2020-11-05 08:59:10

by Alex Shi

Subject: [PATCH v21 07/19] mm: page_idle_get_page() does not need lru_lock

From: Hugh Dickins <[email protected]>

It is necessary for page_idle_get_page() to recheck PageLRU() after
get_page_unless_zero(), but holding lru_lock around that serves no
useful purpose, and adds to lru_lock contention: delete it.

See https://lore.kernel.org/lkml/20150504031722.GA2768@blaptop for the
discussion that led to lru_lock there; but __page_set_anon_rmap() now
uses WRITE_ONCE(), and I see no other risk in page_idle_clear_pte_refs()
using rmap_walk() (beyond the risk of racing PageAnon->PageKsm, mostly
but not entirely prevented by page_count() check in ksm.c's
write_protect_page(): that risk being shared with page_referenced() and
not helped by lru_lock).

Signed-off-by: Hugh Dickins <[email protected]>
Signed-off-by: Alex Shi <[email protected]>
Cc: Andrew Morton <[email protected]>
Cc: Vladimir Davydov <[email protected]>
Cc: Vlastimil Babka <[email protected]>
Cc: Minchan Kim <[email protected]>
Cc: Alex Shi <[email protected]>
Cc: [email protected]
Cc: [email protected]
---
mm/page_idle.c | 4 ----
1 file changed, 4 deletions(-)

diff --git a/mm/page_idle.c b/mm/page_idle.c
index 057c61df12db..64e5344a992c 100644
--- a/mm/page_idle.c
+++ b/mm/page_idle.c
@@ -32,19 +32,15 @@
static struct page *page_idle_get_page(unsigned long pfn)
{
struct page *page = pfn_to_online_page(pfn);
- pg_data_t *pgdat;

if (!page || !PageLRU(page) ||
!get_page_unless_zero(page))
return NULL;

- pgdat = page_pgdat(page);
- spin_lock_irq(&pgdat->lru_lock);
if (unlikely(!PageLRU(page))) {
put_page(page);
page = NULL;
}
- spin_unlock_irq(&pgdat->lru_lock);
return page;
}

--
1.8.3.1

2020-11-10 19:05:17

by Johannes Weiner

Subject: Re: [PATCH v21 07/19] mm: page_idle_get_page() does not need lru_lock

On Thu, Nov 05, 2020 at 04:55:37PM +0800, Alex Shi wrote:
> From: Hugh Dickins <[email protected]>
>
> It is necessary for page_idle_get_page() to recheck PageLRU() after
> get_page_unless_zero(), but holding lru_lock around that serves no
> useful purpose, and adds to lru_lock contention: delete it.
>
> See https://lore.kernel.org/lkml/20150504031722.GA2768@blaptop for the
> discussion that led to lru_lock there; but __page_set_anon_rmap() now
> uses WRITE_ONCE(), and I see no other risk in page_idle_clear_pte_refs()
> using rmap_walk() (beyond the risk of racing PageAnon->PageKsm, mostly
> but not entirely prevented by page_count() check in ksm.c's
> write_protect_page(): that risk being shared with page_referenced() and
> not helped by lru_lock).
>
> Signed-off-by: Hugh Dickins <[email protected]>
> Signed-off-by: Alex Shi <[email protected]>
> Cc: Andrew Morton <[email protected]>
> Cc: Vladimir Davydov <[email protected]>
> Cc: Vlastimil Babka <[email protected]>
> Cc: Minchan Kim <[email protected]>
> Cc: Alex Shi <[email protected]>
> Cc: [email protected]
> Cc: [email protected]

