mm: vmpressure: don't count userspace-induced reclaim as memory pressure

Commit e22c6ed90aa9 ("mm: memcontrol: don't count limit-setting reclaim as
memory pressure") made sure that memory reclaim that is induced by
userspace (limit-setting, proactive reclaim, ..) is not counted as memory
pressure for the purposes of psi.

Instead of counting psi inside try_to_free_mem_cgroup_pages(), callers
from try_charge() and reclaim_high() wrap the call to
try_to_free_mem_cgroup_pages() with psi handlers.

However, vmpressure is still counted in these cases where reclaim is
directly induced by userspace.  This patch makes sure vmpressure is not
counted in those operations, in the same way as psi.  Since vmpressure
calls need to happen deeper within the reclaim path, the same approach
could not be followed.  Hence, a new "controlled" flag is added to struct
scan_control to flag a reclaim operation that is controlled by userspace.
This flag is set by limit-setting and proactive reclaim operations, and is
used to count vmpressure correctly.

To prevent future divergence of psi and vmpressure, commit e22c6ed90aa9
("mm: memcontrol: don't count limit-setting reclaim as memory pressure")
is effectively reverted and the same flag is used to control psi as well.

Link: https://lkml.kernel.org/r/20220623004452.1217326-1-yosryahmed@google.com
Signed-off-by: Yosry Ahmed <yosryahmed@google.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: David Hildenbrand <david@redhat.com>
Cc: Miaohe Lin <linmiaohe@huawei.com>
Cc: NeilBrown <neilb@suse.de>
Cc: Alistair Popple <apopple@nvidia.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: Peter Xu <peterx@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

diff --git a/include/linux/swap.h b/include/linux/swap.h
index 0c0fed1b348f20cc04e3c8c3fc8e0672bacca7c7..5a6766e417afe8c254d55e3cc893c7528184420c 100644 (file)
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -411,10 +411,13 @@ extern void lru_cache_add_inactive_or_unevictable(struct page *page,
 extern unsigned long zone_reclaimable_pages(struct zone *zone);
 extern unsigned long try_to_free_pages(struct zonelist *zonelist, int order,
                                        gfp_t gfp_mask, nodemask_t *mask);
+
+#define MEMCG_RECLAIM_MAY_SWAP (1 << 1)
+#define MEMCG_RECLAIM_CONTROLLED (1 << 2)
 extern unsigned long try_to_free_mem_cgroup_pages(struct mem_cgroup *memcg,
                                                  unsigned long nr_pages,
                                                  gfp_t gfp_mask,
-                                                 bool may_swap);
+                                                 unsigned int reclaim_options);
 extern unsigned long mem_cgroup_shrink_node(struct mem_cgroup *mem,
                                                gfp_t gfp_mask, bool noswap,
                                                pg_data_t *pgdat,

diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 803dbdf5f2338b1c4d903701a2de5f8f004d1c54..35f83822ab03aae09cb12fca8fe9a867d668fee8 100644 (file)
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -2453,20 +2453,16 @@ static unsigned long reclaim_high(struct mem_cgroup *memcg,
                                  gfp_t gfp_mask)
 {
        unsigned long nr_reclaimed = 0;
+       unsigned int reclaim_options = MEMCG_RECLAIM_MAY_SWAP;
 
        do {
-               unsigned long pflags;
-
                if (page_counter_read(&memcg->memory) <=
                    READ_ONCE(memcg->memory.high))
                        continue;
-
                memcg_memory_event(memcg, MEMCG_HIGH);
-
-               psi_memstall_enter(&pflags);
                nr_reclaimed += try_to_free_mem_cgroup_pages(memcg, nr_pages,
-                                                            gfp_mask, true);
-               psi_memstall_leave(&pflags);
+                                                            gfp_mask,
+                                                            reclaim_options);
        } while ((memcg = parent_mem_cgroup(memcg)) &&
                 !mem_cgroup_is_root(memcg));
 
