freebsd-dev/sys/kern/kern_linker.c
1998-01-01 08:56:24 +00:00

574 lines
13 KiB
C

/*-
* Copyright (c) 1997 Doug Rabson
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
*
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
* $Id: kern_linker.c,v 1.5 1997/12/12 04:00:59 dyson Exp $
*/
#include <sys/param.h>
#include <sys/kernel.h>
#include <sys/systm.h>
#include <sys/malloc.h>
#include <sys/sysproto.h>
#include <sys/sysent.h>
#include <sys/proc.h>
#include <sys/lock.h>
#include <machine/cpu.h>
#include <sys/module.h>
#include <sys/linker.h>
#include <sys/unistd.h>
MALLOC_DEFINE(M_LINKER, "kld", "kernel linker");
linker_file_t linker_current_file;
static struct lock lock; /* lock for the file list */
static linker_class_list_t classes;
static linker_file_list_t files;
static int next_file_id = 1;
static void
linker_init(void* arg)
{
lockinit(&lock, PVM, "klink", 0, 0);
TAILQ_INIT(&classes);
TAILQ_INIT(&files);
}
SYSINIT(linker, SI_SUB_KMEM, SI_ORDER_SECOND, linker_init, 0);
int
linker_add_class(const char* desc, void* priv,
struct linker_class_ops* ops)
{
linker_class_t lc;
lc = malloc(sizeof(struct linker_class), M_LINKER, M_NOWAIT);
if (!lc)
return ENOMEM;
lc->desc = desc;
lc->priv = priv;
lc->ops = ops;
TAILQ_INSERT_HEAD(&classes, lc, link);
return 0;
}
static void
linker_file_sysinit(linker_file_t lf)
{
struct linker_set* sysinits;
struct sysinit** sipp;
struct sysinit** xipp;
struct sysinit* save;
linker_current_file = lf;
KLD_DPF(FILE, ("linker_file_sysinit: calling SYSINITs for %s\n",
lf->filename));
sysinits = (struct linker_set*)
linker_file_lookup_symbol(lf, "sysinit_set", 0);
if (!sysinits)
return;
/*
* Perform a bubble sort of the system initialization objects by
* their subsystem (primary key) and order (secondary key).
*
* Since some things care about execution order, this is the
* operation which ensures continued function.
*/
for( sipp = (struct sysinit **)sysinits->ls_items; *sipp; sipp++) {
for( xipp = sipp + 1; *xipp; xipp++) {
if( (*sipp)->subsystem < (*xipp)->subsystem ||
( (*sipp)->subsystem == (*xipp)->subsystem &&
(*sipp)->order < (*xipp)->order))
continue; /* skip*/
save = *sipp;
*sipp = *xipp;
*xipp = save;
}
}
/*
* Traverse the (now) ordered list of system initialization tasks.
* Perform each task, and continue on to the next task.
*
* The last item on the list is expected to be the scheduler,
* which will not return.
*/
for( sipp = (struct sysinit **)sysinits->ls_items; *sipp; sipp++) {
if( (*sipp)->subsystem == SI_SUB_DUMMY)
continue; /* skip dummy task(s)*/
switch( (*sipp)->type) {
case SI_TYPE_DEFAULT:
/* no special processing*/
(*((*sipp)->func))( (*sipp)->udata);
break;
case SI_TYPE_KTHREAD:
#if !defined(SMP)
/* kernel thread*/
if (fork1(&proc0, RFFDG|RFPROC|RFMEM))
panic("fork kernel thread");
cpu_set_fork_handler(pfind(proc0.p_retval[0]),
(*sipp)->func, (*sipp)->udata);
break;
#endif
case SI_TYPE_KPROCESS:
/* kernel thread*/
if (fork1(&proc0, RFFDG|RFPROC))
panic("fork kernel process");
cpu_set_fork_handler(pfind(proc0.p_retval[0]),
(*sipp)->func, (*sipp)->udata);
break;
default:
panic( "linker_file_sysinit: unrecognized init type");
}
}
}
int
linker_load_file(const char* filename, linker_file_t* result)
{
linker_class_t lc;
linker_file_t lf;
int error = 0;
lf = linker_find_file_by_name(filename);
if (lf) {
KLD_DPF(FILE, ("linker_load_file: file %s is already loaded, incrementing refs\n", filename));
*result = lf;
lf->refs++;
goto out;
}
lf = NULL;
for (lc = TAILQ_FIRST(&classes); lc; lc = TAILQ_NEXT(lc, link)) {
KLD_DPF(FILE, ("linker_load_file: trying to load %s as %s\n",
filename, lc->desc));
if (error = lc->ops->load_file(filename, &lf))
goto out;
if (lf) {
linker_file_sysinit(lf);
