mm: thp: grab the lock before manipulation defer list
Subsystems (4)
PATCH v3 1/1 mm: thp: grab the lock before manipulation defer list
As all the other places, we grab the lock before manipulate the defer list. Current implementation may face a race condition.
For example, the potential race would be:
CPU1 CPU2
mem_cgroup_move_account deferred_split_huge_page
list_empty
lock
list_empty
list_add_tail
unlock
lock
# list_empty might not hold anymore
list_add_tail
unlock
When this sequence happens, the list_add_tail() in mem_cgroup_move_account() corrupt the list since which is already been added to some split_queue in split_huge_page_to_list().
Besides this, David Rientjes points out the split_queue_len would be in a wrong state, which would be a significant issue for shrinkers.
---
v3:
* remove all review/ack tag since rewrite the changelog
* use deferred_split_huge_page as the example of race
* add cc stable 5.4+ tag as suggested by David Rientjes
v2:
* move check on compound outside suggested by Alexander
* an example of the race condition, suggested by Michal
--- mm/memcontrol.c | 18 +++++++++++------- 1 file changed, 11 insertions(+), 7 deletions(-)
mm/memcontrol.c +11 -7
@@ -5360,10 +5360,12 @@ static int mem_cgroup_move_account(struct page *page, }#ifdef CONFIG_TRANSPARENT_HUGEPAGE -if (compound && !list_empty (page_deferred_list (page))) { +if (compound) {spin_lock (&from->deferred_split_queue.split_queue_lock); -list_del_init (page_deferred_list (page)); - from->deferred_split_queue.split_queue_len--; +if (!list_empty (page_deferred_list (page))) { +list_del_init (page_deferred_list (page)); + from->deferred_split_queue.split_queue_len--; + }spin_unlock (&from->deferred_split_queue.split_queue_lock); }#endif
@@ -5377,11 +5379,13 @@ static int mem_cgroup_move_account(struct page *page, page->mem_cgroup = to;#ifdef CONFIG_TRANSPARENT_HUGEPAGE -if (compound &&list_empty (page_deferred_list (page))) { +if (compound) {spin_lock (&to->deferred_split_queue.split_queue_lock); -list_add_tail (page_deferred_list (page), - &to->deferred_split_queue.split_queue); - to->deferred_split_queue.split_queue_len++; +if (list_empty (page_deferred_list (page))) { +list_add_tail (page_deferred_list (page), + &to->deferred_split_queue.split_queue); + to->deferred_split_queue.split_queue_len++; + }spin_unlock (&to->deferred_split_queue.split_queue_lock); }#endif
Signature · 1 line
2.17.1
Quoted 70 lines · Wei Yang
As all the other places, we grab the lock before manipulate the defer list. Current implementation may face a race condition.
For example, the potential race would be:
CPU1 CPU2
mem_cgroup_move_accountdeferred_split_huge_page
list_empty
lock
list_empty
list_add_tail
unlock
lock
#list_emptymight not hold anymore
list_add_tail
unlockWhen this sequence happens, the
list_add_tail()inmem_cgroup_move_account()corrupt the list since which is already been added to somesplit_queueinsplit_huge_page_to_list().Besides this, David Rientjes points out the
split_queue_lenwould be in a wrong state, which would be a significant issue for shrinkers.Fixes 87eaceb3faa5 ("mm: thp: make deferred split shrinker memcg aware")Cc <[email protected]> [5.4+]---
v3:
* remove all review/ack tag since rewrite the changelog
* usedeferred_split_huge_pageas the example of race
* add cc stable 5.4+ tag as suggested by David Rientjesv2:
* move check on compound outside suggested by Alexander
* an example of the race condition, suggested by Michal--- mm/memcontrol.c | 18 +++++++++++------- 1 file changed, 11 insertions(+), 7 deletions(-)mm/memcontrol.c +6
-5@@ -5360,10 +5360,12 @@ static int mem_cgroup_move_account(struct page *page, }#ifdef CONFIG_TRANSPARENT_HUGEPAGE -if (compound && !list_empty (page_deferred_list (page))) { +if (compound) {spin_lock (&from->deferred_split_queue.split_queue_lock); -list_del_init (page_deferred_list (page)); - from->deferred_split_queue.split_queue_len--; +if (!list_empty (page_deferred_list (page))) { +list_del_init (page_deferred_list (page)); + from->deferred_split_queue.split_queue_len--; + }spin_unlock (&from->deferred_split_queue.split_queue_lock); }#endif @@ -5377,11 +5379,13 @@ static int mem_cgroup_move_account(struct page *page, page->mem_cgroup = to;#ifdef CONFIG_TRANSPARENT_HUGEPAGE -if (compound &&list_empty (page_deferred_list (page))) { +if (compound) {spin_lock (&to->deferred_split_queue.split_queue_lock); - list_add_tail(page_deferred_list(page),
mm/memcontrol.c static int mem_cgroup_move_account(struct page *page, · line 5391
Quoted 10 lines
mm/memcontrol.c +5
-2- &to->deferred_split_queue.split_queue); - to->deferred_split_queue.split_queue_len++; +if (list_empty (page_deferred_list (page))) { +list_add_tail (page_deferred_list (page), + &to->deferred_split_queue.split_queue); + to->deferred_split_queue.split_queue_len++; + } spin_unlock(&to->deferred_split_queue.split_queue_lock); }#endif
The patch looks OK for me. But there is another question. I forget, why we unconditionally add a page with empty deferred list to deferred_split_queue. Shouldn't we also check that it was initially in the list? Something like:
mm/memcontrol.c +3 -1
@@ -5289,6 +5289,7 @@ static int mem_cgroup_move_account(struct page *page,struct pglist_data *pgdat;unsigned long flags;unsigned int nr_pages = compound ?hpage_nr_pages (page) :1 ; +bool split = false;int ret;bool anon;
@@ -5346,6 +5347,7 @@ static int mem_cgroup_move_account(struct page *page,if (!list_empty (page_deferred_list (page))) {list_del_init (page_deferred_list (page)); from->deferred_split_queue.split_queue_len--; + split = true; }spin_unlock (&from->deferred_split_queue.split_queue_lock); }
@@ -5360,7 +5362,7 @@ static int mem_cgroup_move_account(struct page *page, page->mem_cgroup = to;#ifdef CONFIG_TRANSPARENT_HUGEPAGE -if (compound) { +if (compound && split) {spin_lock (&to->deferred_split_queue.split_queue_lock);if (list_empty (page_deferred_list (page))) { list_add_tail(page_deferred_list(page),
Quoted 65 lines · Kirill Tkhai
Quoted 28 lines
mm/memcontrol.c +6