@@ -2710,9 +2706,8 @@ static int try_charge_memcg(struct mem_cgroup *memcg, gfp_t gfp_mask,
        struct page_counter *counter;
        unsigned long nr_reclaimed;
        bool passed_oom = false;
-       bool may_swap = true;
+       unsigned int reclaim_options = MEMCG_RECLAIM_MAY_SWAP;
        bool drained = false;
-       unsigned long pflags;
 
 retry:
        if (consume_stock(memcg, nr_pages))
@@ -2727,7 +2722,7 @@ retry:
                mem_over_limit = mem_cgroup_from_counter(counter, memory);
        } else {
                mem_over_limit = mem_cgroup_from_counter(counter, memsw);
-               may_swap = false;
+               reclaim_options &= ~MEMCG_RECLAIM_MAY_SWAP;
        }
 
        if (batch > nr_pages) {
@@ -2752,10 +2747,8 @@ retry:
 
        memcg_memory_event(mem_over_limit, MEMCG_MAX);
 
-       psi_memstall_enter(&pflags);
        nr_reclaimed = try_to_free_mem_cgroup_pages(mem_over_limit, nr_pages,
-                                                   gfp_mask, may_swap);
-       psi_memstall_leave(&pflags);
+                                                   gfp_mask, reclaim_options);
 
        if (mem_cgroup_margin(mem_over_limit) >= nr_pages)
                goto retry;
@@ -3530,7 +3523,9 @@ static int mem_cgroup_resize_max(struct mem_cgroup *memcg,
        int ret;
        bool limits_invariant;
        struct page_counter *counter = memsw ? &memcg->memsw : &memcg->memory;
+       unsigned int reclaim_options = memsw ? 0 : MEMCG_RECLAIM_MAY_SWAP;
 
+       reclaim_options |= MEMCG_RECLAIM_CONTROLLED;
        do {
                if (signal_pending(current)) {
                        ret = -EINTR;
@@ -3564,7 +3559,7 @@ static int mem_cgroup_resize_max(struct mem_cgroup *memcg,
                }
 
                if (!try_to_free_mem_cgroup_pages(memcg, 1,
-                                       GFP_KERNEL, !memsw)) {
+                                       GFP_KERNEL, reclaim_options)) {
                        ret = -EBUSY;
                        break;
                }
@@ -3663,6 +3658,9 @@ unsigned long mem_cgroup_soft_limit_reclaim(pg_data_t *pgdat, int order,
 static int mem_cgroup_force_empty(struct mem_cgroup *memcg)
 {
        int nr_retries = MAX_RECLAIM_RETRIES;
+       unsigned int reclaim_options;
+
+       reclaim_options = MEMCG_RECLAIM_CONTROLLED | MEMCG_RECLAIM_MAY_SWAP;
 
        /* we call try-to-free pages for make this cgroup empty */
        lru_add_drain_all();
@@ -3674,7 +3672,8 @@ static int mem_cgroup_force_empty(struct mem_cgroup *memcg)
                if (signal_pending(current))
                        return -EINTR;
 
-               if (!try_to_free_mem_cgroup_pages(memcg, 1, GFP_KERNEL, true))
+               if (!try_to_free_mem_cgroup_pages(memcg, 1, GFP_KERNEL,
+                                                 reclaim_options))
                        nr_retries--;
        }
 
@@ -6436,6 +6435,7 @@ static ssize_t memory_high_write(struct kernfs_open_file *of,
        unsigned int nr_retries = MAX_RECLAIM_RETRIES;
        bool drained = false;
        unsigned long high;
+       unsigned int reclaim_options;
        int err;
 
        buf = strstrip(buf);
@@ -6444,6 +6444,7 @@ static ssize_t memory_high_write(struct kernfs_open_file *of,
                return err;
 
        page_counter_set_high(&memcg->memory, high);
+       reclaim_options = MEMCG_RECLAIM_CONTROLLED | MEMCG_RECLAIM_MAY_SWAP;
 
        for (;;) {
                unsigned long nr_pages = page_counter_read(&memcg->memory);
@@ -6462,7 +6463,7 @@ static ssize_t memory_high_write(struct kernfs_open_file *of,
                }
 
                reclaimed = try_to_free_mem_cgroup_pages(memcg, nr_pages - high,
-                                                        GFP_KERNEL, true);
+                                               GFP_KERNEL, reclaim_options);
 
                if (!reclaimed && !nr_retries--)
                        break;
@@ -6485,6 +6486,7 @@ static ssize_t memory_max_write(struct kernfs_open_file *of,
        unsigned int nr_reclaims = MAX_RECLAIM_RETRIES;
        bool drained = false;
        unsigned long max;
+       unsigned int reclaim_options;
        int err;
 
        buf = strstrip(buf);
@@ -6493,6 +6495,7 @@ static ssize_t memory_max_write(struct kernfs_open_file *of,
                return err;
 
        xchg(&memcg->memory.max, max);
+       reclaim_options = MEMCG_RECLAIM_CONTROLLED | MEMCG_RECLAIM_MAY_SWAP;
 
        for (;;) {
                unsigned long nr_pages = page_counter_read(&memcg->memory);
@@ -6511,7 +6514,7 @@ static ssize_t memory_max_write(struct kernfs_open_file *of,
 
                if (nr_reclaims) {
                        if (!try_to_free_mem_cgroup_pages(memcg, nr_pages - max,
-                                                         GFP_KERNEL, true))
+                                               GFP_KERNEL, reclaim_options))
                                nr_reclaims--;
                        continue;
                }
@@ -6640,6 +6643,7 @@ static ssize_t memory_reclaim(struct kernfs_open_file *of, char *buf,
        struct mem_cgroup *memcg = mem_cgroup_from_css(of_css(of));
        unsigned int nr_retries = MAX_RECLAIM_RETRIES;
        unsigned long nr_to_reclaim, nr_reclaimed = 0;
+       unsigned int reclaim_options;
        int err;
 
        buf = strstrip(buf);
@@ -6647,6 +6651,7 @@ static ssize_t memory_reclaim(struct kernfs_open_file *of, char *buf,
        if (err)
                return err;
 
+       reclaim_options = MEMCG_RECLAIM_CONTROLLED | MEMCG_RECLAIM_MAY_SWAP;
        while (nr_reclaimed < nr_to_reclaim) {
                unsigned long reclaimed;
 
@@ -6663,7 +6668,7 @@ static ssize_t memory_reclaim(struct kernfs_open_file *of, char *buf,
 
                reclaimed = try_to_free_mem_cgroup_pages(memcg,
                                                nr_to_reclaim - nr_reclaimed,
-                                               GFP_KERNEL, true);
+                                               GFP_KERNEL, reclaim_options);
 
                if (!reclaimed && !nr_retries--)
                        return -EAGAIN;

diff --git a/mm/vmscan.c b/mm/vmscan.c
index 893b881711cbf2c977872c3353afeaf06d33247c..d1e437deb5c4423ad2e627a753e3f6419f723ebf 100644 (file)
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -101,6 +101,9 @@ struct scan_control {
        /* Can pages be swapped as part of reclaim? */
        unsigned int may_swap:1;
 
+       /* Reclaim is controlled by userspace */
+       unsigned int controlled:1;
+
        /*
         * Cgroup memory below memory.low is protected as long as we
         * don't threaten to OOM. If any cgroup is reclaimed at
@@ -3175,9 +3178,10 @@ static void shrink_node_memcgs(pg_data_t *pgdat, struct scan_control *sc)
                            sc->priority);
 
