Re: [v4 PATCH 08/11] mm: vmscan: use per memcg nr_deferred of shrinker

On Mon, Jan 25, 2021 at 1:35 AM Kirill Tkhai [off-list ref] wrote:

On 22.01.2021 02:06, Yang Shi wrote:

quoted

Use per memcg's nr_deferred for memcg aware shrinkers.  The shrinker's nr_deferred
will be used in the following cases:
    1. Non memcg aware shrinkers
    2. !CONFIG_MEMCG
    3. memcg is disabled by boot parameter

Signed-off-by: Yang Shi <redacted>
---
 mm/vmscan.c | 81 +++++++++++++++++++++++++++++++++++++++++++++--------
 1 file changed, 69 insertions(+), 12 deletions(-)

diff --git a/mm/vmscan.c b/mm/vmscan.c
index 722aa71b13b2..d8e77ea13815 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c

@@ -359,6 +359,27 @@ static void unregister_memcg_shrinker(struct shrinker *shrinker)
      up_write(&shrinker_rwsem);
 }

+static long count_nr_deferred_memcg(int nid, struct shrinker *shrinker,
+                                 struct mem_cgroup *memcg)
+{
+     struct shrinker_info *info;
+
+     info = rcu_dereference_protected(memcg->nodeinfo[nid]->shrinker_info,
+                                      true);

Since now these rcu_dereference_protected() are in separate functions and there is
no taking a lock near them, it seems it would be better to underling the desired
lock with rcu_dereference_protected(, lockdep_assert_held(lock_you_need_here_locked));

Sure. Will incorporate in v5. BTW I noticed using
lockdep_assert_held() in the parameter of the functions will result in
compilation failure with gcc 10.0.1 (shipped with Fedora 32), but fine
with gcc 8.3.1.

In file included from ./include/linux/rbtree.h:22,
                 from ./include/linux/mm_types.h:10,
                 from ./include/linux/mmzone.h:21,
                 from ./include/linux/gfp.h:6,
                 from ./include/linux/mm.h:10,
                 from mm/vmscan.c:15:
mm/vmscan.c: In function ‘shrinker_info_protected’:
./include/linux/lockdep.h:386:34: error: expected expression before ‘do’
  386 | #define lockdep_assert_held(l)   do { (void)(l); } while (0)
      |                                  ^~
./include/linux/rcupdate.h:337:52: note: in definition of macro
‘RCU_LOCKDEP_WARN’
  337 | #define RCU_LOCKDEP_WARN(c, s) do { } while (0 && (c))
      |                                                    ^
./include/linux/rcupdate.h:554:2: note: in expansion of macro
‘__rcu_dereference_protected’
  554 |  __rcu_dereference_protected((p), (c), __rcu)
      |  ^~~~~~~~~~~~~~~~~~~~~~~~~~~
mm/vmscan.c:389:9: note: in expansion of macro ‘rcu_dereference_protected’
  389 |  return rcu_dereference_protected(memcg->nodeinfo[nid]->shrinker_info,
      |         ^~~~~~~~~~~~~~~~~~~~~~~~~
mm/vmscan.c:390:7: note: in expansion of macro ‘lockdep_assert_held’
  390 |       lockdep_assert_held(&shrinker_rwsem));
      |       ^~~~~~~~~~~~~~~~~~~


I didn't dig into the root cause. Just use lockdep_is_held() instead
of lockdep_assert_held().

quoted

+     return atomic_long_xchg(&info->nr_deferred[shrinker->id], 0);
+}
+
+static long set_nr_deferred_memcg(long nr, int nid, struct shrinker *shrinker,
+                               struct mem_cgroup *memcg)
+{
+     struct shrinker_info *info;
+
+     info = rcu_dereference_protected(memcg->nodeinfo[nid]->shrinker_info,
+                                      true);
+
+     return atomic_long_add_return(nr, &info->nr_deferred[shrinker->id]);
+}
+
 static bool cgroup_reclaim(struct scan_control *sc)
 {
      return sc->target_mem_cgroup;