-5@@ -5360,10 +5360,12 @@ static int mem_cgroup_move_account(struct page *page, }#ifdef CONFIG_TRANSPARENT_HUGEPAGE -if (compound && !list_empty (page_deferred_list (page))) { +if (compound) {spin_lock (&from->deferred_split_queue.split_queue_lock); -list_del_init (page_deferred_list (page)); - from->deferred_split_queue.split_queue_len--; +if (!list_empty (page_deferred_list (page))) { +list_del_init (page_deferred_list (page)); + from->deferred_split_queue.split_queue_len--; + }spin_unlock (&from->deferred_split_queue.split_queue_lock); }#endif @@ -5377,11 +5379,13 @@ static int mem_cgroup_move_account(struct page *page, page->mem_cgroup = to;#ifdef CONFIG_TRANSPARENT_HUGEPAGE -if (compound &&list_empty (page_deferred_list (page))) { +if (compound) {spin_lock (&to->deferred_split_queue.split_queue_lock); - list_add_tail(page_deferred_list(page),mm/memcontrol.c static int mem_cgroup_move_account(struct page *page, · line 5391
Quoted 10 lines
mm/memcontrol.c +5
-2- &to->deferred_split_queue.split_queue); - to->deferred_split_queue.split_queue_len++; +if (list_empty (page_deferred_list (page))) { +list_add_tail (page_deferred_list (page), + &to->deferred_split_queue.split_queue); + to->deferred_split_queue.split_queue_len++; + } spin_unlock(&to->deferred_split_queue.split_queue_lock); }#endif The patch looks OK for me. But there is another question. I forget, why we unconditionally add a page with empty deferred list to
deferred_split_queue. Shouldn't we also check that it was initially in the list? Something like:mm/memcontrol.c +2
-0@@ -5289,6 +5289,7 @@ static int mem_cgroup_move_account(struct page *page,struct pglist_data *pgdat;unsigned long flags;unsigned int nr_pages = compound ?hpage_nr_pages (page) :1 ; +bool split = false;int ret;bool anon;@@ -5346,6 +5347,7 @@ static int mem_cgroup_move_account(struct page *page,if (!list_empty (page_deferred_list (page))) {list_del_init (page_deferred_list (page)); from->deferred_split_queue.split_queue_len--; + split = true; }spin_unlock (&from->deferred_split_queue.split_queue_lock); }
mm/memcontrol.c static int mem_cgroup_move_account(struct page *page, · line 5368
Quoted 8 lines
mm/memcontrol.c +1
-1@@ -5360,7 +5362,7 @@ static int mem_cgroup_move_account(struct page *page, page->mem_cgroup = to;#ifdef CONFIG_TRANSPARENT_HUGEPAGE -if (compound) { +if (compound && split) {spin_lock (&to->deferred_split_queue.split_queue_lock);if (list_empty (page_deferred_list (page))) { list_add_tail(page_deferred_list(page),
I think that's a good point, especially considering that the current code appears to unconditionally place any compound page on the deferred split queue of the destination memcg. The correct list that it should appear on, I believe, depends on whether the pmd has been split for the process being moved: note the MC_TARGET_PAGE caveat in mem_cgroup_move_charge_pte_range() that does not move the charge for compound pages with split pmds. So when mem_cgroup_move_account() is called with compound == true, we're moving the charge of the entire compound page: why would it appear on that memcg's deferred split queue?
Quoted 77 lines · David Rientjes
Quoted 65 lines · Kirill Tkhai
Quoted 28 lines
mm/memcontrol.c +6
-5@@ -5360,10 +5360,12 @@ static int mem_cgroup_move_account(struct page *page, }#ifdef CONFIG_TRANSPARENT_HUGEPAGE -if (compound && !list_empty (page_deferred_list (page))) { +if (compound) {spin_lock (&from->deferred_split_queue.split_queue_lock); -list_del_init (page_deferred_list (page)); - from->deferred_split_queue.split_queue_len--; +if (!list_empty (page_deferred_list (page))) { +list_del_init (page_deferred_list (page)); + from->deferred_split_queue.split_queue_len--; + }spin_unlock (&from->deferred_split_queue.split_queue_lock); }#endif @@ -5377,11 +5379,13 @@ static int mem_cgroup_move_account(struct page *page, page->mem_cgroup = to;#ifdef CONFIG_TRANSPARENT_HUGEPAGE -if (compound &&list_empty (page_deferred_list (page))) { +if (compound) {spin_lock (&to->deferred_split_queue.split_queue_lock); - list_add_tail(page_deferred_list(page),mm/memcontrol.c static int mem_cgroup_move_account(struct page *page, · line 5391
Quoted 10 lines
mm/memcontrol.c +5
-2- &to->deferred_split_queue.split_queue); - to->deferred_split_queue.split_queue_len++; +if (list_empty (page_deferred_list (page))) { +list_add_tail (page_deferred_list (page), + &to->deferred_split_queue.split_queue); + to->deferred_split_queue.split_queue_len++; + } spin_unlock(&to->deferred_split_queue.split_queue_lock); }#endif The patch looks OK for me. But there is another question. I forget, why we unconditionally add a page with empty deferred list to
deferred_split_queue. Shouldn't we also check that it was initially in the list? Something like:mm/memcontrol.c +2
-0@@ -5289,6 +5289,7 @@ static int mem_cgroup_move_account(struct page *page,struct pglist_data *pgdat;unsigned long flags;unsigned int nr_pages = compound ?hpage_nr_pages (page) :1 ; +bool split = false;int ret;bool anon;@@ -5346,6 +5347,7 @@ static int mem_cgroup_move_account(struct page *page,if (!list_empty (page_deferred_list (page))) {list_del_init (page_deferred_list (page)); from->deferred_split_queue.split_queue_len--; + split = true; }spin_unlock (&from->deferred_split_queue.split_queue_lock); }mm/memcontrol.c static int mem_cgroup_move_account(struct page *page, · line 5368
Quoted 8 lines
mm/memcontrol.c +1
-1@@ -5360,7 +5362,7 @@ static int mem_cgroup_move_account(struct page *page, page->mem_cgroup = to;#ifdef CONFIG_TRANSPARENT_HUGEPAGE -if (compound) { +if (compound && split) {spin_lock (&to->deferred_split_queue.split_queue_lock);if (list_empty (page_deferred_list (page))) { list_add_tail(page_deferred_list(page),I think that's a good point, especially considering that the current code
Quoted 8 lines
appears to unconditionally place any compound page on the deferred split queue of the destination memcg. The correct list that it should appear on, I believe, depends on whether the pmd has been split for the process being moved: note the
MC_TARGET_PAGEcaveat inmem_cgroup_move_charge_pte_range()that does not move the charge for compound pages with split pmds. So whenmem_cgroup_move_account()is called with compound == true, we're moving the charge of the entire compound page: why would it appear on that memcg's deferred split queue?
