* [PATCH] mm: Simplify shrink_inactive_list()
@ 2019-04-10 15:07 Kirill Tkhai
2019-04-11 22:13 ` Daniel Jordan
0 siblings, 1 reply; 11+ messages in thread
From: Kirill Tkhai @ 2019-04-10 15:07 UTC (permalink / raw)
To: akpm, mhocko, hannes, dave, ktkhai, linux-mm
This merges together duplicating patterns of code.
Changes in enum vm_event_item is made to underline
that *_DIRECT and *_KSWAPD must differ by 1.
Signed-off-by: Kirill Tkhai <ktkhai@virtuozzo.com>
---
include/linux/vm_event_item.h | 4 ++--
mm/vmscan.c | 31 +++++++++----------------------
2 files changed, 11 insertions(+), 24 deletions(-)
diff --git a/include/linux/vm_event_item.h b/include/linux/vm_event_item.h
index 47a3441cf4c4..8f1403e692a2 100644
--- a/include/linux/vm_event_item.h
+++ b/include/linux/vm_event_item.h
@@ -31,9 +31,9 @@ enum vm_event_item { PGPGIN, PGPGOUT, PSWPIN, PSWPOUT,
PGLAZYFREED,
PGREFILL,
PGSTEAL_KSWAPD,
- PGSTEAL_DIRECT,
+ PGSTEAL_DIRECT = PGSTEAL_KSWAPD + 1,
PGSCAN_KSWAPD,
- PGSCAN_DIRECT,
+ PGSCAN_DIRECT = PGSCAN_KSWAPD + 1,
PGSCAN_DIRECT_THROTTLE,
#ifdef CONFIG_NUMA
PGSCAN_ZONE_RECLAIM_FAILED,
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 836b28913bd7..f8ac0825d1c7 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1907,6 +1907,7 @@ shrink_inactive_list(unsigned long nr_to_scan, struct lruvec *lruvec,
unsigned long nr_taken;
struct reclaim_stat stat;
int file = is_file_lru(lru);
+ int is_direct = !current_is_kswapd();
struct pglist_data *pgdat = lruvec_pgdat(lruvec);
struct zone_reclaim_stat *reclaim_stat = &lruvec->reclaim_stat;
bool stalled = false;
@@ -1934,17 +1935,10 @@ shrink_inactive_list(unsigned long nr_to_scan, struct lruvec *lruvec,
__mod_node_page_state(pgdat, NR_ISOLATED_ANON + file, nr_taken);
reclaim_stat->recent_scanned[file] += nr_taken;
- if (current_is_kswapd()) {
- if (global_reclaim(sc))
- __count_vm_events(PGSCAN_KSWAPD, nr_scanned);
- count_memcg_events(lruvec_memcg(lruvec), PGSCAN_KSWAPD,
- nr_scanned);
- } else {
- if (global_reclaim(sc))
- __count_vm_events(PGSCAN_DIRECT, nr_scanned);
- count_memcg_events(lruvec_memcg(lruvec), PGSCAN_DIRECT,
- nr_scanned);
- }
+ if (global_reclaim(sc))
+ __count_vm_events(PGSCAN_KSWAPD + is_direct, nr_scanned);
+ __count_memcg_events(lruvec_memcg(lruvec), PGSCAN_KSWAPD + is_direct,
+ nr_scanned);
spin_unlock_irq(&pgdat->lru_lock);
if (nr_taken == 0)
@@ -1955,17 +1949,10 @@ shrink_inactive_list(unsigned long nr_to_scan, struct lruvec *lruvec,
spin_lock_irq(&pgdat->lru_lock);
- if (current_is_kswapd()) {
- if (global_reclaim(sc))
- __count_vm_events(PGSTEAL_KSWAPD, nr_reclaimed);
- count_memcg_events(lruvec_memcg(lruvec), PGSTEAL_KSWAPD,
- nr_reclaimed);
- } else {
- if (global_reclaim(sc))
- __count_vm_events(PGSTEAL_DIRECT, nr_reclaimed);
- count_memcg_events(lruvec_memcg(lruvec), PGSTEAL_DIRECT,
- nr_reclaimed);
- }
+ if (global_reclaim(sc))
+ __count_vm_events(PGSTEAL_KSWAPD + is_direct, nr_reclaimed);
+ __count_memcg_events(lruvec_memcg(lruvec), PGSTEAL_KSWAPD + is_direct,
+ nr_reclaimed);
reclaim_stat->recent_rotated[0] = stat.nr_activate[0];
reclaim_stat->recent_rotated[1] = stat.nr_activate[1];
