kernel_optimize_test/security/selinux/ss/conditional.c

758 lines
17 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0-only
/* Authors: Karl MacMillan <kmacmillan@tresys.com>
* Frank Mayer <mayerf@tresys.com>
*
* Copyright (C) 2003 - 2004 Tresys Technology, LLC
*/
#include <linux/kernel.h>
#include <linux/errno.h>
#include <linux/string.h>
#include <linux/spinlock.h>
#include <linux/slab.h>
#include "security.h"
#include "conditional.h"
selinux: extended permissions for ioctls Add extended permissions logic to selinux. Extended permissions provides additional permissions in 256 bit increments. Extend the generic ioctl permission check to use the extended permissions for per-command filtering. Source/target/class sets including the ioctl permission may additionally include a set of commands. Example: allowxperm <source> <target>:<class> ioctl unpriv_app_socket_cmds auditallowxperm <source> <target>:<class> ioctl priv_gpu_cmds Where unpriv_app_socket_cmds and priv_gpu_cmds are macros representing commonly granted sets of ioctl commands. When ioctl commands are omitted only the permissions are checked. This feature is intended to provide finer granularity for the ioctl permission that may be too imprecise. For example, the same driver may use ioctls to provide important and benign functionality such as driver version or socket type as well as dangerous capabilities such as debugging features, read/write/execute to physical memory or access to sensitive data. Per-command filtering provides a mechanism to reduce the attack surface of the kernel, and limit applications to the subset of commands required. The format of the policy binary has been modified to include ioctl commands, and the policy version number has been incremented to POLICYDB_VERSION_XPERMS_IOCTL=30 to account for the format change. The extended permissions logic is deliberately generic to allow components to be reused e.g. netlink filters Signed-off-by: Jeff Vander Stoep <jeffv@google.com> Acked-by: Nick Kralevich <nnk@google.com> Signed-off-by: Paul Moore <pmoore@redhat.com>
2015-07-11 05:19:56 +08:00
#include "services.h"
/*
* cond_evaluate_expr evaluates a conditional expr
* in reverse polish notation. It returns true (1), false (0),
* or undefined (-1). Undefined occurs when the expression
* exceeds the stack depth of COND_EXPR_MAXDEPTH.
*/
static int cond_evaluate_expr(struct policydb *p, struct cond_expr *expr)
{
u32 i;
int s[COND_EXPR_MAXDEPTH];
int sp = -1;
if (expr->len == 0)
return -1;
for (i = 0; i < expr->len; i++) {
struct cond_expr_node *node = &expr->nodes[i];
switch (node->expr_type) {
case COND_BOOL:
if (sp == (COND_EXPR_MAXDEPTH - 1))
return -1;
sp++;
s[sp] = p->bool_val_to_struct[node->bool - 1]->state;
break;
case COND_NOT:
if (sp < 0)
return -1;
s[sp] = !s[sp];
break;
case COND_OR:
if (sp < 1)
return -1;
sp--;
s[sp] |= s[sp + 1];
break;
case COND_AND:
if (sp < 1)
return -1;
sp--;
s[sp] &= s[sp + 1];
break;
case COND_XOR:
if (sp < 1)
return -1;
sp--;
s[sp] ^= s[sp + 1];
break;
case COND_EQ:
if (sp < 1)
return -1;
sp--;
s[sp] = (s[sp] == s[sp + 1]);
break;
case COND_NEQ:
if (sp < 1)
return -1;
sp--;
s[sp] = (s[sp] != s[sp + 1]);
break;
default:
return -1;
}
}
return s[0];
}
/*
* evaluate_cond_node evaluates the conditional stored in
* a struct cond_node and if the result is different than the
* current state of the node it sets the rules in the true/false
* list appropriately. If the result of the expression is undefined
* all of the rules are disabled for safety.
*/
static void evaluate_cond_node(struct policydb *p, struct cond_node *node)
{
struct avtab_node *avnode;
int new_state;
u32 i;
new_state = cond_evaluate_expr(p, &node->expr);
if (new_state != node->cur_state) {
node->cur_state = new_state;
if (new_state == -1)
pr_err("SELinux: expression result was undefined - disabling all rules.\n");
/* turn the rules on or off */
for (i = 0; i < node->true_list.len; i++) {
avnode = node->true_list.nodes[i];
if (new_state <= 0)
avnode->key.specified &= ~AVTAB_ENABLED;
else
avnode->key.specified |= AVTAB_ENABLED;
}
for (i = 0; i < node->false_list.len; i++) {
avnode = node->false_list.nodes[i];
/* -1 or 1 */
if (new_state)
avnode->key.specified &= ~AVTAB_ENABLED;
else
avnode->key.specified |= AVTAB_ENABLED;
}
}
}
void evaluate_cond_nodes(struct policydb *p)
{
u32 i;
for (i = 0; i < p->cond_list_len; i++)
evaluate_cond_node(p, &p->cond_list[i]);
}
void cond_policydb_init(struct policydb *p)
{
p->bool_val_to_struct = NULL;
p->cond_list = NULL;
p->cond_list_len = 0;
avtab_init(&p->te_cond_avtab);
}
static void cond_node_destroy(struct cond_node *node)
{
kfree(node->expr.nodes);
/* the avtab_ptr_t nodes are destroyed by the avtab */
kfree(node->true_list.nodes);
kfree(node->false_list.nodes);
