| 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
| ||
|
Date: Tue, 15 Mar 2011 21:50:42 -0400
From: George Spelvin <linux@...izon.com>
To: penberg@...helsinki.fi, herbert@...dor.apana.org.au,
mpm@...enic.com
Cc: linux-mm@...ck.org, linux-kernel@...r.kernel.org, linux@...izon.com
Subject: [PATCH 8/8] mm/slub.c: Enable Kconfig control.
This introduces three CONFIG_ options:
* CONFIG_RAND_MOD, controlling the drivers/char/random.c support code
* CONFIG_SLUB_RANDOMIZE, enabling the support code
* CONFIG_SLUB_RANDOMIZE_BY_DEFAULT
Randomization may be enabled in three ways:
* By writing 1 to /sys/kernel/slab/$NAME/randomize
* By using the "r" flag in the slub_debug option
* By enabling CONFIG_SLUB_RANDOMIZE_BY_DEFAULT
The feature is independent of SLUB_DEBUG.
---
drivers/char/Kconfig | 4 ++++
drivers/char/random.c | 6 ++++++
include/linux/random.h | 2 ++
include/linux/slab.h | 5 +++++
init/Kconfig | 37 +++++++++++++++++++++++++++++++++++--
mm/slub.c | 21 ++++++++++++++++++++-
6 files changed, 72 insertions(+), 3 deletions(-)
diff --git a/drivers/char/Kconfig b/drivers/char/Kconfig
index b7980a83..d62a188 100644
--- a/drivers/char/Kconfig
+++ b/drivers/char/Kconfig
@@ -797,6 +797,10 @@ config NWFLASH
If you're not sure, say N.
+# selected by SLUB_RANDOMIZE if needed.
+config RAND_MOD
+ boolean
+
source "drivers/char/hw_random/Kconfig"
config NVRAM
diff --git a/drivers/char/random.c b/drivers/char/random.c
index fc36a98..cf7d71e 100644
--- a/drivers/char/random.c
+++ b/drivers/char/random.c
@@ -1626,8 +1626,10 @@ EXPORT_SYMBOL(secure_dccp_sequence_number);
*/
struct cpu_random {
u32 hash[4];
+#ifdef CONFIG_RAND_MOD
u32 lim, x;
int avail; /* Trailing bytes of hash[] available for seed */
+#endif
};
DEFINE_PER_CPU(struct cpu_random, get_random_int_data);
static u32 __get_random_int(u32 *hash)
@@ -1648,11 +1650,14 @@ unsigned int get_random_int(void)
struct cpu_random *r = &get_cpu_var(get_random_int_data);
u32 ret = __get_random_int(r->hash);
+#ifdef CONFIG_RAND_MOD
r->avail = 8;
+#endif
put_cpu_var(r);
return ret;
}
+#ifdef CONFIG_RAND_MOD
struct cpu_random *
get_random_mod_start(void)
{
@@ -1712,6 +1717,7 @@ get_random_mod_stop(struct cpu_random *r)
{
put_cpu_var(r);
}
+#endif
/*
* randomize_range() returns a start address such that
diff --git a/include/linux/random.h b/include/linux/random.h
index 2e1c227..769b2f6 100644
--- a/include/linux/random.h
+++ b/include/linux/random.h
@@ -84,10 +84,12 @@ unsigned long randomize_range(unsigned long start, unsigned long end, unsigned l
* They use per-CPU data, so preemption is disabled in the _start
* function and re-enabled in _stop.
*/
+#ifdef CONFIG_RAND_MOD
struct cpu_random; /* Opaque to acllers of this interface */
struct cpu_random *get_random_mod_start(void);
unsigned get_random_mod(struct cpu_random *r, unsigned m);
void get_random_mod_stop(struct cpu_random *r);
+#endif
u32 random32(void);
void srandom32(u32 seed);
diff --git a/include/linux/slab.h b/include/linux/slab.h
index 8e812f1..45931f8 100644
--- a/include/linux/slab.h
+++ b/include/linux/slab.h
@@ -80,7 +80,12 @@
#define SLAB_RECLAIM_ACCOUNT 0x00020000UL /* Objects are reclaimable */
#define SLAB_TEMPORARY SLAB_RECLAIM_ACCOUNT /* Objects are short-lived */
+#ifdef CONFIG_SLUB_RANDOMIZE
+/* Currently only supported by SLUB, this could be generalized if useful. */
#define SLAB_RANDOMIZE 0x04000000UL /* Randomize allocation order */
+#else
+#define SLAB_RANDOMIZE 0UL
+#endif
/*
* ZERO_SIZE_PTR will be returned for zero sized kmalloc requests.
