| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-Id: <1282293963-27807-9-git-send-email-mitr@redhat.com>
Date: Fri, 20 Aug 2010 10:45:51 +0200
From: Miloslav Trmač <mitr@...hat.com>
To: Herbert Xu <herbert@...dor.hengli.com.au>
Cc: linux-crypto@...r.kernel.org,
Nikos Mavrogiannopoulos <n.mavrogiannopoulos@...il.com>,
Neil Horman <nhorman@...hat.com>, linux-kernel@...r.kernel.org,
Miloslav Trmač <mitr@...hat.com>
Subject: [PATCH 08/19] Add per-process and per-user limits
Right now only key objects, not crypto sessions, are limited.
---
crypto/userspace/Makefile | 2 +-
crypto/userspace/ncr-limits.c | 247 +++++++++++++++++++++++++++++++++++++++++
2 files changed, 248 insertions(+), 1 deletions(-)
create mode 100644 crypto/userspace/ncr-limits.c
diff --git a/crypto/userspace/Makefile b/crypto/userspace/Makefile
index a37969b..3dfbc39 100644
--- a/crypto/userspace/Makefile
+++ b/crypto/userspace/Makefile
@@ -1,6 +1,6 @@
ccflags-y += -I$(src)/libtommath -I$(src)/libtomcrypt/headers -I$(src) -DLTC_SOURCE
-cryptodev-objs := cryptodev_main.o cryptodev_cipher.o utils.o
+cryptodev-objs := cryptodev_main.o cryptodev_cipher.o ncr-limits.o utils.o
obj-$(CONFIG_CRYPTO_USERSPACE) += cryptodev.o
diff --git a/crypto/userspace/ncr-limits.c b/crypto/userspace/ncr-limits.c
new file mode 100644
index 0000000..345cca7
--- /dev/null
+++ b/crypto/userspace/ncr-limits.c
@@ -0,0 +1,247 @@
+/*
+ * New driver for /dev/crypto device (aka CryptoDev)
+
+ * Copyright (c) 2010 Katholieke Universiteit Leuven
+ *
+ * Author: Nikos Mavrogiannopoulos <nmav@...tls.org>
+ *
+ * This file is part of linux cryptodev.
+ *
+ * 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., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
+ */
+
+#include <linux/hash.h>
+#include <linux/mutex.h>
+#include <linux/types.h>
+#include <linux/mm.h>
+#include <linux/ncr.h>
+#include <linux/slab.h>
+#include <linux/highmem.h>
+#include <linux/random.h>
+#include <asm/atomic.h>
+#include <linux/version.h>
+#include <linux/file.h>
+#include <linux/cred.h>
+#include "ncr-int.h"
+
+/* arbitrary now */
+static unsigned int max_per_user[] = {
+ [LIMIT_TYPE_KEY] = 128,
+};
+
+static unsigned int max_per_process[] = {
+ [LIMIT_TYPE_KEY] = 64,
+};
+
+struct limit_user_item_st {
+ struct hlist_node hlist;
+ uid_t uid;
+ atomic_t cnt[NUM_LIMIT_TYPES];
+};
+
+struct limit_process_item_st {
+ struct hlist_node hlist;
+ pid_t pid;
+ atomic_t cnt[NUM_LIMIT_TYPES];
+};
+
+static struct mutex user_limit_mutex;
+#define USER_LIMIT_HASH_BITS 7
+#define USER_LIMIT_TABLE_SIZE (1 << USER_LIMIT_HASH_BITS)
+static struct hlist_head user_limit_table[USER_LIMIT_TABLE_SIZE];
+
+static struct hlist_head *user_limit_hash(uid_t uid)
+{
+ return &user_limit_table[hash_long(uid, USER_LIMIT_HASH_BITS)];
+}
+
+static struct mutex process_limit_mutex;
+#define PROCESS_LIMIT_HASH_BITS 9
+#define PROCESS_LIMIT_TABLE_SIZE (1 << PROCESS_LIMIT_HASH_BITS)
+static struct hlist_head process_limit_table[PROCESS_LIMIT_TABLE_SIZE];
+
+static struct hlist_head *process_limit_hash(pid_t pid)
+{
+ return &process_limit_table[hash_long(pid, PROCESS_LIMIT_HASH_BITS)];
+}
+
+void ncr_limits_init(void)
+{
+ size_t i;
+
+ mutex_init(&user_limit_mutex);
+ for (i = 0; i < USER_LIMIT_TABLE_SIZE; i++)
+ INIT_HLIST_HEAD(&user_limit_table[i]);
+
+ mutex_init(&process_limit_mutex);
+ for (i = 0; i < PROCESS_LIMIT_TABLE_SIZE; i++)
+ INIT_HLIST_HEAD(&process_limit_table[i]);
+}
+
+void ncr_limits_deinit(void)
+{