}
static void cond_list_destroy(struct policydb *p)
{
u32 i;
for (i = 0; i < p->cond_list_len; i++)
cond_node_destroy(&p->cond_list[i]);
kfree(p->cond_list);
}
void cond_policydb_destroy(struct policydb *p)
{
kfree(p->bool_val_to_struct);
avtab_destroy(&p->te_cond_avtab);
cond_list_destroy(p);
}
int cond_init_bool_indexes(struct policydb *p)
{
kfree(p->bool_val_to_struct);
p->bool_val_to_struct = kmalloc_array(p->p_bools.nprim,
sizeof(*p->bool_val_to_struct),
GFP_KERNEL);
if (!p->bool_val_to_struct)
return -ENOMEM;
return 0;
}
int cond_destroy_bool(void *key, void *datum, void *p)
{
kfree(key);
kfree(datum);
return 0;
}
int cond_index_bool(void *key, void *datum, void *datap)
{
struct policydb *p;
struct cond_bool_datum *booldatum;
booldatum = datum;
p = datap;
if (!booldatum->value || booldatum->value > p->p_bools.nprim)
return -EINVAL;
p->sym_val_to_name[SYM_BOOLS][booldatum->value - 1] = key;
p->bool_val_to_struct[booldatum->value - 1] = booldatum;
return 0;
}
static int bool_isvalid(struct cond_bool_datum *b)
{
if (!(b->state == 0 || b->state == 1))
return 0;
return 1;
}
int cond_read_bool(struct policydb *p, struct symtab *s, void *fp)
{
char *key = NULL;
struct cond_bool_datum *booldatum;
__le32 buf[3];
u32 len;
int rc;
booldatum = kzalloc(sizeof(*booldatum), GFP_KERNEL);
if (!booldatum)
return -ENOMEM;
rc = next_entry(buf, fp, sizeof(buf));
if (rc)
goto err;
booldatum->value = le32_to_cpu(buf[0]);
booldatum->state = le32_to_cpu(buf[1]);
rc = -EINVAL;
if (!bool_isvalid(booldatum))
goto err;
len = le32_to_cpu(buf[2]);
if (((len == 0) || (len == (u32)-1)))
goto err;
rc = -ENOMEM;
key = kmalloc(len + 1, GFP_KERNEL);
if (!key)
goto err;
rc = next_entry(key, fp, len);
if (rc)
goto err;
key[len] = '\0';
rc = symtab_insert(s, key, booldatum);
if (rc)
goto err;
return 0;
err:
cond_destroy_bool(key, booldatum, NULL);
return rc;
}
struct cond_insertf_data {
struct policydb *p;
struct avtab_node **dst;
struct cond_av_list *other;
};
static int cond_insertf(struct avtab *a, struct avtab_key *k, struct avtab_datum *d, void *ptr)
{
struct cond_insertf_data *data = ptr;
struct policydb *p = data->p;
struct cond_av_list *other = data->other;
struct avtab_node *node_ptr;
u32 i;
bool found;
/*
* For type rules we have to make certain there aren't any
* conflicting rules by searching the te_avtab and the
* cond_te_avtab.
*/
if (k->specified & AVTAB_TYPE) {
if (avtab_search(&p->te_avtab, k)) {
pr_err("SELinux: type rule already exists outside of a conditional.\n");
return -EINVAL;
}
/*
* If we are reading the false list other will be a pointer to
* the true list. We can have duplicate entries if there is only
* 1 other entry and it is in our true list.
*
* If we are reading the true list (other == NULL) there shouldn't
* be any other entries.
*/
if (other) {
node_ptr = avtab_search_node(&p->te_cond_avtab, k);
if (node_ptr) {
if (avtab_search_node_next(node_ptr, k->specified)) {
pr_err("SELinux: too many conflicting type rules.\n");
return -EINVAL;
}
found = false;
for (i = 0; i < other->len; i++) {
if (other->nodes[i] == node_ptr) {
found = true;
break;
}
}
if (!found) {
pr_err("SELinux: conflicting type rules.\n");
return -EINVAL;
}
}
} else {
if (avtab_search(&p->te_cond_avtab, k)) {
pr_err("SELinux: conflicting type rules when adding type rule for true.\n");
return -EINVAL;
}
}
}
node_ptr = avtab_insert_nonunique(&p->te_cond_avtab, k, d);
if (!node_ptr) {
pr_err("SELinux: could not insert rule.\n");
return -ENOMEM;
}
*data->dst = node_ptr;
return 0;
}
static int cond_read_av_list(struct policydb *p, void *fp,
struct cond_av_list *list,
struct cond_av_list *other)
{
int rc;
__le32 buf[1];
u32 i, len;
struct cond_insertf_data data;
rc = next_entry(buf, fp, sizeof(u32));
if (rc)
return rc;
len = le32_to_cpu(buf[0]);
if (len == 0)
return 0;
list->nodes = kcalloc(len, sizeof(*list->nodes), GFP_KERNEL);
if (!list->nodes)
return -ENOMEM;
data.p = p;
data.other = other;
for (i = 0; i < len; i++) {
data.dst = &list->nodes[i];
rc = avtab_read_item(&p->te_cond_avtab, fp, p, cond_insertf,
&data);
if (rc) {
kfree(list->nodes);
list->nodes = NULL;
return rc;
}
}
list->len = len;
return 0;
}
static int expr_node_isvalid(struct policydb *p, struct cond_expr_node *expr)
{
if (expr->expr_type <= 0 || expr->expr_type > COND_LAST) {
pr_err("SELinux: conditional expressions uses unknown operator.\n");
return 0;
}
if (expr->bool > p->p_bools.nprim) {
pr_err("SELinux: conditional expressions uses unknown bool.\n");
return 0;
}
return 1;
}
static int cond_read_node(struct policydb *p, struct cond_node *node, void *fp)
{
__le32 buf[2];
u32 i, len;
int rc;
rc = next_entry(buf, fp, sizeof(u32) * 2);
if (rc)
return rc;
node->cur_state = le32_to_cpu(buf[0]);
/* expr */
len = le32_to_cpu(buf[1]);
