NAME <name> [ , <base> ]
- The result is going to be <name>.EXE
+ The result is going to be <name>.EXE
LIBRARY <name> [ , <base> ]
- The result is going to be <name>.DLL
+ The result is going to be <name>.DLL
EXPORTS ( <name1> [ = <name2> ] [ @ <integer> ] [ NONAME ] [CONSTANT] ) *
- Declares name1 as an exported symbol from the
- DLL, with optional ordinal number <integer>
+ Declares name1 as an exported symbol from the
+ DLL, with optional ordinal number <integer>
IMPORTS ( [ <name> = ] <name> . <name> ) *
- Ignored for compatibility
+ Ignored for compatibility
DESCRIPTION <string>
- Puts <string> into output .exp file in the .rdata section
+ Puts <string> into output .exp file in the .rdata section
[STACKSIZE|HEAPSIZE] <number-reserve> [ , <number-commit> ]
- Generates --stack|--heap <number-reserve>,<number-commit>
- in the output .drective section. The linker will
- see this and act upon it.
+ Generates --stack|--heap <number-reserve>,<number-commit>
+ in the output .drective section. The linker will
+ see this and act upon it.
[CODE|DATA] <attr>+
SECTIONS ( <sectionname> <attr>+ )*
-
<attr> = READ | WRITE | EXECUTE | SHARED
-
Generates --attr <sectionname> <attr> in the output
- .drective section. The linker will see this and act
- upon it.
+ <attr> = READ | WRITE | EXECUTE | SHARED
+ Generates --attr <sectionname> <attr> in the output
+ .drective section. The linker will see this and act
+ upon it.
A -export:<name> in a .drective section in an input .o or .a
- The program generates output files with the prefix supplied
- on the command line, or in the def file, or taken from the first
- supplied argument.
+ The program generates output files with the prefix supplied
+ on the command line, or in the def file, or taken from the first
+ supplied argument.
- The output files are <prefix>.exp.s and <prefix>.lib.s
+ The output files are <prefix>-exp.s and <prefix>-lib.s
- The .exp.s file contains the information necessary to export
- the routines in the DLL. The .lib.s file contains the information
- necessary to use the DLL's routines from a referencing program.
+ The .exp.s file contains the information necessary to export
+ the routines in the DLL. The .lib.s file contains the information
+ necessary to use the DLL's routines from a referencing program.
-Example:
+ Example:
-file1.c:
- asm (".section .drectve");
- asm (".ascii \"-export:adef\"");
+ file1.c:
+ asm (".section .drectve");
+ asm (".ascii \"-export:adef\"");
- adef(char *s)
- {
- printf("hello from the dll %s\n",s);
- }
+ adef(char *s)
+ {
+ printf("hello from the dll %s\n",s);
+ }
- bdef(char *s)
- {
- printf("hello from the dll and the other entry point %s\n",s);
- }
+ bdef(char *s)
+ {
+ printf("hello from the dll and the other entry point %s\n",s);
+ }
-file2.c:
- asm (".section .drectve");
- asm (".ascii \"-export:cdef\"");
- asm (".ascii \"-export:ddef\"");
- cdef(char *s)
- {
- printf("hello from the dll %s\n",s);
- }
+ file2.c:
+ asm (".section .drectve");
+ asm (".ascii \"-export:cdef\"");
+ asm (".ascii \"-export:ddef\"");
+ cdef(char *s)
+ {
+ printf("hello from the dll %s\n",s);
+ }
- ddef(char *s)
- {
- printf("hello from the dll and the other entry point %s\n",s);
- }
+ ddef(char *s)
+ {
+ printf("hello from the dll and the other entry point %s\n",s);
+ }
- printf()
- {
- return 9;
- }
+ printf()
+ {
+ return 9;
+ }
-main.c
+ main.c
- main()
- {
- cdef();
- }
+ main()
+ {
+ cdef();
+ }
-thedll.def
+ thedll.def
- LIBRARY thedll
- HEAPSIZE 0x40000, 0x2000
- EXPORTS bdef @ 20
- cdef @ 30 NONAME
+ LIBRARY thedll
+ HEAPSIZE 0x40000, 0x2000
+ EXPORTS bdef @ 20
+ cdef @ 30 NONAME
- SECTIONS donkey READ WRITE
- aardvark EXECUTE
+ SECTIONS donkey READ WRITE
+ aardvark EXECUTE
-# compile up the parts of the dll
+ # compile up the parts of the dll
-gcc -c file1.c
-gcc -c file2.c
+ gcc -c file1.c
+ gcc -c file2.c
-# put them in a library (you don't have to, you
-# could name all the .os on the dlltool line)
+ # put them in a library (you don't have to, you
+ # could name all the .os on the dlltool line)
-ar qcv thedll.in file1.o file2.o
-ranlib thedll.in
+ ar qcv thedll.in file1.o file2.o
+ ranlib thedll.in
-# run this tool over the library and the def file
