* [PATCH] char random: fix boot id uniqueness race
@ 2012-01-30 4:40 Mathieu Desnoyers
2012-01-31 15:59 ` Ted Ts'o
0 siblings, 1 reply; 3+ messages in thread
From: Mathieu Desnoyers @ 2012-01-30 4:40 UTC (permalink / raw)
To: Matt Mackall; +Cc: Theodore Ts'o, Greg Kroah-Hartman, linux-kernel
The proc file /proc/sys/kernel/random/boot_id can be read concurrently
by user-space processes. If two (or more) user-space processes
concurrently read boot_id when sysctl_bootid is not yet assigned, a race
can occur making boot_id differ between the reads. Because the whole
point of the boot id is to be unique across a kernel execution, fix this
by protecting this operation with a mutex, and introduce a
boot_id_generated flag, along with appropriate memory barriers, to let
the fast-path know if the boot ID has been generated without having to
hold the mutex.
I propose this approach rather than setting it up within an initcall(),
because letting execution randomness add to entropy before populating
the boot id seems to be a wanted property. Also, populating it lazily
rather than at boot time only makes the performance hit be taken when
boot_id is being read.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
CC: "Theodore Ts'o" <tytso@mit.edu>
CC: Matt Mackall <mpm@selenic.com>
CC: Greg Kroah-Hartman <greg@kroah.com>
---
drivers/char/random.c | 20 +++++++++++++++++---
1 file changed, 17 insertions(+), 3 deletions(-)
Index: linux-2.6-lttng/drivers/char/random.c
===================================================================
--- linux-2.6-lttng.orig/drivers/char/random.c
+++ linux-2.6-lttng/drivers/char/random.c
@@ -1231,6 +1231,8 @@ static int min_read_thresh = 8, min_writ
static int max_read_thresh = INPUT_POOL_WORDS * 32;
static int max_write_thresh = INPUT_POOL_WORDS * 32;
static char sysctl_bootid[16];
+static int boot_id_generated;
+static DEFINE_MUTEX(boot_id_mutex);
/*
* These functions is used to return both the bootid UUID, and random
@@ -1250,10 +1252,22 @@ static int proc_do_uuid(ctl_table *table
uuid = table->data;
if (!uuid) {
uuid = tmp_uuid;
- uuid[8] = 0;
- }
- if (uuid[8] == 0)
generate_random_uuid(uuid);
+ } else {
+ if (unlikely(!ACCESS_ONCE(boot_id_generated))) {
+ mutex_lock(&boot_id_mutex);
+ if (!boot_id_generated) {
+ generate_random_uuid(uuid);
+ /* Store uuid before boot_id_generated. */
+ smp_wmb();
+ boot_id_generated = 1;
+ }
+ mutex_unlock(&boot_id_mutex);
+ } else {
+ /* Load boot_id_generated before uuid */
+ smp_rmb();
+ }
+ }
sprintf(buf, "%pU", uuid);
--
Mathieu Desnoyers
Operating System Efficiency R&D Consultant
EfficiOS Inc.
http://www.efficios.com
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH] char random: fix boot id uniqueness race
2012-01-30 4:40 [PATCH] char random: fix boot id uniqueness race Mathieu Desnoyers
@ 2012-01-31 15:59 ` Ted Ts'o
2012-01-31 16:25 ` Mathieu Desnoyers
0 siblings, 1 reply; 3+ messages in thread
From: Ted Ts'o @ 2012-01-31 15:59 UTC (permalink / raw)
To: Mathieu Desnoyers; +Cc: Matt Mackall, Greg Kroah-Hartman, linux-kernel
On Sun, Jan 29, 2012 at 11:40:12PM -0500, Mathieu Desnoyers wrote:
> Index: linux-2.6-lttng/drivers/char/random.c
> ===================================================================
> --- linux-2.6-lttng.orig/drivers/char/random.c
> +++ linux-2.6-lttng/drivers/char/random.c
> @@ -1231,6 +1231,8 @@ static int min_read_thresh = 8, min_writ
> static int max_read_thresh = INPUT_POOL_WORDS * 32;
> static int max_write_thresh = INPUT_POOL_WORDS * 32;
> static char sysctl_bootid[16];
> +static int boot_id_generated;
> +static DEFINE_MUTEX(boot_id_mutex);
My preference would be to put these statics in proc_do_uuid(), but
that's arguably a nit.
