NetBSD/sys/kern/kern_module.c

668 lines
15 KiB
C

/* $NetBSD: kern_module.c,v 1.9 2008/03/02 11:18:43 jmmv Exp $ */
/*-
* Copyright (c) 2008 The NetBSD Foundation, Inc.
* 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.
* 3. All advertising materials mentioning features or use of this software
* must display the following acknowledgement:
* This product includes software developed by the NetBSD
* Foundation, Inc. and its contributors.
* 4. Neither the name of The NetBSD Foundation nor the names of its
* contributors may be used to endorse or promote products derived
* from this software without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. 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 FOUNDATION 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.
*/
/*
* Kernel module support.
*
* XXX Deps for loadable modules don't work, because we must load the
* module in order to find out which modules it requires. Linking may
* fail because of missing symbols.
*/
#include <sys/cdefs.h>
__KERNEL_RCSID(0, "$NetBSD: kern_module.c,v 1.9 2008/03/02 11:18:43 jmmv Exp $");
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/fcntl.h>
#include <sys/proc.h>
#include <sys/kauth.h>
#include <sys/kobj.h>
#include <sys/kmem.h>
#include <sys/module.h>
#include <sys/syscall.h>
#include <sys/syscallargs.h>
#include <uvm/uvm_extern.h>
#include <machine/stdarg.h>
#ifndef LKM /* XXX */
struct vm_map *lkm_map;
#endif
struct modlist module_list = TAILQ_HEAD_INITIALIZER(module_list);
struct modlist module_bootlist = TAILQ_HEAD_INITIALIZER(module_bootlist);
u_int module_count;
kmutex_t module_lock;
/* Ensure that the kernel's link set isn't empty. */
static modinfo_t module_dummy;
__link_set_add_rodata(modules, module_dummy);
static module_t *module_lookup(const char *);
static int module_do_load(const char *, bool, int, prop_dictionary_t,
module_t **);
static int module_do_unload(const char *);
static void module_error(const char *, ...);
static int module_do_builtin(const char *, module_t **);
/*
* module_error:
*
* Utility function: log an error.
*/
static void
module_error(const char *fmt, ...)
{
va_list ap;
va_start(ap, fmt);
printf("WARNING: module error: ");
vprintf(fmt, ap);
printf("\n");
va_end(ap);
}
/*
* module_init:
*
* Initialize the module subsystem.
*/
void
module_init(void)
{
extern struct vm_map *lkm_map;
if (lkm_map == NULL)
lkm_map = kernel_map;
mutex_init(&module_lock, MUTEX_DEFAULT, IPL_NONE);
}
/*
* module_init_class:
*
* Initialize all built-in and pre-loaded modules of the
* specified class.
*/
void
module_init_class(modclass_t class)
{
__link_set_decl(modules, modinfo_t);
modinfo_t *const *mip, *mi;
module_t *mod;
mutex_enter(&module_lock);
/*
* Builtins first. These can't depend on pre-loaded modules.
*/
__link_set_foreach(mip, modules) {
mi = *mip;
if (mi == &module_dummy) {
continue;
}
if (class != MODULE_CLASS_ANY && class != mi->mi_class) {
continue;
}
(void)module_do_builtin(mi->mi_name, NULL);
}
/*
* Now preloaded modules. These will be pulled off the
* list as we call module_do_load();
*/
while ((mod = TAILQ_FIRST(&module_bootlist)) != NULL) {
module_do_load(mod->mod_info->mi_name, false, 0,
NULL, NULL);
}
mutex_exit(&module_lock);
}
/*
* module_jettison:
*
* Return memory used by pre-loaded modules to the freelist.
*/
void
module_jettison(void)
{
/* XXX nothing yet */
}
/*
* module_load:
*
* Load a single module from the file system.
*/
int
module_load(const char *filename, int flags, prop_dictionary_t props)
{
int error;
mutex_enter(&module_lock);
error = module_do_load(filename, false, flags, props, NULL);
mutex_exit(&module_lock);
return error;
}
/*
* module_unload:
*
* Find and unload a module by name.