-./dlltool -o thedll -d thedll.def thedll.in
+ # run this tool over the library and the def file
+ ./dlltool -o thedll -d thedll.def thedll.in
-# build the export table for the dll
-as -o thedll.exp thedll.exp.s
-# build the dll with the library with file1.o, file2.o and the export table
-ld -o thedll.dll thedll.exp thedll.in
+ # build the export table for the dll
+ as -o thedll.exp thedll-exp.s
+ # build the dll with the library with file1.o, file2.o and the export table
+ ld -o thedll.dll thedll.exp thedll.in
-# build the import table for the executable
-as -o thedll.lib thedll.lib.s
+ # build the import table for the executable
+ as -o thedll.lib thedll-lib.s
-# build the mainline
-gcc -c themain.c
+ # build the mainline
+ gcc -c themain.c
-# link the executable with the import library
-ld -e main -Tthemain.ld -o themain.exe themain.o thedll.lib
+ # link the executable with the import library
+ ld -e main -Tthemain.ld -o themain.exe themain.o thedll.lib
+ */
-*/
-
+#define PAGE_SIZE 4096
+#define PAGE_MASK (-PAGE_SIZE)
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
char *strdup ();
static int machine;
-
+int suckunderscore;
+static int verbose;
+FILE *base_file;
#ifdef DLLTOOL_ARM
-static char *mname = "arm";
+static char *mname = "arm";
#endif
#ifdef DLLTOOL_I386
-static char *mname = "i386";
+static char *mname = "i386";
#endif
+#define PATHMAX 250 /* What's the right name for this ? */
+char outfile[PATHMAX];
struct mac
+ {
+ char *type;
+ char *how_byte;
+ char *how_short;
+ char *how_long;
+ char *how_asciz;
+ char *how_comment;
+ char *how_jump;
+ char *how_global;
+ char *how_space;
+ }
+mtable[]
+=
{
- char *type;
- char *how_byte;
- char *how_short;
- char *how_long;
- char *how_asciz;
- char *how_comment;
- char *how_jump;
- char *how_global;
- char *how_space;
-} mtable[]
-= {{"arm",".byte",".short",".long",".asciz","@","bl",".global",".space"},
- {"i386",".byte",".short",".long",".asciz",";","jmp",".global",".space"},
- 0};
+ {
+ "arm", ".byte", ".short", ".long", ".asciz", "@", "ldr\tip,[pc]\n\tldr\tpc,[ip]\n\t.long", ".global", ".space"
+ }
+ ,
+ {
+ "i386", ".byte", ".short", ".long", ".asciz", "#", "jmp *", ".global", ".space"
+ }
+ ,
+ 0
+};
#define ASM_BYTE mtable[machine].how_byte
#define ASM_SHORT mtable[machine].how_short
#define ASM_GLOBAL mtable[machine].how_global
#define ASM_SPACE mtable[machine].how_space
-#define PATHMAX 250 /* What's the right name for this ? */
+
static char **oav;
int i;
FILE *f = fopen (name, "r");
if (!f)
{
- fprintf (stderr, "Can't open def file %s\n", name);
+ fprintf (stderr, "%s: Can't open def file %s\n", program_name, name);
exit (1);
}
{
char *text;
struct dlist *next;
-} dlist_type;
+}
+dlist_type;
typedef struct export
-{
- char *name;
- char *internal_name;
- int ordinal;
- int constant;
- int noname;
- struct export *next;
-}
+ {
+ char *name;
+ char *internal_name;
+ int ordinal;
+ int constant;
+ int noname;
+ struct export *next;
+ }
export_type;
static char *d_name; /* Arg to NAME or LIBRARY */
static int d_ord; /* Base ordinal index */
static export_type *d_exports; /*list of exported functions */
static char *d_suffix = "dll";
-static dlist_type *d_list; /* Descriptions */
-static dlist_type *a_list; /* Stuff to go in directives */
+static dlist_type *d_list; /* Descriptions */
+static dlist_type *a_list; /* Stuff to go in directives */
static int d_is_dll;
static int d_is_exe;
yyerror ()
{
- fprintf (stderr, "Syntax error in def file %s:%d\n",
- def_file, linenumber);
+ fprintf (stderr, "%s: Syntax error in def file %s:%d\n",
+ program_name, def_file, linenumber);
}
void
d_nfuncs++;
}
+
void
def_name (name, base)
char *name;
int base;
{
- printf ("NAME %s base %x\n", name, base);
+ if (verbose)
+ fprintf (stderr, "%s NAME %s base %x\n", program_name, name, base);
if (d_is_dll)
{
fprintf (stderr, "Can't have LIBRARY and NAME\n");
char *name;
int base;
{
- printf ("LIBRARY %s base %x\n", name, base);
+ if (verbose)
+ printf ("%s: LIBRARY %s base %x\n", program_name, name, base);
if (d_is_exe)
{
- fprintf (stderr, "Can't have LIBRARY and NAME\n");
+ fprintf (stderr, "%s: Can't have LIBRARY and NAME\n", program_name);
}
d_name = name;
if (strchr (d_name, '.'))
