LinuxLists.cc - [PATCH -next 0/2] Make memory reclamation measurable

2023-12-12 03:27:09

Subject: [PATCH -next 0/2] Make memory reclamation measurable

From: cuibixuan <[email protected]>

When the system memory is low, kswapd reclaims the memory. The key steps
of memory reclamation include
1.shrink_lruvec
* shrink_active_list, moves folios from the active LRU to the inactive LRU
* shrink_inactive_list, shrink lru from inactive LRU list
2.shrink_slab
* shrinker->count_objects(), calculates the freeable memory
* shrinker->scan_objects(), reclaims the slab memory

The existing tracers in the vmscan are as follows:

--do_try_to_free_pages
--shrink_zones
--trace_mm_vmscan_node_reclaim_begin (tracer)
--shrink_node
--shrink_node_memcgs
--trace_mm_vmscan_memcg_shrink_begin (tracer)
--shrink_lruvec
--shrink_list
--shrink_active_list
--trace_mm_vmscan_lru_shrink_active (tracer)
--shrink_inactive_list
--trace_mm_vmscan_lru_shrink_inactive (tracer)
--shrink_active_list
--shrink_slab
--do_shrink_slab
--shrinker->count_objects()
--trace_mm_shrink_slab_start (tracer)
--shrinker->scan_objects()
--trace_mm_shrink_slab_end (tracer)
--trace_mm_vmscan_memcg_shrink_end (tracer)
--trace_mm_vmscan_node_reclaim_end (tracer)

If we get the duration and quantity of shrink lru and slab,
then we can measure the memory recycling, as follows

Measuring memory reclamation with bpf:
LRU FILE:
CPU COMM ShrinkActive(us) ShrinkInactive(us) Reclaim(page)
7 kswapd0 26 51 32
7 kswapd0 52 47 13
SLAB:
CPU COMM OBJ_NAME Count_Dur(us) Freeable(page) Scan_Dur(us) Reclaim(page)
1 kswapd0 super_cache_scan.cfi_jt 2 341 3225 128
7 kswapd0 super_cache_scan.cfi_jt 0 2247 8524 1024
7 kswapd0 super_cache_scan.cfi_jt 2367 0 0 0

For this, add the new tracer to shrink_active_list/shrink_inactive_list
and shrinker->count_objects().

cuibixuan (2):
mm: shrinker: add new event to trace shrink count
mm: vmscan: add new event to trace shrink lru

include/trace/events/vmscan.h | 87 ++++++++++++++++++++++++++++++++++-
mm/shrinker.c | 4 ++
mm/vmscan.c | 8 +++-
3 files changed, 95 insertions(+), 4 deletions(-)

--
2.39.0

2023-12-12 03:27:13

by Bixuan Cui

[permalink] [raw]

Subject: [PATCH -next 1/2] mm: shrinker: add new event to trace shrink count

From: cuibixuan <[email protected]>

do_shrink_slab() calculates the freeable memory through shrinker->count_objects(),
and then reclaims the memory through shrinker->scan_objects(). When reclaiming
memory, shrinker->count_objects() takes a certain amount of time:

Fun spend(us)
ext4_es_count 4302
ext4_es_scan 12
super_cache_count 4195
super_cache_scan 2103

Therefore, adding the trace event to count_objects() can more accurately
obtain the time taken for slab memory recycling.

Example of output:
kswapd0-103 [003] ..... 1098.317942: mm_shrink_count_start: kfree_rcu_shrink_count.cfi_jt+0x0/0x8 00000000c540ff51: nid: 0
kswapd0-103 [003] ..... 1098.317951: mm_shrink_count_end: kfree_rcu_shrink_count.cfi_jt+0x0/0x8 00000000c540ff51: nid: 0 freeable:36

Signed-off-by: Bixuan Cui <[email protected]>
---
include/trace/events/vmscan.h | 49 +++++++++++++++++++++++++++++++++++
mm/shrinker.c | 4 +++
2 files changed, 53 insertions(+)

diff --git a/include/trace/events/vmscan.h b/include/trace/events/vmscan.h
index 1a488c30afa5..406faa5591c1 100644
--- a/include/trace/events/vmscan.h
+++ b/include/trace/events/vmscan.h
@@ -196,6 +196,55 @@ DEFINE_EVENT(mm_vmscan_direct_reclaim_end_template, mm_vmscan_memcg_softlimit_re
);
#endif /* CONFIG_MEMCG */