Well, Kirill's change is easy to understand, while your statement here is a bit hard for me. Seems I lack some knowledge about cgroup. I am sorry about this. :-(
Signature · 2 lines
Wei Yang Help you, Help me
Michal Hocko
maintainerQuoted 77 lines · David Rientjes
Quoted 65 lines · Kirill Tkhai
Quoted 28 lines
mm/memcontrol.c +6
-5@@ -5360,10 +5360,12 @@ static int mem_cgroup_move_account(struct page *page, }#ifdef CONFIG_TRANSPARENT_HUGEPAGE -if (compound && !list_empty (page_deferred_list (page))) { +if (compound) {spin_lock (&from->deferred_split_queue.split_queue_lock); -list_del_init (page_deferred_list (page)); - from->deferred_split_queue.split_queue_len--; +if (!list_empty (page_deferred_list (page))) { +list_del_init (page_deferred_list (page)); + from->deferred_split_queue.split_queue_len--; + }spin_unlock (&from->deferred_split_queue.split_queue_lock); }#endif @@ -5377,11 +5379,13 @@ static int mem_cgroup_move_account(struct page *page, page->mem_cgroup = to;#ifdef CONFIG_TRANSPARENT_HUGEPAGE -if (compound &&list_empty (page_deferred_list (page))) { +if (compound) {spin_lock (&to->deferred_split_queue.split_queue_lock); - list_add_tail(page_deferred_list(page),mm/memcontrol.c static int mem_cgroup_move_account(struct page *page, · line 5391
Quoted 10 lines
mm/memcontrol.c +5
-2- &to->deferred_split_queue.split_queue); - to->deferred_split_queue.split_queue_len++; +if (list_empty (page_deferred_list (page))) { +list_add_tail (page_deferred_list (page), + &to->deferred_split_queue.split_queue); + to->deferred_split_queue.split_queue_len++; + } spin_unlock(&to->deferred_split_queue.split_queue_lock); }#endif The patch looks OK for me. But there is another question. I forget, why we unconditionally add a page with empty deferred list to
deferred_split_queue. Shouldn't we also check that it was initially in the list? Something like:mm/memcontrol.c +2
-0@@ -5289,6 +5289,7 @@ static int mem_cgroup_move_account(struct page *page,struct pglist_data *pgdat;unsigned long flags;unsigned int nr_pages = compound ?hpage_nr_pages (page) :1 ; +bool split = false;int ret;bool anon;@@ -5346,6 +5347,7 @@ static int mem_cgroup_move_account(struct page *page,if (!list_empty (page_deferred_list (page))) {list_del_init (page_deferred_list (page)); from->deferred_split_queue.split_queue_len--; + split = true; }spin_unlock (&from->deferred_split_queue.split_queue_lock); }mm/memcontrol.c static int mem_cgroup_move_account(struct page *page, · line 5368
Quoted 8 lines
mm/memcontrol.c +1
-1@@ -5360,7 +5362,7 @@ static int mem_cgroup_move_account(struct page *page, page->mem_cgroup = to;#ifdef CONFIG_TRANSPARENT_HUGEPAGE -if (compound) { +if (compound && split) {spin_lock (&to->deferred_split_queue.split_queue_lock);if (list_empty (page_deferred_list (page))) { list_add_tail(page_deferred_list(page),I think that's a good point, especially considering that the current code
Quoted 8 lines
appears to unconditionally place any compound page on the deferred split queue of the destination memcg. The correct list that it should appear on, I believe, depends on whether the pmd has been split for the process being moved: note the
MC_TARGET_PAGEcaveat inmem_cgroup_move_charge_pte_range()that does not move the charge for compound pages with split pmds. So whenmem_cgroup_move_account()is called with compound == true, we're moving the charge of the entire compound page: why would it appear on that memcg's deferred split queue?
I believe Kirill asked how do we know that the page should be actually added to the deferred list just from the list_empty check. In other words what if the page hasn't been split at all?
Signature · 2 lines
Michal Hocko SUSE Labs
Quoted 87 lines · Michal Hocko
Quoted 77 lines · David Rientjes
Quoted 65 lines · Kirill Tkhai
Quoted 28 lines
mm/memcontrol.c +6
-5@@ -5360,10 +5360,12 @@ static int (struct page *page, }#ifdef CONFIG_TRANSPARENT_HUGEPAGE -if (compound && !list_empty (page_deferred_list (page))) { +if (compound) {spin_lock (&from->deferred_split_queue.split_queue_lock); -list_del_init (page_deferred_list (page)); - from->deferred_split_queue.split_queue_len--; +if (!list_empty (page_deferred_list (page))) { +list_del_init (page_deferred_list (page)); + from->deferred_split_queue.split_queue_len--; + }spin_unlock (&from->deferred_split_queue.split_queue_lock); }#endif @@ -5377,11 +5379,13 @@ static int mem_cgroup_move_account(struct page *page, page->mem_cgroup = to;#ifdef CONFIG_TRANSPARENT_HUGEPAGE -if (compound &&list_empty (page_deferred_list (page))) { +if (compound) {spin_lock (&to->deferred_split_queue.split_queue_lock); - list_add_tail(page_deferred_list(page),mm/memcontrol.c static int mem_cgroup_move_account(struct page *page, · line 5391
Quoted 10 lines
mm/memcontrol.c +5
-2- &to->deferred_split_queue.split_queue); - to->deferred_split_queue.split_queue_len++; +if (list_empty (page_deferred_list (page))) { +list_add_tail (page_deferred_list (page), + &to->deferred_split_queue.split_queue); + to->deferred_split_queue.split_queue_len++; + } spin_unlock(&to->deferred_split_queue.split_queue_lock); }#endif The patch looks OK for me. But there is another question. I forget, why we unconditionally add a page with empty deferred list to
deferred_split_queue. Shouldn't we also check that it was initially in the list? Something like:mm/memcontrol.c +2
-0@@ -5289,6 +5289,7 @@ static int mem_cgroup_move_account(struct page *page,struct pglist_data *pgdat;unsigned long flags;unsigned int nr_pages = compound ?hpage_nr_pages (page) :1 ; +bool split = false;int ret;bool anon;@@ -5346,6 +5347,7 @@ static int mem_cgroup_move_account(struct page *page,if (!list_empty (page_deferred_list (page))) {list_del_init (page_deferred_list (page)); from->deferred_split_queue.split_queue_len--; + split = true; }spin_unlock (&from->deferred_split_queue.split_queue_lock); }mm/memcontrol.c static int mem_cgroup_move_account(struct page *page, · line 5368
Quoted 8 lines
mm/memcontrol.c +1
-1@@ -5360,7 +5362,7 @@ static int mem_cgroup_move_account(struct page *page, page->mem_cgroup = to;#ifdef CONFIG_TRANSPARENT_HUGEPAGE -if (compound) { +if (compound && split) {spin_lock (&to->deferred_split_queue.split_queue_lock);if (list_empty (page_deferred_list (page))) { list_add_tail(page_deferred_list(page),I think that's a good point, especially considering that the current code
Quoted 8 lines
appears to unconditionally place any compound page on the deferred split queue of the destination memcg. The correct list that it should appear on, I believe, depends on whether the pmd has been split for the process being moved: note the
MC_TARGET_PAGEcaveat inmem_cgroup_move_charge_pte_range()that does not move the charge for compound pages with split pmds. So whenmem_cgroup_move_account()is called with compound == true, we're moving the charge of the entire compound page: why would it appear on that memcg's deferred split queue?
Quoted 3 lines
I believe Kirill asked how do we know that the page should be actually added to the deferred list just from the
list_emptycheck. In other words what if the page hasn't been split at all?
Yes, I'm talking about this. Function mem_cgroup_move_account() adds every huge page to the deferred list, while we need to do that only for pages, which are queued for splitting...
Quoted 10 lines · Kirill Tkhai
Quoted 10 lines
Quoted 9 lines
I think that's a good point, especially considering that the current code appears to unconditionally place any compound page on the deferred split queue of the destination memcg. The correct list that it should appear on, I believe, depends on whether the pmd has been split for the process being moved: note the
MC_TARGET_PAGEcaveat inmem_cgroup_move_charge_pte_range()that does not move the charge for compound pages with split pmds. So whenmem_cgroup_move_account()is called with compound == true, we're moving the charge of the entire compound page: why would it appear on that memcg's deferred split queue?
Quoted 8 lines
Quoted 3 lines
I believe Kirill asked how do we know that the page should be actually added to the deferred list just from the
list_emptycheck. In other words what if the page hasn't been split at all?Yes, I'm talking about this. Function
mem_cgroup_move_account()adds every huge page to the deferred list, while we need to do that only for pages, which are queued for splitting...
Yup, and that appears broken before Wei's patch. Since we only migrate charges of entire compound pages (we have a mapping pmd, the underlying page cannot be split), it should not appear on the deferred split queue for any memcg, right?