^ permalink raw reply related [flat|nested] 11+ messages in thread* Re: [PATCH] mm: Simplify shrink_inactive_list()
2019-04-10 15:07 [PATCH] mm: Simplify shrink_inactive_list() Kirill Tkhai
@ 2019-04-11 22:13 ` Daniel Jordan
2019-04-12 0:05 ` Baoquan He
0 siblings, 1 reply; 11+ messages in thread
From: Daniel Jordan @ 2019-04-11 22:13 UTC (permalink / raw)
To: Kirill Tkhai; +Cc: akpm, mhocko, hannes, dave, linux-mm
On Wed, Apr 10, 2019 at 06:07:04PM +0300, Kirill Tkhai wrote:
> @@ -1934,17 +1935,10 @@ shrink_inactive_list(unsigned long nr_to_scan, struct lruvec *lruvec,
> __mod_node_page_state(pgdat, NR_ISOLATED_ANON + file, nr_taken);
> reclaim_stat->recent_scanned[file] += nr_taken;
>
> - if (current_is_kswapd()) {
> - if (global_reclaim(sc))
> - __count_vm_events(PGSCAN_KSWAPD, nr_scanned);
> - count_memcg_events(lruvec_memcg(lruvec), PGSCAN_KSWAPD,
> - nr_scanned);
> - } else {
> - if (global_reclaim(sc))
> - __count_vm_events(PGSCAN_DIRECT, nr_scanned);
> - count_memcg_events(lruvec_memcg(lruvec), PGSCAN_DIRECT,
> - nr_scanned);
> - }
> + if (global_reclaim(sc))
> + __count_vm_events(PGSCAN_KSWAPD + is_direct, nr_scanned);
> + __count_memcg_events(lruvec_memcg(lruvec), PGSCAN_KSWAPD + is_direct,
> + nr_scanned);
Nice to avoid duplication like this, but now it takes looking at
vm_event_item.h to understand that (PGSCAN_KSWAPD + is_direct) might mean
PGSCAN_DIRECT.
What about this pattern for each block instead, which makes the stat used
explicit and avoids the header change?
stat = current_is_kswapd() ? PG*_KSWAPD : PG*_DIRECT;
^ permalink raw reply [flat|nested] 11+ messages in thread* Re: [PATCH] mm: Simplify shrink_inactive_list()
2019-04-11 22:13 ` Daniel Jordan
@ 2019-04-12 0:05 ` Baoquan He
2019-04-12 10:52 ` Kirill Tkhai
2019-04-12 10:55 ` [PATCH v2] " Kirill Tkhai
0 siblings, 2 replies; 11+ messages in thread
From: Baoquan He @ 2019-04-12 0:05 UTC (permalink / raw)
To: Daniel Jordan; +Cc: Kirill Tkhai, akpm, mhocko, hannes, dave, linux-mm
On 04/11/19 at 06:13pm, Daniel Jordan wrote:
> On Wed, Apr 10, 2019 at 06:07:04PM +0300, Kirill Tkhai wrote:
> > @@ -1934,17 +1935,10 @@ shrink_inactive_list(unsigned long nr_to_scan, struct lruvec *lruvec,
> > __mod_node_page_state(pgdat, NR_ISOLATED_ANON + file, nr_taken);
> > reclaim_stat->recent_scanned[file] += nr_taken;
> >
> > - if (current_is_kswapd()) {
> > - if (global_reclaim(sc))
> > - __count_vm_events(PGSCAN_KSWAPD, nr_scanned);
> > - count_memcg_events(lruvec_memcg(lruvec), PGSCAN_KSWAPD,
> > - nr_scanned);
> > - } else {
> > - if (global_reclaim(sc))
> > - __count_vm_events(PGSCAN_DIRECT, nr_scanned);
> > - count_memcg_events(lruvec_memcg(lruvec), PGSCAN_DIRECT,
> > - nr_scanned);
> > - }
> > + if (global_reclaim(sc))
> > + __count_vm_events(PGSCAN_KSWAPD + is_direct, nr_scanned);
> > + __count_memcg_events(lruvec_memcg(lruvec), PGSCAN_KSWAPD + is_direct,
> > + nr_scanned);
>
> Nice to avoid duplication like this, but now it takes looking at
> vm_event_item.h to understand that (PGSCAN_KSWAPD + is_direct) might mean
> PGSCAN_DIRECT.
>
> What about this pattern for each block instead, which makes the stat used
> explicit and avoids the header change?