node->expr.nodes = kcalloc(len, sizeof(*node->expr.nodes), GFP_KERNEL);
if (!node->expr.nodes)
return -ENOMEM;
node->expr.len = len;
for (i = 0; i < len; i++) {
struct cond_expr_node *expr = &node->expr.nodes[i];
rc = next_entry(buf, fp, sizeof(u32) * 2);
if (rc)
return rc;
expr->expr_type = le32_to_cpu(buf[0]);
expr->bool = le32_to_cpu(buf[1]);
if (!expr_node_isvalid(p, expr))
return -EINVAL;
}
rc = cond_read_av_list(p, fp, &node->true_list, NULL);
if (rc)
return rc;
return cond_read_av_list(p, fp, &node->false_list, &node->true_list);
}
int cond_read_list(struct policydb *p, void *fp)
{
__le32 buf[1];
u32 i, len;
int rc;
rc = next_entry(buf, fp, sizeof(buf));
if (rc)
return rc;
len = le32_to_cpu(buf[0]);
p->cond_list = kcalloc(len, sizeof(*p->cond_list), GFP_KERNEL);
if (!p->cond_list)
return -ENOMEM;
rc = avtab_alloc(&(p->te_cond_avtab), p->te_avtab.nel);
if (rc)
goto err;
p->cond_list_len = len;
for (i = 0; i < len; i++) {
rc = cond_read_node(p, &p->cond_list[i], fp);
if (rc)
goto err;
}
return 0;
err:
cond_list_destroy(p);
p->cond_list = NULL;
return rc;
}
int cond_write_bool(void *vkey, void *datum, void *ptr)
{
char *key = vkey;
struct cond_bool_datum *booldatum = datum;
struct policy_data *pd = ptr;
void *fp = pd->fp;
__le32 buf[3];
u32 len;
int rc;
len = strlen(key);
buf[0] = cpu_to_le32(booldatum->value);
buf[1] = cpu_to_le32(booldatum->state);
buf[2] = cpu_to_le32(len);
rc = put_entry(buf, sizeof(u32), 3, fp);
if (rc)
return rc;
rc = put_entry(key, 1, len, fp);
if (rc)
return rc;
return 0;
}
/*
* cond_write_cond_av_list doesn't write out the av_list nodes.
* Instead it writes out the key/value pairs from the avtab. This
* is necessary because there is no way to uniquely identifying rules
* in the avtab so it is not possible to associate individual rules
* in the avtab with a conditional without saving them as part of
* the conditional. This means that the avtab with the conditional
* rules will not be saved but will be rebuilt on policy load.
*/
static int cond_write_av_list(struct policydb *p,
struct cond_av_list *list, struct policy_file *fp)
{
__le32 buf[1];
u32 i;
int rc;
buf[0] = cpu_to_le32(list->len);
rc = put_entry(buf, sizeof(u32), 1, fp);
if (rc)
return rc;
for (i = 0; i < list->len; i++) {
rc = avtab_write_item(p, list->nodes[i], fp);
if (rc)
return rc;
}
return 0;
}
static int cond_write_node(struct policydb *p, struct cond_node *node,
struct policy_file *fp)
{
__le32 buf[2];
int rc;
u32 i;
buf[0] = cpu_to_le32(node->cur_state);
rc = put_entry(buf, sizeof(u32), 1, fp);
if (rc)
return rc;
buf[0] = cpu_to_le32(node->expr.len);
rc = put_entry(buf, sizeof(u32), 1, fp);
if (rc)
return rc;
for (i = 0; i < node->expr.len; i++) {
buf[0] = cpu_to_le32(node->expr.nodes[i].expr_type);
buf[1] = cpu_to_le32(node->expr.nodes[i].bool);
rc = put_entry(buf, sizeof(u32), 2, fp);
if (rc)
return rc;
}
rc = cond_write_av_list(p, &node->true_list, fp);
if (rc)
return rc;
rc = cond_write_av_list(p, &node->false_list, fp);
if (rc)
return rc;
return 0;
}
int cond_write_list(struct policydb *p, void *fp)
{
u32 i;
__le32 buf[1];
int rc;
buf[0] = cpu_to_le32(p->cond_list_len);
rc = put_entry(buf, sizeof(u32), 1, fp);
if (rc)
return rc;
for (i = 0; i < p->cond_list_len; i++) {
rc = cond_write_node(p, &p->cond_list[i], fp);
if (rc)
return rc;
}
return 0;
}
selinux: extended permissions for ioctls Add extended permissions logic to selinux. Extended permissions provides additional permissions in 256 bit increments. Extend the generic ioctl permission check to use the extended permissions for per-command filtering. Source/target/class sets including the ioctl permission may additionally include a set of commands. Example: allowxperm <source> <target>:<class> ioctl unpriv_app_socket_cmds auditallowxperm <source> <target>:<class> ioctl priv_gpu_cmds Where unpriv_app_socket_cmds and priv_gpu_cmds are macros representing commonly granted sets of ioctl commands. When ioctl commands are omitted only the permissions are checked. This feature is intended to provide finer granularity for the ioctl permission that may be too imprecise. For example, the same driver may use ioctls to provide important and benign functionality such as driver version or socket type as well as dangerous capabilities such as debugging features, read/write/execute to physical memory or access to sensitive data. Per-command filtering provides a mechanism to reduce the attack surface of the kernel, and limit applications to the subset of commands required. The format of the policy binary has been modified to include ioctl commands, and the policy version number has been incremented to POLICYDB_VERSION_XPERMS_IOCTL=30 to account for the format change. The extended permissions logic is deliberately generic to allow components to be reused e.g. netlink filters Signed-off-by: Jeff Vander Stoep <jeffv@google.com> Acked-by: Nick Kralevich <nnk@google.com> Signed-off-by: Paul Moore <pmoore@redhat.com>