Quoted 21 lines · David Rientjes
Quoted 10 lines · Kirill Tkhai
Quoted 10 lines
Quoted 9 lines
I think that's a good point, especially considering that the current code appears to unconditionally place any compound page on the deferred split queue of the destination memcg. The correct list that it should appear on, I believe, depends on whether the pmd has been split for the process being moved: note the
MC_TARGET_PAGEcaveat inmem_cgroup_move_charge_pte_range()that does not move the charge for compound pages with split pmds. So whenmem_cgroup_move_account()is called with compound == true, we're moving the charge of the entire compound page: why would it appear on that memcg's deferred split queue?Quoted 8 lines
Quoted 3 lines
I believe Kirill asked how do we know that the page should be actually added to the deferred list just from the
list_emptycheck. In other words what if the page hasn't been split at all?Yes, I'm talking about this. Function
mem_cgroup_move_account()adds every huge page to the deferred list, while we need to do that only for pages, which are queued for splitting...
Quoted 4 lines
Yup, and that appears broken before Wei's patch. Since we only migrate charges of entire compound pages (we have a mapping pmd, the underlying page cannot be split), it should not appear on the deferred split queue for any memcg, right?
Hm. Can't a huge page be mapped in two tasks:
1)the first task unmapped a part of page and initiated splitting, 2)the second task still refers the whole page,
then we move account for the second task?
Michal Hocko
maintainerQuoted 22 lines · Kirill Tkhai
Quoted 21 lines · David Rientjes
Quoted 18 lines · Kirill Tkhai
Quoted 10 lines
Quoted 9 lines
I think that's a good point, especially considering that the current code appears to unconditionally place any compound page on the deferred split queue of the destination memcg. The correct list that it should appear on, I believe, depends on whether the pmd has been split for the process being moved: note the
MC_TARGET_PAGEcaveat inmem_cgroup_move_charge_pte_range()that does not move the charge for compound pages with split pmds. So whenmem_cgroup_move_account()is called with compound == true, we're moving the charge of the entire compound page: why would it appear on that memcg's deferred split queue?Quoted 3 lines
I believe Kirill asked how do we know that the page should be actually added to the deferred list just from the
list_emptycheck. In other words what if the page hasn't been split at all?Yes, I'm talking about this. Function
mem_cgroup_move_account()adds every huge page to the deferred list, while we need to do that only for pages, which are queued for splitting...
Quoted 6 lines
Quoted 4 lines
Yup, and that appears broken before Wei's patch. Since we only migrate charges of entire compound pages (we have a mapping pmd, the underlying page cannot be split), it should not appear on the deferred split queue for any memcg, right?
Hm. Can't a huge page be mapped in two tasks:
It can but it will get charged to only of the initially. I haven't checked the THP code in that aspect but from what I remember subpages shouldn't refer to different memcgs. Kirill Shutemov?
Signature · 2 lines
Michal Hocko SUSE Labs
Quoted 87 lines · Michal Hocko
Quoted 77 lines · David Rientjes
Quoted 65 lines · Kirill Tkhai
Quoted 28 lines
mm/memcontrol.c +6
-5@@ -5360,10 +5360,12 @@ static int mem_cgroup_move_account(struct page *page, }#ifdef CONFIG_TRANSPARENT_HUGEPAGE -if (compound && !list_empty (page_deferred_list (page))) { +if (compound) {spin_lock (&from->deferred_split_queue.split_queue_lock); -list_del_init (page_deferred_list (page)); - from->deferred_split_queue.split_queue_len--; +if (!list_empty (page_deferred_list (page))) { +list_del_init (page_deferred_list (page)); + from->deferred_split_queue.split_queue_len--; + }spin_unlock (&from->deferred_split_queue.split_queue_lock); }#endif @@ -5377,11 +5379,13 @@ static int mem_cgroup_move_account(struct page *page, page->mem_cgroup = to;#ifdef CONFIG_TRANSPARENT_HUGEPAGE -if (compound &&list_empty (page_deferred_list (page))) { +if (compound) {spin_lock (&to->deferred_split_queue.split_queue_lock); - list_add_tail(page_deferred_list(page),mm/memcontrol.c static int mem_cgroup_move_account(struct page *page, · line 5391
Quoted 10 lines
mm/memcontrol.c +5
-2- &to->deferred_split_queue.split_queue); - to->deferred_split_queue.split_queue_len++; +if (list_empty (page_deferred_list (page))) { +list_add_tail (page_deferred_list (page), + &to->deferred_split_queue.split_queue); + to->deferred_split_queue.split_queue_len++; + } spin_unlock(&to->deferred_split_queue.split_queue_lock); }#endif The patch looks OK for me. But there is another question. I forget, why we unconditionally add a page with empty deferred list to
deferred_split_queue. Shouldn't we also check that it was initially in the list? Something like:mm/memcontrol.c +2
-0@@ -5289,6 +5289,7 @@ static int mem_cgroup_move_account(struct page *page,struct pglist_data *pgdat;unsigned long flags;unsigned int nr_pages = compound ?hpage_nr_pages (page) :1 ; +bool split = false;int ret;bool anon;@@ -5346,6 +5347,7 @@ static int mem_cgroup_move_account(struct page *page,if (!list_empty (page_deferred_list (page))) {list_del_init (page_deferred_list (page)); from->deferred_split_queue.split_queue_len--; + split = true; }spin_unlock (&from->deferred_split_queue.split_queue_lock); }mm/memcontrol.c static int mem_cgroup_move_account(struct page *page, · line 5368
Quoted 8 lines
mm/memcontrol.c +1
-1@@ -5360,7 +5362,7 @@ static int mem_cgroup_move_account(struct page *page, page->mem_cgroup = to;#ifdef CONFIG_TRANSPARENT_HUGEPAGE -if (compound) { +if (compound && split) {spin_lock (&to->deferred_split_queue.split_queue_lock);if (list_empty (page_deferred_list (page))) { list_add_tail(page_deferred_list(page),I think that's a good point, especially considering that the current code
Quoted 8 lines
appears to unconditionally place any compound page on the deferred split queue of the destination memcg. The correct list that it should appear on, I believe, depends on whether the pmd has been split for the process being moved: note the
MC_TARGET_PAGEcaveat inmem_cgroup_move_charge_pte_range()that does not move the charge for compound pages with split pmds. So whenmem_cgroup_move_account()is called with compound == true, we're moving the charge of the entire compound page: why would it appear on that memcg's deferred split queue?
Quoted 4 lines
I believe Kirill asked how do we know that the page should be actually added to the deferred list just from the
list_emptycheck. In other words what if the page hasn't been split at all?
Right, and I don't think that it necessarily is and the second conditional in Wei's patch will always succeed unless we have raced. That patch is for a lock concern but I think Kirill's question has uncovered something more interesting.
Kirill S would definitely be best to answer Kirill T's question, but from my understanding when mem_cgroup_move_account() is called with compound == true that we always have an intact pmd (we never migrate partial page charges for pages on the deferred split queue with the current charge migration implementation) and thus the underlying page is not eligible to be split and shouldn't be on the deferred split queue.
In other words, a page being on the deferred split queue for a memcg should only happen when it is charged to that memcg. (This wasn't the case when we only had per-node split queues.) I think that's currently broken in mem_cgroup_move_account() before Wei's patch.