>
> stat = current_is_kswapd() ? PG*_KSWAPD : PG*_DIRECT;
Yeah, looks nice. Maybe name it as item or event since we have had stat
locally defined as "struct reclaim_stat stat".
enum vm_event_item item;
...
item = current_is_kswapd() ? PG*_KSWAPD : PG*_DIRECT;
^ permalink raw reply [flat|nested] 11+ messages in thread* Re: [PATCH] mm: Simplify shrink_inactive_list()
2019-04-12 0:05 ` Baoquan He
@ 2019-04-12 10:52 ` Kirill Tkhai
2019-04-12 10:55 ` [PATCH v2] " Kirill Tkhai
1 sibling, 0 replies; 11+ messages in thread
From: Kirill Tkhai @ 2019-04-12 10:52 UTC (permalink / raw)
To: Baoquan He, Daniel Jordan; +Cc: akpm, mhocko, hannes, dave, linux-mm
On 12.04.2019 03:05, Baoquan He wrote:
> On 04/11/19 at 06:13pm, Daniel Jordan wrote:
>> On Wed, Apr 10, 2019 at 06:07:04PM +0300, Kirill Tkhai wrote:
>>> @@ -1934,17 +1935,10 @@ shrink_inactive_list(unsigned long nr_to_scan, struct lruvec *lruvec,
>>> __mod_node_page_state(pgdat, NR_ISOLATED_ANON + file, nr_taken);
>>> reclaim_stat->recent_scanned[file] += nr_taken;
>>>
>>> - if (current_is_kswapd()) {
>>> - if (global_reclaim(sc))
>>> - __count_vm_events(PGSCAN_KSWAPD, nr_scanned);
>>> - count_memcg_events(lruvec_memcg(lruvec), PGSCAN_KSWAPD,
>>> - nr_scanned);
>>> - } else {
>>> - if (global_reclaim(sc))
>>> - __count_vm_events(PGSCAN_DIRECT, nr_scanned);
>>> - count_memcg_events(lruvec_memcg(lruvec), PGSCAN_DIRECT,
>>> - nr_scanned);
>>> - }
>>> + if (global_reclaim(sc))
>>> + __count_vm_events(PGSCAN_KSWAPD + is_direct, nr_scanned);
>>> + __count_memcg_events(lruvec_memcg(lruvec), PGSCAN_KSWAPD + is_direct,
>>> + nr_scanned);
>>
>> Nice to avoid duplication like this, but now it takes looking at
>> vm_event_item.h to understand that (PGSCAN_KSWAPD + is_direct) might mean
>> PGSCAN_DIRECT.
>>
>> What about this pattern for each block instead, which makes the stat used
>> explicit and avoids the header change?
>>
>> stat = current_is_kswapd() ? PG*_KSWAPD : PG*_DIRECT;
>
> Yeah, looks nice. Maybe name it as item or event since we have had stat
> locally defined as "struct reclaim_stat stat".
>
> enum vm_event_item item;
> ...
> item = current_is_kswapd() ? PG*_KSWAPD : PG*_DIRECT;
Sounds good.
^ permalink raw reply [flat|nested] 11+ messages in thread* [PATCH v2] mm: Simplify shrink_inactive_list()
2019-04-12 0:05 ` Baoquan He
2019-04-12 10:52 ` Kirill Tkhai
@ 2019-04-12 10:55 ` Kirill Tkhai
2019-04-12 11:31 ` Michal Hocko
1 sibling, 1 reply; 11+ messages in thread
From: Kirill Tkhai @ 2019-04-12 10:55 UTC (permalink / raw)
To: Baoquan He, Daniel Jordan; +Cc: akpm, mhocko, hannes, dave, linux-mm
This merges together duplicating patterns of code.
Also, replace count_memcg_events() with its
irq-careless namesake.
Signed-off-by: Kirill Tkhai <ktkhai@virtuozzo.com>
v2: Introduce local variable.
