[PATCH 1/3] mm-unstable: Multi-gen LRU: Fix per-zone reclaim

[PATCH 1/3] mm-unstable: Multi-gen LRU: Fix per-zone reclaim

[Kalesh Singh]

MGLRU has a LRU list for each zone for each type (anon/file) in each
generation:

	long nr_pages[MAX_NR_GENS][ANON_AND_FILE][MAX_NR_ZONES];

The min_seq (oldest generation) can progress independently for each
type but the max_seq (youngest generation) is shared for both anon and
file. This is to maintain a common frame of reference.

In order for eviction to advance the min_seq of a type, all the per-zone
lists in the oldest generation of that type must be empty.

The eviction logic only considers pages from eligible zones for
eviction or promotion.

    scan_folios() {
	...
	for (zone = sc->reclaim_idx; zone >= 0; zone--)  {
	    ...
	    sort_folio(); 	// Promote
	    ...
	    isolate_folio(); 	// Evict
	}
	...
    }

Consider the system has the movable zone configured and default 4
generations. The current state of the system is as shown below
(only illustrating one type for simplicity):

Type: ANON

	Zone    DMA32     Normal    Movable    Device

	Gen 0       0          0        4GB         0

	Gen 1       0        1GB        1MB         0

	Gen 2     1MB        4GB        1MB         0

	Gen 3     1MB        1MB        1MB         0

Now consider there is a GFP_KERNEL allocation request (eligible zone
index <= Normal), evict_folios() will return without doing any work
since there are no pages to scan in the eligible zones of the oldest
generation. Reclaim won't make progress until triggered from a ZONE_MOVABLE
allocation request; which may not happen soon if there is a lot of free
memory in the movable zone. This can lead to OOM kills, although there
is 1GB pages in the Normal zone of Gen 1 that we have not yet tried to
reclaim.

This issue is not seen in the conventional active/inactive LRU since
there are no per-zone lists.

If there are no (not enough) folios to scan in the eligible zones, move
folios from ineligible zone (zone_index > reclaim_index) to the next
generation. This allows for the progression of min_seq and reclaiming
from the next generation (Gen 1).

Qualcomm, Mediatek and raspberrypi [1] discovered this issue independently.

[1] https://github.com/raspberrypi/linux/issues/5395

Cc: Yu Zhao <yuzhao@google.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Reported-by: Charan Teja Kalla <quic_charante@quicinc.com>
Reported-by: Lecopzer Chen <lecopzer.chen@mediatek.com>
Signed-off-by: Kalesh Singh <kaleshsingh@google.com>
---
 mm/vmscan.c | 18 ++++++++++++++----
 1 file changed, 14 insertions(+), 4 deletions(-)

diff --git a/mm/vmscan.c b/mm/vmscan.c
index 4039620d30fe..489a4fc7d9b1 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -4889,7 +4889,8 @@ static int lru_gen_memcg_seg(struct lruvec *lruvec)
  *                          the eviction
  ******************************************************************************/
 
-static bool sort_folio(struct lruvec *lruvec, struct folio *folio, int tier_idx)
+static bool sort_folio(struct lruvec *lruvec, struct folio *folio, struct scan_control *sc,
+		       int tier_idx)
 {
 	bool success;
 	int gen = folio_lru_gen(folio);
@@ -4939,6 +4940,13 @@ static bool sort_folio(struct lruvec *lruvec, struct folio *folio, int tier_idx)
 		return true;
 	}
 
+	/* ineligible */
+	if (zone > sc->reclaim_idx) {
+		gen = folio_inc_gen(lruvec, folio, false);
+		list_move_tail(&folio->lru, &lrugen->folios[gen][type][zone]);
+		return true;
+	}
+
 	/* waiting for writeback */
 	if (folio_test_locked(folio) || folio_test_writeback(folio) ||
 	    (type == LRU_GEN_FILE && folio_test_dirty(folio))) {
@@ -4987,7 +4995,8 @@ static bool isolate_folio(struct lruvec *lruvec, struct folio *folio, struct sca
 static int scan_folios(struct lruvec *lruvec, struct scan_control *sc,
 		       int type, int tier, struct list_head *list)
 {
-	int gen, zone;
+	int i;
+	int gen;
 	enum vm_event_item item;
 	int sorted = 0;
 	int scanned = 0;
@@ -5003,9 +5012,10 @@ static int scan_folios(struct lruvec *lruvec, struct scan_control *sc,
 
 	gen = lru_gen_from_seq(lrugen->min_seq[type]);
 