Quoted 106 lines · David Rientjes
Quoted 87 lines · Michal Hocko
Quoted 77 lines · David Rientjes
Quoted 65 lines · Kirill Tkhai
Quoted 28 lines
mm/memcontrol.c +6
-5@@ -5360,10 +5360,12 @@ static int mem_cgroup_move_account(struct page *page, }#ifdef CONFIG_TRANSPARENT_HUGEPAGE -if (compound && !list_empty (page_deferred_list (page))) { +if (compound) {spin_lock (&from->deferred_split_queue.split_queue_lock); -list_del_init (page_deferred_list (page)); - from->deferred_split_queue.split_queue_len--; +if (!list_empty (page_deferred_list (page))) { +list_del_init (page_deferred_list (page)); + from->deferred_split_queue.split_queue_len--; + }spin_unlock (&from->deferred_split_queue.split_queue_lock); }#endif @@ -5377,11 +5379,13 @@ static int mem_cgroup_move_account(struct page *page, page->mem_cgroup = to;#ifdef CONFIG_TRANSPARENT_HUGEPAGE -if (compound &&list_empty (page_deferred_list (page))) { +if (compound) {spin_lock (&to->deferred_split_queue.split_queue_lock); - list_add_tail(page_deferred_list(page),mm/memcontrol.c static int mem_cgroup_move_account(struct page *page, · line 5391
Quoted 10 lines
mm/memcontrol.c +5
-2- &to->deferred_split_queue.split_queue); - to->deferred_split_queue.split_queue_len++; +if (list_empty (page_deferred_list (page))) { +list_add_tail (page_deferred_list (page), + &to->deferred_split_queue.split_queue); + to->deferred_split_queue.split_queue_len++; + } spin_unlock(&to->deferred_split_queue.split_queue_lock); }#endif The patch looks OK for me. But there is another question. I forget, why we unconditionally add a page with empty deferred list to
deferred_split_queue. Shouldn't we also check that it was initially in the list? Something like:mm/memcontrol.c +2
-0@@ -5289,6 +5289,7 @@ static int mem_cgroup_move_account(struct page *page,struct pglist_data *pgdat;unsigned long flags;unsigned int nr_pages = compound ?hpage_nr_pages (page) :1 ; +bool split = false;int ret;bool anon;@@ -5346,6 +5347,7 @@ static int mem_cgroup_move_account(struct page *page,if (!list_empty (page_deferred_list (page))) {list_del_init (page_deferred_list (page)); from->deferred_split_queue.split_queue_len--; + split = true; }spin_unlock (&from->deferred_split_queue.split_queue_lock); }mm/memcontrol.c static int mem_cgroup_move_account(struct page *page, · line 5368
Quoted 8 lines
mm/memcontrol.c +1
-1@@ -5360,7 +5362,7 @@ static int mem_cgroup_move_account(struct page *page, page->mem_cgroup = to;#ifdef CONFIG_TRANSPARENT_HUGEPAGE -if (compound) { +if (compound && split) {spin_lock (&to->deferred_split_queue.split_queue_lock);if (list_empty (page_deferred_list (page))) { list_add_tail(page_deferred_list(page),I think that's a good point, especially considering that the current code
Quoted 8 lines
appears to unconditionally place any compound page on the deferred split queue of the destination memcg. The correct list that it should appear on, I believe, depends on whether the pmd has been split for the process being moved: note the
MC_TARGET_PAGEcaveat inmem_cgroup_move_charge_pte_range()that does not move the charge for compound pages with split pmds. So whenmem_cgroup_move_account()is called with compound == true, we're moving the charge of the entire compound page: why would it appear on that memcg's deferred split queue?Quoted 4 lines
I believe Kirill asked how do we know that the page should be actually added to the deferred list just from the
list_emptycheck. In other words what if the page hasn't been split at all?Right, and I don't think that it necessarily is and the second conditional in Wei's patch will always succeed unless we have raced. That patch is for a lock concern but I think Kirill's question has uncovered something more interesting.
Kirill S would definitely be best to answer Kirill T's question, but from my understanding when
mem_cgroup_move_account()is called with compound == true that we always have an intact pmd (we never migrate partial page charges for pages on the deferred split queue with the current charge migration implementation) and thus the underlying page is not eligible to be split and shouldn't be on the deferred split queue.
Quoted 4 lines
In other words, a page being on the deferred split queue for a memcg should only happen when it is charged to that memcg. (This wasn't the case when we only had per-node split queues.) I think that's currently broken in
mem_cgroup_move_account()before Wei's patch.
Right. It's broken indeed.
We are dealing with anon page here. And it cannot be on deferred list as long as it's mapped with PMD. We cannot get compound == true && !list_empty() on the (first) enter to the function. Any PMD-mapped page will be put onto deferred by the function. This is wrong.
The fix is not obvious.
This comment got in mem_cgroup_move_charge_pte_range() my attention:
/*
* We can have a part of the split pmd here. Moving it
* can be done but it would be too convoluted so simply
* ignore such a partial THP and keep it in original
* memcg. There should be somebody mapping the head. */
That's exactly the case we care about: PTE-mapped THP that has to be split under load. We don't move charge of them between memcgs and therefore we should not move the page to different memcg.
I guess this will do the trick :P
mm/memcontrol.c +0 -18
@@ -5359,14 +5359,6 @@ static int mem_cgroup_move_account(struct page *page,__mod_lruvec_state (to_vec,NR_WRITEBACK , nr_pages); } -#ifdef CONFIG_TRANSPARENT_HUGEPAGE -if (compound && !list_empty (page_deferred_list (page))) { -spin_lock (&from->deferred_split_queue.split_queue_lock); -list_del_init (page_deferred_list (page)); - from->deferred_split_queue.split_queue_len--; -spin_unlock (&from->deferred_split_queue.split_queue_lock); - } -#endif /* * It is safe to change page->mem_cgroup here because the page * is referenced, charged, and isolated - we can't race with
@@ -5376,16 +5368,6 @@ static int mem_cgroup_move_account(struct page *page,/* caller should have done css_get */ page->mem_cgroup = to; -#ifdef CONFIG_TRANSPARENT_HUGEPAGE -if (compound &&list_empty (page_deferred_list (page))) { -spin_lock (&to->deferred_split_queue.split_queue_lock); -list_add_tail (page_deferred_list (page), - &to->deferred_split_queue.split_queue); - to->deferred_split_queue.split_queue_len++; -spin_unlock (&to->deferred_split_queue.split_queue_lock); - } -#endif -spin_unlock_irqrestore (&from->move_lock, flags); ret =0 ;
Signature · 1 line
Kirill A. Shutemov
Quoted 67 lines · Kirill A. Shutemov
Quoted 12 lines
Right, and I don't think that it necessarily is and the second conditional in Wei's patch will always succeed unless we have raced. That patch is for a lock concern but I think Kirill's question has uncovered something more interesting.
Kirill S would definitely be best to answer Kirill T's question, but from my understanding when
mem_cgroup_move_account()is called with compound == true that we always have an intact pmd (we never migrate partial page charges for pages on the deferred split queue with the current charge migration implementation) and thus the underlying page is not eligible to be split and shouldn't be on the deferred split queue.Quoted 4 lines
In other words, a page being on the deferred split queue for a memcg should only happen when it is charged to that memcg. (This wasn't the case when we only had per-node split queues.) I think that's currently broken in
mem_cgroup_move_account()before Wei's patch.Right. It's broken indeed.
We are dealing with anon page here. And it cannot be on deferred list as long as it's mapped with PMD. We cannot get compound == true && !list_empty() on the (first) enter to the function. Any PMD-mapped page will be put onto deferred by the function. This is wrong.
The fix is not obvious.