---
mm/vmscan.c | 31 +++++++++----------------------
1 file changed, 9 insertions(+), 22 deletions(-)
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 836b28913bd7..d96efff59a11 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1907,6 +1907,7 @@ shrink_inactive_list(unsigned long nr_to_scan, struct lruvec *lruvec,
unsigned long nr_taken;
struct reclaim_stat stat;
int file = is_file_lru(lru);
+ enum vm_event_item item;
struct pglist_data *pgdat = lruvec_pgdat(lruvec);
struct zone_reclaim_stat *reclaim_stat = &lruvec->reclaim_stat;
bool stalled = false;
@@ -1934,17 +1935,10 @@ shrink_inactive_list(unsigned long nr_to_scan, struct lruvec *lruvec,
__mod_node_page_state(pgdat, NR_ISOLATED_ANON + file, nr_taken);
reclaim_stat->recent_scanned[file] += nr_taken;
- if (current_is_kswapd()) {
- if (global_reclaim(sc))
- __count_vm_events(PGSCAN_KSWAPD, nr_scanned);
- count_memcg_events(lruvec_memcg(lruvec), PGSCAN_KSWAPD,
- nr_scanned);
- } else {
- if (global_reclaim(sc))
- __count_vm_events(PGSCAN_DIRECT, nr_scanned);
- count_memcg_events(lruvec_memcg(lruvec), PGSCAN_DIRECT,
- nr_scanned);
- }
+ item = current_is_kswapd() ? PGSCAN_KSWAPD : PGSCAN_DIRECT;
+ if (global_reclaim(sc))
+ __count_vm_events(item, nr_scanned);
+ __count_memcg_events(lruvec_memcg(lruvec), item, nr_scanned);
spin_unlock_irq(&pgdat->lru_lock);
if (nr_taken == 0)
@@ -1955,17 +1949,10 @@ shrink_inactive_list(unsigned long nr_to_scan, struct lruvec *lruvec,
spin_lock_irq(&pgdat->lru_lock);
- if (current_is_kswapd()) {
- if (global_reclaim(sc))
- __count_vm_events(PGSTEAL_KSWAPD, nr_reclaimed);
- count_memcg_events(lruvec_memcg(lruvec), PGSTEAL_KSWAPD,
- nr_reclaimed);
- } else {
- if (global_reclaim(sc))
- __count_vm_events(PGSTEAL_DIRECT, nr_reclaimed);
- count_memcg_events(lruvec_memcg(lruvec), PGSTEAL_DIRECT,
- nr_reclaimed);
- }
+ item = current_is_kswapd() ? PGSTEAL_KSWAPD : PGSTEAL_DIRECT;
+ if (global_reclaim(sc))
+ __count_vm_events(item, nr_reclaimed);
+ __count_memcg_events(lruvec_memcg(lruvec), item, nr_reclaimed);
reclaim_stat->recent_rotated[0] = stat.nr_activate[0];
reclaim_stat->recent_rotated[1] = stat.nr_activate[1];
^ permalink raw reply related [flat|nested] 11+ messages in thread* Re: [PATCH v2] mm: Simplify shrink_inactive_list()
2019-04-12 10:55 ` [PATCH v2] " Kirill Tkhai
@ 2019-04-12 11:31 ` Michal Hocko
2019-04-12 11:33 ` Kirill Tkhai
0 siblings, 1 reply; 11+ messages in thread
From: Michal Hocko @ 2019-04-12 11:31 UTC (permalink / raw)
To: Kirill Tkhai; +Cc: Baoquan He, Daniel Jordan, akpm, hannes, dave, linux-mm
On Fri 12-04-19 13:55:59, Kirill Tkhai wrote:
> This merges together duplicating patterns of code.
OK, this looks better than the previous version
> Also, replace count_memcg_events() with its
> irq-careless namesake.
Why?
> Signed-off-by: Kirill Tkhai <ktkhai@virtuozzo.com>
>
> v2: Introduce local variable.
> ---
> mm/vmscan.c | 31 +++++++++----------------------
> 1 file changed, 9 insertions(+), 22 deletions(-)
>
> diff --git a/mm/vmscan.c b/mm/vmscan.c
> index 836b28913bd7..d96efff59a11 100644
> --- a/mm/vmscan.c
> +++ b/mm/vmscan.c
> @@ -1907,6 +1907,7 @@ shrink_inactive_list(unsigned long nr_to_scan, struct lruvec *lruvec,
> unsigned long nr_taken;
> struct reclaim_stat stat;
> int file = is_file_lru(lru);
> + enum vm_event_item item;
> struct pglist_data *pgdat = lruvec_pgdat(lruvec);
> struct zone_reclaim_stat *reclaim_stat = &lruvec->reclaim_stat;
> bool stalled = false;
> @@ -1934,17 +1935,10 @@ shrink_inactive_list(unsigned long nr_to_scan, struct lruvec *lruvec,
> __mod_node_page_state(pgdat, NR_ISOLATED_ANON + file, nr_taken);
> reclaim_stat->recent_scanned[file] += nr_taken;
>
> - if (current_is_kswapd()) {
> - if (global_reclaim(sc))
> - __count_vm_events(PGSCAN_KSWAPD, nr_scanned);
> - count_memcg_events(lruvec_memcg(lruvec), PGSCAN_KSWAPD,
> - nr_scanned);
> - } else {
> - if (global_reclaim(sc))
> - __count_vm_events(PGSCAN_DIRECT, nr_scanned);
> - count_memcg_events(lruvec_memcg(lruvec), PGSCAN_DIRECT,
> - nr_scanned);
> - }
> + item = current_is_kswapd() ? PGSCAN_KSWAPD : PGSCAN_DIRECT;
> + if (global_reclaim(sc))
> + __count_vm_events(item, nr_scanned);
> + __count_memcg_events(lruvec_memcg(lruvec), item, nr_scanned);
> spin_unlock_irq(&pgdat->lru_lock);
>
> if (nr_taken == 0)
> @@ -1955,17 +1949,10 @@ shrink_inactive_list(unsigned long nr_to_scan, struct lruvec *lruvec,
>
> spin_lock_irq(&pgdat->lru_lock);
>
> - if (current_is_kswapd()) {
> - if (global_reclaim(sc))
> - __count_vm_events(PGSTEAL_KSWAPD, nr_reclaimed);
> - count_memcg_events(lruvec_memcg(lruvec), PGSTEAL_KSWAPD,
> - nr_reclaimed);
> - } else {
> - if (global_reclaim(sc))
> - __count_vm_events(PGSTEAL_DIRECT, nr_reclaimed);
> - count_memcg_events(lruvec_memcg(lruvec), PGSTEAL_DIRECT,
> - nr_reclaimed);
> - }
> + item = current_is_kswapd() ? PGSTEAL_KSWAPD : PGSTEAL_DIRECT;
> + if (global_reclaim(sc))
> + __count_vm_events(item, nr_reclaimed);
> + __count_memcg_events(lruvec_memcg(lruvec), item, nr_reclaimed);
> reclaim_stat->recent_rotated[0] = stat.nr_activate[0];
> reclaim_stat->recent_rotated[1] = stat.nr_activate[1];
>
--
Michal Hocko
SUSE Labs
^ permalink raw reply [flat|nested] 11+ messages in thread* Re: [PATCH v2] mm: Simplify shrink_inactive_list()
2019-04-12 11:31 ` Michal Hocko
@ 2019-04-12 11:33 ` Kirill Tkhai
2019-04-12 12:05 ` Michal Hocko
0 siblings, 1 reply; 11+ messages in thread
From: Kirill Tkhai @ 2019-04-12 11:33 UTC (permalink / raw)
To: Michal Hocko; +Cc: Baoquan He, Daniel Jordan, akpm, hannes, dave, linux-mm
On 12.04.2019 14:31, Michal Hocko wrote:
> On Fri 12-04-19 13:55:59, Kirill Tkhai wrote:
>> This merges together duplicating patterns of code.
>
> OK, this looks better than the previous version
>
>> Also, replace count_memcg_events() with its
>> irq-careless namesake.
>
> Why?
Since interrupts are already disabled, and there is
no a sense to disable them twice.
>
>> Signed-off-by: Kirill Tkhai <ktkhai@virtuozzo.com>
>>
>> v2: Introduce local variable.
>> ---
>> mm/vmscan.c | 31 +++++++++----------------------
>> 1 file changed, 9 insertions(+), 22 deletions(-)
>>
>> diff --git a/mm/vmscan.c b/mm/vmscan.c
>> index 836b28913bd7..d96efff59a11 100644
>> --- a/mm/vmscan.c
>> +++ b/mm/vmscan.c
>> @@ -1907,6 +1907,7 @@ shrink_inactive_list(unsigned long nr_to_scan, struct lruvec *lruvec,
>> unsigned long nr_taken;
>> struct reclaim_stat stat;
>> int file = is_file_lru(lru);
>> + enum vm_event_item item;
>> struct pglist_data *pgdat = lruvec_pgdat(lruvec);
>> struct zone_reclaim_stat *reclaim_stat = &lruvec->reclaim_stat;
>> bool stalled = false;
>> @@ -1934,17 +1935,10 @@ shrink_inactive_list(unsigned long nr_to_scan, struct lruvec *lruvec,
>> __mod_node_page_state(pgdat, NR_ISOLATED_ANON + file, nr_taken);
>> reclaim_stat->recent_scanned[file] += nr_taken;
>>
>> - if (current_is_kswapd()) {