2015-07-11 05:19:56 +08:00
void cond_compute_xperms(struct avtab *ctab, struct avtab_key *key,
struct extended_perms_decision *xpermd)
{
struct avtab_node *node;
if (!ctab || !key || !xpermd)
return;
for (node = avtab_search_node(ctab, key); node;
node = avtab_search_node_next(node, key->specified)) {
if (node->key.specified & AVTAB_ENABLED)
services_compute_xperms_decision(xpermd, node);
}
return;
}
/* Determine whether additional permissions are granted by the conditional
* av table, and if so, add them to the result
*/
selinux: extended permissions for ioctls Add extended permissions logic to selinux. Extended permissions provides additional permissions in 256 bit increments. Extend the generic ioctl permission check to use the extended permissions for per-command filtering. Source/target/class sets including the ioctl permission may additionally include a set of commands. Example: allowxperm <source> <target>:<class> ioctl unpriv_app_socket_cmds auditallowxperm <source> <target>:<class> ioctl priv_gpu_cmds Where unpriv_app_socket_cmds and priv_gpu_cmds are macros representing commonly granted sets of ioctl commands. When ioctl commands are omitted only the permissions are checked. This feature is intended to provide finer granularity for the ioctl permission that may be too imprecise. For example, the same driver may use ioctls to provide important and benign functionality such as driver version or socket type as well as dangerous capabilities such as debugging features, read/write/execute to physical memory or access to sensitive data. Per-command filtering provides a mechanism to reduce the attack surface of the kernel, and limit applications to the subset of commands required. The format of the policy binary has been modified to include ioctl commands, and the policy version number has been incremented to POLICYDB_VERSION_XPERMS_IOCTL=30 to account for the format change. The extended permissions logic is deliberately generic to allow components to be reused e.g. netlink filters Signed-off-by: Jeff Vander Stoep <jeffv@google.com> Acked-by: Nick Kralevich <nnk@google.com> Signed-off-by: Paul Moore <pmoore@redhat.com>
2015-07-11 05:19:56 +08:00
void cond_compute_av(struct avtab *ctab, struct avtab_key *key,
struct av_decision *avd, struct extended_perms *xperms)
{
struct avtab_node *node;
if (!ctab || !key || !avd)
return;
for (node = avtab_search_node(ctab, key); node;
node = avtab_search_node_next(node, key->specified)) {
if ((u16)(AVTAB_ALLOWED|AVTAB_ENABLED) ==
(node->key.specified & (AVTAB_ALLOWED|AVTAB_ENABLED)))
selinux: extended permissions for ioctls Add extended permissions logic to selinux. Extended permissions provides additional permissions in 256 bit increments. Extend the generic ioctl permission check to use the extended permissions for per-command filtering. Source/target/class sets including the ioctl permission may additionally include a set of commands. Example: allowxperm <source> <target>:<class> ioctl unpriv_app_socket_cmds auditallowxperm <source> <target>:<class> ioctl priv_gpu_cmds Where unpriv_app_socket_cmds and priv_gpu_cmds are macros representing commonly granted sets of ioctl commands. When ioctl commands are omitted only the permissions are checked. This feature is intended to provide finer granularity for the ioctl permission that may be too imprecise. For example, the same driver may use ioctls to provide important and benign functionality such as driver version or socket type as well as dangerous capabilities such as debugging features, read/write/execute to physical memory or access to sensitive data. Per-command filtering provides a mechanism to reduce the attack surface of the kernel, and limit applications to the subset of commands required. The format of the policy binary has been modified to include ioctl commands, and the policy version number has been incremented to POLICYDB_VERSION_XPERMS_IOCTL=30 to account for the format change. The extended permissions logic is deliberately generic to allow components to be reused e.g. netlink filters Signed-off-by: Jeff Vander Stoep <jeffv@google.com> Acked-by: Nick Kralevich <nnk@google.com> Signed-off-by: Paul Moore <pmoore@redhat.com>
2015-07-11 05:19:56 +08:00
avd->allowed |= node->datum.u.data;
if ((u16)(AVTAB_AUDITDENY|AVTAB_ENABLED) ==
(node->key.specified & (AVTAB_AUDITDENY|AVTAB_ENABLED)))
/* Since a '0' in an auditdeny mask represents a
* permission we do NOT want to audit (dontaudit), we use
* the '&' operand to ensure that all '0's in the mask
* are retained (much unlike the allow and auditallow cases).