This comment got in
mem_cgroup_move_charge_pte_range()my attention:/*
* We can have a part of the split pmd here. Moving it
* can be done but it would be too convoluted so simply
* ignore such a partial THP and keep it in original
* memcg. There should be somebody mapping the head. */That's exactly the case we care about: PTE-mapped THP that has to be split under load. We don't move charge of them between memcgs and therefore we should not move the page to different memcg.
I guess this will do the trick :P
mm/memcontrol.c +0
-11@@ -5359,14 +5359,6 @@ static int mem_cgroup_move_account(struct page *page,__mod_lruvec_state (to_vec,NR_WRITEBACK , nr_pages); } -#ifdef CONFIG_TRANSPARENT_HUGEPAGE -if (compound && !list_empty (page_deferred_list (page))) { -spin_lock (&from->deferred_split_queue.split_queue_lock); -list_del_init (page_deferred_list (page)); - from->deferred_split_queue.split_queue_len--; -spin_unlock (&from->deferred_split_queue.split_queue_lock); - } -#endif /* * It is safe to change page->mem_cgroup here because the page * is referenced, charged, and isolated - we can't race with @@ -5376,16 +5368,6 @@ static int mem_cgroup_move_account(struct page *page,/* caller should have done css_get */ page->mem_cgroup = to; -#ifdef CONFIG_TRANSPARENT_HUGEPAGE -if (compound &&list_empty (page_deferred_list (page))) { -spin_lock (&to->deferred_split_queue.split_queue_lock);
mm/memcontrol.c static int mem_cgroup_move_account(struct page *page, · line 5373
Quoted 10 lines
mm/memcontrol.c +0
-7-list_add_tail (page_deferred_list (page), - &to->deferred_split_queue.split_queue); - to->deferred_split_queue.split_queue_len++; - spin_unlock(&to->deferred_split_queue.split_queue_lock); - } -#endif -spin_unlock_irqrestore (&from->move_lock, flags); ret =0 ;
Yeah, this is what I was thinking as well. When PageTransHuge(page) == true and there's a mapping pmd, the charge gets moved but the page shouldn't appear on any deferred split queue; when there isn't a mapped pmd, it should already be on a queue but the charge doesn't get moved so no change in which queue is needed.
There was no deferred split handling in mem_cgroup_move_account() needed for per-node deferred split queues either so this is purely an issue for commit 87eaceb3faa5 ("mm: thp: make deferred split shrinker memcg aware") so I think we need your patch and it should be annotated for stable 5.4+.
Quoted 103 lines · Kirill A. Shutemov
Quoted 102 lines · David Rientjes
Quoted 84 lines · Michal Hocko
Quoted 67 lines · David Rientjes
Quoted 65 lines · Kirill Tkhai
Quoted 28 lines
mm/memcontrol.c +6
-5@@ -5360,10 +5360,12 @@ static int mem_cgroup_move_account(struct page *page, }#ifdef CONFIG_TRANSPARENT_HUGEPAGE -if (compound && !list_empty (page_deferred_list (page))) { +if (compound) {spin_lock (&from->deferred_split_queue.split_queue_lock); -list_del_init (page_deferred_list (page)); - from->deferred_split_queue.split_queue_len--; +if (!list_empty (page_deferred_list (page))) { +list_del_init (page_deferred_list (page)); + from->deferred_split_queue.split_queue_len--; + }spin_unlock (&from->deferred_split_queue.split_queue_lock); }#endif @@ -5377,11 +5379,13 @@ static int mem_cgroup_move_account(struct page *page, page->mem_cgroup = to;#ifdef CONFIG_TRANSPARENT_HUGEPAGE -if (compound &&list_empty (page_deferred_list (page))) { +if (compound) {spin_lock (&to->deferred_split_queue.split_queue_lock); - list_add_tail(page_deferred_list(page),mm/memcontrol.c static int mem_cgroup_move_account(struct page *page, · line 5391
Quoted 10 lines
mm/memcontrol.c +5
-2- &to->deferred_split_queue.split_queue); - to->deferred_split_queue.split_queue_len++; +if (list_empty (page_deferred_list (page))) { +list_add_tail (page_deferred_list (page), + &to->deferred_split_queue.split_queue); + to->deferred_split_queue.split_queue_len++; + } spin_unlock(&to->deferred_split_queue.split_queue_lock); }#endif The patch looks OK for me. But there is another question. I forget, why we unconditionally add a page with empty deferred list to
deferred_split_queue. Shouldn't we also check that it was initially in the list? Something like:mm/memcontrol.c +2
-0@@ -5289,6 +5289,7 @@ static int mem_cgroup_move_account(struct page *page,struct pglist_data *pgdat;unsigned long flags;unsigned int nr_pages = compound ?hpage_nr_pages (page) :1 ; +bool split = false;int ret;bool anon;@@ -5346,6 +5347,7 @@ static int mem_cgroup_move_account(struct page *page,if (!list_empty (page_deferred_list (page))) {list_del_init (page_deferred_list (page)); from->deferred_split_queue.split_queue_len--; + split = true; }spin_unlock (&from->deferred_split_queue.split_queue_lock); }@@ -5360,7 +5362,7 @@ static int mem_cgroup_move_account(struct page *page, page->mem_cgroup = to;Quoted 16 lines
mm/memcontrol.c static int mem_cgroup_move_account(struct page *page, · line 5368
Quoted 7 lines
mm/memcontrol.c +1
-1#ifdef CONFIG_TRANSPARENT_HUGEPAGE -if (compound) { +if (compound && split) {spin_lock (&to->deferred_split_queue.split_queue_lock);if (list_empty (page_deferred_list (page))) { list_add_tail(page_deferred_list(page),I think that's a good point, especially considering that the current code appears to unconditionally place any compound page on the deferred split queue of the destination memcg. The correct list that it should appear on, I believe, depends on whether the pmd has been split for the process being moved: note the
MC_TARGET_PAGEcaveat inmem_cgroup_move_charge_pte_range()that does not move the charge for compound pages with split pmds. So whenmem_cgroup_move_account()is called with compound == true, we're moving the charge of the entire compound page: why would it appear on that memcg's deferred split queue?Quoted 4 lines
I believe Kirill asked how do we know that the page should be actually added to the deferred list just from the
list_emptycheck. In other words what if the page hasn't been split at all?Right, and I don't think that it necessarily is and the second conditional in Wei's patch will always succeed unless we have raced. That patch is for a lock concern but I think Kirill's question has uncovered something more interesting.
Kirill S would definitely be best to answer Kirill T's question, but from my understanding when
mem_cgroup_move_account()is called with compound == true that we always have an intact pmd (we never migrate partial page charges for pages on the deferred split queue with the current charge migration implementation) and thus the underlying page is not eligible to be split and shouldn't be on the deferred split queue.
Quoted 5 lines
Quoted 4 lines
In other words, a page being on the deferred split queue for a memcg should only happen when it is charged to that memcg. (This wasn't the case when we only had per-node split queues.) I think that's currently broken in
mem_cgroup_move_account()before Wei's patch.Right. It's broken indeed.
Hmm... Yes, definitely. I wasn't realized this at the first place.
Quoted 17 lines
We are dealing with anon page here. And it cannot be on deferred list as long as it's mapped with PMD. We cannot get compound == true && !list_empty() on the (first) enter to the function. Any PMD-mapped page will be put onto deferred by the function. This is wrong.
The fix is not obvious.