>> - if (global_reclaim(sc))
>> - __count_vm_events(PGSCAN_KSWAPD, nr_scanned);
>> - count_memcg_events(lruvec_memcg(lruvec), PGSCAN_KSWAPD,
>> - nr_scanned);
>> - } else {
>> - if (global_reclaim(sc))
>> - __count_vm_events(PGSCAN_DIRECT, nr_scanned);
>> - count_memcg_events(lruvec_memcg(lruvec), PGSCAN_DIRECT,
>> - nr_scanned);
>> - }
>> + item = current_is_kswapd() ? PGSCAN_KSWAPD : PGSCAN_DIRECT;
>> + if (global_reclaim(sc))
>> + __count_vm_events(item, nr_scanned);
>> + __count_memcg_events(lruvec_memcg(lruvec), item, nr_scanned);
>> spin_unlock_irq(&pgdat->lru_lock);
>>
>> if (nr_taken == 0)
>> @@ -1955,17 +1949,10 @@ shrink_inactive_list(unsigned long nr_to_scan, struct lruvec *lruvec,
>>
>> spin_lock_irq(&pgdat->lru_lock);
>>
>> - if (current_is_kswapd()) {
>> - if (global_reclaim(sc))
>> - __count_vm_events(PGSTEAL_KSWAPD, nr_reclaimed);
>> - count_memcg_events(lruvec_memcg(lruvec), PGSTEAL_KSWAPD,
>> - nr_reclaimed);
>> - } else {
>> - if (global_reclaim(sc))
>> - __count_vm_events(PGSTEAL_DIRECT, nr_reclaimed);
>> - count_memcg_events(lruvec_memcg(lruvec), PGSTEAL_DIRECT,
>> - nr_reclaimed);
>> - }
>> + item = current_is_kswapd() ? PGSTEAL_KSWAPD : PGSTEAL_DIRECT;
>> + if (global_reclaim(sc))
>> + __count_vm_events(item, nr_reclaimed);
>> + __count_memcg_events(lruvec_memcg(lruvec), item, nr_reclaimed);
>> reclaim_stat->recent_rotated[0] = stat.nr_activate[0];
>> reclaim_stat->recent_rotated[1] = stat.nr_activate[1];
>>
>
^ permalink raw reply [flat|nested] 11+ messages in thread* Re: [PATCH v2] mm: Simplify shrink_inactive_list()
2019-04-12 11:33 ` Kirill Tkhai
@ 2019-04-12 12:05 ` Michal Hocko
2019-04-12 12:10 ` [PATCH v3] " Kirill Tkhai
0 siblings, 1 reply; 11+ messages in thread
From: Michal Hocko @ 2019-04-12 12:05 UTC (permalink / raw)
To: Kirill Tkhai; +Cc: Baoquan He, Daniel Jordan, akpm, hannes, dave, linux-mm
On Fri 12-04-19 14:33:27, Kirill Tkhai wrote:
> On 12.04.2019 14:31, Michal Hocko wrote:
> > On Fri 12-04-19 13:55:59, Kirill Tkhai wrote:
> >> This merges together duplicating patterns of code.
> >
> > OK, this looks better than the previous version
> >
> >> Also, replace count_memcg_events() with its
> >> irq-careless namesake.
> >
> > Why?
>
> Since interrupts are already disabled, and there is
> no a sense to disable them twice.
right and changelog could have mentioned that...
> >> Signed-off-by: Kirill Tkhai <ktkhai@virtuozzo.com>
Anyway, forgot to add
Acked-by: Michal Hocko <mhocko@suse.com>
> >>
> >> v2: Introduce local variable.
> >> ---
> >> mm/vmscan.c | 31 +++++++++----------------------
> >> 1 file changed, 9 insertions(+), 22 deletions(-)
> >>
> >> diff --git a/mm/vmscan.c b/mm/vmscan.c
> >> index 836b28913bd7..d96efff59a11 100644
> >> --- a/mm/vmscan.c
> >> +++ b/mm/vmscan.c
> >> @@ -1907,6 +1907,7 @@ shrink_inactive_list(unsigned long nr_to_scan, struct lruvec *lruvec,
> >> unsigned long nr_taken;
> >> struct reclaim_stat stat;
> >> int file = is_file_lru(lru);
> >> + enum vm_event_item item;
> >> struct pglist_data *pgdat = lruvec_pgdat(lruvec);
> >> struct zone_reclaim_stat *reclaim_stat = &lruvec->reclaim_stat;
> >> bool stalled = false;
> >> @@ -1934,17 +1935,10 @@ shrink_inactive_list(unsigned long nr_to_scan, struct lruvec *lruvec,
> >> __mod_node_page_state(pgdat, NR_ISOLATED_ANON + file, nr_taken);
> >> reclaim_stat->recent_scanned[file] += nr_taken;
> >>
> >> - if (current_is_kswapd()) {