*
diff --git a/init/Kconfig b/init/Kconfig
index be788c0..7ccb2ea 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -1180,8 +1180,41 @@ config SLUB_DEBUG
help
SLUB has extensive debug support features. Disabling these can
result in significant savings in code size. This also disables
- SLUB sysfs support. /sys/slab will not exist and there will be
- no support for cache validation etc.
+ SLUB sysfs support. /sys/kernel/slab will not exist and there
+ will be no support for cache validation etc.
+
+config SLUB_RANDOMIZE
+ default n
+ bool "Enable SLUB randomization support"
+ depends on SLUB
+ select CONFIG_RAND_MOD
+ help
+ This feature randomizes the order of allocation of kernel
+ objects within pages. To be precise, it shuffles the initial
+ free list of each newly allocated slab page in random order.
+
+ The intention is to impede buffer overrun attacks against kernel
+ objects. It will not always work; In many cases, the attacker
+ can simply allocate hundreds of objects before triggering the
+ overrun, and test them all to find which was corrupted.
+ (See discussion at http://marc.info/?t=129917479800001)
+
+ This only enables the code; randomization must be switched
+ on for each slab via /sys/kernel/slab, the slub_debug kernel
+ command line ('r' flag), or CONFIG_SLUB_RANDOMIZE_BY_DEFAULT.
+
+ This is for testing and the very paranoid. It imposes some
+ performance cost, and the kernel should not have bugs which
+ allow such attacks anyway. Say N unless you understand all
+ of this and still want the feature.
+
+config SLUB_RANDOMIZE_BY_DEFAULT
+ default n
+ bool "Enable SLUB randomization by default"
+ depends on SLUB_RANDOMIZE
+ help
+ Randomize all SLUB allocations starting at boot. This can
+ still be turned off on a per-slab basis using /sys/kernel/slab.
config COMPAT_BRK
bool "Disable heap randomization"
diff --git a/mm/slub.c b/mm/slub.c
index 4ba1db4..867ca4f 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -1031,6 +1031,10 @@ static int __init setup_slub_debug(char *str)
case 'a':
slub_debug |= SLAB_FAILSLAB;
break;
+ case 'r':
+ /* Allowed (but ignored) even if not configured */
+ slub_debug |= SLAB_RANDOMIZE;
+ break;
default:
printk(KERN_ERR "slub_debug option '%c' "
"unknown. skipped\n", *str);
@@ -1181,11 +1185,15 @@ static void setup_object(struct kmem_cache *s, void *object)
s->ctor(object);
}
+#ifdef CONFIG_SLUB_RANDOMIZE
/*
* Initialize a slab's free list in random order, to make
* buffer overrun attacks harder. Using a (moderately) secure
* random number generator, this ensures an attacker can't
* figure out which other object an overrun will hit.
+ *
+ * (This should be eliminated as dead code if CONFIG_SLUB_RANDOMIZE
+ * is not set.)
*/
static void *
setup_slab_randomized(struct kmem_cache *s, void *start, int count)
@@ -1211,6 +1219,10 @@ setup_slab_randomized(struct kmem_cache *s, void *start, int count)
return start;
}
+#else
+/* Should be eliminated as dead code, anyway */
+#define setup_slab_randomized(s, start, count) (start)
+#endif
static struct page *new_slab(struct kmem_cache *s, gfp_t flags, int node)
{
@@ -2377,6 +2389,9 @@ static int kmem_cache_open(struct kmem_cache *s,
s->ctor = ctor;
s->objsize = size;
s->align = align;
+#ifdef CONFIG_SLUB_RANDOMIZE_BY_DEFAULT
+ flags |= SLAB_RANDOMIZE;
+#endif
s->flags = kmem_cache_flags(size, flags, name, ctor);
if (!calculate_sizes(s, -1))
@@ -4192,6 +4207,7 @@ static ssize_t failslab_store(struct kmem_cache *s, const char *buf,
SLAB_ATTR(failslab);
#endif
+#if CONFIG_SLUB_RANDOMIZE
static ssize_t randomize_show(struct kmem_cache *s, char *buf)
{
return flag_show(s, buf, SLAB_RANDOMIZE);
@@ -4203,6 +4219,7 @@ static ssize_t randomize_store(struct kmem_cache *s,
return flag_store(s, buf, length, SLAB_RANDOMIZE);
}
SLAB_ATTR(randomize);
+#endif
static ssize_t shrink_show(struct kmem_cache *s, char *buf)
{
@@ -4336,8 +4353,10 @@ static struct attribute *slab_attrs[] = {
&hwcache_align_attr.attr,
&reclaim_account_attr.attr,
&destroy_by_rcu_attr.attr,
- &randomize_attr.attr,
&shrink_attr.attr,
+#ifdef CONFIG_SLUB_RANDOMIZE
+ &randomize_attr.attr,
+#endif
#ifdef CONFIG_SLUB_DEBUG
&total_objects_attr.attr,
&slabs_attr.attr,
--
1.7.4.1
--
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