+struct limit_process_item_st* pitem;
+struct limit_user_item_st* uitem;
+struct hlist_node *pos, *tmp;
+size_t i;
+
+ mutex_lock(&user_limit_mutex);
+ for (i = 0; i < USER_LIMIT_TABLE_SIZE; i++) {
+ hlist_for_each_entry_safe(uitem, pos, tmp, &user_limit_table[i],
+ hlist) {
+ hlist_del(&uitem->hlist);
+ kfree(uitem);
+ }
+ }
+ mutex_unlock(&user_limit_mutex);
+
+ mutex_lock(&process_limit_mutex);
+ for (i = 0; i < PROCESS_LIMIT_TABLE_SIZE; i++) {
+ hlist_for_each_entry_safe(pitem, pos, tmp,
+ &process_limit_table[i], hlist) {
+ hlist_del(&pitem->hlist);
+ kfree(pitem);
+ }
+ }
+ mutex_unlock(&process_limit_mutex);
+
+}
+
+int ncr_limits_add_and_check(uid_t uid, pid_t pid, limits_type_t type)
+{
+struct limit_process_item_st* pitem;
+struct limit_user_item_st* uitem;
+struct hlist_head *user_head, *process_head;
+struct hlist_node *pos;
+int add = 1;
+int ret;
+ BUG_ON(type >= NUM_LIMIT_TYPES);
+
+ user_head = user_limit_hash(uid);
+ mutex_lock(&user_limit_mutex);
+ hlist_for_each_entry(uitem, pos, user_head, hlist) {
+ if (uitem->uid == uid) {
+ add = 0;
+
+ if (atomic_add_unless(&uitem->cnt[type], 1, max_per_user[type])==0) {
+ err();
+ mutex_unlock(&user_limit_mutex);
+ return -EPERM;
+ }
+ break;
+ }
+ }
+
+ if (add) {
+ size_t i;
+
+ uitem = kmalloc( sizeof(*uitem), GFP_KERNEL);
+ if (uitem == NULL) {
+ err();
+ mutex_unlock(&user_limit_mutex);
+ return -ENOMEM;
+ }
+ uitem->uid = uid;
+ for (i = 0; i < NUM_LIMIT_TYPES; i++)
+ atomic_set(&uitem->cnt[i], 0);
+ atomic_set(&uitem->cnt[type], 1);
+
+ hlist_add_head(&uitem->hlist, user_head);
+ }
+ mutex_unlock(&user_limit_mutex);
+
+ add = 1;
+ /* check process limits */
+ process_head = process_limit_hash(uid);
+ mutex_lock(&process_limit_mutex);
+ hlist_for_each_entry(pitem, pos, process_head, hlist) {
+ if (pitem->pid == pid) {
+ add = 0;
+ if (atomic_add_unless(&pitem->cnt[type], 1, max_per_process[type])==0) {
+ err();
+ mutex_unlock(&process_limit_mutex);
+
+ ret = -EPERM;
+ goto restore_user;
+ }
+ break;
+ }
+ }
+
+
+ if (add) {
+ size_t i;
+
+ pitem = kmalloc(sizeof(*pitem), GFP_KERNEL);
+ if (pitem == NULL) {
+ err();
+ mutex_unlock(&process_limit_mutex);
+ ret = -ENOMEM;
+ goto restore_user;
+ }
+ pitem->pid = pid;
+ for (i = 0; i < NUM_LIMIT_TYPES; i++)
+ atomic_set(&pitem->cnt[i], 0);
+ atomic_set(&pitem->cnt[type], 1);
+
+ hlist_add_head(&pitem->hlist, process_head);
+ }
+ mutex_unlock(&process_limit_mutex);
+
+ return 0;
+
+restore_user:
+ mutex_lock(&user_limit_mutex);
+ hlist_for_each_entry(uitem, pos, user_head, hlist) {
+ if (uitem->uid == uid) {
+ atomic_dec(&uitem->cnt[type]);
+ break;
+ }
+ }
+ mutex_unlock(&user_limit_mutex);
+ return ret;
+}
+
+void ncr_limits_remove(uid_t uid, pid_t pid, limits_type_t type)
+{
+struct limit_process_item_st* pitem;
+struct limit_user_item_st* uitem;
+struct hlist_head *hhead;
+struct hlist_node *pos;
+
+ BUG_ON(type >= NUM_LIMIT_TYPES);
+ hhead = user_limit_hash(uid);
+ mutex_lock(&user_limit_mutex);
+ hlist_for_each_entry(uitem, pos, hhead, hlist) {
+ if (uitem->uid == uid) {
+ atomic_dec(&uitem->cnt[type]);
+ break;
+ }
+ }
+ mutex_unlock(&user_limit_mutex);
+
+ /* check process limits */
+ hhead = process_limit_hash(uid);
+ mutex_lock(&process_limit_mutex);
+ hlist_for_each_entry(pitem, pos, hhead, hlist) {
+ if (pitem->pid == pid) {
+ atomic_dec(&pitem->cnt[type]);
+ break;
+ }
+ }
+ mutex_unlock(&process_limit_mutex);
+
+ return;
+}
--
1.7.2.1
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists