mirrors/binutils-gdb
1
0
Fork 0
mirror of https://sourceware.org/git/binutils-gdb.git synced 2024-11-25 02:53:48 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity
65fd87bce3
binutils-gdb/ld/deffilep.y
Nick Clifton a6483292c2 Added year 2000 copyright notice
2000-01-04 23:33:29 +00:00

1025 lines
23 KiB
Plaintext
Raw Blame History

%{ /* deffilep.y - parser for .def files */
/* Copyright (C) 1995, 1997, 1998, 1999, 2000 Free Software Foundation, Inc.
This file is part of GNU Binutils.
This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 2 of the License, or
(at your option) any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program; if not, write to the Free Software
Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. */
#include <stdio.h>
#include <ctype.h>
#include "libiberty.h"
#include "bfd.h"
#include "sysdep.h"
#include "ld.h"
#include "ldmisc.h"
#include "deffile.h"
#define TRACE 0
#define ROUND_UP(a, b) (((a)+((b)-1))&~((b)-1))
/* Remap normal yacc parser interface names (yyparse, yylex, yyerror, etc),
as well as gratuitiously global symbol names, so we can have multiple
yacc generated parsers in ld. Note that these are only the variables
produced by yacc. If other parser generators (bison, byacc, etc) produce
additional global names that conflict at link time, then those parser
generators need to be fixed instead of adding those names to this list. */
#define yymaxdepth def_maxdepth
#define yyparse def_parse
#define yylex def_lex
#define yyerror def_error
#define yylval def_lval
#define yychar def_char
#define yydebug def_debug
#define yypact def_pact
#define yyr1 def_r1
#define yyr2 def_r2
#define yydef def_def
#define yychk def_chk
#define yypgo def_pgo
#define yyact def_act
#define yyexca def_exca
#define yyerrflag def_errflag
#define yynerrs def_nerrs
#define yyps def_ps
#define yypv def_pv
#define yys def_s
#define yy_yys def_yys
#define yystate def_state
#define yytmp def_tmp
#define yyv def_v
#define yy_yyv def_yyv
#define yyval def_val
#define yylloc def_lloc
#define yyreds def_reds /* With YYDEBUG defined */
#define yytoks def_toks /* With YYDEBUG defined */
#define yylhs def_yylhs
#define yylen def_yylen
#define yydefred def_yydefred
#define yydgoto def_yydgoto
#define yysindex def_yysindex
#define yyrindex def_yyrindex
#define yygindex def_yygindex
#define yytable def_yytable
#define yycheck def_yycheck
static int def_lex ();
static void def_description PARAMS ((const char *));
static void def_exports PARAMS ((const char *, const char *, int, int));
static void def_heapsize PARAMS ((int, int));
static void def_import
PARAMS ((const char *, const char *, const char *, const char *, int));
static void def_library PARAMS ((const char *, int));
static void def_name PARAMS ((const char *, int));
static void def_section PARAMS ((const char *, int));
static void def_section_alt PARAMS ((const char *, const char *));
static void def_stacksize PARAMS ((int, int));
static void def_version PARAMS ((int, int));
static void def_directive PARAMS ((char *));
static int def_parse PARAMS ((void));
static int def_error PARAMS ((const char *));
static int def_lex PARAMS ((void));
static int lex_forced_token = 0;
static const char *lex_parse_string = 0;
static const char *lex_parse_string_end = 0;
%}
%union {
char *id;
int number;
};
%token NAME, LIBRARY, DESCRIPTION, STACKSIZE, HEAPSIZE, CODE, DATAU, DATAL
%token SECTIONS, EXPORTS, IMPORTS, VERSIONK, BASE, CONSTANTU, CONSTANTL