+TRACE_EVENT(mm_shrink_count_start,
+ TP_PROTO(struct shrinker *shr, struct shrink_control *sc),
+
+ TP_ARGS(shr, sc),
+
+ TP_STRUCT__entry(
+ __field(struct shrinker *, shr)
+ __field(void *, shrink)
+ __field(int, nid)
+ ),
+
+ TP_fast_assign(
+ __entry->shr = shr;
+ __entry->shrink = shr->count_objects;
+ __entry->nid = sc->nid;
+ ),
+
+ TP_printk("%pS %p: nid: %d",
+ __entry->shrink,
+ __entry->shr,
+ __entry->nid)
+);
+
+TRACE_EVENT(mm_shrink_count_end,
+ TP_PROTO(struct shrinker *shr, struct shrink_control *sc, long freeable),
+
+ TP_ARGS(shr, sc, freeable),
+
+ TP_STRUCT__entry(
+ __field(struct shrinker *, shr)
+ __field(void *, shrink)
+ __field(int, nid)
+ __field(long, freeable)
+ ),
+
+ TP_fast_assign(
+ __entry->shr = shr;
+ __entry->shrink = shr->count_objects;
+ __entry->nid = sc->nid;
+ __entry->freeable = freeable;
+ ),
+
+ TP_printk("%pS %p: nid: %d freeable:%ld",
+ __entry->shrink,
+ __entry->shr,
+ __entry->nid,
+ __entry->freeable)
+);
+
TRACE_EVENT(mm_shrink_slab_start,
TP_PROTO(struct shrinker *shr, struct shrink_control *sc,
long nr_objects_to_shrink, unsigned long cache_items,
diff --git a/mm/shrinker.c b/mm/shrinker.c
index dd91eab43ed3..d0c7bf61db61 100644
--- a/mm/shrinker.c
+++ b/mm/shrinker.c
@@ -379,7 +379,11 @@ static unsigned long do_shrink_slab(struct shrink_control *shrinkctl,
: SHRINK_BATCH;
long scanned = 0, next_deferred;

+ trace_mm_shrink_count_start(shrinker, shrinkctl);
+
freeable = shrinker->count_objects(shrinker, shrinkctl);
+
+ trace_mm_shrink_count_end(shrinker, shrinkctl, freeable);
if (freeable == 0 || freeable == SHRINK_EMPTY)
return freeable;

--
2.39.0

2023-12-12 03:27:26

by Bixuan Cui

[permalink] [raw]

Subject: [PATCH -next 2/2] mm: vmscan: add new event to trace shrink lru

From: cuibixuan <[email protected]>

Add a new event to calculate the shrink_inactive_list()/shrink_active_list()
execution time.

Example of output:
kswapd0-103 [007] ..... 1098.353020: mm_vmscan_lru_shrink_active_start: nid=0
kswapd0-103 [007] ..... 1098.353040: mm_vmscan_lru_shrink_active_end: nid=0 nr_taken=32 nr_active=0 nr_deactivated=32 nr_referenced=0 priority=6 flags=RECLAIM_WB_FILE|RECLAIM_WB_ASYNC
kswapd0-103 [007] ..... 1098.353040: mm_vmscan_lru_shrink_inactive_start: nid=0
kswapd0-103 [007] ..... 1098.353094: mm_vmscan_lru_shrink_inactive_end: nid=0 nr_scanned=32 nr_reclaimed=0 nr_dirty=0 nr_writeback=0 nr_congested=0 nr_immediate=0 nr_activate_anon=0 nr_activate_file=0 nr_ref_keep=32 nr_unmap_fail=0 priority=6 flags=RECLAIM_WB_ANON|RECLAIM_WB_ASYNC
kswapd0-103 [007] ..... 1098.353094: mm_vmscan_lru_shrink_inactive_start: nid=0
kswapd0-103 [007] ..... 1098.353162: mm_vmscan_lru_shrink_inactive_end: nid=0 nr_scanned=32 nr_reclaimed=21 nr_dirty=0 nr_writeback=0 nr_congested=0 nr_immediate=0 nr_activate_anon=0 nr_activate_file=0 nr_ref_keep=11 nr_unmap_fail=0 priority=6 flags=RECLAIM_WB_FILE|RECLAIM_WB_ASYNC