*/
int
module_unload(const char *name)
{
int error;
mutex_enter(&module_lock);
error = module_do_unload(name);
mutex_exit(&module_lock);
return error;
}
/*
* module_lookup:
*
* Look up a module by name.
*/
module_t *
module_lookup(const char *name)
{
module_t *mod;
KASSERT(mutex_owned(&module_lock));
TAILQ_FOREACH(mod, &module_list, mod_chain) {
if (strcmp(mod->mod_info->mi_name, name) == 0) {
break;
}
}
return mod;
}
/*
* module_hold:
*
* Add a single reference to a module. It's the caller's
* responsibility to ensure that the reference is dropped
* later.
*/
int
module_hold(const char *name)
{
module_t *mod;
mutex_enter(&module_lock);
mod = module_lookup(name);
if (mod == NULL) {
mutex_exit(&module_lock);
return ENOENT;
}
mod->mod_refcnt++;
mutex_exit(&module_lock);
return 0;
}
/*
* module_rele:
*
* Release a reference acquired with module_hold().
*/
void
module_rele(const char *name)
{
module_t *mod;
mutex_enter(&module_lock);
mod = module_lookup(name);
if (mod == NULL) {
mutex_exit(&module_lock);
panic("module_rele: gone");
}
mod->mod_refcnt--;
mutex_exit(&module_lock);
}
/*
* module_do_builtin:
*
* Initialize a single module from the list of modules that are
* built into the kernel (linked into the kernel image).
*/
static int
module_do_builtin(const char *name, module_t **modp)
{
__link_set_decl(modules, modinfo_t);
modinfo_t *const *mip;
const char *p, *s;
char buf[MAXMODNAME];
modinfo_t *mi;
module_t *mod, *mod2;
size_t len;
int error, i;
KASSERT(mutex_owned(&module_lock));
/*
* Check to see if already loaded.
*/
if ((mod = module_lookup(name)) != NULL) {
if (modp != NULL) {
*modp = mod;
}
return 0;
}
/*
* Search the list to see if we have a module by this name.
*/
error = ENOENT;
__link_set_foreach(mip, modules) {
mi = *mip;
if (mi == &module_dummy) {
continue;
}
if (strcmp(mi->mi_name, name) == 0) {
error = 0;
break;
}
}
if (error != 0) {
return error;
}
/*
* Initialize pre-requisites.
*/
mod = kmem_zalloc(sizeof(*mod), KM_SLEEP);
if (mod == NULL) {
return ENOMEM;
}
if (modp != NULL) {
*modp = mod;
}
if (mi->mi_required != NULL) {
for (s = mi->mi_required; *s != '\0'; s = p) {
if (*s == ',')
s++;
p = s;
while (*p != '\0' && *p != ',')
p++;
len = min(p - s + 1, sizeof(buf));
strlcpy(buf, s, len);
if (buf[0] == '\0')
break;
if (mod->mod_nrequired == MAXMODDEPS - 1) {
module_error("too many required modules");
kmem_free(mod, sizeof(*mod));
return EINVAL;
}
error = module_do_builtin(buf, &mod2);
if (error != 0) {
kmem_free(mod, sizeof(*mod));
return error;
}
mod->mod_required[mod->mod_nrequired++] = mod2;
}
}
/*
* Try to initialize the module.
*/
error = (*mi->mi_modcmd)(MODULE_CMD_INIT, NULL);
if (error != 0) {
module_error("builtin module `%s' "
"failed to init", mi->mi_name);
kmem_free(mod, sizeof(*mod));
return error;
}
mod->mod_info = mi;
mod->mod_source = MODULE_SOURCE_KERNEL;
module_count++;
TAILQ_INSERT_TAIL(&module_list, mod, mod_chain);
/*
* If that worked, count dependencies.
*/
for (i = 0; i < mod->mod_nrequired; i++) {
mod->mod_required[i]->mod_refcnt++;
}
return 0;
}
/*
* module_do_load:
*
* Helper routine: load a module from the file system, or one
* pushed by the boot loader.
*/
static int
module_do_load(const char *filename, bool isdep, int flags,
prop_dictionary_t props, module_t **modp)
{
static TAILQ_HEAD(,module) pending = TAILQ_HEAD_INITIALIZER(pending);
static int depth;
const int maxdepth = 6;
modinfo_t *mi;
module_t *mod, *mod2;
char buf[MAXMODNAME];
const char *s, *p;
void *addr;
size_t size;
int error;
size_t len;
u_int i;
bool closed = false;
KASSERT(mutex_owned(&module_lock));
error = 0;
/*
* Avoid recursing too far.
*/
if (++depth > maxdepth) {
module_error("too many required modules");
depth--;
return EMLINK;
}
/*
* Load the module and link. Before going to the file system,
* scan the list of modules loaded by the boot loader. Just
* before init is started the list of modules loaded at boot
* will be purged. Before init is started we can assume that
* `filename' is a module name and not a path name.
*/
TAILQ_FOREACH(mod, &module_bootlist, mod_chain) {
if (strcmp(mod->mod_info->mi_name, filename) == 0) {
TAILQ_REMOVE(&module_bootlist, mod, mod_chain);
break;
}
}
if (mod == NULL) {
mod = kmem_zalloc(sizeof(*mod), KM_SLEEP);
if (mod == NULL) {
depth--;
return ENOMEM;
}
error = kobj_open_file(&mod->mod_kobj, filename);
if (error != 0) {
kmem_free(mod, sizeof(*mod));
depth--;
module_error("unable to open object file");
return error;
}
error = kobj_load(mod->mod_kobj);
if (error != 0) {
kobj_close(mod->mod_kobj);
kmem_free(mod, sizeof(*mod));
depth--;
module_error("unable to load kernel object");
return error;
}
mod->mod_source = MODULE_SOURCE_FILESYS;
}
TAILQ_INSERT_TAIL(&pending, mod, mod_chain);
/*
* Find module info record and check compatibility.
*/
error = kobj_find_section(mod->mod_kobj, "link_set_modules",
&addr, &size);
if (error != 0) {
module_error("`link_set_modules' section not present");
goto fail;
}
if (size != sizeof(modinfo_t **)) {
module_error("`link_set_modules' section wrong size");
goto fail;
}
mod->mod_info = *(modinfo_t **)addr;
mi = mod->mod_info;
if (strlen(mi->mi_name) >= MAXMODNAME) {
error = EINVAL;
module_error("module name too long");
goto fail;
}
/*
* If loading a dependency, `filename' is a plain module name.
* The name must match.
*/
if (isdep && strcmp(mi->mi_name, filename) != 0) {
error = ENOENT;
goto fail;
}
/*
* Check to see if the module is already loaded. If so, we may
* have been recursively called to handle a dependency, so be sure
* to set modp.
*/
if ((mod2 = module_lookup(mi->mi_name)) != NULL) {
if (modp != NULL)
*modp = mod2;
error = EEXIST;
goto fail;
}
/*
* Block circular dependencies.
*/
TAILQ_FOREACH(mod2, &pending, mod_chain) {
if (mod == mod2) {
continue;
}
if (strcmp(mod2->mod_info->mi_name, mi->mi_name) == 0) {
error = EDEADLK;
module_error("circular dependency detected");
goto fail;
}
}
/*
* Pass proper name to kobj. This will register the module
* with the ksyms framework.
*/
error = kobj_set_name(mod->mod_kobj, mi->mi_name);
if (error != 0) {
module_error("unable to set name");
goto fail;
}
/*
* Close the kobj before handling dependencies since we're done
* with it and don't want to open an already locked file if a
* circular dependency exists.
*/
kobj_close(mod->mod_kobj);
closed = true;
/*
* Now try to load any requisite modules.
*/
if (mi->mi_required != NULL) {
for (s = mi->mi_required; *s != '\0'; s = p) {
if (*s == ',')
s++;
p = s;
while (*p != '\0' && *p != ',')
p++;
len = p - s + 1;
if (len >= MAXMODNAME) {
error = EINVAL;
module_error("required module name too long");
goto fail;
}
strlcpy(buf, s, len);
if (buf[0] == '\0')
break;
if (mod->mod_nrequired == MAXMODDEPS - 1) {
error = EINVAL;
module_error("too many required modules");
goto fail;
}
if (strcmp(buf, mi->mi_name) == 0) {
error = EDEADLK;
module_error("self-dependency detected");
goto fail;
}
error = module_do_load(buf, true, flags, NULL,
&mod->mod_required[mod->mod_nrequired++]);
if (error != 0 && error != EEXIST)
goto fail;
}
}
/*
* We loaded all needed modules successfully: initialize.
*/
error = (*mi->mi_modcmd)(MODULE_CMD_INIT, props);
if (error != 0) {
module_error("modctl function returned error %d", error);
goto fail;
}
/*
* Good, the module loaded successfully. Put it onto the
* list and add references to its requisite modules.
*/
module_count++;
if (!closed)
kobj_close(mod->mod_kobj);
TAILQ_REMOVE(&pending, mod, mod_chain);
TAILQ_INSERT_TAIL(&module_list, mod, mod_chain);
for (i = 0; i < mod->mod_nrequired; i++) {
KASSERT(mod->mod_required[i] != NULL);
mod->mod_required[i]->mod_refcnt++;
}
if (modp != NULL) {
*modp = mod;
}
depth--;
return 0;
fail:
if (!closed)
kobj_close(mod->mod_kobj);
TAILQ_REMOVE(&pending, mod, mod_chain);
kobj_unload(mod->mod_kobj);
kmem_free(mod, sizeof(*mod));
depth--;
return error;
}
/*
* module_do_unload:
*
* Helper routine: do the dirty work of unloading a module.
*/
static int
module_do_unload(const char *name)
{
module_t *mod;
int error;
u_int i;
KASSERT(mutex_owned(&module_lock));
mod = module_lookup(name);
if (mod == NULL) {
return ENOENT;
}
if (mod->mod_refcnt != 0 || mod->mod_source == MODULE_SOURCE_KERNEL) {
return EBUSY;
}
error = (*mod->mod_info->mi_modcmd)(MODULE_CMD_FINI, NULL);
if (error != 0) {
return error;
}
module_count--;
TAILQ_REMOVE(&module_list, mod, mod_chain);
for (i = 0; i < mod->mod_nrequired; i++) {
mod->mod_required[i]->mod_refcnt--;
}
if (mod->mod_kobj != NULL) {
kobj_unload(mod->mod_kobj);
}
kmem_free(mod, sizeof(*mod));
return 0;
}
/*
* module_prime:
*
* Push a module loaded by the bootloader onto our internal
* list.
*/
int
module_prime(void *base, size_t size)
{
module_t *mod;
int error;
mod = kmem_zalloc(sizeof(*mod), KM_SLEEP);
if (mod == NULL) {
return ENOMEM;
}
mod->mod_source = MODULE_SOURCE_BOOT;
error = kobj_open_mem(&mod->mod_kobj, base, size);
if (error == 0) {
kmem_free(mod, sizeof(*mod));
module_error("unable to open object pushed by boot loader");
return error;
}
error = kobj_load(mod->mod_kobj);
if (error != 0) {
kmem_free(mod, sizeof(*mod));
module_error("unable to push object pushed by boot loader");
return error;
}
mod->mod_source = MODULE_SOURCE_FILESYS;
TAILQ_INSERT_TAIL(&module_bootlist, mod, mod_chain);
return 0;
}