%token PRIVATEU, PRIVATEL
%token READ WRITE EXECUTE SHARED NONAMEU NONAMEL DIRECTIVE
%token <id> ID
%token <number> NUMBER
%type <number> opt_base opt_ordinal
%type <number> attr attr_list opt_number exp_opt_list exp_opt
%type <id> opt_name opt_equal_name
%%
start: start command
| command
;
command:
NAME opt_name opt_base { def_name ($2, $3); }
| LIBRARY opt_name opt_base { def_library ($2, $3); }
| DESCRIPTION ID { def_description ($2);}
| STACKSIZE NUMBER opt_number { def_stacksize ($2, $3);}
| HEAPSIZE NUMBER opt_number { def_heapsize ($2, $3);}
| CODE attr_list { def_section ("CODE", $2);}
| DATAU attr_list { def_section ("DATA", $2);}
| SECTIONS seclist
| EXPORTS explist
| IMPORTS implist
| VERSIONK NUMBER { def_version ($2, 0);}
| VERSIONK NUMBER '.' NUMBER { def_version ($2, $4);}
| DIRECTIVE ID { def_directive ($2);}
;
explist:
/* EMPTY */
| expline
| explist expline
;
expline:
/* The opt_comma is necessary to support both the usual
DEF file syntax as well as .drectve syntax which
mandates <expsym>,<expoptlist>. */
ID opt_equal_name opt_ordinal opt_comma exp_opt_list
{ def_exports ($1, $2, $3, $5); }
;
exp_opt_list:
/* The opt_comma is necessary to support both the usual
DEF file syntax as well as .drectve syntax which
allows for comma separated opt list. */
exp_opt opt_comma exp_opt_list { $$ = $1 | $3; }
| { $$ = 0; }
;
exp_opt:
NONAMEU { $$ = 1; }
| NONAMEL { $$ = 1; }
| CONSTANTU { $$ = 2; }
| CONSTANTL { $$ = 2; }
| DATAU { $$ = 4; }
| DATAL { $$ = 4; }
| PRIVATEU { $$ = 8; }
| PRIVATEL { $$ = 8; }
;
implist:
implist impline
| impline
;
impline:
ID '=' ID '.' ID '.' ID { def_import ($1, $3, $5, $7, -1); }
| ID '=' ID '.' ID '.' NUMBER { def_import ($1, $3, $5, 0, $7); }
| ID '=' ID '.' ID { def_import ($1, $3, 0, $5, -1); }
| ID '=' ID '.' NUMBER { def_import ($1, $3, 0, 0, $5); }
| ID '.' ID '.' ID { def_import ( 0, $1, $3, $5, -1); }
| ID '.' ID { def_import ( 0, $1, 0, $3, -1); }
;
seclist:
seclist secline
| secline
;
secline:
ID attr_list { def_section ($1, $2);}
| ID ID { def_section_alt ($1, $2);}
;
attr_list:
attr_list opt_comma attr { $$ = $1 | $3; }
| attr { $$ = $1; }
;
opt_comma:
','
|
;
opt_number: ',' NUMBER { $$=$2;}
| { $$=-1;}
;
attr:
READ { $$ = 1;}
| WRITE { $$ = 2;}
| EXECUTE { $$=4;}
| SHARED { $$=8;}
;
opt_name: ID { $$ = $1; }
| ID '.' ID
{
char * name = xmalloc (strlen ($1) + 1 + strlen ($3) + 1);
sprintf (name, "%s.%s", $1, $3);
$$ = name;
}
| { $$ = ""; }
;
opt_ordinal:
'@' NUMBER { $$ = $2;}
| { $$ = -1;}
;
opt_equal_name:
'=' ID { $$ = $2; }
| { $$ = 0; }
;
opt_base: BASE '=' NUMBER { $$ = $3;}
| { $$ = 0;}
;
%%
/*****************************************************************************
API
*****************************************************************************/
static FILE *the_file;
static const char *def_filename;
static int linenumber;
static def_file *def;
static int saw_newline;
struct directive
{
struct directive *next;
char *name;
int len;
};
static struct directive *directives = 0;
def_file *
def_file_empty ()
{
def_file *rv = (def_file *) xmalloc (sizeof (def_file));
memset (rv, 0, sizeof (def_file));
rv->is_dll = -1;
rv->base_address = (bfd_vma) (-1);
rv->stack_reserve = rv->stack_commit = -1;
rv->heap_reserve = rv->heap_commit = -1;
rv->version_major = rv->version_minor = -1;
return rv;
}
def_file *
def_file_parse (filename, add_to)
const char *filename;
def_file *add_to;
{
struct directive *d;
the_file = fopen (filename, "r");
def_filename = filename;
linenumber = 1;
