* is safe to take the lu_sites_guard lock.
*
* Ideally we should accurately return the remaining number of cached
- * objects without taking the lu_sites_guard lock, but this is not
+ * objects without taking the lu_sites_guard lock, but this is not
* possible in the current implementation.
*/
static int lu_cache_shrink(SHRINKER_ARGS(sc, nr_to_scan, gfp_mask))