-	for (zone = sc->reclaim_idx; zone >= 0; zone--) {
+	for (i = MAX_NR_ZONES; i > 0; i--) {
 		LIST_HEAD(moved);
 		int skipped = 0;
+		int zone = (sc->reclaim_idx + i) % MAX_NR_ZONES;
 		struct list_head *head = &lrugen->folios[gen][type][zone];
 
 		while (!list_empty(head)) {
@@ -5019,7 +5029,7 @@ static int scan_folios(struct lruvec *lruvec, struct scan_control *sc,
 
 			scanned += delta;
 
-			if (sort_folio(lruvec, folio, tier))
+			if (sort_folio(lruvec, folio, sc, tier))
 				sorted += delta;
 			else if (isolate_folio(lruvec, folio, sc)) {
 				list_add(&folio->lru, list);
-- 
2.41.0.255.g8b1d071c50-goog

[PATCH 2/3] mm-unstable: Multi-gen LRU: Avoid race in inc_min_seq()

[Kalesh Singh]

inc_max_seq() will try to inc_min_seq() if nr_gens == MAX_NR_GENS. This
is because the generations are reused (the last oldest now empty
generation will become the next youngest generation).

inc_min_seq() is retried until successful, dropping the lru_lock
and yielding the CPU on each failure, and retaking the lock before
trying again:

        while (!inc_min_seq(lruvec, type, can_swap)) {
                spin_unlock_irq(&lruvec->lru_lock);
                cond_resched();
                spin_lock_irq(&lruvec->lru_lock);
        }

However, the initial condition that required incrementing the min_seq
(nr_gens == MAX_NR_GENS) is not retested. This can change by another
call to inc_max_seq() from run_aging() with force_scan=true from the
debugfs interface.

Since the eviction stalls when the nr_gens == MIN_NR_GENS, avoid
unnecessarily incrementing the min_seq by rechecking the number of
generations before each attempt.

This issue was uncovered in previous discussion on the list by Yu Zhao
and Aneesh Kumar [1].

[1] https://lore.kernel.org/linux-mm/CAOUHufbO7CaVm=xjEb1avDhHVvnC8pJmGyKcFf2iY_dpf+zR3w@mail.gmail.com/

Cc: Yu Zhao <yuzhao@google.com>
Cc: Aneesh Kumar K V <aneesh.kumar@linux.ibm.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Kalesh Singh <kaleshsingh@google.com>
---
 mm/vmscan.c | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

diff --git a/mm/vmscan.c b/mm/vmscan.c
index 489a4fc7d9b1..6eecd291756c 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -4439,7 +4439,7 @@ static void inc_max_seq(struct lruvec *lruvec, bool can_swap, bool force_scan)
 	int prev, next;
 	int type, zone;
 	struct lru_gen_folio *lrugen = &lruvec->lrugen;
-
+restart:
 	spin_lock_irq(&lruvec->lru_lock);
 
 	VM_WARN_ON_ONCE(!seq_is_valid(lruvec));
@@ -4450,11 +4450,12 @@ static void inc_max_seq(struct lruvec *lruvec, bool can_swap, bool force_scan)
 
 		VM_WARN_ON_ONCE(!force_scan && (type == LRU_GEN_FILE || can_swap));
 
-		while (!inc_min_seq(lruvec, type, can_swap)) {
-			spin_unlock_irq(&lruvec->lru_lock);
-			cond_resched();
-			spin_lock_irq(&lruvec->lru_lock);
-		}
+		if (inc_min_seq(lruvec, type, can_swap))
+			continue;
+
+		spin_unlock_irq(&lruvec->lru_lock);
+		cond_resched();
+		goto restart;
 	}
 
 	/*
-- 
2.41.0.255.g8b1d071c50-goog

[PATCH 3/3] mm-unstable: Multi-gen LRU: Fix can_swap in lru_gen_look_around()

[Kalesh Singh]

walk->can_swap might be invalid since it's not guaranteed to be
initialized for the particular lruvec. Instead deduce it from the folio
type (anon/file).