if (!the_file)
{
perror (filename);
return 0;
}
if (add_to)
{
def = add_to;
}
else
{
def = def_file_empty ();
}
saw_newline = 1;
if (def_parse ())
{
def_file_free (def);
fclose (the_file);
return 0;
}
fclose (the_file);
for (d = directives; d; d = d->next)
{
#if TRACE
printf ("Adding directive %08x `%s'\n", d->name, d->name);
#endif
def_file_add_directive (def, d->name, d->len);
}
return def;
}
void
def_file_free (def)
def_file *def;
{
int i;
if (!def)
return;
if (def->name)
free (def->name);
if (def->description)
free (def->description);
if (def->section_defs)
{
for (i = 0; i < def->num_section_defs; i++)
{
if (def->section_defs[i].name)
free (def->section_defs[i].name);
if (def->section_defs[i].class)
free (def->section_defs[i].class);
}
free (def->section_defs);
}
if (def->exports)
{
for (i = 0; i < def->num_exports; i++)
{
if (def->exports[i].internal_name
&& def->exports[i].internal_name != def->exports[i].name)
free (def->exports[i].internal_name);
if (def->exports[i].name)
free (def->exports[i].name);
}
free (def->exports);
}
if (def->imports)
{
for (i = 0; i < def->num_imports; i++)
{
if (def->imports[i].internal_name
&& def->imports[i].internal_name != def->imports[i].name)
free (def->imports[i].internal_name);
if (def->imports[i].name)
free (def->imports[i].name);
}
free (def->imports);
}
while (def->modules)
{
def_file_module *m = def->modules;
def->modules = def->modules->next;
free (m);
}
free (def);
}
#ifdef DEF_FILE_PRINT
void
def_file_print (file, def)
FILE *file;
def_file *def;
{
int i;
fprintf (file, ">>>> def_file at 0x%08x\n", def);
if (def->name)
fprintf (file, " name: %s\n", def->name ? def->name : "(unspecified)");
if (def->is_dll != -1)
fprintf (file, " is dll: %s\n", def->is_dll ? "yes" : "no");
if (def->base_address != (bfd_vma) (-1))
fprintf (file, " base address: 0x%08x\n", def->base_address);
if (def->description)
fprintf (file, " description: `%s'\n", def->description);
if (def->stack_reserve != -1)
fprintf (file, " stack reserve: 0x%08x\n", def->stack_reserve);
if (def->stack_commit != -1)
fprintf (file, " stack commit: 0x%08x\n", def->stack_commit);
if (def->heap_reserve != -1)
fprintf (file, " heap reserve: 0x%08x\n", def->heap_reserve);
if (def->heap_commit != -1)
fprintf (file, " heap commit: 0x%08x\n", def->heap_commit);
if (def->num_section_defs > 0)
{
fprintf (file, " section defs:\n");
for (i = 0; i < def->num_section_defs; i++)
{
fprintf (file, " name: `%s', class: `%s', flags:",
def->section_defs[i].name, def->section_defs[i].class);
if (def->section_defs[i].flag_read)
fprintf (file, " R");
if (def->section_defs[i].flag_write)
fprintf (file, " W");
if (def->section_defs[i].flag_execute)
fprintf (file, " X");
if (def->section_defs[i].flag_shared)
fprintf (file, " S");
fprintf (file, "\n");
}
}
if (def->num_exports > 0)
{
fprintf (file, " exports:\n");
for (i = 0; i < def->num_exports; i++)
{
fprintf (file, " name: `%s', int: `%s', ordinal: %d, flags:",
def->exports[i].name, def->exports[i].internal_name,
def->exports[i].ordinal);
if (def->exports[i].flag_private)
fprintf (file, " P");
if (def->exports[i].flag_constant)
fprintf (file, " C");
if (def->exports[i].flag_noname)
fprintf (file, " N");
if (def->exports[i].flag_data)
fprintf (file, " D");
fprintf (file, "\n");
}
}
if (def->num_imports > 0)
{
fprintf (file, " imports:\n");
for (i = 0; i < def->num_imports; i++)
{
fprintf (file, " int: %s, from: `%s', name: `%s', ordinal: %d\n",
def->imports[i].internal_name,