> >> - if (global_reclaim(sc))
> >> - __count_vm_events(PGSCAN_KSWAPD, nr_scanned);
> >> - count_memcg_events(lruvec_memcg(lruvec), PGSCAN_KSWAPD,
> >> - nr_scanned);
> >> - } else {
> >> - if (global_reclaim(sc))
> >> - __count_vm_events(PGSCAN_DIRECT, nr_scanned);
> >> - count_memcg_events(lruvec_memcg(lruvec), PGSCAN_DIRECT,
> >> - nr_scanned);
> >> - }
> >> + item = current_is_kswapd() ? PGSCAN_KSWAPD : PGSCAN_DIRECT;
> >> + if (global_reclaim(sc))
> >> + __count_vm_events(item, nr_scanned);
> >> + __count_memcg_events(lruvec_memcg(lruvec), item, nr_scanned);
> >> spin_unlock_irq(&pgdat->lru_lock);
> >>
> >> if (nr_taken == 0)
> >> @@ -1955,17 +1949,10 @@ shrink_inactive_list(unsigned long nr_to_scan, struct lruvec *lruvec,
> >>
> >> spin_lock_irq(&pgdat->lru_lock);
> >>
> >> - if (current_is_kswapd()) {
> >> - if (global_reclaim(sc))
> >> - __count_vm_events(PGSTEAL_KSWAPD, nr_reclaimed);
> >> - count_memcg_events(lruvec_memcg(lruvec), PGSTEAL_KSWAPD,
> >> - nr_reclaimed);
> >> - } else {
> >> - if (global_reclaim(sc))
> >> - __count_vm_events(PGSTEAL_DIRECT, nr_reclaimed);
> >> - count_memcg_events(lruvec_memcg(lruvec), PGSTEAL_DIRECT,
> >> - nr_reclaimed);
> >> - }
> >> + item = current_is_kswapd() ? PGSTEAL_KSWAPD : PGSTEAL_DIRECT;
> >> + if (global_reclaim(sc))
> >> + __count_vm_events(item, nr_reclaimed);
> >> + __count_memcg_events(lruvec_memcg(lruvec), item, nr_reclaimed);
> >> reclaim_stat->recent_rotated[0] = stat.nr_activate[0];
> >> reclaim_stat->recent_rotated[1] = stat.nr_activate[1];
> >>
> >
>
--
Michal Hocko
SUSE Labs
^ permalink raw reply [flat|nested] 11+ messages in thread* [PATCH v3] mm: Simplify shrink_inactive_list()
2019-04-12 12:05 ` Michal Hocko
@ 2019-04-12 12:10 ` Kirill Tkhai
2019-04-12 14:16 ` Daniel Jordan
2019-04-12 15:16 ` Johannes Weiner
0 siblings, 2 replies; 11+ messages in thread
From: Kirill Tkhai @ 2019-04-12 12:10 UTC (permalink / raw)
To: Michal Hocko; +Cc: Baoquan He, Daniel Jordan, akpm, hannes, dave, linux-mm
This merges together duplicating patterns of code.
Also, replace count_memcg_events() with its
irq-careless namesake, because they are already
called in interrupts disabled context.
Signed-off-by: Kirill Tkhai <ktkhai@virtuozzo.com>
Acked-by: Michal Hocko <mhocko@suse.com>
v3: Advance changelog.
v2: Introduce local variable.
---
mm/vmscan.c | 31 +++++++++----------------------
1 file changed, 9 insertions(+), 22 deletions(-)
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 836b28913bd7..d96efff59a11 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1907,6 +1907,7 @@ shrink_inactive_list(unsigned long nr_to_scan, struct lruvec *lruvec,
unsigned long nr_taken;
struct reclaim_stat stat;
int file = is_file_lru(lru);
+ enum vm_event_item item;
struct pglist_data *pgdat = lruvec_pgdat(lruvec);
struct zone_reclaim_stat *reclaim_stat = &lruvec->reclaim_stat;
bool stalled = false;
@@ -1934,17 +1935,10 @@ shrink_inactive_list(unsigned long nr_to_scan, struct lruvec *lruvec,
__mod_node_page_state(pgdat, NR_ISOLATED_ANON + file, nr_taken);
reclaim_stat->recent_scanned[file] += nr_taken;
- if (current_is_kswapd()) {
- if (global_reclaim(sc))
- __count_vm_events(PGSCAN_KSWAPD, nr_scanned);
- count_memcg_events(lruvec_memcg(lruvec), PGSCAN_KSWAPD,
- nr_scanned);
- } else {
- if (global_reclaim(sc))
- __count_vm_events(PGSCAN_DIRECT, nr_scanned);
- count_memcg_events(lruvec_memcg(lruvec), PGSCAN_DIRECT,
- nr_scanned);
- }