                /* Record the group's reclaim efficiency */
-               vmpressure(sc->gfp_mask, memcg, false,
-                          sc->nr_scanned - scanned,
-                          sc->nr_reclaimed - reclaimed);
+               if (!sc->controlled)
+                       vmpressure(sc->gfp_mask, memcg, false,
+                                  sc->nr_scanned - scanned,
+                                  sc->nr_reclaimed - reclaimed);
 
        } while ((memcg = mem_cgroup_iter(target_memcg, memcg, NULL)));
 }
@@ -3300,9 +3304,10 @@ again:
        }
 
        /* Record the subtree's reclaim efficiency */
-       vmpressure(sc->gfp_mask, sc->target_mem_cgroup, true,
-                  sc->nr_scanned - nr_scanned,
-                  sc->nr_reclaimed - nr_reclaimed);
+       if (!sc->controlled)
+               vmpressure(sc->gfp_mask, sc->target_mem_cgroup, true,
+                          sc->nr_scanned - nr_scanned,
+                          sc->nr_reclaimed - nr_reclaimed);
 
        if (sc->nr_reclaimed - nr_reclaimed)
                reclaimable = true;
@@ -3584,8 +3589,9 @@ retry:
                __count_zid_vm_events(ALLOCSTALL, sc->reclaim_idx, 1);
 
        do {
-               vmpressure_prio(sc->gfp_mask, sc->target_mem_cgroup,
-                               sc->priority);
+               if (!sc->controlled)
+                       vmpressure_prio(sc->gfp_mask, sc->target_mem_cgroup,
+                                       sc->priority);
                sc->nr_scanned = 0;
                shrink_zones(zonelist, sc);
 
@@ -3875,10 +3881,12 @@ unsigned long mem_cgroup_shrink_node(struct mem_cgroup *memcg,
 unsigned long try_to_free_mem_cgroup_pages(struct mem_cgroup *memcg,
                                           unsigned long nr_pages,
                                           gfp_t gfp_mask,
-                                          bool may_swap)
+                                          unsigned int reclaim_options)
 {
        unsigned long nr_reclaimed;
+       unsigned long pflags;
        unsigned int noreclaim_flag;
+       bool controlled_reclaim = reclaim_options & MEMCG_RECLAIM_CONTROLLED;
        struct scan_control sc = {
                .nr_to_reclaim = max(nr_pages, SWAP_CLUSTER_MAX),
                .gfp_mask = (current_gfp_context(gfp_mask) & GFP_RECLAIM_MASK) |
@@ -3888,7 +3896,8 @@ unsigned long try_to_free_mem_cgroup_pages(struct mem_cgroup *memcg,
                .priority = DEF_PRIORITY,
                .may_writepage = !laptop_mode,
                .may_unmap = 1,
-               .may_swap = may_swap,
+               .may_swap = !!(reclaim_options & MEMCG_RECLAIM_MAY_SWAP),
+               .controlled = controlled_reclaim,
        };
        /*
         * Traverse the ZONELIST_FALLBACK zonelist of the current node to put
@@ -3898,12 +3907,19 @@ unsigned long try_to_free_mem_cgroup_pages(struct mem_cgroup *memcg,
        struct zonelist *zonelist = node_zonelist(numa_node_id(), sc.gfp_mask);
 
        set_task_reclaim_state(current, &sc.reclaim_state);
+
        trace_mm_vmscan_memcg_reclaim_begin(0, sc.gfp_mask);
+
+       if (!controlled_reclaim)
+               psi_memstall_enter(&pflags);
        noreclaim_flag = memalloc_noreclaim_save();
 
        nr_reclaimed = do_try_to_free_pages(zonelist, &sc);
 
        memalloc_noreclaim_restore(noreclaim_flag);
+       if (!controlled_reclaim)
+               psi_memstall_leave(&pflags);
+
        trace_mm_vmscan_memcg_reclaim_end(nr_reclaimed);
        set_task_reclaim_state(current, NULL);