Acked-by: Johannes Weiner <[email protected]>

2020-11-11 08:21:53

by huang ying

Subject: Re: [PATCH v21 07/19] mm: page_idle_get_page() does not need lru_lock

On Thu, Nov 5, 2020 at 4:56 PM Alex Shi <[email protected]> wrote:
>
> From: Hugh Dickins <[email protected]>
>
> It is necessary for page_idle_get_page() to recheck PageLRU() after
> get_page_unless_zero(), but holding lru_lock around that serves no
> useful purpose, and adds to lru_lock contention: delete it.
>
> See https://lore.kernel.org/lkml/20150504031722.GA2768@blaptop for the
> discussion that led to lru_lock there; but __page_set_anon_rmap() now
> uses WRITE_ONCE(), and I see no other risk in page_idle_clear_pte_refs()
> using rmap_walk() (beyond the risk of racing PageAnon->PageKsm, mostly
> but not entirely prevented by page_count() check in ksm.c's
> write_protect_page(): that risk being shared with page_referenced() and
> not helped by lru_lock).
>
> Signed-off-by: Hugh Dickins <[email protected]>
> Signed-off-by: Alex Shi <[email protected]>
> Cc: Andrew Morton <[email protected]>
> Cc: Vladimir Davydov <[email protected]>
> Cc: Vlastimil Babka <[email protected]>
> Cc: Minchan Kim <[email protected]>
> Cc: Alex Shi <[email protected]>
> Cc: [email protected]
> Cc: [email protected]
> ---
> mm/page_idle.c | 4 ----
> 1 file changed, 4 deletions(-)
>
> diff --git a/mm/page_idle.c b/mm/page_idle.c
> index 057c61df12db..64e5344a992c 100644
> --- a/mm/page_idle.c
> +++ b/mm/page_idle.c
> @@ -32,19 +32,15 @@
> static struct page *page_idle_get_page(unsigned long pfn)
> {
> struct page *page = pfn_to_online_page(pfn);
> - pg_data_t *pgdat;
>
> if (!page || !PageLRU(page) ||
> !get_page_unless_zero(page))
> return NULL;
>
> - pgdat = page_pgdat(page);
> - spin_lock_irq(&pgdat->lru_lock);

get_page_unless_zero() is a full memory barrier. But do we need a
compiler barrier here to prevent the compiler to cache PageLRU()
results here? Otherwise looks OK to me,

Acked-by: "Huang, Ying" <[email protected]>

Best Regards,
Huang, Ying

> if (unlikely(!PageLRU(page))) {
> put_page(page);
> page = NULL;
> }
> - spin_unlock_irq(&pgdat->lru_lock);
> return page;
> }
>
> --
> 1.8.3.1
>
>

2020-11-11 12:54:03

by Vlastimil Babka

Subject: Re: [PATCH v21 07/19] mm: page_idle_get_page() does not need lru_lock

On 11/11/20 9:17 AM, huang ying wrote:
> On Thu, Nov 5, 2020 at 4:56 PM Alex Shi <[email protected]> wrote:
>>
>> From: Hugh Dickins <[email protected]>
>>
>> It is necessary for page_idle_get_page() to recheck PageLRU() after
>> get_page_unless_zero(), but holding lru_lock around that serves no
>> useful purpose, and adds to lru_lock contention: delete it.
>>
>> See https://lore.kernel.org/lkml/20150504031722.GA2768@blaptop for the
>> discussion that led to lru_lock there; but __page_set_anon_rmap() now
>> uses WRITE_ONCE(), and I see no other risk in page_idle_clear_pte_refs()
>> using rmap_walk() (beyond the risk of racing PageAnon->PageKsm, mostly
>> but not entirely prevented by page_count() check in ksm.c's
>> write_protect_page(): that risk being shared with page_referenced() and
>> not helped by lru_lock).
>>
>> Signed-off-by: Hugh Dickins <[email protected]>
>> Signed-off-by: Alex Shi <[email protected]>
>> Cc: Andrew Morton <[email protected]>
>> Cc: Vladimir Davydov <[email protected]>
>> Cc: Vlastimil Babka <[email protected]>
>> Cc: Minchan Kim <[email protected]>
>> Cc: Alex Shi <[email protected]>
>> Cc: [email protected]
>> Cc: [email protected]
>> ---
>> mm/page_idle.c | 4 ----
>> 1 file changed, 4 deletions(-)
>>
>> diff --git a/mm/page_idle.c b/mm/page_idle.c
>> index 057c61df12db..64e5344a992c 100644
>> --- a/mm/page_idle.c
>> +++ b/mm/page_idle.c
>> @@ -32,19 +32,15 @@
>> static struct page *page_idle_get_page(unsigned long pfn)
>> {
>> struct page *page = pfn_to_online_page(pfn);
>> - pg_data_t *pgdat;
>>
>> if (!page || !PageLRU(page) ||
>> !get_page_unless_zero(page))
>> return NULL;
>>
>> - pgdat = page_pgdat(page);
>> - spin_lock_irq(&pgdat->lru_lock);
>
> get_page_unless_zero() is a full memory barrier. But do we need a
> compiler barrier here to prevent the compiler to cache PageLRU()
> results here? Otherwise looks OK to me,

I think the compiler barrier is also implied by the full memory barrier and
prevents the caching.

Acked-by: Vlastimil Babka <[email protected]>

> Acked-by: "Huang, Ying" <[email protected]>
>
> Best Regards,
> Huang, Ying
>
>> if (unlikely(!PageLRU(page))) {
>> put_page(page);
>> page = NULL;
>> }
>> - spin_unlock_irq(&pgdat->lru_lock);
>> return page;
>> }
>>
>> --
>> 1.8.3.1
>>
>>
>