+ item = current_is_kswapd() ? PGSCAN_KSWAPD : PGSCAN_DIRECT;
+ if (global_reclaim(sc))
+ __count_vm_events(item, nr_scanned);
+ __count_memcg_events(lruvec_memcg(lruvec), item, nr_scanned);
spin_unlock_irq(&pgdat->lru_lock);
if (nr_taken == 0)
@@ -1955,17 +1949,10 @@ shrink_inactive_list(unsigned long nr_to_scan, struct lruvec *lruvec,
spin_lock_irq(&pgdat->lru_lock);
- if (current_is_kswapd()) {
- if (global_reclaim(sc))
- __count_vm_events(PGSTEAL_KSWAPD, nr_reclaimed);
- count_memcg_events(lruvec_memcg(lruvec), PGSTEAL_KSWAPD,
- nr_reclaimed);
- } else {
- if (global_reclaim(sc))
- __count_vm_events(PGSTEAL_DIRECT, nr_reclaimed);
- count_memcg_events(lruvec_memcg(lruvec), PGSTEAL_DIRECT,
- nr_reclaimed);
- }
+ item = current_is_kswapd() ? PGSTEAL_KSWAPD : PGSTEAL_DIRECT;
+ if (global_reclaim(sc))
+ __count_vm_events(item, nr_reclaimed);
+ __count_memcg_events(lruvec_memcg(lruvec), item, nr_reclaimed);
reclaim_stat->recent_rotated[0] = stat.nr_activate[0];
reclaim_stat->recent_rotated[1] = stat.nr_activate[1];
^ permalink raw reply related [flat|nested] 11+ messages in thread* Re: [PATCH v3] mm: Simplify shrink_inactive_list()
2019-04-12 12:10 ` [PATCH v3] " Kirill Tkhai
@ 2019-04-12 14:16 ` Daniel Jordan
2019-04-12 15:16 ` Johannes Weiner
1 sibling, 0 replies; 11+ messages in thread
From: Daniel Jordan @ 2019-04-12 14:16 UTC (permalink / raw)
To: Kirill Tkhai
Cc: Michal Hocko, Baoquan He, Daniel Jordan, akpm, hannes, dave, linux-mm
On Fri, Apr 12, 2019 at 03:10:01PM +0300, Kirill Tkhai wrote:
> This merges together duplicating patterns of code.
> Also, replace count_memcg_events() with its
> irq-careless namesake, because they are already
> called in interrupts disabled context.
>
> Signed-off-by: Kirill Tkhai <ktkhai@virtuozzo.com>
> Acked-by: Michal Hocko <mhocko@suse.com>
Looks good!
Reviewed-by: Daniel Jordan <daniel.m.jordan@oracle.com>
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [PATCH v3] mm: Simplify shrink_inactive_list()
2019-04-12 12:10 ` [PATCH v3] " Kirill Tkhai
2019-04-12 14:16 ` Daniel Jordan
@ 2019-04-12 15:16 ` Johannes Weiner
1 sibling, 0 replies; 11+ messages in thread
From: Johannes Weiner @ 2019-04-12 15:16 UTC (permalink / raw)
To: Kirill Tkhai
Cc: Michal Hocko, Baoquan He, Daniel Jordan, akpm, dave, linux-mm
On Fri, Apr 12, 2019 at 03:10:01PM +0300, Kirill Tkhai wrote:
> This merges together duplicating patterns of code.
> Also, replace count_memcg_events() with its
> irq-careless namesake, because they are already
> called in interrupts disabled context.
>
> Signed-off-by: Kirill Tkhai <ktkhai@virtuozzo.com>
> Acked-by: Michal Hocko <mhocko@suse.com>
>
> v3: Advance changelog.
> v2: Introduce local variable.
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
^ permalink raw reply [flat|nested] 11+ messages in thread
end of thread, other threads:[~2019-04-12 15:17 UTC | newest]
Thread overview: 11+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-04-10 15:07 [PATCH] mm: Simplify shrink_inactive_list() Kirill Tkhai
2019-04-11 22:13 ` Daniel Jordan
2019-04-12 0:05 ` Baoquan He
2019-04-12 10:52 ` Kirill Tkhai
2019-04-12 10:55 ` [PATCH v2] " Kirill Tkhai
2019-04-12 11:31 ` Michal Hocko
2019-04-12 11:33 ` Kirill Tkhai
2019-04-12 12:05 ` Michal Hocko
2019-04-12 12:10 ` [PATCH v3] " Kirill Tkhai
2019-04-12 14:16 ` Daniel Jordan
2019-04-12 15:16 ` Johannes Weiner
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.