def_description (desc)
char *desc;
{
- dlist_type *d = (dlist_type *)xmalloc(sizeof(dlist_type));
+ dlist_type *d = (dlist_type *) xmalloc (sizeof (dlist_type));
d->text = strdup (desc);
d->next = d_list;
d_list = d;
}
-void new_directive(dir)
-char *dir;
+void
+new_directive (dir)
+ char *dir;
{
- dlist_type *d = (dlist_type *)xmalloc(sizeof(dlist_type));
+ dlist_type *d = (dlist_type *) xmalloc (sizeof (dlist_type));
d->text = strdup (dir);
d->next = a_list;
a_list = d;
int commit;
{
char b[200];
- if (commit>0)
- sprintf (b,"-stack 0x%x,0x%x ", reserve, commit);
- else
- sprintf (b,"-stack 0x%x ", reserve);
- new_directive (strdup(b));
+ if (commit > 0)
+ sprintf (b, "-stack 0x%x,0x%x ", reserve, commit);
+ else
+ sprintf (b, "-stack 0x%x ", reserve);
+ new_directive (strdup (b));
}
void
int commit;
{
char b[200];
- if (commit>0)
- sprintf (b,"-heap 0x%x,0x%x ", reserve, commit);
- else
- sprintf (b,"-heap 0x%x ", reserve);
- new_directive (strdup(b));
+ if (commit > 0)
+ sprintf (b, "-heap 0x%x,0x%x ", reserve, commit);
+ else
+ sprintf (b, "-heap 0x%x ", reserve);
+ new_directive (strdup (b));
}
char *module;
char *entry;
{
- fprintf (stderr, "IMPORTS are ignored");
+ if (verbose)
+ fprintf (stderr, "%s: IMPORTS are ignored", program_name);
}
void
int attr;
{
char buf[200];
- char atts[5];
+ char atts[5];
char *d = atts;
if (attr & 1)
- *d++= 'R';
+ *d++ = 'R';
if (attr & 2)
*d++ = 'W';
*d++ = 'S';
*d++ = 0;
sprintf (buf, "-attr %s %s", name, atts);
- new_directive (strdup(buf));
+ new_directive (strdup (buf));
}
void
def_code (attr)
int attr;
{
-def_section ("CODE", attr);
+ def_section ("CODE", attr);
}
void
def_data (attr)
int attr;
{
- def_section ("DATA",attr);
+ def_section ("DATA", attr);
}
/**********************************************************************/
+/* read in and block out the base relocations */
+static void
+basenames (abfd)
+ bfd *abfd;
+{
+
+
+
+
+}
+
void
scan_open_obj_file (abfd)
bfd *abfd;
char *p;
char *e;
bfd_get_section_contents (abfd, s, buf, 0, size);
- printf ("Sucking in info from %s\n",
- bfd_get_filename (abfd));
+ if (verbose)
+ fprintf (stderr, "%s: Sucking in info from %s\n",
+ program_name,
+ bfd_get_filename (abfd));
/* Search for -export: strings */
p = buf;
}
free (buf);
}
+
+ basenames (abfd);
+
+ if (verbose)
+ fprintf (stderr, "%s: Done readin\n",
+ program_name);
+
}
if (!f)
{
- fprintf (stderr, "Unable to open object file %s\n", filename);
+ fprintf (stderr, "%s: Unable to open object file %s\n",
+ program_name,
+ filename);
exit (1);
}
if (bfd_check_format (f, bfd_archive))
void
dump_def_info (f)
-FILE *f;
+ FILE *f;
{
int i;
export_type *exp;
- fprintf(f,"%s ", ASM_C);
- for (i= 0; oav[i]; i++)
- fprintf(f,"%s ", oav[i]);
- fprintf(f,"\n");
+ fprintf (f, "%s ", ASM_C);
+ for (i = 0; oav[i]; i++)
+ fprintf (f, "%s ", oav[i]);
+ fprintf (f, "\n");
for (i = 0, exp = d_exports; exp; i++, exp = exp->next)
{
fprintf (f, "%s %d = %s %s @ %d %s%s\n",
ASM_C,
i,
- exp->name,
- exp->internal_name,
+ exp->name,
+ exp->internal_name,
exp->ordinal,
exp->noname ? "NONAME " : "",
exp->constant ? "CONSTANT" : "");
}
/* Generate the .exp file */
+int
+sfunc (a, b)
+ long *a;
+ long *b;
+{
+ return *a - *b;
+}
+
+static void
+flush_page (f, need, page_addr, on_page)
+ FILE *f;
+ long *need;
