[PATCH percpu/for-3.19 1/2] percpu_ref: remove unnecessary ACCESS_ONCE() in percpu_ref_tryget_live()

[Tejun Heo]

__ref_is_percpu() needs ACCESS_ONCE() on @ref->percpu_count_ptr
because the value is tested for !__PERCPU_REF_ATOMIC, which may be set
asynchronously, and then used as a pointer.  If the compiler generates
a separate fetch when using it as a pointer, __PERCPU_REF_ATOMIC may
be set in between contaminating the pointer value.

percpu_ref_tryget_live() also uses ACCESS_ONCE() to test
__PERCPU_REF_DEAD; however, there's no reason for this.  I just copied
ACCESS_ONCE() usage blindly from __ref_is_percpu().  All it does is
confusing people trying to understand what's going on.

This patch removes the unnecessary ACCESS_ONCE() usage from
percpu_ref_tryget_live() and adds a comment explaining why
__ref_is_percpu() needs it.

Signed-off-by: Tejun Heo <tj@xxxxxxxxxx>
Cc: Kent Overstreet <kmo@xxxxxxxxxxxxx>
---
 include/linux/percpu-refcount.h |   14 ++++++++++++--
 1 file changed, 12 insertions(+), 2 deletions(-)

--- a/include/linux/percpu-refcount.h
+++ b/include/linux/percpu-refcount.h
@@ -128,7 +128,17 @@ static inline void percpu_ref_kill(struc
 static inline bool __ref_is_percpu(struct percpu_ref *ref,
 					  unsigned long __percpu **percpu_countp)
 {
-	unsigned long percpu_ptr = ACCESS_ONCE(ref->percpu_count_ptr);
+	unsigned long percpu_ptr;
+
+	/*
+	 * The value of @ref->percpu_count_ptr is tested for
+	 * !__PERCPU_REF_ATOMIC, which may be set asynchronously, and then
+	 * used as a pointer.  If the compiler generates a separate fetch
+	 * when using it as a pointer, __PERCPU_REF_ATOMIC may be set in
+	 * between contaminating the pointer value.  Make sure that the
+	 * value is fetched once.
+	 */
+	percpu_ptr = ACCESS_ONCE(ref->percpu_count_ptr);
 
 	/* paired with smp_store_release() in percpu_ref_reinit() */
 	smp_read_barrier_depends();
@@ -215,7 +225,7 @@ static inline bool percpu_ref_tryget_liv
 	if (__ref_is_percpu(ref, &percpu_count)) {
 		this_cpu_inc(*percpu_count);
 		ret = true;
-	} else if (!(ACCESS_ONCE(ref->percpu_count_ptr) & __PERCPU_REF_DEAD)) {
+	} else if (!(ref->percpu_count_ptr & __PERCPU_REF_DEAD)) {
 		ret = atomic_long_inc_not_zero(&ref->count);
 	}
 
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/