> + } else {
> + if (unlikely(!ACCESS_ONCE(boot_id_generated))) {
> + mutex_lock(&boot_id_mutex);
> + if (!boot_id_generated) {
> + generate_random_uuid(uuid);
> + /* Store uuid before boot_id_generated. */
> + smp_wmb();
> + boot_id_generated = 1;
> + }
> + mutex_unlock(&boot_id_mutex);
> + } else {
> + /* Load boot_id_generated before uuid */
> + smp_rmb();
> + }
> + }
I don't believe the smp_wmb() and smp_rmb() are necessary here; the
mutex_lock() and mutex_unlock() should put the necessary memory
barriers in place.
Regards,
- Ted
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH] char random: fix boot id uniqueness race
2012-01-31 15:59 ` Ted Ts'o
@ 2012-01-31 16:25 ` Mathieu Desnoyers
0 siblings, 0 replies; 3+ messages in thread
From: Mathieu Desnoyers @ 2012-01-31 16:25 UTC (permalink / raw)
To: Ted Ts'o, Matt Mackall, Greg Kroah-Hartman, linux-kernel
Cc: Paul E. McKenney
* Ted Ts'o (tytso@mit.edu) wrote:
> On Sun, Jan 29, 2012 at 11:40:12PM -0500, Mathieu Desnoyers wrote:
> > Index: linux-2.6-lttng/drivers/char/random.c
> > ===================================================================
> > --- linux-2.6-lttng.orig/drivers/char/random.c
> > +++ linux-2.6-lttng/drivers/char/random.c
> > @@ -1231,6 +1231,8 @@ static int min_read_thresh = 8, min_writ
> > static int max_read_thresh = INPUT_POOL_WORDS * 32;
> > static int max_write_thresh = INPUT_POOL_WORDS * 32;
> > static char sysctl_bootid[16];
> > +static int boot_id_generated;
> > +static DEFINE_MUTEX(boot_id_mutex);
Hi Ted,
>
> My preference would be to put these statics in proc_do_uuid(), but
> that's arguably a nit.
No problem, can do.
>
> > + } else {
> > + if (unlikely(!ACCESS_ONCE(boot_id_generated))) {
> > + mutex_lock(&boot_id_mutex);
> > + if (!boot_id_generated) {
> > + generate_random_uuid(uuid);
> > + /* Store uuid before boot_id_generated. */
> > + smp_wmb();
> > + boot_id_generated = 1;
> > + }
> > + mutex_unlock(&boot_id_mutex);
> > + } else {
> > + /* Load boot_id_generated before uuid */
> > + smp_rmb();
> > + }
> > + }
>
> I don't believe the smp_wmb() and smp_rmb() are necessary here; the
> mutex_lock() and mutex_unlock() should put the necessary memory
> barriers in place.
The need for memory barriers is a consequence of letting the fast-path
run without holding this mutex.
Here is the race dealt with by the smp_rmb()/smp_wmb(). I'm showing the
result of reversed write order here:
CPU A CPU B
Load boot_id_generated
(test -> false)
mutex_lock(&boot_id_mutex)
(implied memory barrier
with acquire semantic)
Load boot_id_generated again
(test -> false)
boot_id_generated = 1
(both the compiler and
CPU are free to reorder
the boot_id_generated
store before uuid stores)
Load boot_id_generated
(test -> true)
Load uuid content
(races with generate_random_uuid:
result either 0 or corrupted)
Return corrupted uuid.
generate_random_uuid(uuid)
mutex_unlock(&boot_id_mutex)
I prefer not requiring the fast-path to take a mutex, because this
would transform a read-mostly operation into an operation that
requires cache-line exchanges (the mutex). However, if we want the
fast-path to be mutex-free, we need to enforce order with
memory barriers: smp_rmb on the read-side, smp_wmb on the
update-side. Failure to do so leads to the race shown above,
where a corrupted boot_id can be returned.
Please let me know if there are aspects of your question I can
address better,
Thanks,
Mathieu
>
> Regards,
>
> - Ted
--
Mathieu Desnoyers
Operating System Efficiency R&D Consultant
EfficiOS Inc.
http://www.efficios.com
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2012-01-31 16:25 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2012-01-30 4:40 [PATCH] char random: fix boot id uniqueness race Mathieu Desnoyers
2012-01-31 15:59 ` Ted Ts'o
2012-01-31 16:25 ` Mathieu Desnoyers
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).