+ long page_addr;
+ int on_page;
+{
+ int i;
+ /* Flush this page */
+ fprintf (f, "\t%s\t0x%x\t%s Starting RVA for chunk\n",
+ ASM_LONG,
+ page_addr,
+ ASM_C);
+ fprintf (f, "\t%s\t0x%x\t%s Size of block\n",
+ ASM_LONG,
+ (on_page * 2) + (on_page & 1) * 2 + 8,
+ ASM_C);
+ for (i = 0; i < on_page; i++)
+ {
+ fprintf (f, "\t%s\t0x%x\n", ASM_SHORT, need[i] - page_addr | 0x3000);
+ }
+ /* And padding */
+ if (on_page & 1)
+ fprintf (f, "\t%s\t0x%x\n", ASM_SHORT, 0 | 0x0000);
+
+}
+
void
gen_exp_file ()
{
FILE *f;
- char outfile[PATHMAX];
int i;
export_type *exp;
dlist_type *dl;
- sprintf (outfile, "%s.exp.s", outfile_prefix);
+ int had_noname = 0;
+
+ sprintf (outfile, "%s-exp.s", outfile_prefix);
+
+ if (verbose)
+ fprintf (stderr, "%s: Generate exp file %s\n",
+ program_name, outfile_prefix);
f = fopen (outfile, "w");
if (!f)
{
- fprintf (stderr, "Unable to open output file %s\n", outfile);
+ fprintf (stderr, "%s: Unable to open output file %s\n", program_name, outfile);
exit (1);
}
+ if (verbose)
+ {
+ fprintf (stderr, "%s: Opened file %s\n",
+ program_name, outfile);
+ }
+
dump_def_info (f);
fprintf (f, "\t.section .edata\n\n");
fprintf (f, "\t%s 0 %s Allways 0\n", ASM_LONG, ASM_C);
i = d_ord;
for (exp = d_exports; exp; exp = exp->next)
{
+#if 0
+ /* This seems necessary in the doc, but in real
+ life it's not used.. */
if (exp->ordinal != i)
{
- fprintf (f, "\t%s\t%d\t@ %d..%d missing\n", ASM_SPACE,
+ fprintf (f, "%s\t%s\t%d\t@ %d..%d missing\n", ASM_C, ASM_SPACE,
(exp->ordinal - i) * 4,
i, exp->ordinal - 1);
i = exp->ordinal;
}
+#endif
fprintf (f, "\t%s %s\t%s %d\n", ASM_LONG, exp->internal_name, ASM_C, exp->ordinal);
i++;
}
fprintf (f, "anames:\n");
for (i = 0, exp = d_exports; exp; i++, exp = exp->next)
- if (exp->noname)
- fprintf (f, "\t%s 0\t%sNoname\n", ASM_LONG, ASM_C);
- else
- fprintf (f, "\t%s n%d\n", ASM_LONG, i);
+ {
+ if (exp->noname)
+ {
+ had_noname = 1;
+ fprintf (f, "\t%s nNoname\n", ASM_LONG, ASM_C);
+ }
+ else
+ {
+ fprintf (f, "\t%s n%d\n", ASM_LONG, i);
+ }
+ }
fprintf (f, "anords:\n");
for (exp = d_exports; exp; exp = exp->next)
else
fprintf (f, "n%d: %s \"%s\"\n", i, ASM_TEXT, exp->name);
+ if (had_noname)
+ fprintf (f, "nNoname: %s \"__noname__\"\n", ASM_TEXT);
if (a_list)
{
- fprintf(f,"\t.section .drectve\n");
+ fprintf (f, "\t.section .drectve\n");
for (dl = a_list; dl; dl = dl->next)
{
- fprintf (f,"\t%s\t\"%s\"\n", ASM_TEXT, dl->text);
+ fprintf (f, "\t%s\t\"%s\"\n", ASM_TEXT, dl->text);
}
}
- if (d_list)
+ if (d_list)
{
- fprintf(f,"\t.section .rdata\n");
+ fprintf (f, "\t.section .rdata\n");
for (dl = d_list; dl; dl = dl->next)
{
char *p;
be quote characters in the string */
l = 0;
- for (p = dl->text; *p; p++) {
- if (l == 0)
- fprintf(f,"\t%s\t", ASM_BYTE);
- else
- fprintf(f,",");
- fprintf(f,"%d", *p);
- if (p[1] == 0) {
- fprintf(f,",0\n");
- break;
+ for (p = dl->text; *p; p++)
+ {
+ if (l == 0)
+ fprintf (f, "\t%s\t", ASM_BYTE);
+ else
+ fprintf (f, ",");
+ fprintf (f, "%d", *p);
+ if (p[1] == 0)
+ {
+ fprintf (f, ",0\n");
+ break;
+ }
+ if (++l == 10)
+ {
+ fprintf (f, "\n");
+ l = 0;
+ }
}
- if (++l == 10) {
- fprintf(f,"\n");
- l = 0;
+ }
+ }
+
+ /* Dump the reloc section if a base file is provided */
+ if (base_file)
+ {
+ int addr;
+ long need[PAGE_SIZE];
+ long page_addr;