Signed-off-by: Bixuan Cui <[email protected]>
---
include/trace/events/vmscan.h | 38 +++++++++++++++++++++++++++++++++--
mm/vmscan.c | 8 ++++++--
2 files changed, 42 insertions(+), 4 deletions(-)

diff --git a/include/trace/events/vmscan.h b/include/trace/events/vmscan.h
index 406faa5591c1..9809d158f968 100644
--- a/include/trace/events/vmscan.h
+++ b/include/trace/events/vmscan.h
@@ -395,7 +395,24 @@ TRACE_EVENT(mm_vmscan_write_folio,
show_reclaim_flags(__entry->reclaim_flags))
);

-TRACE_EVENT(mm_vmscan_lru_shrink_inactive,
+TRACE_EVENT(mm_vmscan_lru_shrink_inactive_start,
+
+ TP_PROTO(int nid),
+
+ TP_ARGS(nid),
+
+ TP_STRUCT__entry(
+ __field(int, nid)
+ ),
+
+ TP_fast_assign(
+ __entry->nid = nid;
+ ),
+
+ TP_printk("nid=%d", __entry->nid)
+);
+
+TRACE_EVENT(mm_vmscan_lru_shrink_inactive_end,

TP_PROTO(int nid,
unsigned long nr_scanned, unsigned long nr_reclaimed,
@@ -446,7 +463,24 @@ TRACE_EVENT(mm_vmscan_lru_shrink_inactive,
show_reclaim_flags(__entry->reclaim_flags))
);

-TRACE_EVENT(mm_vmscan_lru_shrink_active,
+TRACE_EVENT(mm_vmscan_lru_shrink_active_start,
+
+ TP_PROTO(int nid),
+
+ TP_ARGS(nid),
+
+ TP_STRUCT__entry(
+ __field(int, nid)
+ ),
+
+ TP_fast_assign(
+ __entry->nid = nid;
+ ),
+
+ TP_printk("nid=%d", __entry->nid)
+);
+
+TRACE_EVENT(mm_vmscan_lru_shrink_active_end,

TP_PROTO(int nid, unsigned long nr_taken,
unsigned long nr_active, unsigned long nr_deactivated,
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 4e3b835c6b4a..73e690b3ce68 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1906,6 +1906,8 @@ static unsigned long shrink_inactive_list(unsigned long nr_to_scan,
struct pglist_data *pgdat = lruvec_pgdat(lruvec);
bool stalled = false;

+ trace_mm_vmscan_lru_shrink_inactive_start(pgdat->node_id);
+
while (unlikely(too_many_isolated(pgdat, file, sc))) {
if (stalled)
return 0;
@@ -1990,7 +1992,7 @@ static unsigned long shrink_inactive_list(unsigned long nr_to_scan,
if (file)
sc->nr.file_taken += nr_taken;

- trace_mm_vmscan_lru_shrink_inactive(pgdat->node_id,
+ trace_mm_vmscan_lru_shrink_inactive_end(pgdat->node_id,
nr_scanned, nr_reclaimed, &stat, sc->priority, file);
return nr_reclaimed;
}
@@ -2028,6 +2030,8 @@ static void shrink_active_list(unsigned long nr_to_scan,
int file = is_file_lru(lru);
struct pglist_data *pgdat = lruvec_pgdat(lruvec);

+ trace_mm_vmscan_lru_shrink_active_start(pgdat->node_id);
+
lru_add_drain();

spin_lock_irq(&lruvec->lru_lock);
@@ -2107,7 +2111,7 @@ static void shrink_active_list(unsigned long nr_to_scan,
lru_note_cost(lruvec, file, 0, nr_rotated);
mem_cgroup_uncharge_list(&l_active);
free_unref_page_list(&l_active);
- trace_mm_vmscan_lru_shrink_active(pgdat->node_id, nr_taken, nr_activate,
+ trace_mm_vmscan_lru_shrink_active_end(pgdat->node_id, nr_taken, nr_activate,
nr_deactivate, nr_rotated, sc->priority, file);
}