Cc: Yu Zhao <yuzhao@google.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Kalesh Singh <kaleshsingh@google.com>
---
 mm/vmscan.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/mm/vmscan.c b/mm/vmscan.c
index 6eecd291756c..b4329f93a682 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -4656,6 +4656,7 @@ void lru_gen_look_around(struct page_vma_mapped_walk *pvmw)
 	pte_t *pte = pvmw->pte;
 	unsigned long addr = pvmw->address;
 	struct folio *folio = pfn_folio(pvmw->pfn);
+	bool can_swap = !folio_is_file_lru(folio);
 	struct mem_cgroup *memcg = folio_memcg(folio);
 	struct pglist_data *pgdat = folio_pgdat(folio);
 	struct lruvec *lruvec = mem_cgroup_lruvec(memcg, pgdat);
@@ -4704,7 +4705,7 @@ void lru_gen_look_around(struct page_vma_mapped_walk *pvmw)
 		if (!pte_young(ptent))
 			continue;
 
-		folio = get_pfn_folio(pfn, memcg, pgdat, !walk || walk->can_swap);
+		folio = get_pfn_folio(pfn, memcg, pgdat, can_swap);
 		if (!folio)
 			continue;
 
-- 
2.41.0.255.g8b1d071c50-goog

Re: [PATCH 1/3] mm-unstable: Multi-gen LRU: Fix per-zone reclaim

[Yu Zhao]

On Tue, Aug 1, 2023 at 6:19 PM Kalesh Singh <kaleshsingh@google.com> wrote:
>
> MGLRU has a LRU list for each zone for each type (anon/file) in each
> generation:
>
>         long nr_pages[MAX_NR_GENS][ANON_AND_FILE][MAX_NR_ZONES];
>
> The min_seq (oldest generation) can progress independently for each
> type but the max_seq (youngest generation) is shared for both anon and
> file. This is to maintain a common frame of reference.
>
> In order for eviction to advance the min_seq of a type, all the per-zone
> lists in the oldest generation of that type must be empty.
>
> The eviction logic only considers pages from eligible zones for
> eviction or promotion.
>
>     scan_folios() {
>         ...
>         for (zone = sc->reclaim_idx; zone >= 0; zone--)  {
>             ...
>             sort_folio();       // Promote
>             ...
>             isolate_folio();    // Evict
>         }
>         ...
>     }
>
> Consider the system has the movable zone configured and default 4
> generations. The current state of the system is as shown below
> (only illustrating one type for simplicity):
>
> Type: ANON
>
>         Zone    DMA32     Normal    Movable    Device
>
>         Gen 0       0          0        4GB         0
>
>         Gen 1       0        1GB        1MB         0
>
>         Gen 2     1MB        4GB        1MB         0
>
>         Gen 3     1MB        1MB        1MB         0
>
> Now consider there is a GFP_KERNEL allocation request (eligible zone
> index <= Normal), evict_folios() will return without doing any work
> since there are no pages to scan in the eligible zones of the oldest
> generation. Reclaim won't make progress until triggered from a ZONE_MOVABLE
> allocation request; which may not happen soon if there is a lot of free
> memory in the movable zone. This can lead to OOM kills, although there
> is 1GB pages in the Normal zone of Gen 1 that we have not yet tried to
> reclaim.
>
> This issue is not seen in the conventional active/inactive LRU since
> there are no per-zone lists.
>
> If there are no (not enough) folios to scan in the eligible zones, move
> folios from ineligible zone (zone_index > reclaim_index) to the next
> generation. This allows for the progression of min_seq and reclaiming
> from the next generation (Gen 1).
>
> Qualcomm, Mediatek and raspberrypi [1] discovered this issue independently.
>
> [1] https://github.com/raspberrypi/linux/issues/5395
>
> Cc: Yu Zhao <yuzhao@google.com>
> Cc: Andrew Morton <akpm@linux-foundation.org>
> Reported-by: Charan Teja Kalla <quic_charante@quicinc.com>
> Reported-by: Lecopzer Chen <lecopzer.chen@mediatek.com>
> Signed-off-by: Kalesh Singh <kaleshsingh@google.com>

LGTM. But I think we need the Fixes tag and Cc stable.

Re: [PATCH 1/3] mm-unstable: Multi-gen LRU: Fix per-zone reclaim

[Kalesh Singh]

On Tue, Aug 1, 2023 at 7:08 PM Yu Zhao <yuzhao@google.com> wrote:
>
> On Tue, Aug 1, 2023 at 6:19 PM Kalesh Singh <kaleshsingh@google.com> wrote:
> >
> > MGLRU has a LRU list for each zone for each type (anon/file) in each
> > generation:
> >
> >         long nr_pages[MAX_NR_GENS][ANON_AND_FILE][MAX_NR_ZONES];
> >
> > The min_seq (oldest generation) can progress independently for each
> > type but the max_seq (youngest generation) is shared for both anon and
> > file. This is to maintain a common frame of reference.
> >
> > In order for eviction to advance the min_seq of a type, all the per-zone
> > lists in the oldest generation of that type must be empty.
> >
> > The eviction logic only considers pages from eligible zones for
> > eviction or promotion.
> >
> >     scan_folios() {
> >         ...
> >         for (zone = sc->reclaim_idx; zone >= 0; zone--)  {
> >             ...
> >             sort_folio();       // Promote
> >             ...
> >             isolate_folio();    // Evict
> >         }
> >         ...
> >     }
> >
> > Consider the system has the movable zone configured and default 4
> > generations. The current state of the system is as shown below
> > (only illustrating one type for simplicity):
> >
> > Type: ANON
> >
> >         Zone    DMA32     Normal    Movable    Device
> >
> >         Gen 0       0          0        4GB         0
> >
> >         Gen 1       0        1GB        1MB         0
> >
> >         Gen 2     1MB        4GB        1MB         0
> >
> >         Gen 3     1MB        1MB        1MB         0
> >
> > Now consider there is a GFP_KERNEL allocation request (eligible zone
> > index <= Normal), evict_folios() will return without doing any work
> > since there are no pages to scan in the eligible zones of the oldest
> > generation. Reclaim won't make progress until triggered from a ZONE_MOVABLE
> > allocation request; which may not happen soon if there is a lot of free
> > memory in the movable zone. This can lead to OOM kills, although there
> > is 1GB pages in the Normal zone of Gen 1 that we have not yet tried to
> > reclaim.
> >
> > This issue is not seen in the conventional active/inactive LRU since
> > there are no per-zone lists.
> >
> > If there are no (not enough) folios to scan in the eligible zones, move
> > folios from ineligible zone (zone_index > reclaim_index) to the next
> > generation. This allows for the progression of min_seq and reclaiming
> > from the next generation (Gen 1).
> >
> > Qualcomm, Mediatek and raspberrypi [1] discovered this issue independently.
> >
> > [1] https://github.com/raspberrypi/linux/issues/5395
> >
> > Cc: Yu Zhao <yuzhao@google.com>
> > Cc: Andrew Morton <akpm@linux-foundation.org>
> > Reported-by: Charan Teja Kalla <quic_charante@quicinc.com>
> > Reported-by: Lecopzer Chen <lecopzer.chen@mediatek.com>
> > Signed-off-by: Kalesh Singh <kaleshsingh@google.com>
>
> LGTM. But I think we need the Fixes tag and Cc stable.

I've reposted the patches with fixes tag and stable cc'ed at:
https://lore.kernel.org/lkml/20230802025606.346758-1-kaleshsingh@google.com/#t

Thanks,
Kalesh