL; dst->symbol.flags |= BSF_NOT_AT_END; #endif /* Fall through to next case */ #endif case C_EXT: #ifdef RS6000COFF_C case C_HIDEXT: #endif if ((src->u.syment.n_scnum) == 0) { if ((src->u.syment.n_value) == 0) { dst->symbol.section = &bfd_und_section; dst->symbol.value = 0; } else { dst->symbol.section = &bfd_com_section; dst->symbol.value = (src->u.syment.n_value); } } else { /* Base the value as an index from the base of the section */ dst->symbol.flags = BSF_EXPORT | BSF_GLOBAL; dst->symbol.value = src->u.syment.n_value - dst->symbol.section->vma; if (ISFCN ((src->u.syment.n_type))) { /* A function ext does not go at the end of a file */ dst->symbol.flags |= BSF_NOT_AT_END; } } #ifdef RS6000COFF_C /* If this symbol has a csect aux of type LD, the scnlen field is actually the index of the containing csect symbol. We need to pointerize it. */ if (src->u.syment.n_numaux > 0) { combined_entry_type *aux; aux = src + src->u.syment.n_numaux - 1; if (SMTYP_SMTYP (aux->u.auxent.x_csect.x_smtyp) == XTY_LD) { aux->u.auxent.x_csect.x_scnlen.p = native_symbols + aux->u.auxent.x_csect.x_scnlen.l; aux->fix_scnlen = 1; } } #endif break; case C_STAT: /* static */ #ifdef I960 case C_LEAFSTAT: /* static leaf procedure */ #endif case C_LABEL: /* label */ if (src->u.syment.n_scnum == -2) dst->symbol.flags = BSF_DEBUGGING; else dst->symbol.flags = BSF_LOCAL; /* Base the value as an index from the base of the section, if there is one */ if (dst->symbol.section) dst->symbol.value = (src->u.syment.n_value) - dst->symbol.section->vma; else dst->symbol.value = (src->u.syment.n_value); break; case C_MOS: /* member of structure */ case C_EOS: /* end of structure */ #ifdef NOTDEF /* C_AUTOARG has the same value */ #ifdef C_GLBLREG case C_GLBLREG: /* A29k-specific storage class */ #endif #endif case C_REGPARM: /* register parameter */ case C_REG: /* register variable */ #ifdef C_AUTOARG case C_AUTOARG: /* 960-specific storage class */ #endif case C_TPDEF: /* type definition */ case C_ARG: case C_AUTO: /* automatic variable */ case C_FIELD: /* bit field */ case C_ENTAG: /* enumeration tag */ case C_MOE: /* member of enumeration */ case C_MOU: /* member of union */ case C_UNTAG: /* union tag */ dst->symbol.flags = BSF_DEBUGGING; dst->symbol.value = (src->u.syment.n_value); break; case C_FILE: /* file name */ case C_STRTAG: /* structure tag */ #ifdef RS6000COFF_C case C_BINCL: /* beginning of include file */ case C_EINCL: /* ending of include file */ case C_GSYM: case C_LSYM: case C_PSYM: case C_RSYM: case C_RPSYM: case C_STSYM: case C_DECL: case C_ENTRY: case C_FUN: case C_ESTAT: #endif dst->symbol.flags = BSF_DEBUGGING; dst->symbol.value = (src->u.syment.n_value); break; #ifdef RS6000COFF_C case C_BSTAT: dst->symbol.flags = BSF_DEBUGGING; dst->symbol.value = src->u.syment.n_value; /* The value is actually a symbol index. Save a pointer to the symbol instead of the index. FIXME: This should use a union. */ src->u.syment.n_value = (long) (native_symbols + src->u.syment.n_value); src->fix_value = 1; break; #endif case C_BLOCK: /* ".bb" or ".eb" */ case C_FCN: /* ".bf" or ".ef" */ case C_EFCN: /* physical end of function */ dst->symbol.flags = BSF_LOCAL; /* Base the value as an index from the base of the section */ dst->symbol.value = (src->u.syment.n_value) - dst->symbol.section->vma; break; case C_NULL: case C_EXTDEF: /* external definition */ case C_ULABEL: /* undefined label */ case C_USTATIC: /* undefined static */ case C_LINE: /* line # reformatted as symbol table entry */ case C_ALIAS: /* duplicate tag */ case C_HIDDEN: /* ext symbol in dmert public lib */ default: fprintf (stderr, "Unrecognized storage class %d (assuming debugging)\n for %s symbol `%s'\n", src->u.syment.n_sclass, dst->symbol.section->name, dst->symbol.name); /* abort();*/ dst->symbol.flags = BSF_DEBUGGING; dst->symbol.value = (src->u.syment.n_value); break; } /* BFD_ASSERT(dst->symbol.flags != 0);*/ dst->native = src; dst->symbol.udata = 0; dst->lineno = (alent *) NULL; this_index += (src->u.syment.n_numaux) + 1; dst++; number_of_symbols++; } /* walk the native symtab */ } /* bfdize the native symtab */ obj_symbols (abfd) = cached_area; obj_raw_syments (abfd) = native_symbols; obj_conv_table_size (abfd) = bfd_get_symcount (abfd); bfd_get_symcount (abfd) = number_of_symbols; obj_convert (abfd) = table_ptr; /* Slurp the line tables for each section too */ { asection *p; p = abfd->sections; while (p) { coff_slurp_line_table (abfd, p); p = p->next; } } return true; } /* coff_slurp_symbol_table() */ /* SUBSUBSECTION Reading relocations Coff relocations are easily transformed into the internal BFD form (@code{arelent}). Reading a coff relocation table is done in the following stages: o Read the entire coff relocation table into memory. o Process each relocation in turn; first swap it from the external to the internal form. o Turn the symbol referenced in the relocation's symbol index into a pointer into the canonical symbol table. This table is the same as the one returned by a call to @code{bfd_canonicalize_symtab}. The back end will call that routine and save the result if a canonicalization hasn't been done. o The reloc index is turned into a pointer to a howto structure, in a back end specific way. For instance, the 386 and 960 use the @code{r_type} to directly produce an index into a howto table vector; the 88k subtracts a number from the @code{r_type} field and creates an addend field. */ #ifndef CALC_ADDEND #define CALC_ADDEND(abfd, ptr, reloc, cache_ptr) \ { \ coff_symbol_type *coffsym = (coff_symbol_type *) NULL; \ if (ptr && bfd_asymbol_bfd (ptr) != abfd) \ coffsym = (obj_symbols (abfd) \ + (cache_ptr->sym_ptr_ptr - symbols)); \ else if (ptr) \ coffsym = coff_symbol_from (abfd, ptr); \ if (coffsym != (coff_symbol_type *) NULL \ && coffsym->native->u.syment.n_scnum == 0) \ cache_ptr->addend = 0; \ else if (ptr && bfd_asymbol_bfd (ptr) == abfd \ && ptr->section != (asection *) NULL) \ cache_ptr->addend = - (ptr->section->vma + ptr->value); \ else \ cache_ptr->addend = 0; \ } #endif static boolean coff_slurp_reloc_table (abfd, asect, symbols) bfd * abfd; sec_ptr asect; asymbol ** symbols; { RELOC *native_relocs; arelent *reloc_cache; arelent *cache_ptr; unsigned int idx; if (asect->relocation) return true; if (asect->reloc_count == 0) return true; if (asect->flags & SEC_CONSTRUCTOR) return true; if (!coff_slurp_symbol_table (abfd)) return false; native_relocs = (RELOC *) buy_and_read (abfd, asect->rel_filepos, SEEK_SET, (size_t) (RELSZ * asect->reloc_count)); reloc_cache = (arelent *) bfd_alloc (abfd, (size_t) (asect->reloc_count * sizeof (arelent))); if (reloc_cache == NULL) { bfd_set_error (bfd_error_no_memory); return false; } for (idx = 0; idx < asect->reloc_count; idx++) { #ifdef RELOC_PROCESSING struct internal_reloc dst; struct external_reloc *src; cache_ptr = reloc_cache + idx; src = native_relocs + idx; bfd_swap_reloc_in (abfd, src, &dst); RELOC_PROCESSING (cache_ptr, &dst, symbols, abfd, asect); #else struct internal_reloc dst; asymbol *ptr; struct external_reloc *src; cache_ptr = reloc_cache + idx; src = native_relocs + idx; bfd_swap_reloc_in (abfd, src, &dst); cache_ptr->address = dst.r_vaddr; if (dst.r_symndx != -1) { /* @@ Should never be greater than count of symbols! */ if (dst.r_symndx >= obj_conv_table_size (abfd)) abort (); cache_ptr->sym_ptr_ptr = symbols + obj_convert (abfd)[dst.r_symndx]; ptr = *(cache_ptr->sym_ptr_ptr); } else { cache_ptr->sym_ptr_ptr = bfd_abs_section.symbol_ptr_ptr; ptr = 0; } /* The symbols definitions that we have read in have been relocated as if their sections started at 0. But the offsets refering to the symbols in the raw data have not been modified, so we have to have a negative addend to compensate. Note that symbols which used to be common must be left alone */ /* Calculate any reloc addend by looking at the symbol */ CALC_ADDEND (abfd, ptr, dst, cache_ptr); cache_ptr->address -= asect->vma; /* !! cache_ptr->section = (asection *) NULL;*/ /* Fill in the cache_ptr->howto field from dst.r_type */ RTYPE2HOWTO (cache_ptr, &dst); #endif } asect->relocation = reloc_cache; return true; } /* This is stupid. This function should be a boolean predicate. */ static long coff_canonicalize_reloc (abfd, section, relptr, symbols) bfd * abfd; sec_ptr section; arelent ** relptr; asymbol ** symbols; { arelent *tblptr = section->relocation; unsigned int count = 0; if (section->flags & SEC_CONSTRUCTOR) { /* this section has relocs made up by us, they are not in the file, so take them out of their chain and place them into the data area provided */ arelent_chain *chain = section->constructor_chain; for (count = 0; count < section->reloc_count; count++) { *relptr++ = &chain->relent; chain = chain->next; } } else { if (! coff_slurp_reloc_table (abfd, section, symbols)) return -1; tblptr = section->relocation; for (; count++ < section->reloc_count;) *relptr++ = tblptr++; } *relptr = 0; return section->reloc_count; } #ifdef GNU960 file_ptr coff_sym_filepos (abfd) bfd *abfd; { return obj_sym_filepos (abfd); } #endif #ifndef coff_reloc16_estimate #define coff_reloc16_estimate dummy_reloc16_estimate static int dummy_reloc16_estimate (abfd, input_section, reloc, shrink, link_info) bfd *abfd; asection *input_section; arelent *reloc; unsigned int shrink; struct bfd_link_info *link_info; { abort (); } #endif #ifndef coff_reloc16_extra_cases #define coff_reloc16_extra_cases dummy_reloc16_extra_cases /* This works even if abort is not declared in any header file. */ static void dummy_reloc16_extra_cases (abfd, link_info, link_order, reloc, data, src_ptr, dst_ptr) bfd *abfd; struct bfd_link_info *link_info; struct bfd_link_order *link_order; arelent *reloc; bfd_byte *data; unsigned int *src_ptr; unsigned int *dst_ptr; { fprintf (stderr, "%s\n", reloc->howto->name); abort (); } #endif static CONST bfd_coff_backend_data bfd_coff_std_swap_table = { coff_swap_aux_in, coff_swap_sym_in, coff_swap_lineno_in, coff_swap_aux_out, coff_swap_sym_out, coff_swap_lineno_out, coff_swap_reloc_out, coff_swap_filehdr_out, coff_swap_aouthdr_out, coff_swap_scnhdr_out, FILHSZ, AOUTSZ, SCNHSZ, SYMESZ, AUXESZ, LINESZ, #ifdef COFF_LONG_FILENAMES true, #else false, #endif coff_swap_filehdr_in, coff_swap_aouthdr_in, coff_swap_scnhdr_in, coff_bad_format_hook, coff_set_arch_mach_hook, coff_mkobject_hook, styp_to_sec_flags, coff_make_section_hook, coff_set_alignment_hook, coff_slurp_symbol_table, symname_in_debug_hook, coff_reloc16_extra_cases, coff_reloc16_estimate }; #define coff_close_and_cleanup _bfd_generic_close_and_cleanup #define coff_bfd_free_cached_info _bfd_generic_bfd_free_cached_info #define coff_get_section_contents _bfd_generic_get_section_contents #define coff_bfd_copy_private_section_data \ _bfd_generic_bfd_copy_private_section_data #define coff_bfd_copy_private_bfd_data _bfd_generic_bfd_copy_private_bfd_data #ifndef coff_bfd_is_local_label #define coff_bfd_is_local_label bfd_generic_is_local_label #endif /* The reloc lookup routine must be supplied by each individual COFF backend. */ #ifndef coff_bfd_reloc_type_lookup #define coff_bfd_reloc_type_lookup _bfd_norelocs_bfd_reloc_type_lookup #endif #define coff_bfd_get_relocated_section_contents \ bfd_generic_get_relocated_section_contents #define coff_bfd_relax_section bfd_generic_relax_section #define coff_bfd_link_hash_table_create _bfd_generic_link_hash_table_create #define coff_bfd_link_add_symbols _bfd_generic_link_add_symbols #define coff_bfd_final_link _bfd_generic_final_link