--
2.39.0

2023-12-13 03:03:51

by Andrew Morton

[permalink] [raw]

Subject: Re: [PATCH -next 2/2] mm: vmscan: add new event to trace shrink lru

On Mon, 11 Dec 2023 19:26:40 -0800 Bixuan Cui <[email protected]> wrote:

> -TRACE_EVENT(mm_vmscan_lru_shrink_inactive,
> +TRACE_EVENT(mm_vmscan_lru_shrink_inactive_start,

Current kernels have a call to trace_mm_vmscan_lru_shrink_inactive() in
evict_folios(), so this renaming broke the build.

2023-12-13 06:31:16

by Bixuan Cui

[permalink] [raw]

Subject: Re: [PATCH -next 2/2] mm: vmscan: add new event to trace shrink lru

在 2023/12/13 11:03, Andrew Morton 写道:
>> -TRACE_EVENT(mm_vmscan_lru_shrink_inactive,
>> +TRACE_EVENT(mm_vmscan_lru_shrink_inactive_start,
> Current kernels have a call to trace_mm_vmscan_lru_shrink_inactive() in
> evict_folios(), so this renaming broke the build.
Sorry, I did not enable CONFIG_LRU_GEN when compiling and testing.
I will double check my patches.

Thanks
Bixuan Cui

2023-12-14 14:29:56

by kernel test robot

[permalink] [raw]