*/
selinux: extended permissions for ioctls Add extended permissions logic to selinux. Extended permissions provides additional permissions in 256 bit increments. Extend the generic ioctl permission check to use the extended permissions for per-command filtering. Source/target/class sets including the ioctl permission may additionally include a set of commands. Example: allowxperm <source> <target>:<class> ioctl unpriv_app_socket_cmds auditallowxperm <source> <target>:<class> ioctl priv_gpu_cmds Where unpriv_app_socket_cmds and priv_gpu_cmds are macros representing commonly granted sets of ioctl commands. When ioctl commands are omitted only the permissions are checked. This feature is intended to provide finer granularity for the ioctl permission that may be too imprecise. For example, the same driver may use ioctls to provide important and benign functionality such as driver version or socket type as well as dangerous capabilities such as debugging features, read/write/execute to physical memory or access to sensitive data. Per-command filtering provides a mechanism to reduce the attack surface of the kernel, and limit applications to the subset of commands required. The format of the policy binary has been modified to include ioctl commands, and the policy version number has been incremented to POLICYDB_VERSION_XPERMS_IOCTL=30 to account for the format change. The extended permissions logic is deliberately generic to allow components to be reused e.g. netlink filters Signed-off-by: Jeff Vander Stoep <jeffv@google.com> Acked-by: Nick Kralevich <nnk@google.com> Signed-off-by: Paul Moore <pmoore@redhat.com>
2015-07-11 05:19:56 +08:00
avd->auditdeny &= node->datum.u.data;
if ((u16)(AVTAB_AUDITALLOW|AVTAB_ENABLED) ==
(node->key.specified & (AVTAB_AUDITALLOW|AVTAB_ENABLED)))
selinux: extended permissions for ioctls Add extended permissions logic to selinux. Extended permissions provides additional permissions in 256 bit increments. Extend the generic ioctl permission check to use the extended permissions for per-command filtering. Source/target/class sets including the ioctl permission may additionally include a set of commands. Example: allowxperm <source> <target>:<class> ioctl unpriv_app_socket_cmds auditallowxperm <source> <target>:<class> ioctl priv_gpu_cmds Where unpriv_app_socket_cmds and priv_gpu_cmds are macros representing commonly granted sets of ioctl commands. When ioctl commands are omitted only the permissions are checked. This feature is intended to provide finer granularity for the ioctl permission that may be too imprecise. For example, the same driver may use ioctls to provide important and benign functionality such as driver version or socket type as well as dangerous capabilities such as debugging features, read/write/execute to physical memory or access to sensitive data. Per-command filtering provides a mechanism to reduce the attack surface of the kernel, and limit applications to the subset of commands required. The format of the policy binary has been modified to include ioctl commands, and the policy version number has been incremented to POLICYDB_VERSION_XPERMS_IOCTL=30 to account for the format change. The extended permissions logic is deliberately generic to allow components to be reused e.g. netlink filters Signed-off-by: Jeff Vander Stoep <jeffv@google.com> Acked-by: Nick Kralevich <nnk@google.com> Signed-off-by: Paul Moore <pmoore@redhat.com>
2015-07-11 05:19:56 +08:00
avd->auditallow |= node->datum.u.data;
if (xperms && (node->key.specified & AVTAB_ENABLED) &&
selinux: extended permissions for ioctls Add extended permissions logic to selinux. Extended permissions provides additional permissions in 256 bit increments. Extend the generic ioctl permission check to use the extended permissions for per-command filtering. Source/target/class sets including the ioctl permission may additionally include a set of commands. Example: allowxperm <source> <target>:<class> ioctl unpriv_app_socket_cmds auditallowxperm <source> <target>:<class> ioctl priv_gpu_cmds Where unpriv_app_socket_cmds and priv_gpu_cmds are macros representing commonly granted sets of ioctl commands. When ioctl commands are omitted only the permissions are checked. This feature is intended to provide finer granularity for the ioctl permission that may be too imprecise. For example, the same driver may use ioctls to provide important and benign functionality such as driver version or socket type as well as dangerous capabilities such as debugging features, read/write/execute to physical memory or access to sensitive data. Per-command filtering provides a mechanism to reduce the attack surface of the kernel, and limit applications to the subset of commands required. The format of the policy binary has been modified to include ioctl commands, and the policy version number has been incremented to POLICYDB_VERSION_XPERMS_IOCTL=30 to account for the format change. The extended permissions logic is deliberately generic to allow components to be reused e.g. netlink filters Signed-off-by: Jeff Vander Stoep <jeffv@google.com> Acked-by: Nick Kralevich <nnk@google.com> Signed-off-by: Paul Moore <pmoore@redhat.com>
2015-07-11 05:19:56 +08:00
(node->key.specified & AVTAB_XPERMS))
services_compute_xperms_drivers(xperms, node);
}
}
