1) reading nf_conntrack_locks_all needs ACQUIRE memory
ordering.
In addition, to simplify backporting, the patch also adds:
2) For some architectures the ACQUIRE during spin_lock only applies to
loading the lock, not to storing the lock state. E.g. see
commit 51d7d5205d33 ("powerpc: Add smp_mb() to arch_spin_is_locked()")
nf_conntrack_lock() does not handle this correctly:
/* 1) Acquire the lock */
spin_lock(lock);
while (unlikely(nf_conntrack_locks_all)) {
spin_unlock(lock);
Thus a memory barrier might be missing between spin_lock and reading
nf_conntrack_locks_all.
3) Between the write of nf_conntrack_locks_all and spin_unlock_wait(),
a memory barrier might be required.
As improvement:
4) Minor issue: If there would be many nf_conntrack_all_lock() callers,
then nf_conntrack_lock() would loop forever.
Therefore: Change nf_conntrack_lock and nf_conntract_lock_all() to the
approach used by ipc/sem.c:
1) add smp_load_acquire()
2) add smb_mb() after spin_lock()
Note: redundant if spin_unlock_wait() is implemented as spin_lock();
spin_unlock().
3) add smb_rmb() after spin_unlock_wait()
Note: redundant after
commit 2c6100227116 ("locking/qspinlock: Fix spin_unlock_wait() some more")
4) for nf_conntrack_lock, use spin_lock(&global_lock) instead of
spin_unlock_wait(&global_lock) and loop backward.
5) use smp_store_mb() instead of a raw smp_mb()
As the minimal bugfix, it might be sufficient just to add the
smp_load_acquire(), but then it must be checked first if all
updates to qspinlock were backported.
Fixes: b16c29191dc8
Signed-off-by: Manfred Spraul <manfred@xxxxxxxxxxxxxxxx>
Cc: <stable@xxxxxxxxxxxxxxx>
Cc: Sasha Levin <sasha.levin@xxxxxxxxxx>
Cc: Pablo Neira Ayuso <pablo@xxxxxxxxxxxxx>
Cc: netfilter-devel@xxxxxxxxxxxxxxx
---
net/netfilter/nf_conntrack_core.c | 41 ++++++++++++++++++++++++++-------------
1 file changed, 27 insertions(+), 14 deletions(-)
diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c
index 7d90a5d..3847f09 100644
--- a/net/netfilter/nf_conntrack_core.c
+++ b/net/netfilter/nf_conntrack_core.c
@@ -79,20 +79,29 @@ static __read_mostly bool nf_conntrack_locks_all;
void nf_conntrack_lock(spinlock_t *lock) __acquires(lock)
{
+ /* 1) Acquire the lock */
spin_lock(lock);
- while (unlikely(nf_conntrack_locks_all)) {
- spin_unlock(lock);
- /*
- * Order the 'nf_conntrack_locks_all' load vs. the
- * spin_unlock_wait() loads below, to ensure
- * that 'nf_conntrack_locks_all_lock' is indeed held:
- */
- smp_rmb(); /* spin_lock(&nf_conntrack_locks_all_lock) */
- spin_unlock_wait(&nf_conntrack_locks_all_lock);
- spin_lock(lock);
- }
+ /* 2) Order storing the lock and reading nf_conntrack_locks_all */
+ smp_mb();
+
+ /* 3) read nf_conntrack_locks_all, with ACQUIRE semantics */
+ if (likely(smp_load_acquire(&nf_conntrack_locks_all) == false))
+ return;
+
+ /* fast path failed, unlock */
+ spin_unlock(lock);
+
+ /* Slow path 1) get global lock */
+ spin_lock(&nf_conntrack_locks_all_lock);
+
+ /* Slow path 2) get the lock we want */
+ spin_lock(lock);
+
+ /* Slow path 3) release the global lock */
+ spin_unlock(&nf_conntrack_locks_all_lock);
}
+
EXPORT_SYMBOL_GPL(nf_conntrack_lock);
static void nf_conntrack_double_unlock(unsigned int h1, unsigned int h2)
@@ -132,19 +141,23 @@ static void nf_conntrack_all_lock(void)
int i;
spin_lock(&nf_conntrack_locks_all_lock);
- nf_conntrack_locks_all = true;
/*
- * Order the above store of 'nf_conntrack_locks_all' against
+ * Order the store of 'nf_conntrack_locks_all' against
* the spin_unlock_wait() loads below, such that if
* nf_conntrack_lock() observes 'nf_conntrack_locks_all'
* we must observe nf_conntrack_locks[] held:
*/
- smp_mb(); /* spin_lock(&nf_conntrack_locks_all_lock) */
+ smp_store_mb(nf_conntrack_locks_all, true);
for (i = 0; i < CONNTRACK_LOCKS; i++) {
spin_unlock_wait(&nf_conntrack_locks[i]);
}
+ /* spin_unlock_wait() is at least a control barrier.
+ * Add smp_rmb() to upgrade it to an ACQUIRE, it must
+ * pair with the spin_unlock(&nf_conntrack_locks[])
+ */
+ smp_rmb();
}
static void nf_conntrack_all_unlock(void)
--
2.7.4
--
To unsubscribe from this list: send the line "unsubscribe stable" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html