def->imports[i].module,
def->imports[i].name,
def->imports[i].ordinal);
}
}
if (def->version_major != -1)
fprintf (file, " version: %d.%d\n", def->version_major, def->version_minor);
fprintf (file, "<<<< def_file at 0x%08x\n", def);
}
#endif
def_file_export *
def_file_add_export (def, external_name, internal_name, ordinal)
def_file *def;
const char *external_name;
const char *internal_name;
int ordinal;
{
def_file_export *e;
int max_exports = ROUND_UP(def->num_exports, 32);
if (def->num_exports >= max_exports)
{
max_exports = ROUND_UP(def->num_exports+1, 32);
if (def->exports)
def->exports = (def_file_export *) xrealloc (def->exports, max_exports * sizeof (def_file_export));
else
def->exports = (def_file_export *) xmalloc (max_exports * sizeof (def_file_export));
}
e = def->exports + def->num_exports;
memset (e, 0, sizeof (def_file_export));
if (internal_name && !external_name)
external_name = internal_name;
if (external_name && !internal_name)
internal_name = external_name;
e->name = xstrdup (external_name);
e->internal_name = xstrdup (internal_name);
e->ordinal = ordinal;
def->num_exports++;
return e;
}
static def_file_module *
def_stash_module (def, name)
def_file *def;
char *name;
{
def_file_module *s;
for (s=def->modules; s; s=s->next)
if (strcmp (s->name, name) == 0)
return s;
s = (def_file_module *) xmalloc (sizeof (def_file_module) + strlen (name));
s->next = def->modules;
def->modules = s;
s->user_data = 0;
strcpy (s->name, name);
return s;
}
def_file_import *
def_file_add_import (def, name, module, ordinal, internal_name)
def_file *def;
const char *name;
const char *module;
int ordinal;
const char *internal_name;
{
def_file_import *i;
int max_imports = ROUND_UP(def->num_imports, 16);
if (def->num_imports >= max_imports)
{
max_imports = ROUND_UP(def->num_imports+1, 16);
if (def->imports)
def->imports = (def_file_import *) xrealloc (def->imports, max_imports * sizeof (def_file_import));
else
def->imports = (def_file_import *) xmalloc (max_imports * sizeof (def_file_import));
}
i = def->imports + def->num_imports;
memset (i, 0, sizeof (def_file_import));
if (name)
i->name = xstrdup (name);
if (module)
i->module = def_stash_module(def, module);
i->ordinal = ordinal;
if (internal_name)
i->internal_name = xstrdup (internal_name);
else
i->internal_name = i->name;
def->num_imports++;
return i;
}
struct
{
char *param;
int token;
}
diropts[] =
{
{ "-heap", HEAPSIZE },
{ "-stack", STACKSIZE },
{ "-attr", SECTIONS },
{ "-export", EXPORTS },
{ 0, 0 }
};
void
def_file_add_directive (my_def, param, len)
def_file *my_def;
const char *param;
int len;
{
def_file *save_def = def;
const char *pend = param + len;
const char *tend = param;
int i;
def = my_def;
while (param < pend)
{
while (param < pend && isspace (*param))
param++;
for (tend = param + 1;
tend < pend && !(isspace (tend[-1]) && *tend == '-');
tend++);
for (i = 0; diropts[i].param; i++)
{
int len = strlen (diropts[i].param);
if (tend - param >= len
&& strncmp (param, diropts[i].param, len) == 0
&& (param[len] == ':' || param[len] == ' '))
{
lex_parse_string_end = tend;
lex_parse_string = param + len + 1;
lex_forced_token = diropts[i].token;
saw_newline = 0;
def_parse ();
break;
}
}
if (!diropts[i].param)
{
/* xgettext:c-format */
einfo (_("Warning: .drectve `%.*s' unrecognized\n"),
tend - param, param);
}
lex_parse_string = 0;
param = tend;
}
def = save_def;
}
/*****************************************************************************
Parser Callbacks