*result = lf;
goto out;
}
}
error = ENOEXEC; /* format not recognised */
out:
return error;
}
linker_file_t
linker_find_file_by_name(const char* filename)
{
linker_file_t lf = 0;
lockmgr(&lock, LK_SHARED, 0, curproc);
for (lf = TAILQ_FIRST(&files); lf; lf = TAILQ_NEXT(lf, link))
if (!strcmp(lf->filename, filename))
break;
lockmgr(&lock, LK_RELEASE, 0, curproc);
return lf;
}
linker_file_t
linker_find_file_by_id(int fileid)
{
linker_file_t lf = 0;
lockmgr(&lock, LK_SHARED, 0, curproc);
for (lf = TAILQ_FIRST(&files); lf; lf = TAILQ_NEXT(lf, link))
if (lf->id == fileid)
break;
lockmgr(&lock, LK_RELEASE, 0, curproc);
return lf;
}
linker_file_t
linker_make_file(const char* filename, void* priv, struct linker_file_ops* ops)
{
linker_file_t lf = 0;
int namelen;
KLD_DPF(FILE, ("linker_make_file: new file, filename=%s\n", filename));
lockmgr(&lock, LK_EXCLUSIVE|LK_RETRY, 0, curproc);
namelen = strlen(filename) + 1;
lf = malloc(sizeof(struct linker_file) + namelen, M_LINKER, M_WAITOK);
if (!lf)
goto out;
lf->refs = 1;
lf->userrefs = 0;
lf->filename = (char*) (lf + 1);
strcpy(lf->filename, filename);
lf->id = next_file_id++;
lf->ndeps = 0;
lf->deps = NULL;
STAILQ_INIT(&lf->common);
TAILQ_INIT(&lf->modules);
lf->priv = priv;
lf->ops = ops;
TAILQ_INSERT_TAIL(&files, lf, link);
out:
lockmgr(&lock, LK_RELEASE, 0, curproc);
return lf;
}
int
linker_file_unload(linker_file_t file)
{
module_t mod, next;
struct common_symbol* cp;
int error = 0;
int i;
KLD_DPF(FILE, ("linker_file_unload: lf->refs=%d\n", lf->refs));
lockmgr(&lock, LK_EXCLUSIVE|LK_RETRY, 0, curproc);
if (file->refs == 1) {
KLD_DPF(FILE, ("linker_file_unload: file is unloading, informing modules\n"));
/*
* Inform any modules associated with this file.
*/
for (mod = TAILQ_FIRST(&file->modules); mod; mod = next) {
next = module_getfnext(mod);
/*
* Give the module a chance to veto the unload.
*/
if (error = module_unload(mod)) {
KLD_DPF(FILE, ("linker_file_unload: module %x vetoes unload\n",
mod));
lockmgr(&lock, LK_RELEASE, 0, curproc);
goto out;
}
module_release(mod);
}
}
file->refs--;
if (file->refs > 0) {
lockmgr(&lock, LK_RELEASE, 0, curproc);
goto out;
}
TAILQ_REMOVE(&files, file, link);
lockmgr(&lock, LK_RELEASE, 0, curproc);
for (i = 0; i < file->ndeps; i++)
linker_file_unload(file->deps[i]);
free(file->deps, M_LINKER);
for (cp = STAILQ_FIRST(&file->common); cp;
cp = STAILQ_FIRST(&file->common)) {
STAILQ_REMOVE(&file->common, cp, common_symbol, link);
free(cp, M_LINKER);
}
file->ops->unload(file);
free(file, M_LINKER);
out:
return error;
}
int
linker_file_add_dependancy(linker_file_t file, linker_file_t dep)
{
linker_file_t* newdeps;
newdeps = malloc((file->ndeps + 1) * sizeof(linker_file_t*),
M_LINKER, M_WAITOK);
if (newdeps == NULL)
return ENOMEM;
if (file->deps) {
bcopy(file->deps, newdeps, file->ndeps * sizeof(linker_file_t*));
free(file->deps, M_LINKER);
}
file->deps = newdeps;
file->deps[file->ndeps] = dep;
file->ndeps++;
return 0;
}
caddr_t
linker_file_lookup_symbol(linker_file_t file, const char* name, int deps)
{
caddr_t address;
size_t size;
size_t common_size = 0;
int i;
KLD_DPF(SYM, ("linker_file_lookup_symbol: file=%x, name=%s, deps=%d",
file, name, deps));
if (file->ops->lookup_symbol(file, name, &address, &size) == 0)
if (address == 0)
/*
* For commons, first look them up in the dependancies and
* only allocate space if not found there.
*/
common_size = size;
else
return address;
if (deps)
for (i = 0; i < file->ndeps; i++) {
address = linker_file_lookup_symbol(file->deps[i], name, 0);
if (address)
return address;
}
if (common_size > 0) {
/*
* This is a common symbol which was not found in the
* dependancies. We maintain a simple common symbol table in
* the file object.