@@ -397,6 +418,18 @@ static void unregister_memcg_shrinker(struct shrinker *shrinker)
 {
 }

+static long count_nr_deferred_memcg(int nid, struct shrinker *shrinker,
+                                 struct mem_cgroup *memcg)
+{
+     return 0;
+}
+
+static long set_nr_deferred_memcg(long nr, int nid, struct shrinker *shrinker,
+                               struct mem_cgroup *memcg)
+{
+     return 0;
+}
+
 static bool cgroup_reclaim(struct scan_control *sc)
 {
      return false;

@@ -408,6 +441,39 @@ static bool writeback_throttling_sane(struct scan_control *sc)
 }
 #endif

+static long count_nr_deferred(struct shrinker *shrinker,
+                           struct shrink_control *sc)
+{
+     int nid = sc->nid;
+
+     if (!(shrinker->flags & SHRINKER_NUMA_AWARE))
+             nid = 0;
+
+     if (sc->memcg &&
+         (shrinker->flags & SHRINKER_MEMCG_AWARE))
+             return count_nr_deferred_memcg(nid, shrinker,
+                                            sc->memcg);
+
+     return atomic_long_xchg(&shrinker->nr_deferred[nid], 0);
+}
+
+
+static long set_nr_deferred(long nr, struct shrinker *shrinker,
+                         struct shrink_control *sc)
+{
+     int nid = sc->nid;
+
+     if (!(shrinker->flags & SHRINKER_NUMA_AWARE))
+             nid = 0;
+
+     if (sc->memcg &&
+         (shrinker->flags & SHRINKER_MEMCG_AWARE))
+             return set_nr_deferred_memcg(nr, nid, shrinker,
+                                          sc->memcg);
+
+     return atomic_long_add_return(nr, &shrinker->nr_deferred[nid]);
+}
+
 /*
  * This misses isolated pages which are not accounted for to save counters.
  * As the data only determines if reclaim or compaction continues, it is

@@ -544,14 +610,10 @@ static unsigned long do_shrink_slab(struct shrink_control *shrinkctl,
      long freeable;
      long nr;
      long new_nr;
-     int nid = shrinkctl->nid;
      long batch_size = shrinker->batch ? shrinker->batch
                                        : SHRINK_BATCH;
      long scanned = 0, next_deferred;

-     if (!(shrinker->flags & SHRINKER_NUMA_AWARE))
-             nid = 0;
-
      freeable = shrinker->count_objects(shrinker, shrinkctl);
      if (freeable == 0 || freeable == SHRINK_EMPTY)
              return freeable;

@@ -561,7 +623,7 @@ static unsigned long do_shrink_slab(struct shrink_control *shrinkctl,
       * and zero it so that other concurrent shrinker invocations
       * don't also do this scanning work.
       */
-     nr = atomic_long_xchg(&shrinker->nr_deferred[nid], 0);
+     nr = count_nr_deferred(shrinker, shrinkctl);

      total_scan = nr;
      if (shrinker->seeks) {

@@ -652,14 +714,9 @@ static unsigned long do_shrink_slab(struct shrink_control *shrinkctl,
              next_deferred = 0;
      /*
       * move the unused scan count back into the shrinker in a
-      * manner that handles concurrent updates. If we exhausted the
-      * scan, there is no need to do an update.
+      * manner that handles concurrent updates.
       */
-     if (next_deferred > 0)
-             new_nr = atomic_long_add_return(next_deferred,
-                                             &shrinker->nr_deferred[nid]);
-     else
-             new_nr = atomic_long_read(&shrinker->nr_deferred[nid]);
+     new_nr = set_nr_deferred(next_deferred, shrinker, shrinkctl);

      trace_mm_shrink_slab_end(shrinker, shrinkctl->nid, freed, nr, new_nr, total_scan);
      return freed;