This comment got in
mem_cgroup_move_charge_pte_range()my attention:/*
* We can have a part of the split pmd here. Moving it
* can be done but it would be too convoluted so simply
* ignore such a partial THP and keep it in original
* memcg. There should be somebody mapping the head. */
Quoted 5 lines
That's exactly the case we care about: PTE-mapped THP that has to be split under load. We don't move charge of them between memcgs and therefore we should not move the page to different memcg.
I guess this will do the trick :P
It seems correct to me. In addition, memcg move charge just move PMD mapped THP, the THP should be never on the deferred split queue of "from" if it is PMD mapped, so actually we don't have to move it to the deferred split queue of "to".
Quoted 37 lines
mm/memcontrol.c +0
-18@@ -5359,14 +5359,6 @@ static int mem_cgroup_move_account(struct page *page,__mod_lruvec_state (to_vec,NR_WRITEBACK , nr_pages); } -#ifdef CONFIG_TRANSPARENT_HUGEPAGE -if (compound && !list_empty (page_deferred_list (page))) { -spin_lock (&from->deferred_split_queue.split_queue_lock); -list_del_init (page_deferred_list (page)); - from->deferred_split_queue.split_queue_len--; -spin_unlock (&from->deferred_split_queue.split_queue_lock); - } -#endif /* * It is safe to change page->mem_cgroup here because the page * is referenced, charged, and isolated - we can't race with @@ -5376,16 +5368,6 @@ static int mem_cgroup_move_account(struct page *page,/* caller should have done css_get */ page->mem_cgroup = to; -#ifdef CONFIG_TRANSPARENT_HUGEPAGE -if (compound &&list_empty (page_deferred_list (page))) { -spin_lock (&to->deferred_split_queue.split_queue_lock); -list_add_tail (page_deferred_list (page), - &to->deferred_split_queue.split_queue); - to->deferred_split_queue.split_queue_len++; -spin_unlock (&to->deferred_split_queue.split_queue_lock); - } -#endif -spin_unlock_irqrestore (&from->move_lock, flags); ret =0 ;
Quoted 131 lines · Yang Shi
Quoted 99 lines · Kirill A. Shutemov
Quoted 98 lines · David Rientjes
Quoted 80 lines · Michal Hocko
Quoted 71 lines · David Rientjes
Quoted 60 lines · Kirill Tkhai
Quoted 26 lines
mm/memcontrol.c +6
-5@@ -5360,10 +5360,12 @@ static int mem_cgroup_move_account(struct page *page, }#ifdef CONFIG_TRANSPARENT_HUGEPAGE -if (compound && !list_empty (page_deferred_list (page))) { +if (compound) {spin_lock (&from->deferred_split_queue.split_queue_lock); -list_del_init (page_deferred_list (page)); - from->deferred_split_queue.split_queue_len--; +if (!list_empty (page_deferred_list (page))) { +list_del_init (page_deferred_list (page)); + from->deferred_split_queue.split_queue_len--; + }spin_unlock (&from->deferred_split_queue.split_queue_lock); }#endif @@ -5377,11 +5379,13 @@ static int mem_cgroup_move_account(struct page *page, page->mem_cgroup = to;#ifdef CONFIG_TRANSPARENT_HUGEPAGE -if (compound &&list_empty (page_deferred_list (page))) { +if (compound) {spin_lock (&to->deferred_split_queue.split_queue_lock); - list_add_tail(page_deferred_list(page),mm/memcontrol.c static int mem_cgroup_move_account(struct page *page, · line 5390
Quoted 10 lines
mm/memcontrol.c +5
-2- &to->deferred_split_queue.split_queue); - to->deferred_split_queue.split_queue_len++; +if (list_empty (page_deferred_list (page))) { +list_add_tail (page_deferred_list (page), + &to->deferred_split_queue.split_queue); + to->deferred_split_queue.split_queue_len++; + } spin_unlock(&to->deferred_split_queue.split_queue_lock); }#endif The patch looks OK for me. But there is another question. I forget, why we unconditionally add a page with empty deferred list to
deferred_split_queue. Shouldn't we also check that it was initially in the list? Something like:mm/memcontrol.c +2
-0@@ -5289,6 +5289,7 @@ static int mem_cgroup_move_account(struct page *page,struct pglist_data *pgdat;unsigned long flags;unsigned int nr_pages = compound ?hpage_nr_pages (page) :1 ; +bool split = false;int ret;bool anon;@@ -5346,6 +5347,7 @@ static int mem_cgroup_move_account(struct page *page,if (!list_empty (page_deferred_list (page))) {list_del_init (page_deferred_list (page)); from->deferred_split_queue.split_queue_len--; + split = true; }spin_unlock (&from->deferred_split_queue.split_queue_lock); }mm/memcontrol.c static int mem_cgroup_move_account(struct page *page, · line 5368
Quoted 8 lines
mm/memcontrol.c +1
-1@@ -5360,7 +5362,7 @@ static int mem_cgroup_move_account(struct page *page, page->mem_cgroup = to;#ifdef CONFIG_TRANSPARENT_HUGEPAGE -if (compound) { +if (compound && split) {spin_lock (&to->deferred_split_queue.split_queue_lock);if (list_empty (page_deferred_list (page))) { list_add_tail(page_deferred_list(page),I think that's a good point, especially considering that the current code
Quoted 8 lines
appears to unconditionally place any compound page on the deferred split queue of the destination memcg. The correct list that it should appear on, I believe, depends on whether the pmd has been split for the process being moved: note the
MC_TARGET_PAGEcaveat inmem_cgroup_move_charge_pte_range()that does not move the charge for compound pages with split pmds. So whenmem_cgroup_move_account()is called with compound == true, we're moving the charge of the entire compound page: why would it appear on that memcg's deferred split queue?Quoted 4 lines
I believe Kirill asked how do we know that the page should be actually added to the deferred list just from the
list_emptycheck. In other words what if the page hasn't been split at all?Right, and I don't think that it necessarily is and the second conditional in Wei's patch will always succeed unless we have raced. That patch is for a lock concern but I think Kirill's question has uncovered something more interesting.
Kirill S would definitely be best to answer Kirill T's question, but from my understanding when
mem_cgroup_move_account()is called with compound == true that we always have an intact pmd (we never migrate partial page charges for pages on the deferred split queue with the current charge migration implementation) and thus the underlying page is not eligible to be split and shouldn't be on the deferred split queue.Quoted 5 lines
Quoted 4 lines
In other words, a page being on the deferred split queue for a memcg should only happen when it is charged to that memcg. (This wasn't the case when we only had per-node split queues.) I think that's currently broken in
mem_cgroup_move_account()before Wei's patch.Right. It's broken indeed.
Hmm... Yes, definitely. I wasn't realized this at the first place.
Quoted 21 lines
We are dealing with anon page here. And it cannot be on deferred list as long as it's mapped with PMD. We cannot get compound == true && !list_empty() on the (first) enter to the function. Any PMD-mapped page will be put onto deferred by the function. This is wrong.
The fix is not obvious.
This comment got in
mem_cgroup_move_charge_pte_range()my attention:/*
* We can have a part of the split pmd here. Moving it
* can be done but it would be too convoluted so simply
* ignore such a partial THP and keep it in original
* memcg. There should be somebody mapping the head. */That's exactly the case we care about: PTE-mapped THP that has to be split under load. We don't move charge of them between memcgs and therefore we should not move the page to different memcg.
Quoted 7 lines
Quoted 1 line
I guess this will do the trick :P
It seems correct to me. In addition, memcg move charge just move PMD mapped THP, the THP should be never on the deferred split queue of "from" if it is PMD mapped, so actually we don't have to move it to the deferred split queue of "to".