*****************************************************************************/
static void
def_name (name, base)
const char *name;
int base;
{
if (def->name)
free (def->name);
def->name = xstrdup (name);
def->base_address = base;
def->is_dll = 0;
}
static void
def_library (name, base)
const char *name;
int base;
{
if (def->name)
free (def->name);
def->name = xstrdup (name);
def->base_address = base;
def->is_dll = 1;
}
static void
def_description (text)
const char *text;
{
int len = def->description ? strlen (def->description) : 0;
len += strlen (text) + 1;
if (def->description)
{
def->description = (char *) xrealloc (def->description, len);
strcat (def->description, text);
}
else
{
def->description = (char *) xmalloc (len);
strcpy (def->description, text);
}
}
static void
def_stacksize (reserve, commit)
int reserve;
int commit;
{
def->stack_reserve = reserve;
def->stack_commit = commit;
}
static void
def_heapsize (reserve, commit)
int reserve;
int commit;
{
def->heap_reserve = reserve;
def->heap_commit = commit;
}
static void
def_section (name, attr)
const char *name;
int attr;
{
def_file_section *s;
int max_sections = ROUND_UP(def->num_section_defs, 4);
if (def->num_section_defs >= max_sections)
{
max_sections = ROUND_UP(def->num_section_defs+1, 4);
if (def->section_defs)
def->section_defs = (def_file_section *) xrealloc (def->section_defs, max_sections * sizeof (def_file_import));
else
def->section_defs = (def_file_section *) xmalloc (max_sections * sizeof (def_file_import));
}
s = def->section_defs + def->num_section_defs;
memset (s, 0, sizeof (def_file_section));
s->name = xstrdup (name);
if (attr & 1)
s->flag_read = 1;
if (attr & 2)
s->flag_write = 1;
if (attr & 4)
s->flag_execute = 1;
if (attr & 8)
s->flag_shared = 1;
def->num_section_defs++;
}
static void
def_section_alt (name, attr)
const char *name;
const char *attr;
{
int aval = 0;
for (; *attr; attr++)
{
switch (*attr)
{
case 'R':
case 'r':
aval |= 1;
break;
case 'W':
case 'w':
aval |= 2;
break;
case 'X':
case 'x':
aval |= 4;
break;
case 'S':
case 's':
aval |= 8;
break;
}
}
def_section (name, aval);
}
static void
def_exports (external_name, internal_name, ordinal, flags)
const char *external_name;
const char *internal_name;
int ordinal;
int flags;
{
def_file_export *dfe;
if (!internal_name && external_name)
internal_name = external_name;
#if TRACE
printf ("def_exports, ext=%s int=%s\n", external_name, internal_name);
#endif
dfe = def_file_add_export (def, external_name, internal_name, ordinal);
if (flags & 1)
dfe->flag_noname = 1;
if (flags & 2)
dfe->flag_constant = 1;
if (flags & 4)
dfe->flag_data = 1;
if (flags & 8)
dfe->flag_private = 1;
}
static void
def_import (internal_name, module, dllext, name, ordinal)
const char *internal_name;
const char *module;
const char *dllext;
const char *name;
int ordinal;
{
char *buf = 0;
if (dllext != NULL)
{
buf = (char *) xmalloc (strlen (module) + strlen (dllext) + 2);
sprintf (buf, "%s.%s", module, dllext);
module = buf;
}
def_file_add_import (def, name, module, ordinal, internal_name);
if (buf)
free (buf);
}
static void
def_version (major, minor)
int major;
int minor;
{
def->version_major = major;
def->version_minor = minor;
}
static void
def_directive (str)
char *str;
{
struct directive *d = (struct directive *) xmalloc (sizeof (struct directive));
d->next = directives;
directives = d;
d->name = xstrdup (str);
d->len = strlen (str);
}
static int
def_error (err)
const char *err;
{
