diff options
author | Frederic Weisbecker <frederic@kernel.org> | 2023-11-15 22:11:28 +0300 |
---|---|---|
committer | Boqun Feng <boqun.feng@gmail.com> | 2024-02-14 18:50:45 +0300 |
commit | 1e8e6951a5774c8dd9d1f14af9c5b7d66130d96f (patch) | |
tree | 85e20d29695f6129b05af15c6bf73ef4d98d803a /kernel/rcu/tree_nocb.h | |
parent | ca16265aaf9d357035000833636dcddbfafacac3 (diff) | |
download | linux-1e8e6951a5774c8dd9d1f14af9c5b7d66130d96f.tar.xz |
rcu/nocb: Remove needless full barrier after callback advancing
A full barrier is issued from nocb_gp_wait() upon callbacks advancing
to order grace period completion with callbacks execution.
However these two events are already ordered by the
smp_mb__after_unlock_lock() barrier within the call to
raw_spin_lock_rcu_node() that is necessary for callbacks advancing to
happen.
The following litmus test shows the kind of guarantee that this barrier
provides:
C smp_mb__after_unlock_lock
{}
// rcu_gp_cleanup()
P0(spinlock_t *rnp_lock, int *gpnum)
{
// Grace period cleanup increase gp sequence number
spin_lock(rnp_lock);
WRITE_ONCE(*gpnum, 1);
spin_unlock(rnp_lock);
}
// nocb_gp_wait()
P1(spinlock_t *rnp_lock, spinlock_t *nocb_lock, int *gpnum, int *cb_ready)
{
int r1;
// Call rcu_advance_cbs() from nocb_gp_wait()
spin_lock(nocb_lock);
spin_lock(rnp_lock);
smp_mb__after_unlock_lock();
r1 = READ_ONCE(*gpnum);
WRITE_ONCE(*cb_ready, 1);
spin_unlock(rnp_lock);
spin_unlock(nocb_lock);
}
// nocb_cb_wait()
P2(spinlock_t *nocb_lock, int *cb_ready, int *cb_executed)
{
int r2;
// rcu_do_batch() -> rcu_segcblist_extract_done_cbs()
spin_lock(nocb_lock);
r2 = READ_ONCE(*cb_ready);
spin_unlock(nocb_lock);
// Actual callback execution
WRITE_ONCE(*cb_executed, 1);
}
P3(int *cb_executed, int *gpnum)
{
int r3;
WRITE_ONCE(*cb_executed, 2);
smp_mb();
r3 = READ_ONCE(*gpnum);
}
exists (1:r1=1 /\ 2:r2=1 /\ cb_executed=2 /\ 3:r3=0) (* Bad outcome. *)
Here the bad outcome only occurs if the smp_mb__after_unlock_lock() is
removed. This barrier orders the grace period completion against
callbacks advancing and even later callbacks invocation, thanks to the
opportunistic propagation via the ->nocb_lock to nocb_cb_wait().
Therefore the smp_mb() placed after callbacks advancing can be safely
removed.
Signed-off-by: Frederic Weisbecker <frederic@kernel.org>
Reviewed-by: Paul E. McKenney <paulmck@kernel.org>
Signed-off-by: Boqun Feng <boqun.feng@gmail.com>
Diffstat (limited to 'kernel/rcu/tree_nocb.h')
-rw-r--r-- | kernel/rcu/tree_nocb.h | 1 |
1 files changed, 0 insertions, 1 deletions
diff --git a/kernel/rcu/tree_nocb.h b/kernel/rcu/tree_nocb.h index 785946834c6b..b2c3145c4c13 100644 --- a/kernel/rcu/tree_nocb.h +++ b/kernel/rcu/tree_nocb.h @@ -779,7 +779,6 @@ static void nocb_gp_wait(struct rcu_data *my_rdp) if (rcu_segcblist_ready_cbs(&rdp->cblist)) { needwake = rdp->nocb_cb_sleep; WRITE_ONCE(rdp->nocb_cb_sleep, false); - smp_mb(); /* CB invocation -after- GP end. */ } else { needwake = false; } |