+ int numbytes;
+ int num_entries;
+ long *copy;
+ int j;
+ int on_page;
+ fprintf (f, "\t.section\t.reloc\n");
+ fseek (base_file, 0, SEEK_END);
+ numbytes = ftell (base_file);
+ fseek (base_file, 0, SEEK_SET);
+ copy = malloc (numbytes);
+ fread (copy, 1, numbytes, base_file);
+ num_entries = numbytes / sizeof (long);
+
+ qsort (copy, num_entries, sizeof (long), sfunc);
+
+ addr = copy[0];
+ page_addr = addr & PAGE_MASK; /* work out the page addr */
+ on_page = 0;
+ for (j = 0; j < num_entries; j++)
+ {
+ addr = copy[j];
+ if ((addr & PAGE_MASK) != page_addr)
+ {
+ flush_page (f, need, page_addr, on_page);
+ on_page = 0;
+ page_addr = addr & PAGE_MASK;
}
- }
+ need[on_page++] = addr;
}
+ flush_page (f, need, page_addr, on_page);
}
+
fclose (f);
}
+static char *
+xlate (char *name)
+{
+ char *p;
+ if (!suckunderscore)
+ return name;
+
+ if (name[0] == '_')
+ name++;
+ p = strchr (name, '@');
+ if (p)
+ *p = 0;
+ return name;
+}
+
/**********************************************************************/
gen_lib_file ()
{
- char outfile[PATHMAX];
int i;
FILE *f;
export_type *exp;
- sprintf (outfile, "%s.lib.s", outfile_prefix);
+ sprintf (outfile, "%s-lib.s", outfile_prefix);
f = fopen (outfile, "w");
if (!f)
for (i = 0, exp = d_exports; exp; i++, exp = exp->next)
{
fprintf (f, "\t%s\t%s\n", ASM_GLOBAL, exp->name);
- fprintf (f, "\t%s\t__imp__%s\n", ASM_GLOBAL, exp->name);
+ fprintf (f, "\t%s\t__imp_%s\n", ASM_GLOBAL, exp->name);
}
for (i = 0, exp = d_exports; exp; i++, exp = exp->next)
{
- fprintf (f, "%s:\t%s\t__imp__%s\n", exp->name, ASM_JUMP, exp->name);
+ fprintf (f, "%s:\t%s\t__imp_%s\n", exp->name, ASM_JUMP, exp->name);
}
fprintf (f, "\t%s\tiname\t%s imported dll's name\n", ASM_LONG, ASM_C);
fprintf (f, "\t%s\tfthunk\t%s pointer to firstthunk\n", ASM_LONG, ASM_C);
+ fprintf (f, "%sStuff for compatibility\n", ASM_C);
+ fprintf (f, "\t.section\t.idata$3\n");
+ fprintf (f, "\t%s\t0\n", ASM_LONG);
+ fprintf (f, "\t%s\t0\n", ASM_LONG);
+ fprintf (f, "\t%s\t0\n", ASM_LONG);
+ fprintf (f, "\t%s\t0\n", ASM_LONG);
+ fprintf (f, "\t%s\t0\n", ASM_LONG);
+ fprintf (f, "\t.section\t.idata$5\n");
+ fprintf (f, "\t%s\t0\n", ASM_LONG);
+
+ fprintf (f, "\t.section\t.idata$4\n");
+ fprintf (f, "\t%s\t0\n", ASM_LONG);
+
fprintf (f, "\n%s Loader modifies this\n", ASM_C);
fprintf (f, "\t.section .idata$5\n");
fprintf (f, "fthunk:\n");
for (i = 0, exp = d_exports; exp; i++, exp = exp->next)
{
- fprintf (f, "__imp__%s:\n", exp->name);
+ fprintf (f, "__imp_%s:\n", exp->name);
fprintf (f, "\t%s\tID%d\n", ASM_LONG, i);
}
+ fprintf (f, "\t%s\t0\n", ASM_LONG);
fprintf (f, "\n%s Hint name array\n", ASM_C);
fprintf (f, "\t.section .idata$4\n");
fprintf (f, "\t%s\tID%d\n", ASM_LONG, i);
}
+ fprintf (f, "\t%s\t0\n", ASM_LONG);
fprintf (f, "%s Hint/name array storage and import dll name\n", ASM_C);
fprintf (f, "\t.section .idata$6\n");
for (i = 0, exp = d_exports; exp; i++, exp = exp->next)
{
fprintf (f, "ID%d:\t%s\t%d\n", i, ASM_SHORT, exp->ordinal);
- fprintf (f, "\t%s\t\"%s\"\n", ASM_TEXT, exp->name);
+ fprintf (f, "\t%s\t\"%s\"\n", ASM_TEXT, xlate (exp->name));
}
-
+ fprintf (f, "\t%s\t0\n", ASM_LONG);
fprintf (f, "iname:\t%s\t\"%s.%s\"\n", ASM_TEXT, outfile_prefix, d_suffix);
fclose (f);
}
export_type *b = d_export_vec[i + 1];
more = 1;
-
- fprintf (stderr, "warning, ignoring duplicate EXPORT %s\n",
- a->name);
+ if (verbose)
+ fprintf (stderr, "Warning, ignoring duplicate EXPORT %s %d,%d\n",
+ a->name,
+ a->ordinal,
+ b->ordinal);
if (a->ordinal != -1
&& b->ordinal != -1)
{
export_type **d_export_vec;
{
int lowest = 0;
+ int unset = 0;
+ char *ptr;
qsort (d_export_vec, d_nfuncs, sizeof (export_type *), pfunc);
- /* fill in the unset ordinals with ones from the minimum */
+ /* fill in the unset ordinals with ones from our range */
+
+ ptr = (char *) malloc (65536);
+
+ memset (ptr, 65536, 0);
+
+ /* Mark in our large vector all the numbers that are taken */
for (i = 0; i < d_nfuncs; i++)
{
- if (d_export_vec[i]->ordinal == -1)
+ if (d_export_vec[i]->ordinal != -1)
{
- d_export_vec[i]->ordinal = lowest++;
+ ptr[d_export_vec[i]->ordinal] = 1;
+ if (lowest == 0)
+ lowest = d_export_vec[i]->ordinal;
}
- else
+ }
+
+ for (i = 0; i < d_nfuncs; i++)
+ {
+ if (d_export_vec[i]->ordinal == -1)
{
- if (lowest == d_export_vec[i]->ordinal)
- {
- fprintf (stderr, "Warning, Duplicate ordinal %s @ %d\n",
- d_export_vec[i]->name,
- d_export_vec[i]->ordinal);
- }
- lowest = d_export_vec[i]->ordinal + 1;
+ int j;
+ for (j = lowest; j < 65536; j++)
+ if (ptr[j] == 0)
+ {
+ ptr[j] = 1;
+ d_export_vec[i]->ordinal = j;
+ goto done;
+ }
+
+ for (j = 1; j < lowest; j++)
+ if (ptr[j] == 0)
+ {
+ ptr[j] = 1;
+ d_export_vec[i]->ordinal = j;
+ goto done;
+ }
+ done:;
+
}
}
+ free (ptr);
+
+ /* And resort */
+
+ qsort (d_export_vec, d_nfuncs, sizeof (export_type *), pfunc);
+
/* Work out the lowest ordinal number */
if (d_export_vec[0])
d_ord = d_export_vec[0]->ordinal;
FILE *file;
int status;
{
- fprintf (file, "Usage %s [-m|--machine machine] [-o outprefix] [-d|--def deffile] [--def deffile]\n", program_name);
+ fprintf (file, "Usage %s <options> <object-files>\n", program_name);
+ fprintf (file, "\t -m <machine> Generate code for <machine>\n");
+ fprintf (file, "\t --machine <machine>\n");
+ fprintf (file, "\t -o <outprefix> Set output prefix\n");
+ fprintf (file, "\t -d <deffile> Name input .def file\n");
+ fprintf (file, "\t --def <deffile> \n");
+ fprintf (file, "\t --base-file <basefile> Read linker generated base file\n");
+ fprintf (file, "\t -b <basefile> \n");
+ fprintf (file, "\t -v Verbose\n");
+
exit (status);
}
static struct option long_options[] =
{
{"def", required_argument, NULL, 'd'},
+ {"underscore", no_argument, NULL, 'u'},
{"help", no_argument, NULL, 'h'},
{"machine", required_argument, NULL, 'm'},
+ {"base-file", required_argument, NULL, 'b'},
0
};
char *firstarg = 0;
program_name = av[0];
oav = av;
-
- while ((c = getopt_long (ac, av, "h?m:o:Dd:", long_options, 0)) != EOF)
+
+ while ((c = getopt_long (ac, av, "vbuh?m:o:Dd:", long_options, 0)) != EOF)
{
switch (c)
{
case 'h':
case '?':
- usage(stderr,0);
+ usage (stderr, 0);
break;
case 'm':
mname = optarg;
case 'o':
outfile_prefix = optarg;
break;
+ case 'v':
+ verbose = 1;
+ break;
case 'D':
yydebug = 1;
break;
+ case 'u':
+ suckunderscore = 1;
+ break;
case 'd':
def_file = optarg;
break;
+ case 'b':
+ base_file = fopen (optarg, "r");
+ if (!base_file)
+ {
+ fprintf (stderr, "%s: Unable to open base-file %s\n",
+ av[0],
+ optarg);
+ exit (1);
+ }
+ break;
default:
usage (stderr, 1);
}
}
- for (i = 0; mtable[i].type; i++)
+ for (i = 0; mtable[i].type; i++)
{
if (strcmp (mtable[i].type, mname) == 0)
break;
}
- if (!mtable[i].type)
+ if (!mtable[i].type)
{
- fprintf(stderr,"Machine not supported\n");
- exit(1);
+ fprintf (stderr, "Machine not supported\n");
+ exit (1);
}
machine = i;
if (def_file)
{
-
process_def_file (def_file);
}
while (optind < ac)
}
outfile_prefix = prefix (outfile_prefix);
+ if (verbose)
+ fprintf (stderr, "%s: Outfile prefix is %s\n",
+ program_name, outfile_prefix);
mangle_defs ();
+
gen_exp_file ();
+
+
gen_lib_file ();
+
return 0;
}
All debugging information is preserved */
#include <bfd.h>
-#include <getopt.h>
-#include <stdio.h>
-#include <time.h>
-#include <libiberty.h>
+#include "sysdep.h"
#include "bucomm.h"
#include "sysroff.h"
#include "coffgrok.h"
+#include <libiberty.h>
+#include <getopt.h>
#include "coff/internal.h"
#include "../bfd/libcoff.h"
#define PROGRAM_VERSION "1.5"
-/*#define FOOP1 1*/
+/*#define FOOP1 1 */
static int sh;
static int h8300;
static bfd *abfd;
static int debug = 0;
static int quick = 0;
-
+static int noprescan = 0;
static struct coff_ofile *tree;
+/* Obsolete ??
+ static int absolute_p;
+ */
-static int absolute_p;
static int segmented_p;
static int code;
}
static void
-wr_un (ptr, sfile, first)
+wr_un (ptr, sfile, first, nsecs)
struct coff_ofile *ptr;
struct coff_sfile *sfile;
int first;
+ int nsecs;
{
struct IT_un un;
else
un.format = FORMAT_OM;
un.spare1 = 0;
+
+
+#if 0
un.nsections = ptr->nsections - 1; /* Don't count the abs section */
+#else
+ /*NEW - only count sections with size */
+ un.nsections = nsecs;
+#endif
+
un.nextdefs = 0;
un.nextrefs = 0;
/* Count all the undefined and defined variables with global scope */
unsigned char stuff[200];
i = 0;
- while (i < section->size)
+ while (i < section->bfd_section->_raw_size)
{
struct IT_ob ob;
int todo = 200; /* Copy in 200 byte lumps */
ob.spare = 0;
- if (i + todo > section->size)
- todo = section->size - i;
+ if (i + todo > section->bfd_section->_raw_size)
+ todo = section->bfd_section->_raw_size - i;
if (first)
{
ob.data.len = todo;
bfd_get_section_contents (abfd, section->bfd_section, stuff, i, todo);
ob.data.data = stuff;
- sysroff_swap_ob_out (file, &ob /*, i + todo < section->size*/ );
+ sysroff_swap_ob_out (file, &ob /*, i + todo < section->size */ );
i += todo;
}
+ /* Now fill the rest with blanks */
+ while (i < section->size)
+ {
+ struct IT_ob ob;
+ int todo = 200; /* Copy in 200 byte lumps */
+ ob.spare = 0;
+ if (i + todo > section->size)
+ todo = section->size - i;
+ ob.saf = 0;
+
+ ob.cpf = 0; /* Never compress */
+ ob.data.len = todo;
+ memset (stuff, 0, todo);
+ ob.data.data = stuff;
+ sysroff_swap_ob_out (file, &ob);
+ i += todo;
+ }
+ /* Now fill the rest with blanks */
+
}
static void
}
}
-static void
-dty_start ()
-{
- struct IT_dty dty;
- dty.end = 0;
- dty.neg = 0x1001;
- dty.spare = 0;
- sysroff_swap_dty_out (file, &dty);
-}
-
-static void
-dty_stop ()
-{
- struct IT_dty dty;
- dty.end = 0;
- dty.neg = 0x1001;
- dty.end = 1;
- sysroff_swap_dty_out (file, &dty);
-}
-
-
-static void
-dump_tree_structure (sfile, symbol, type, nest)
- struct coff_sfile *sfile;
- struct coff_symbol *symbol;
- struct coff_type *type;
- int nest;
-{
- if (symbol->type->type == coff_function_type)
- {
-
-
- }
-
-}
-
+/* Obsolete ?
+ static void
+ dty_start ()
+ {
+ struct IT_dty dty;
+ dty.end = 0;
+ dty.neg = 0x1001;
+ dty.spare = 0;
+ sysroff_swap_dty_out (file, &dty);
+ }
+
+ static void
+ dty_stop ()
+ {
+ struct IT_dty dty;
+ dty.end = 0;
+ dty.neg = 0x1001;
+ dty.end = 1;
+ sysroff_swap_dty_out (file, &dty);
+ }
+
+
+ static void
+ dump_tree_structure (sfile, symbol, type, nest)
+ struct coff_sfile *sfile;
+ struct coff_symbol *symbol;
+ struct coff_type *type;
+ int nest;
+ {
+ if (symbol->type->type == coff_function_type)
+ {
+
+
+ }
+
+ }
+ */
static void
walk_tree_type (sfile, symbol, type, nest)
break;
case coff_where_member_of_enum:
/* dsy.bitunit = 0;
- dsy.field_len = symbol->type->size;
- dsy.field_off = symbol->where->offset;*/
+ dsy.field_len = symbol->type->size;
+ dsy.field_off = symbol->where->offset; */
break;
case coff_where_register:
case coff_where_unknown:
}
/* Find the offset of the .text section for this sfile in the
- .text section for the output file */
+ .text section for the output file */
static int
find_base (sfile, section)
for all the sections which appear in the output file, even
if there isn't an equivalent one on the input */
-static void
+static int
wr_sc (ptr, sfile)
struct coff_ofile *ptr;
struct coff_sfile *sfile;
{
int i;
+int scount = 0;
/* First work out the total number of sections */
int total_sec = ptr->nsections;
struct coff_symbol *symbol;
struct myinfo *info
- = (struct myinfo *) calloc (total_sec, sizeof (struct myinfo));
+ = (struct myinfo *) calloc (total_sec, sizeof (struct myinfo));
{
sc.contents = CONTENTS_CODE;
}
-
-
- sysroff_swap_sc_out (file, &sc);
-
-
-
+ /* NEW */
+ if (sc.length) {
+ sysroff_swap_sc_out (file, &sc);
+ scount++;
+ }
}
-
+return scount;
}
static void
sfile;
sfile = sfile->next)
{
- wr_un (ptr, sfile, first);
- wr_sc (ptr, sfile);
+ long p1;
+ long p2;
+ int nsecs;
+ p1 = ftell (file);
+ wr_un (ptr, sfile, first, 0);
+ nsecs = wr_sc (ptr, sfile);
+ p2 = ftell (file);
+ fseek (file, p1, 0);
+ wr_un (ptr, sfile, first, nsecs);
+ fseek (file, p2, 0);
wr_er (ptr, sfile, first);
wr_ed (ptr, sfile, first);
first = 0;
if (s->visible->type == coff_vis_common)
{
struct coff_where *w = s->where;
- /* s->visible->type = coff_vis_ext_def; leave it as common */
+ /* s->visible->type = coff_vis_ext_def; leave it as common */
common_section->size = align (common_section->size);
w->offset = common_section->size + common_section->address;
w->section = common_section;
{
{"debug", no_argument, 0, 'd'},
{"quick", no_argument, 0, 'q'},
+ {"noprescan", no_argument, 0, 'n'},
{"help", no_argument, 0, 'h'},
{"version", no_argument, 0, 'V'},
{NULL, no_argument, 0, 0}
program_name = av[0];
xmalloc_set_program_name (program_name);
- while ((opt = getopt_long (ac, av, "dhVq", long_options,
+ while ((opt = getopt_long (ac, av, "dhVqn", long_options,
(int *) NULL))
!= EOF)
{
case 'q':
quick = 1;
break;
+ case 'n':
+ noprescan = 1;
+ break;
case 'd':
debug = 1;
break;
case 'h':
show_help ();
- /*NOTREACHED*/
+ /*NOTREACHED */
case 'V':
printf ("GNU %s version %s\n", program_name, PROGRAM_VERSION);
exit (0);
- /*NOTREACHED*/
+ /*NOTREACHED */
case 0:
break;
default:
show_usage (stderr, 1);
- /*NOTREACHED*/
+ /*NOTREACHED */
}
}
if (!output_file)
{
/* Take a .o off the input file and stick on a .obj. If
- it doesn't end in .o, then stick a .obj on anyway */
+ it doesn't end in .o, then stick a .obj on anyway */
int len = strlen (input_file);
output_file = xmalloc (len + 5);
if (debug)
printf ("ids %d %d\n", base1, base2);
tree = coff_grok (abfd);
- prescan (tree);
+ if (!noprescan)
+ prescan (tree);
wr_module (tree);
return 0;
}
projects / binutils-gdb.git / commitdiff