mm: skip folio reclaim in legacy memcg contexts for deadlockable mappings

Currently in shrink_folio_list(), reclaim for folios under writeback
falls into 3 different cases:
1) Reclaim is encountering an excessive number of folios under
   writeback and this folio has both the writeback and reclaim flags
   set
2) Dirty throttling is enabled (this happens if reclaim through cgroup
   is not enabled, if reclaim through cgroupv2 memcg is enabled, or
   if reclaim is on the root cgroup), or if the folio is not marked for
   immediate reclaim, or if the caller does not have __GFP_FS (or
   __GFP_IO if it's going to swap) set
3) Legacy cgroupv1 encounters a folio that already has the reclaim flag
   set and the caller did not have __GFP_FS (or __GFP_IO if swap) set

In cases 1) and 2), we activate the folio and skip reclaiming it while
in case 3), we wait for writeback to finish on the folio and then try
to reclaim the folio again. In case 3, we wait on writeback because
cgroupv1 does not have dirty folio throttling, as such this is a
mitigation against the case where there are too many folios in writeback
with nothing else to reclaim.

If a filesystem (eg fuse) may deadlock due to reclaim waiting on
writeback, then the filesystem needs to add inefficient messy workarounds
to prevent this. To improve the performance of these filesystems, this
commit adds two things:
a) a AS_WRITEBACK_MAY_DEADLOCK_ON_RECLAIM mapping flag that filesystems
   may set to indicate that reclaim should not wait on writeback
b) if legacy memcg encounters a folio with this
   AS_WRITEBACK_MAY_DEADLOCK_ON_RECLAIM flag set (eg case 3), the folio
   will be activated and skip reclaim (eg default to behavior in case 2)
   instead.

Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Reviewed-by: Shakeel Butt <shakeel.butt@linux.dev>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Acked-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Jingbo Xu <jefflexu@linux.alibaba.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>

diff --git a/include/linux/pagemap.h b/include/linux/pagemap.h
index 26baa78f1ca7d8d04e6f071ef4e0bb150a508a04..e19ff3183bbf020a78f2a4602f6841caf6590003 100644 (file)
--- a/include/linux/pagemap.h
+++ b/include/linux/pagemap.h
@@ -210,6 +210,7 @@ enum mapping_flags {
        AS_STABLE_WRITES = 7,   /* must wait for writeback before modifying
                                   folio contents */
        AS_INACCESSIBLE = 8,    /* Do not attempt direct R/W access to the mapping */
+       AS_WRITEBACK_MAY_DEADLOCK_ON_RECLAIM = 9,
        /* Bits 16-25 are used for FOLIO_ORDER */
        AS_FOLIO_ORDER_BITS = 5,
        AS_FOLIO_ORDER_MIN = 16,
@@ -335,6 +336,16 @@ static inline bool mapping_inaccessible(struct address_space *mapping)
        return test_bit(AS_INACCESSIBLE, &mapping->flags);
 }
 
+static inline void mapping_set_writeback_may_deadlock_on_reclaim(struct address_space *mapping)
+{
+       set_bit(AS_WRITEBACK_MAY_DEADLOCK_ON_RECLAIM, &mapping->flags);
+}
+
+static inline bool mapping_writeback_may_deadlock_on_reclaim(struct address_space *mapping)
+{
+       return test_bit(AS_WRITEBACK_MAY_DEADLOCK_ON_RECLAIM, &mapping->flags);
+}
+
 static inline gfp_t mapping_gfp_mask(struct address_space * mapping)
 {
        return mapping->gfp_mask;

diff --git a/mm/vmscan.c b/mm/vmscan.c
index b620d74b0f66e33ce821260366ec1afde2abcaac..8fc180fbd9443ad2d858fe58781a29a3fb857a3f 100644 (file)
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1187,8 +1187,10 @@ retry:
                 * 2) Global or new memcg reclaim encounters a folio that is
                 *    not marked for immediate reclaim, or the caller does not
                 *    have __GFP_FS (or __GFP_IO if it's simply going to swap,
-                *    not to fs). In this case mark the folio for immediate
-                *    reclaim and continue scanning.
+                *    not to fs), or the folio belongs to a mapping where
+                *    waiting on writeback during reclaim may lead to a deadlock.
+                *    In this case mark the folio for immediate reclaim and
+                *    continue scanning.
                 *
                 *    Require may_enter_fs() because we would wait on fs, which
                 *    may not have submitted I/O yet. And the loop driver might
@@ -1213,6 +1215,8 @@ retry:
                 * takes to write them to disk.
                 */
                if (folio_test_writeback(folio)) {
+                       mapping = folio_mapping(folio);
+
                        /* Case 1 above */
                        if (current_is_kswapd() &&
                            folio_test_reclaim(folio) &&
@@ -1223,7 +1227,9 @@ retry:
                        /* Case 2 above */
                        } else if (writeback_throttling_sane(sc) ||
                            !folio_test_reclaim(folio) ||
-                           !may_enter_fs(folio, sc->gfp_mask)) {
+                           !may_enter_fs(folio, sc->gfp_mask) ||
+                           (mapping &&
+                            mapping_writeback_may_deadlock_on_reclaim(mapping))) {
                                /*
                                 * This is slightly racy -
                                 * folio_end_writeback() might have