selinux: refactor changing booleans Refactor the logic for changing SELinux policy booleans in a similar manner to the refactoring of policy load, thereby reducing the size of the critical section when the policy write-lock is held and making it easier to convert the policy rwlock to RCU in the future. Instead of directly modifying the policydb in place, modify a copy and then swap it into place through a single pointer update. Only fully copy the portions of the policydb that are affected by boolean changes to avoid the full cost of a deep policydb copy. Introduce another level of indirection for the sidtab since changing booleans does not require updating the sidtab, unlike policy load. While we are here, create a common helper for notifying other kernel components and userspace of a policy change and call it from both security_set_bools() and selinux_policy_commit(). Based on an old (2004) patch by Kaigai Kohei [1] to convert the policy rwlock to RCU that was deferred at the time since it did not significantly improve performance and introduced complexity. Peter Enderborg later submitted a patch series to convert to RCU [2] that would have made changing booleans a much more expensive operation by requiring a full policydb_write();policydb_read(); sequence to deep copy the entire policydb and also had concerns regarding atomic allocations. This change is now simplified by the earlier work to encapsulate policy state in the selinux_policy struct and to refactor policy load. After this change, the last major obstacle to converting the policy rwlock to RCU is likely the sidtab live convert support. [1] https://lore.kernel.org/selinux/6e2f9128-e191-ebb3-0e87-74bfccb0767f@tycho.nsa.gov/ [2] https://lore.kernel.org/selinux/20180530141104.28569-1-peter.enderborg@sony.com/ Signed-off-by: Stephen Smalley <stephen.smalley.work@gmail.com> Signed-off-by: Paul Moore <paul@paul-moore.com>
2020-08-12 03:01:56 +08:00
static int cond_dup_av_list(struct cond_av_list *new,
struct cond_av_list *orig,
struct avtab *avtab)
{
u32 i;
memset(new, 0, sizeof(*new));
new->nodes = kcalloc(orig->len, sizeof(*new->nodes), GFP_KERNEL);
if (!new->nodes)
return -ENOMEM;
for (i = 0; i < orig->len; i++) {
selinux: fix cond_list corruption when changing booleans commit d8f5f0ea5b86300390b026b6c6e7836b7150814a upstream. Currently, duplicate_policydb_cond_list() first copies the whole conditional avtab and then tries to link to the correct entries in cond_dup_av_list() using avtab_search(). However, since the conditional avtab may contain multiple entries with the same key, this approach often fails to find the right entry, potentially leading to wrong rules being activated/deactivated when booleans are changed. To fix this, instead start with an empty conditional avtab and add the individual entries one-by-one while building the new av_lists. This approach leads to the correct result, since each entry is present in the av_lists exactly once. The issue can be reproduced with Fedora policy as follows: # sesearch -s ftpd_t -t public_content_rw_t -c dir -p create -A allow ftpd_t non_security_file_type:dir { add_name create getattr ioctl link lock open read remove_name rename reparent rmdir search setattr unlink watch watch_reads write }; [ ftpd_full_access ]:True allow ftpd_t public_content_rw_t:dir { add_name create link remove_name rename reparent rmdir setattr unlink watch watch_reads write }; [ ftpd_anon_write ]:True # setsebool ftpd_anon_write=off ftpd_connect_all_unreserved=off ftpd_connect_db=off ftpd_full_access=off On fixed kernels, the sesearch output is the same after the setsebool command: # sesearch -s ftpd_t -t public_content_rw_t -c dir -p create -A allow ftpd_t non_security_file_type:dir { add_name create getattr ioctl link lock open read remove_name rename reparent rmdir search setattr unlink watch watch_reads write }; [ ftpd_full_access ]:True allow ftpd_t public_content_rw_t:dir { add_name create link remove_name rename reparent rmdir setattr unlink watch watch_reads write }; [ ftpd_anon_write ]:True While on the broken kernels, it will be different: # sesearch -s ftpd_t -t public_content_rw_t -c dir -p create -A allow ftpd_t non_security_file_type:dir { add_name create getattr ioctl link lock open read remove_name rename reparent rmdir search setattr unlink watch watch_reads write }; [ ftpd_full_access ]:True allow ftpd_t non_security_file_type:dir { add_name create getattr ioctl link lock open read remove_name rename reparent rmdir search setattr unlink watch watch_reads write }; [ ftpd_full_access ]:True allow ftpd_t non_security_file_type:dir { add_name create getattr ioctl link lock open read remove_name rename reparent rmdir search setattr unlink watch watch_reads write }; [ ftpd_full_access ]:True While there, also simplify the computation of nslots. This changes the nslots values for nrules 2 or 3 to just two slots instead of 4, which makes the sequence more consistent. Cc: stable@vger.kernel.org Fixes: c7c556f1e81b ("selinux: refactor changing booleans") Signed-off-by: Ondrej Mosnacek <omosnace@redhat.com> Signed-off-by: Paul Moore <paul@paul-moore.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2021-04-02 16:56:19 +08:00
new->nodes[i] = avtab_insert_nonunique(avtab,
&orig->nodes[i]->key,
&orig->nodes[i]->datum);
if (!new->nodes[i])
return -ENOMEM;
selinux: refactor changing booleans Refactor the logic for changing SELinux policy booleans in a similar manner to the refactoring of policy load, thereby reducing the size of the critical section when the policy write-lock is held and making it easier to convert the policy rwlock to RCU in the future. Instead of directly modifying the policydb in place, modify a copy and then swap it into place through a single pointer update. Only fully copy the portions of the policydb that are affected by boolean changes to avoid the full cost of a deep policydb copy. Introduce another level of indirection for the sidtab since changing booleans does not require updating the sidtab, unlike policy load. While we are here, create a common helper for notifying other kernel components and userspace of a policy change and call it from both security_set_bools() and selinux_policy_commit(). Based on an old (2004) patch by Kaigai Kohei [1] to convert the policy rwlock to RCU that was deferred at the time since it did not significantly improve performance and introduced complexity. Peter Enderborg later submitted a patch series to convert to RCU [2] that would have made changing booleans a much more expensive operation by requiring a full policydb_write();policydb_read(); sequence to deep copy the entire policydb and also had concerns regarding atomic allocations. This change is now simplified by the earlier work to encapsulate policy state in the selinux_policy struct and to refactor policy load. After this change, the last major obstacle to converting the policy rwlock to RCU is likely the sidtab live convert support. [1] https://lore.kernel.org/selinux/6e2f9128-e191-ebb3-0e87-74bfccb0767f@tycho.nsa.gov/ [2] https://lore.kernel.org/selinux/20180530141104.28569-1-peter.enderborg@sony.com/ Signed-off-by: Stephen Smalley <stephen.smalley.work@gmail.com> Signed-off-by: Paul Moore <paul@paul-moore.com>
2020-08-12 03:01:56 +08:00
new->len++;
}
return 0;
}
static int duplicate_policydb_cond_list(struct policydb *newp,
struct policydb *origp)
{
int rc, i, j;
selinux: fix cond_list corruption when changing booleans commit d8f5f0ea5b86300390b026b6c6e7836b7150814a upstream. Currently, duplicate_policydb_cond_list() first copies the whole conditional avtab and then tries to link to the correct entries in cond_dup_av_list() using avtab_search(). However, since the conditional avtab may contain multiple entries with the same key, this approach often fails to find the right entry, potentially leading to wrong rules being activated/deactivated when booleans are changed. To fix this, instead start with an empty conditional avtab and add the individual entries one-by-one while building the new av_lists. This approach leads to the correct result, since each entry is present in the av_lists exactly once. The issue can be reproduced with Fedora policy as follows: # sesearch -s ftpd_t -t public_content_rw_t -c dir -p create -A allow ftpd_t non_security_file_type:dir { add_name create getattr ioctl link lock open read remove_name rename reparent rmdir search setattr unlink watch watch_reads write }; [ ftpd_full_access ]:True allow ftpd_t public_content_rw_t:dir { add_name create link remove_name rename reparent rmdir setattr unlink watch watch_reads write }; [ ftpd_anon_write ]:True # setsebool ftpd_anon_write=off ftpd_connect_all_unreserved=off ftpd_connect_db=off ftpd_full_access=off On fixed kernels, the sesearch output is the same after the setsebool command: # sesearch -s ftpd_t -t public_content_rw_t -c dir -p create -A allow ftpd_t non_security_file_type:dir { add_name create getattr ioctl link lock open read remove_name rename reparent rmdir search setattr unlink watch watch_reads write }; [ ftpd_full_access ]:True allow ftpd_t public_content_rw_t:dir { add_name create link remove_name rename reparent rmdir setattr unlink watch watch_reads write }; [ ftpd_anon_write ]:True While on the broken kernels, it will be different: # sesearch -s ftpd_t -t public_content_rw_t -c dir -p create -A allow ftpd_t non_security_file_type:dir { add_name create getattr ioctl link lock open read remove_name rename reparent rmdir search setattr unlink watch watch_reads write }; [ ftpd_full_access ]:True allow ftpd_t non_security_file_type:dir { add_name create getattr ioctl link lock open read remove_name rename reparent rmdir search setattr unlink watch watch_reads write }; [ ftpd_full_access ]:True allow ftpd_t non_security_file_type:dir { add_name create getattr ioctl link lock open read remove_name rename reparent rmdir search setattr unlink watch watch_reads write }; [ ftpd_full_access ]:True While there, also simplify the computation of nslots. This changes the nslots values for nrules 2 or 3 to just two slots instead of 4, which makes the sequence more consistent. Cc: stable@vger.kernel.org Fixes: c7c556f1e81b ("selinux: refactor changing booleans") Signed-off-by: Ondrej Mosnacek <omosnace@redhat.com> Signed-off-by: Paul Moore <paul@paul-moore.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2021-04-02 16:56:19 +08:00
rc = avtab_alloc_dup(&newp->te_cond_avtab, &origp->te_cond_avtab);
selinux: refactor changing booleans Refactor the logic for changing SELinux policy booleans in a similar manner to the refactoring of policy load, thereby reducing the size of the critical section when the policy write-lock is held and making it easier to convert the policy rwlock to RCU in the future. Instead of directly modifying the policydb in place, modify a copy and then swap it into place through a single pointer update. Only fully copy the portions of the policydb that are affected by boolean changes to avoid the full cost of a deep policydb copy. Introduce another level of indirection for the sidtab since changing booleans does not require updating the sidtab, unlike policy load. While we are here, create a common helper for notifying other kernel components and userspace of a policy change and call it from both security_set_bools() and selinux_policy_commit(). Based on an old (2004) patch by Kaigai Kohei [1] to