einfo ("%P: %s:%d: %s\n", def_filename, linenumber, err);
return 0;
}
/*****************************************************************************
Lexical Scanner
*****************************************************************************/
#undef TRACE
#define TRACE 0
/* Never freed, but always reused as needed, so no real leak */
static char *buffer = 0;
static int buflen = 0;
static int bufptr = 0;
static void
put_buf (c)
char c;
{
if (bufptr == buflen)
{
buflen += 50; /* overly reasonable, eh? */
if (buffer)
buffer = (char *) xrealloc (buffer, buflen + 1);
else
buffer = (char *) xmalloc (buflen + 1);
}
buffer[bufptr++] = c;
buffer[bufptr] = 0; /* not optimal, but very convenient */
}
static struct
{
char *name;
int token;
}
tokens[] =
{
{ "BASE", BASE },
{ "CODE", CODE },
{ "CONSTANT", CONSTANTU },
{ "constant", CONSTANTL },
{ "DATA", DATAU },
{ "data", DATAL },
{ "DESCRIPTION", DESCRIPTION },
{ "DIRECTIVE", DIRECTIVE },
{ "EXECUTE", EXECUTE },
{ "EXPORTS", EXPORTS },
{ "HEAPSIZE", HEAPSIZE },
{ "IMPORTS", IMPORTS },
{ "LIBRARY", LIBRARY },
{ "NAME", NAME },
{ "NONAME", NONAMEU },
{ "noname", NONAMEL },
{ "PRIVATE", PRIVATEU },
{ "private", PRIVATEL },
{ "READ", READ },
{ "SECTIONS", SECTIONS },
{ "SEGMENTS", SECTIONS },
{ "SHARED", SHARED },
{ "STACKSIZE", STACKSIZE },
{ "VERSION", VERSIONK },
{ "WRITE", WRITE },
{ 0, 0 }
};
static int
def_getc ()
{
int rv;
if (lex_parse_string)
{
if (lex_parse_string >= lex_parse_string_end)
rv = EOF;
else
rv = *lex_parse_string++;
}
else
{
rv = fgetc (the_file);
}
if (rv == '\n')
saw_newline = 1;
return rv;
}
static int
def_ungetc (c)
int c;
{
if (lex_parse_string)
{
lex_parse_string--;
return c;
}
else
return ungetc (c, the_file);
}
static int
def_lex ()
{
int c, i, q;
if (lex_forced_token)
{
i = lex_forced_token;
lex_forced_token = 0;
#if TRACE
printf ("lex: forcing token %d\n", i);
#endif
return i;
}
c = def_getc ();
/* trim leading whitespace */
while (c != EOF && (c == ' ' || c == '\t') && saw_newline)
c = def_getc ();
if (c == EOF)
{
#if TRACE
printf ("lex: EOF\n");
#endif
return 0;
}
if (saw_newline && c == ';')
{
do
{
c = def_getc ();
}
while (c != EOF && c != '\n');
if (c == '\n')
return def_lex ();
return 0;
}
/* must be something else */
saw_newline = 0;
if (isdigit (c))
{
bufptr = 0;
while (c != EOF && (isxdigit (c) || (c == 'x')))
{
put_buf (c);
c = def_getc ();
}
if (c != EOF)
def_ungetc (c);
yylval.number = strtoul (buffer, 0, 0);
#if TRACE
printf ("lex: `%s' returns NUMBER %d\n", buffer, yylval.number);
#endif
return NUMBER;
}
if (isalpha (c) || strchr ("$:-_?", c))
{
bufptr = 0;
while (c != EOF && (isalnum (c) || strchr ("$:-_?/@", c)))
{
put_buf (c);
c = def_getc ();
}
if (c != EOF)
def_ungetc (c);
for (i = 0; tokens[i].name; i++)
if (strcmp (tokens[i].name, buffer) == 0)
{
#if TRACE
printf ("lex: `%s' is a string token\n", buffer);
#endif
return tokens[i].token;
}
#if TRACE
printf ("lex: `%s' returns ID\n", buffer);
#endif
yylval.id = xstrdup (buffer);
return ID;
}
if (c == '\'' || c == '"')
{
q = c;
c = def_getc ();
bufptr = 0;
while (c != EOF && c != q)
{
put_buf (c);
c = def_getc ();
}
yylval.id = xstrdup (buffer);
#if TRACE
printf ("lex: `%s' returns ID\n", buffer);
#endif
return ID;
}
if (c == '=' || c == '.' || c == '@' || c == ',')
{
#if TRACE
printf ("lex: `%c' returns itself\n", c);
#endif
return c;
}
if (c == '\n')
{
linenumber++;
saw_newline = 1;
}
/*printf ("lex: 0x%02x ignored\n", c); */
return def_lex ();
}
Reference in New Issue View Git Blame Copy Permalink