*/
struct common_symbol* cp;
for (cp = STAILQ_FIRST(&file->common); cp;
cp = STAILQ_NEXT(cp, link))
if (!strcmp(cp->name, name))
return cp->address;
/*
* Round the symbol size up to align.
*/
common_size = (common_size + sizeof(int) - 1) & -sizeof(int);
cp = malloc(sizeof(struct common_symbol)
+ common_size
+ strlen(name) + 1,
M_LINKER, M_WAITOK);
if (!cp)
return 0;
cp->address = (caddr_t) (cp + 1);
cp->name = cp->address + common_size;
strcpy(cp->name, name);
bzero(cp->address, common_size);
STAILQ_INSERT_TAIL(&file->common, cp, link);
return cp->address;
}
return 0;
}
/*
* Syscalls.
*/
int
kldload(struct proc* p, struct kldload_args* uap)
{
char* filename = NULL;
linker_file_t lf;
int error = 0;
p->p_retval[0] = -1;
if (securelevel > 0)
return EPERM;
if (error = suser(p->p_ucred, &p->p_acflag))
return error;
filename = malloc(MAXPATHLEN, M_TEMP, M_WAITOK);
if (error = copyinstr(SCARG(uap, file), filename, MAXPATHLEN, NULL))
goto out;
if (error = linker_load_file(uap->file, &lf))
goto out;
lf->userrefs++;
p->p_retval[0] = lf->id;
out:
if (filename)
free(filename, M_TEMP);
return error;
}
int
kldunload(struct proc* p, struct kldunload_args* uap)
{
linker_file_t lf;
int error = 0;
if (securelevel > 0)
return EPERM;
if (error = suser(p->p_ucred, &p->p_acflag))
return error;
lf = linker_find_file_by_id(SCARG(uap, fileid));
if (lf) {
KLD_DPF(FILE, ("kldunload: lf->userrefs=%d\n", lf->userrefs));
if (lf->userrefs == 0) {
printf("linkerunload: attempt to unload file which was not loaded by user\n");
error = EBUSY;
goto out;
}
lf->userrefs--;
error = linker_file_unload(lf);
} else
error = ENOENT;
out:
return error;
}
int
kldfind(struct proc* p, struct kldfind_args* uap)
{
char* filename = NULL;
linker_file_t lf;
int error = 0;
p->p_retval[0] = -1;
filename = malloc(MAXPATHLEN, M_TEMP, M_WAITOK);
if (error = copyinstr(SCARG(uap, file), filename, MAXPATHLEN, NULL))
goto out;
lf = linker_find_file_by_name(filename);
if (lf)
p->p_retval[0] = lf->id;
else
error = ENOENT;
out:
if (filename)
free(filename, M_TEMP);
return error;
}
int
kldnext(struct proc* p, struct kldnext_args* uap)
{
linker_file_t lf;
int error = 0;
if (SCARG(uap, fileid) == 0) {
if (TAILQ_FIRST(&files))
p->p_retval[0] = TAILQ_FIRST(&files)->id;
else
p->p_retval[0] = 0;
return 0;
}
lf = linker_find_file_by_id(SCARG(uap, fileid));
if (lf) {
if (TAILQ_NEXT(lf, link))
p->p_retval[0] = TAILQ_NEXT(lf, link)->id;
else
p->p_retval[0] = 0;
} else
error = ENOENT;
return error;
}
int
kldstat(struct proc* p, struct kldstat_args* uap)
{
linker_file_t lf;
int error = 0;
int version;
struct kld_file_stat* stat;
int namelen;
lf = linker_find_file_by_id(SCARG(uap, fileid));
if (!lf) {
error = ENOENT;
goto out;
}
stat = SCARG(uap, stat);
/*
* Check the version of the user's structure.
*/
if (error = copyin(&stat->version, &version, sizeof(version)))
goto out;
if (version != sizeof(struct kld_file_stat)) {
error = EINVAL;
goto out;
}
namelen = strlen(lf->filename) + 1;
if (namelen > MAXPATHLEN)
namelen = MAXPATHLEN;
if (error = copyout(lf->filename, &stat->name[0], namelen))
goto out;
if (error = copyout(&lf->refs, &stat->refs, sizeof(int)))
goto out;
if (error = copyout(&lf->id, &stat->id, sizeof(int)))
goto out;
if (error = copyout(&lf->address, &stat->address, sizeof(caddr_t)))
goto out;
if (error = copyout(&lf->size, &stat->size, sizeof(size_t)))
goto out;
p->p_retval[0] = 0;
out:
return error;
}
int
kldfirstmod(struct proc* p, struct kldfirstmod_args* uap)
{
linker_file_t lf;
int error = 0;
lf = linker_find_file_by_id(SCARG(uap, fileid));
if (lf) {
if (TAILQ_FIRST(&lf->modules))
p->p_retval[0] = module_getid(TAILQ_FIRST(&lf->modules));
else
p->p_retval[0] = 0;
} else
error = ENOENT;
return error;
}