Subject: Re: [PATCH -next 2/2] mm: vmscan: add new event to trace shrink lru

Hi Bixuan,

kernel test robot
[auto build test
url: ERROR on next-20231211]
s://github.com/intel-lab-lkp/linux/commits/Bixuan-Cui/mm-shrinker-add-new-event-to-trace-shrink-count/20231212-112824">https://github.com/intel-lab-lkp/linux/commits/Bixuan-Cui/mm-shrinker-add-new-event-to-trace-shrink-count/20231212-112824
https://lore.kernel.org/r/20231212032640.6968-3-cuibixuan%40vivo.com
[PATCH -next 2/2] mm: vmscan: add new event to trace shrink lru
-randconfig-003-20231214 (https://download.01.org/0day-ci/archive/20231214/[email protected]/config)
(Debian 11.3.0-12) 11.3.0
is a W=1 build): (https://download.01.org/0day-ci/archive/20231214/[email protected]/reproduce)
issue in a separate patch/commit (i.e. not just a new version of
kindly add following tags
kernel test robot <[email protected]>
href="https://lore.kernel.org/oe-kbuild-all/202312142212.vbSe7CMs-lkp@intel.com/">https://lore.kernel.org/oe-kbuild-all/[email protected]/
ones prefixed by >>):
In function 'evict_folios':
error: implicit declaration of function 'trace_mm_vmscan_lru_shrink_inactive'; did you mean 'trace_mm_vmscan_lru_shrink_inactive_end'? [-Werror=implicit-function-declaration]
trace_mm_vmscan_lru_shrink_inactive(pgdat->node_id,
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
trace_mm_vmscan_lru_shrink_inactive_end
warnings being treated as errors
2022-09-18 4500
2022-12-21 4501 static int evict_folios(struct lruvec *lruvec, struct scan_control *sc, int swappiness)
2022-09-18 4502 {
2022-09-18 4503 int type;
2022-09-18 4504 int scanned;
2022-09-18 4505 int reclaimed;
2022-09-18 4506 LIST_HEAD(list);
2022-11-15 4507 LIST_HEAD(clean);
2022-09-18 4508 struct folio *folio;
2022-11-15 4509 struct folio *next;
2022-09-18 4510 enum vm_event_item item;
2022-09-18 4511 struct reclaim_stat stat;
2022-09-18 4512 struct lru_gen_mm_walk *walk;
2022-11-15 4513 bool skip_retry = false;
2022-09-18 4514 struct mem_cgroup *memcg = lruvec_memcg(lruvec);
2022-09-18 4515 struct pglist_data *pgdat = lruvec_pgdat(lruvec);
2022-09-18 4516
2022-09-18 4517 spin_lock_irq(&lruvec->lru_lock);
2022-09-18 4518
2022-09-18 4519 scanned = isolate_folios(lruvec, sc, swappiness, &type, &list);
2022-09-18 4520
2022-09-18 4521 scanned += try_to_inc_min_seq(lruvec, swappiness);
2022-09-18 4522
2022-09-18 4523 if (get_nr_gens(lruvec, !swappiness) == MIN_NR_GENS)
2022-09-18 4524 scanned = 0;
2022-09-18 4525
2022-09-18 4526 spin_unlock_irq(&lruvec->lru_lock);
2022-09-18 4527
2022-09-18 4528 if (list_empty(&list))
2022-09-18 4529 return scanned;
2022-11-15 4530 retry:
Matthew Wilcox (Oracle 2022-09-02 4531) reclaimed = shrink_folio_list(&list, pgdat, sc, &stat, false);
2022-11-15 4532 sc->nr_reclaimed += reclaimed;
2023-10-03 @4533 trace_mm_vmscan_lru_shrink_inactive(pgdat->node_id,
2023-10-03 4534 scanned, reclaimed, &stat, sc->priority,
2023-10-03 4535 type ? LRU_INACTIVE_FILE : LRU_INACTIVE_ANON);
2022-11-15 4536
2022-11-15 4537 list_for_each_entry_safe_reverse(folio, next, &list, lru) {
2022-11-15 4538 if (!folio_evictable(folio)) {
2022-11-15 4539 list_del(&folio->lru);
2022-11-15 4540 folio_putback_lru(folio);
2022-11-15 4541 continue;
2022-11-15 4542 }
2022-09-18 4543
2022-11-15 4544 if (folio_test_reclaim(folio) &&
2022-11-15 4545 (folio_test_dirty(folio) || folio_test_writeback(folio))) {
2022-09-18 4546 /* restore LRU_REFS_FLAGS cleared by isolate_folio() */
2022-09-18 4547 if (folio_test_workingset(folio))
2022-09-18 4548 folio_set_referenced(folio);
2022-11-15 4549 continue;
2022-11-15 4550 }
2022-09-18 4551
2022-11-15 4552 if (skip_retry || folio_test_active(folio) || folio_test_referenced(folio) ||
2022-11-15 4553 folio_mapped(folio) || folio_test_locked(folio) ||
2022-11-15 4554 folio_test_dirty(folio) || folio_test_writeback(folio)) {
2022-11-15 4555 /* don't add rejected folios to the oldest generation */
2022-11-15 4556 set_mask_bits(&folio->flags, LRU_REFS_MASK | LRU_REFS_FLAGS,
2022-11-15 4557 BIT(PG_active));
2022-11-15 4558 continue;
2022-11-15 4559 }
2022-11-15 4560
2022-11-15 4561 /* retry folios that may have missed folio_rotate_reclaimable() */
2022-11-15 4562 list_move(&folio->lru, &clean);
2022-11-15 4563 sc->nr_scanned -= folio_nr_pages(folio);
2022-09-18 4564 }
2022-09-18 4565
2022-09-18 4566 spin_lock_irq(&lruvec->lru_lock);
2022-09-18 4567
Matthew Wilcox (Oracle 2022-09-02 4568) move_folios_to_lru(lruvec, &list);
2022-09-18 4569
2022-09-18 4570 walk = current->reclaim_state->mm_walk;
2022-09-18 4571 if (walk && walk->batched)
2022-09-18 4572 reset_batch_size(lruvec, walk);
2022-09-18 4573
2022-10-26 4574 item = PGSTEAL_KSWAPD + reclaimer_offset();
2022-09-18 4575 if (!cgroup_reclaim(sc))
2022-09-18 4576 __count_vm_events(item, reclaimed);
2022-09-18 4577 __count_memcg_events(memcg, item, reclaimed);
2022-09-18 4578 __count_vm_events(PGSTEAL_ANON + type, reclaimed);
2022-09-18 4579
2022-09-18 4580 spin_unlock_irq(&lruvec->lru_lock);
2022-09-18 4581
2022-09-18 4582 mem_cgroup_uncharge_list(&list);
2022-09-18 4583 free_unref_page_list(&list);
2022-09-18 4584
2022-11-15 4585 INIT_LIST_HEAD(&list);
2022-11-15 4586 list_splice_init(&clean, &list);
2022-11-15 4587
2022-11-15 4588 if (!list_empty(&list)) {
2022-11-15 4589 skip_retry = true;
2022-11-15 4590 goto retry;
2022-11-15 4591 }
2022-09-18 4592
2022-09-18 4593 return scanned;
2022-09-18 4594 }
2022-09-18 4595
ub.com/intel/lkp-tests/wiki">https://github.com/intel/lkp-tests/wiki

2023-12-14 16:59:48

by kernel test robot

[permalink] [raw]

Subject: Re: [PATCH -next 2/2] mm: vmscan: add new event to trace shrink lru

Hi Bixuan,

kernel test robot noticed the following build errors:

[auto build test ERROR on next-20231211]

url: https://github.com/intel-lab-lkp/linux/commits/Bixuan-Cui/mm-shrinker-add-new-event-to-trace-shrink-count/20231212-112824
base: next-20231211
patch link: https://lore.kernel.org/r/20231212032640.6968-3-cuibixuan%40vivo.com
patch subject: [PATCH -next 2/2] mm: vmscan: add new event to trace shrink lru
config: i386-randconfig-014-20231214 (https://download.01.org/0day-ci/archive/20231215/[email protected]/config)
compiler: clang version 16.0.4 (https://github.com/llvm/llvm-project.git ae42196bc493ffe877a7e3dff8be32035dea4d07)
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20231215/[email protected]/reproduce)

If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <[email protected]>
| Closes: https://lore.kernel.org/oe-kbuild-all/[email protected]/

All errors (new ones prefixed by >>):

>> mm/vmscan.c:4533:2: error: call to undeclared function 'trace_mm_vmscan_lru_shrink_inactive'; ISO C99 and later do not support implicit function declarations [-Wimplicit-function-declaration]
trace_mm_vmscan_lru_shrink_inactive(pgdat->node_id,
^
mm/vmscan.c:4533:2: note: did you mean 'trace_mm_vmscan_lru_shrink_inactive_end'?
include/trace/events/vmscan.h:415:1: note: 'trace_mm_vmscan_lru_shrink_inactive_end' declared here
TRACE_EVENT(mm_vmscan_lru_shrink_inactive_end,
^
include/linux/tracepoint.h:566:2: note: expanded from macro 'TRACE_EVENT'
DECLARE_TRACE(name, PARAMS(proto), PARAMS(args))
^
include/linux/tracepoint.h:432:2: note: expanded from macro 'DECLARE_TRACE'
__DECLARE_TRACE(name, PARAMS(proto), PARAMS(args), \
^
include/linux/tracepoint.h:255:21: note: expanded from macro '__DECLARE_TRACE'
static inline void trace_##name(proto) \
^
<scratch space>:60:1: note: expanded from here
trace_mm_vmscan_lru_shrink_inactive_end
^
1 error generated.

vim +/trace_mm_vmscan_lru_shrink_inactive +4533 mm/vmscan.c

ac35a490237446 Yu Zhao 2022-09-18 4500
a579086c99ed70 Yu Zhao 2022-12-21 4501 static int evict_folios(struct lruvec *lruvec, struct scan_control *sc, int swappiness)
ac35a490237446 Yu Zhao 2022-09-18 4502 {
ac35a490237446 Yu Zhao 2022-09-18 4503 int type;
ac35a490237446 Yu Zhao 2022-09-18 4504 int scanned;
ac35a490237446 Yu Zhao 2022-09-18 4505 int reclaimed;
ac35a490237446 Yu Zhao 2022-09-18 4506 LIST_HEAD(list);
359a5e1416caaf Yu Zhao 2022-11-15 4507 LIST_HEAD(clean);
ac35a490237446 Yu Zhao 2022-09-18 4508 struct folio *folio;
359a5e1416caaf Yu Zhao 2022-11-15 4509 struct folio *next;
ac35a490237446 Yu Zhao 2022-09-18 4510 enum vm_event_item item;
ac35a490237446 Yu Zhao 2022-09-18 4511 struct reclaim_stat stat;
bd74fdaea14602 Yu Zhao 2022-09-18 4512 struct lru_gen_mm_walk *walk;
359a5e1416caaf Yu Zhao 2022-11-15 4513 bool skip_retry = false;
ac35a490237446 Yu Zhao 2022-09-18 4514 struct mem_cgroup *memcg = lruvec_memcg(lruvec);
ac35a490237446 Yu Zhao 2022-09-18 4515 struct pglist_data *pgdat = lruvec_pgdat(lruvec);
ac35a490237446 Yu Zhao 2022-09-18 4516
ac35a490237446 Yu Zhao 2022-09-18 4517 spin_lock_irq(&lruvec->lru_lock);
ac35a490237446 Yu Zhao 2022-09-18 4518
ac35a490237446 Yu Zhao 2022-09-18 4519 scanned = isolate_folios(lruvec, sc, swappiness, &type, &list);
ac35a490237446 Yu Zhao 2022-09-18 4520
ac35a490237446 Yu Zhao 2022-09-18 4521 scanned += try_to_inc_min_seq(lruvec, swappiness);
ac35a490237446 Yu Zhao 2022-09-18 4522
ac35a490237446 Yu Zhao 2022-09-18 4523 if (get_nr_gens(lruvec, !swappiness) == MIN_NR_GENS)
ac35a490237446 Yu Zhao 2022-09-18 4524 scanned = 0;
ac35a490237446 Yu Zhao 2022-09-18 4525
ac35a490237446 Yu Zhao 2022-09-18 4526 spin_unlock_irq(&lruvec->lru_lock);
ac35a490237446 Yu Zhao 2022-09-18 4527
ac35a490237446 Yu Zhao 2022-09-18 4528 if (list_empty(&list))
ac35a490237446 Yu Zhao 2022-09-18 4529 return scanned;
359a5e1416caaf Yu Zhao 2022-11-15 4530 retry:
49fd9b6df54e61 Matthew Wilcox (Oracle 2022-09-02 4531) reclaimed = shrink_folio_list(&list, pgdat, sc, &stat, false);
359a5e1416caaf Yu Zhao 2022-11-15 4532 sc->nr_reclaimed += reclaimed;
8c2214fc9a470a Jaewon Kim 2023-10-03 @4533 trace_mm_vmscan_lru_shrink_inactive(pgdat->node_id,
8c2214fc9a470a Jaewon Kim 2023-10-03 4534 scanned, reclaimed, &stat, sc->priority,
8c2214fc9a470a Jaewon Kim 2023-10-03 4535 type ? LRU_INACTIVE_FILE : LRU_INACTIVE_ANON);
359a5e1416caaf Yu Zhao 2022-11-15 4536
359a5e1416caaf Yu Zhao 2022-11-15 4537 list_for_each_entry_safe_reverse(folio, next, &list, lru) {
359a5e1416caaf Yu Zhao 2022-11-15 4538 if (!folio_evictable(folio)) {
359a5e1416caaf Yu Zhao 2022-11-15 4539 list_del(&folio->lru);
359a5e1416caaf Yu Zhao 2022-11-15 4540 folio_putback_lru(folio);
359a5e1416caaf Yu Zhao 2022-11-15 4541 continue;
359a5e1416caaf Yu Zhao 2022-11-15 4542 }
ac35a490237446 Yu Zhao 2022-09-18 4543
359a5e1416caaf Yu Zhao 2022-11-15 4544 if (folio_test_reclaim(folio) &&
359a5e1416caaf Yu Zhao 2022-11-15 4545 (folio_test_dirty(folio) || folio_test_writeback(folio))) {
ac35a490237446 Yu Zhao 2022-09-18 4546 /* restore LRU_REFS_FLAGS cleared by isolate_folio() */