Well, I got the point.
Since Kirill S found the correct solution, should I prepare v3 or Kirill will send it?
Quoted 34 lines
Quoted 34 lines
mm/memcontrol.c +0
-18@@ -5359,14 +5359,6 @@ static int mem_cgroup_move_account(struct page *page,__mod_lruvec_state (to_vec,NR_WRITEBACK , nr_pages); } -#ifdef CONFIG_TRANSPARENT_HUGEPAGE -if (compound && !list_empty (page_deferred_list (page))) { -spin_lock (&from->deferred_split_queue.split_queue_lock); -list_del_init (page_deferred_list (page)); - from->deferred_split_queue.split_queue_len--; -spin_unlock (&from->deferred_split_queue.split_queue_lock); - } -#endif /* * It is safe to change page->mem_cgroup here because the page * is referenced, charged, and isolated - we can't race with @@ -5376,16 +5368,6 @@ static int mem_cgroup_move_account(struct page *page,/* caller should have done css_get */ page->mem_cgroup = to; -#ifdef CONFIG_TRANSPARENT_HUGEPAGE -if (compound &&list_empty (page_deferred_list (page))) { -spin_lock (&to->deferred_split_queue.split_queue_lock); -list_add_tail (page_deferred_list (page), - &to->deferred_split_queue.split_queue); - to->deferred_split_queue.split_queue_len++; -spin_unlock (&to->deferred_split_queue.split_queue_lock); - } -#endif -spin_unlock_irqrestore (&from->move_lock, flags); ret =0 ;
Signature · 2 lines
Wei Yang Help you, Help me
Quoted 140 lines · Wei Yang
Quoted 131 lines · Yang Shi
Quoted 99 lines · Kirill A. Shutemov
Quoted 98 lines · David Rientjes
Quoted 80 lines · Michal Hocko
Quoted 71 lines · David Rientjes
Quoted 60 lines · Kirill Tkhai
Quoted 26 lines
mm/memcontrol.c +6
-5@@ -5360,10 +5360,12 @@ static int mem_cgroup_move_account(struct page *page, }#ifdef CONFIG_TRANSPARENT_HUGEPAGE -if (compound && !list_empty (page_deferred_list (page))) { +if (compound) {spin_lock (&from->deferred_split_queue.split_queue_lock); -list_del_init (page_deferred_list (page)); - from->deferred_split_queue.split_queue_len--; +if (!list_empty (page_deferred_list (page))) { +list_del_init (page_deferred_list (page)); + from->deferred_split_queue.split_queue_len--; + }spin_unlock (&from->deferred_split_queue.split_queue_lock); }#endif @@ -5377,11 +5379,13 @@ static int mem_cgroup_move_account(struct page *page, page->mem_cgroup = to;#ifdef CONFIG_TRANSPARENT_HUGEPAGE -if (compound &&list_empty (page_deferred_list (page))) { +if (compound) {spin_lock (&to->deferred_split_queue.split_queue_lock); - list_add_tail(page_deferred_list(page),mm/memcontrol.c static int mem_cgroup_move_account(struct page *page, · line 5390
Quoted 10 lines
mm/memcontrol.c +5
-2- &to->deferred_split_queue.split_queue); - to->deferred_split_queue.split_queue_len++; +if (list_empty (page_deferred_list (page))) { +list_add_tail (page_deferred_list (page), + &to->deferred_split_queue.split_queue); + to->deferred_split_queue.split_queue_len++; + } spin_unlock(&to->deferred_split_queue.split_queue_lock); }#endif The patch looks OK for me. But there is another question. I forget, why we unconditionally add a page with empty deferred list to
deferred_split_queue. Shouldn't we also check that it was initially in the list? Something like:mm/memcontrol.c +2
-0@@ -5289,6 +5289,7 @@ static int mem_cgroup_move_account(struct page *page,struct pglist_data *pgdat;unsigned long flags;unsigned int nr_pages = compound ?hpage_nr_pages (page) :1 ; +bool split = false;int ret;bool anon;@@ -5346,6 +5347,7 @@ static int mem_cgroup_move_account(struct page *page,if (!list_empty (page_deferred_list (page))) {list_del_init (page_deferred_list (page)); from->deferred_split_queue.split_queue_len--; + split = true; }spin_unlock (&from->deferred_split_queue.split_queue_lock); }mm/memcontrol.c static int mem_cgroup_move_account(struct page *page, · line 5368
Quoted 8 lines
mm/memcontrol.c +1
-1@@ -5360,7 +5362,7 @@ static int mem_cgroup_move_account(struct page *page, page->mem_cgroup = to;#ifdef CONFIG_TRANSPARENT_HUGEPAGE -if (compound) { +if (compound && split) {spin_lock (&to->deferred_split_queue.split_queue_lock);if (list_empty (page_deferred_list (page))) { list_add_tail(page_deferred_list(page),I think that's a good point, especially considering that the current code
Quoted 8 lines
appears to unconditionally place any compound page on the deferred split queue of the destination memcg. The correct list that it should appear on, I believe, depends on whether the pmd has been split for the process being moved: note the
MC_TARGET_PAGEcaveat inmem_cgroup_move_charge_pte_range()that does not move the charge for compound pages with split pmds. So whenmem_cgroup_move_account()is called with compound == true, we're moving the charge of the entire compound page: why would it appear on that memcg's deferred split queue?Quoted 4 lines
I believe Kirill asked how do we know that the page should be actually added to the deferred list just from the
list_emptycheck. In other words what if the page hasn't been split at all?Right, and I don't think that it necessarily is and the second conditional in Wei's patch will always succeed unless we have raced. That patch is for a lock concern but I think Kirill's question has uncovered something more interesting.
Kirill S would definitely be best to answer Kirill T's question, but from my understanding when
mem_cgroup_move_account()is called with compound == true that we always have an intact pmd (we never migrate partial page charges for pages on the deferred split queue with the current charge migration implementation) and thus the underlying page is not eligible to be split and shouldn't be on the deferred split queue.Quoted 5 lines
Quoted 4 lines
In other words, a page being on the deferred split queue for a memcg should only happen when it is charged to that memcg. (This wasn't the case when we only had per-node split queues.) I think that's currently broken in
mem_cgroup_move_account()before Wei's patch.Right. It's broken indeed.
Hmm... Yes, definitely. I wasn't realized this at the first place.
Quoted 21 lines
We are dealing with anon page here. And it cannot be on deferred list as long as it's mapped with PMD. We cannot get compound == true && !list_empty() on the (first) enter to the function. Any PMD-mapped page will be put onto deferred by the function. This is wrong.
The fix is not obvious.
This comment got in
mem_cgroup_move_charge_pte_range()my attention:/*
* We can have a part of the split pmd here. Moving it
* can be done but it would be too convoluted so simply
* ignore such a partial THP and keep it in original
* memcg. There should be somebody mapping the head. */That's exactly the case we care about: PTE-mapped THP that has to be split under load. We don't move charge of them between memcgs and therefore we should not move the page to different memcg.
Quoted 7 lines
Quoted 1 line
I guess this will do the trick :P
It seems correct to me. In addition, memcg move charge just move PMD mapped THP, the THP should be never on the deferred split queue of "from" if it is PMD mapped, so actually we don't have to move it to the deferred split queue of "to".
Quoted 4 lines
Well, I got the point.
Since Kirill S found the correct solution, should I prepare v3 or Kirill will send it?
Go ahead. With my Suggested-by.
Signature · 1 line
Kirill A. Shutemov