convert the policy rwlock to RCU that was deferred at the time since it did not significantly improve performance and introduced complexity. Peter Enderborg later submitted a patch series to convert to RCU [2] that would have made changing booleans a much more expensive operation by requiring a full policydb_write();policydb_read(); sequence to deep copy the entire policydb and also had concerns regarding atomic allocations. This change is now simplified by the earlier work to encapsulate policy state in the selinux_policy struct and to refactor policy load. After this change, the last major obstacle to converting the policy rwlock to RCU is likely the sidtab live convert support. [1] https://lore.kernel.org/selinux/6e2f9128-e191-ebb3-0e87-74bfccb0767f@tycho.nsa.gov/ [2] https://lore.kernel.org/selinux/20180530141104.28569-1-peter.enderborg@sony.com/ Signed-off-by: Stephen Smalley <stephen.smalley.work@gmail.com> Signed-off-by: Paul Moore <paul@paul-moore.com>
2020-08-12 03:01:56 +08:00
if (rc)
return rc;
newp->cond_list_len = 0;
newp->cond_list = kcalloc(origp->cond_list_len,
sizeof(*newp->cond_list),
GFP_KERNEL);
if (!newp->cond_list)
goto error;
for (i = 0; i < origp->cond_list_len; i++) {
struct cond_node *newn = &newp->cond_list[i];
struct cond_node *orign = &origp->cond_list[i];
newp->cond_list_len++;
newn->cur_state = orign->cur_state;
newn->expr.nodes = kcalloc(orign->expr.len,
sizeof(*newn->expr.nodes), GFP_KERNEL);
if (!newn->expr.nodes)
goto error;
for (j = 0; j < orign->expr.len; j++)
newn->expr.nodes[j] = orign->expr.nodes[j];
newn->expr.len = orign->expr.len;
rc = cond_dup_av_list(&newn->true_list, &orign->true_list,
&newp->te_cond_avtab);
if (rc)
goto error;
rc = cond_dup_av_list(&newn->false_list, &orign->false_list,
&newp->te_cond_avtab);
if (rc)
goto error;
}
return 0;
error:
avtab_destroy(&newp->te_cond_avtab);
cond_list_destroy(newp);
return -ENOMEM;
}
static int cond_bools_destroy(void *key, void *datum, void *args)
{
/* key was not copied so no need to free here */
kfree(datum);
return 0;
}
static int cond_bools_copy(struct hashtab_node *new, struct hashtab_node *orig, void *args)
{
struct cond_bool_datum *datum;
datum = kmemdup(orig->datum, sizeof(struct cond_bool_datum),
GFP_KERNEL);
selinux: refactor changing booleans Refactor the logic for changing SELinux policy booleans in a similar manner to the refactoring of policy load, thereby reducing the size of the critical section when the policy write-lock is held and making it easier to convert the policy rwlock to RCU in the future. Instead of directly modifying the policydb in place, modify a copy and then swap it into place through a single pointer update. Only fully copy the portions of the policydb that are affected by boolean changes to avoid the full cost of a deep policydb copy. Introduce another level of indirection for the sidtab since changing booleans does not require updating the sidtab, unlike policy load. While we are here, create a common helper for notifying other kernel components and userspace of a policy change and call it from both security_set_bools() and selinux_policy_commit(). Based on an old (2004) patch by Kaigai Kohei [1] to convert the policy rwlock to RCU that was deferred at the time since it did not significantly improve performance and introduced complexity. Peter Enderborg later submitted a patch series to convert to RCU [2] that would have made changing booleans a much more expensive operation by requiring a full policydb_write();policydb_read(); sequence to deep copy the entire policydb and also had concerns regarding atomic allocations. This change is now simplified by the earlier work to encapsulate policy state in the selinux_policy struct and to refactor policy load. After this change, the last major obstacle to converting the policy rwlock to RCU is likely the sidtab live convert support. [1] https://lore.kernel.org/selinux/6e2f9128-e191-ebb3-0e87-74bfccb0767f@tycho.nsa.gov/ [2] https://lore.kernel.org/selinux/20180530141104.28569-1-peter.enderborg@sony.com/ Signed-off-by: Stephen Smalley <stephen.smalley.work@gmail.com> Signed-off-by: Paul Moore <paul@paul-moore.com>
2020-08-12 03:01:56 +08:00
if (!datum)
return -ENOMEM;
new->key = orig->key; /* No need to copy, never modified */
new->datum = datum;
return 0;
}
static int cond_bools_index(void *key, void *datum, void *args)
{
struct cond_bool_datum *booldatum, **cond_bool_array;
booldatum = datum;
cond_bool_array = args;
cond_bool_array[booldatum->value - 1] = booldatum;
return 0;
}
static int duplicate_policydb_bools(struct policydb *newdb,
struct policydb *orig)
{
struct cond_bool_datum **cond_bool_array;
int rc;
cond_bool_array = kmalloc_array(orig->p_bools.nprim,
sizeof(*orig->bool_val_to_struct),
GFP_KERNEL);
if (!cond_bool_array)
return -ENOMEM;
rc = hashtab_duplicate(&newdb->p_bools.table, &orig->p_bools.table,
cond_bools_copy, cond_bools_destroy, NULL);
if (rc) {
kfree(cond_bool_array);
return -ENOMEM;
}
hashtab_map(&newdb->p_bools.table, cond_bools_index, cond_bool_array);
newdb->bool_val_to_struct = cond_bool_array;
newdb->p_bools.nprim = orig->p_bools.nprim;
return 0;
}
void cond_policydb_destroy_dup(struct policydb *p)
{
hashtab_map(&p->p_bools.table, cond_bools_destroy, NULL);
hashtab_destroy(&p->p_bools.table);
cond_policydb_destroy(p);
}
int cond_policydb_dup(struct policydb *new, struct policydb *orig)
{
cond_policydb_init(new);
if (duplicate_policydb_bools(new, orig))
return -ENOMEM;
if (duplicate_policydb_cond_list(new, orig)) {
cond_policydb_destroy_dup(new);
return -ENOMEM;
}
return 0;
}