ac35a490237446 Yu Zhao 2022-09-18 4547 if (folio_test_workingset(folio))
ac35a490237446 Yu Zhao 2022-09-18 4548 folio_set_referenced(folio);
359a5e1416caaf Yu Zhao 2022-11-15 4549 continue;
359a5e1416caaf Yu Zhao 2022-11-15 4550 }
ac35a490237446 Yu Zhao 2022-09-18 4551
359a5e1416caaf Yu Zhao 2022-11-15 4552 if (skip_retry || folio_test_active(folio) || folio_test_referenced(folio) ||
359a5e1416caaf Yu Zhao 2022-11-15 4553 folio_mapped(folio) || folio_test_locked(folio) ||
359a5e1416caaf Yu Zhao 2022-11-15 4554 folio_test_dirty(folio) || folio_test_writeback(folio)) {
359a5e1416caaf Yu Zhao 2022-11-15 4555 /* don't add rejected folios to the oldest generation */
359a5e1416caaf Yu Zhao 2022-11-15 4556 set_mask_bits(&folio->flags, LRU_REFS_MASK | LRU_REFS_FLAGS,
359a5e1416caaf Yu Zhao 2022-11-15 4557 BIT(PG_active));
359a5e1416caaf Yu Zhao 2022-11-15 4558 continue;
359a5e1416caaf Yu Zhao 2022-11-15 4559 }
359a5e1416caaf Yu Zhao 2022-11-15 4560
359a5e1416caaf Yu Zhao 2022-11-15 4561 /* retry folios that may have missed folio_rotate_reclaimable() */
359a5e1416caaf Yu Zhao 2022-11-15 4562 list_move(&folio->lru, &clean);
359a5e1416caaf Yu Zhao 2022-11-15 4563 sc->nr_scanned -= folio_nr_pages(folio);
ac35a490237446 Yu Zhao 2022-09-18 4564 }
ac35a490237446 Yu Zhao 2022-09-18 4565
ac35a490237446 Yu Zhao 2022-09-18 4566 spin_lock_irq(&lruvec->lru_lock);
ac35a490237446 Yu Zhao 2022-09-18 4567
49fd9b6df54e61 Matthew Wilcox (Oracle 2022-09-02 4568) move_folios_to_lru(lruvec, &list);
ac35a490237446 Yu Zhao 2022-09-18 4569
bd74fdaea14602 Yu Zhao 2022-09-18 4570 walk = current->reclaim_state->mm_walk;
bd74fdaea14602 Yu Zhao 2022-09-18 4571 if (walk && walk->batched)
bd74fdaea14602 Yu Zhao 2022-09-18 4572 reset_batch_size(lruvec, walk);
bd74fdaea14602 Yu Zhao 2022-09-18 4573
57e9cc50f4dd92 Johannes Weiner 2022-10-26 4574 item = PGSTEAL_KSWAPD + reclaimer_offset();
ac35a490237446 Yu Zhao 2022-09-18 4575 if (!cgroup_reclaim(sc))
ac35a490237446 Yu Zhao 2022-09-18 4576 __count_vm_events(item, reclaimed);
ac35a490237446 Yu Zhao 2022-09-18 4577 __count_memcg_events(memcg, item, reclaimed);
ac35a490237446 Yu Zhao 2022-09-18 4578 __count_vm_events(PGSTEAL_ANON + type, reclaimed);
ac35a490237446 Yu Zhao 2022-09-18 4579
ac35a490237446 Yu Zhao 2022-09-18 4580 spin_unlock_irq(&lruvec->lru_lock);
ac35a490237446 Yu Zhao 2022-09-18 4581
ac35a490237446 Yu Zhao 2022-09-18 4582 mem_cgroup_uncharge_list(&list);
ac35a490237446 Yu Zhao 2022-09-18 4583 free_unref_page_list(&list);
ac35a490237446 Yu Zhao 2022-09-18 4584
359a5e1416caaf Yu Zhao 2022-11-15 4585 INIT_LIST_HEAD(&list);
359a5e1416caaf Yu Zhao 2022-11-15 4586 list_splice_init(&clean, &list);
359a5e1416caaf Yu Zhao 2022-11-15 4587
359a5e1416caaf Yu Zhao 2022-11-15 4588 if (!list_empty(&list)) {
359a5e1416caaf Yu Zhao 2022-11-15 4589 skip_retry = true;
359a5e1416caaf Yu Zhao 2022-11-15 4590 goto retry;
359a5e1416caaf Yu Zhao 2022-11-15 4591 }
ac35a490237446 Yu Zhao 2022-09-18 4592
ac35a490237446 Yu Zhao 2022-09-18 4593 return scanned;
ac35a490237446 Yu Zhao 2022-09-18 4594 }
ac35a490237446 Yu Zhao 2022-09-18 4595

--
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki