* [Patch v3] mm: thp: grab the lock before manipulation defer list @ 2020-01-16 1:31 Wei Yang 2020-01-16 9:35 ` Kirill Tkhai 0 siblings, 1 reply; 15+ messages in thread From: Wei Yang @ 2020-01-16 1:31 UTC (permalink / raw) To: hannes, mhocko, vdavydov.dev, akpm, ktkhai, kirill.shutemov, yang.shi Cc: cgroups, linux-mm, linux-kernel, alexander.duyck, rientjes, Wei Yang, stable 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. Fixes: 87eaceb3faa5 ("mm: thp: make deferred split shrinker memcg aware") Signed-off-by: Wei Yang <richardw.yang@linux.intel.com> Cc: <stable@vger.kernel.org> [5.4+] --- 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(-) diff --git a/mm/memcontrol.c b/mm/memcontrol.c index c5b5f74cfd4d..6450bbe394e2 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -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 -- 2.17.1 ^ permalink raw reply related [flat|nested] 15+ messages in thread
* Re: [Patch v3] mm: thp: grab the lock before manipulation defer list 2020-01-16 1:31 [Patch v3] mm: thp: grab the lock before manipulation defer list Wei Yang @ 2020-01-16 9:35 ` Kirill Tkhai 2020-01-16 22:01 ` David Rientjes 0 siblings, 1 reply; 15+ messages in thread From: Kirill Tkhai @ 2020-01-16 9:35 UTC (permalink / raw) To: Wei Yang, hannes, mhocko, vdavydov.dev, akpm, kirill.shutemov, yang.shi Cc: cgroups, linux-mm, linux-kernel, alexander.duyck, rientjes, stable On 16.01.2020 04:31, Wei Yang wrote: > 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. > > Fixes: 87eaceb3faa5 ("mm: thp: make deferred split shrinker memcg aware") > > Signed-off-by: Wei Yang <richardw.yang@linux.intel.com> > Cc: <stable@vger.kernel.org> [5.4+] > > --- > 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(-) > > diff --git a/mm/memcontrol.c b/mm/memcontrol.c > index c5b5f74cfd4d..6450bbe394e2 100644 > --- a/mm/memcontrol.c > +++ b/mm/memcontrol.c > @@ -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 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: diff --git a/mm/memcontrol.c b/mm/memcontrol.c index d4394ae4e5be..0be0136adaa6 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -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), ^ permalink raw reply related [flat|nested] 15+ messages in thread
* Re: [Patch v3] mm: thp: grab the lock before manipulation defer list 2020-01-16 9:35 ` Kirill Tkhai @ 2020-01-16 22:01 ` David Rientjes 2020-01-17 0:47 ` Wei Yang 2020-01-17 9:10 ` Michal Hocko 0 siblings, 2 replies; 15+ messages in thread From: David Rientjes @ 2020-01-16 22:01 UTC (permalink / raw) To: Kirill Tkhai Cc: Wei Yang, hannes, mhocko, vdavydov.dev, akpm, kirill.shutemov, yang.shi, cgroups, linux-mm, linux-kernel, alexander.duyck, stable On Thu, 16 Jan 2020, Kirill Tkhai wrote: > > diff --git a/mm/memcontrol.c b/mm/memcontrol.c > > index c5b5f74cfd4d..6450bbe394e2 100644 > > --- a/mm/memcontrol.c > > +++ b/mm/memcontrol.c > > @@ -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 > > 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: > > diff --git a/mm/memcontrol.c b/mm/memcontrol.c > index d4394ae4e5be..0be0136adaa6 100644 > --- a/mm/memcontrol.c > +++ b/mm/memcontrol.c > @@ -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), > 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? ^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: [Patch v3] mm: thp: grab the lock before manipulation defer list 2020-01-16 22:01 ` David Rientjes @ 2020-01-17 0:47 ` Wei Yang 2020-01-17 9:10 ` Michal Hocko 1 sibling, 0 replies; 15+ messages in thread From: Wei Yang @ 2020-01-17 0:47 UTC (permalink / raw) To: David Rientjes Cc: Kirill Tkhai, Wei Yang, hannes, mhocko, vdavydov.dev, akpm, kirill.shutemov, yang.shi, cgroups, linux-mm, linux-kernel, alexander.duyck, stable On Thu, Jan 16, 2020 at 02:01:59PM -0800, David Rientjes wrote: >On Thu, 16 Jan 2020, Kirill Tkhai wrote: > >> > diff --git a/mm/memcontrol.c b/mm/memcontrol.c >> > index c5b5f74cfd4d..6450bbe394e2 100644 >> > --- a/mm/memcontrol.c >> > +++ b/mm/memcontrol.c >> > @@ -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 >> >> 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: >> >> diff --git a/mm/memcontrol.c b/mm/memcontrol.c >> index d4394ae4e5be..0be0136adaa6 100644 >> --- a/mm/memcontrol.c >> +++ b/mm/memcontrol.c >> @@ -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), >> > >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? 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. :-( -- Wei Yang Help you, Help me ^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: [Patch v3] mm: thp: grab the lock before manipulation defer list 2020-01-16 22:01 ` David Rientjes 2020-01-17 0:47 ` Wei Yang @ 2020-01-17 9:10 ` Michal Hocko 2020-01-17 9:26 ` Kirill Tkhai 2020-01-17 9:31 ` David Rientjes 1 sibling, 2 replies; 15+ messages in thread From: Michal Hocko @ 2020-01-17 9:10 UTC (permalink / raw) To: David Rientjes Cc: Kirill Tkhai, Wei Yang, hannes, vdavydov.dev, akpm, kirill.shutemov, yang.shi, cgroups, linux-mm, linux-kernel, alexander.duyck, stable On Thu 16-01-20 14:01:59, David Rientjes wrote: > On Thu, 16 Jan 2020, Kirill Tkhai wrote: > > > > diff --git a/mm/memcontrol.c b/mm/memcontrol.c > > > index c5b5f74cfd4d..6450bbe394e2 100644 > > > --- a/mm/memcontrol.c > > > +++ b/mm/memcontrol.c > > > @@ -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 > > > > 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: > > > > diff --git a/mm/memcontrol.c b/mm/memcontrol.c > > index d4394ae4e5be..0be0136adaa6 100644 > > --- a/mm/memcontrol.c > > +++ b/mm/memcontrol.c > > @@ -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), > > > > 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? 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? -- Michal Hocko SUSE Labs ^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: [Patch v3] mm: thp: grab the lock before manipulation defer list 2020-01-17 9:10 ` Michal Hocko @ 2020-01-17 9:26 ` Kirill Tkhai 2020-01-17 9:32 ` David Rientjes 2020-01-17 9:31 ` David Rientjes 1 sibling, 1 reply; 15+ messages in thread From: Kirill Tkhai @ 2020-01-17 9:26 UTC (permalink / raw) To: Michal Hocko, David Rientjes Cc: Wei Yang, hannes, vdavydov.dev, akpm, kirill.shutemov, yang.shi, cgroups, linux-mm, linux-kernel, alexander.duyck, stable On 17.01.2020 12:10, Michal Hocko wrote: > On Thu 16-01-20 14:01:59, David Rientjes wrote: >> On Thu, 16 Jan 2020, Kirill Tkhai wrote: >> >>>> diff --git a/mm/memcontrol.c b/mm/memcontrol.c >>>> index c5b5f74cfd4d..6450bbe394e2 100644 >>>> --- a/mm/memcontrol.c >>>> +++ b/mm/memcontrol.c >>>> @@ -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), >>>> - &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: >>> >>> diff --git a/mm/memcontrol.c b/mm/memcontrol.c >>> index d4394ae4e5be..0be0136adaa6 100644 >>> --- a/mm/memcontrol.c >>> +++ b/mm/memcontrol.c >>> @@ -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), >>> >> >> 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? > > 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? 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... ^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: [Patch v3] mm: thp: grab the lock before manipulation defer list 2020-01-17 9:26 ` Kirill Tkhai @ 2020-01-17 9:32 ` David Rientjes 2020-01-17 9:42 ` Kirill Tkhai 0 siblings, 1 reply; 15+ messages in thread From: David Rientjes @ 2020-01-17 9:32 UTC (permalink / raw) To: Kirill Tkhai Cc: Michal Hocko, Wei Yang, hannes, vdavydov.dev, akpm, kirill.shutemov, yang.shi, cgroups, linux-mm, linux-kernel, alexander.duyck, stable On Fri, 17 Jan 2020, Kirill Tkhai wrote: > >> 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? > > > > 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? > > 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? ^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: [Patch v3] mm: thp: grab the lock before manipulation defer list 2020-01-17 9:32 ` David Rientjes @ 2020-01-17 9:42 ` Kirill Tkhai 2020-01-17 11:59 ` Michal Hocko 0 siblings, 1 reply; 15+ messages in thread From: Kirill Tkhai @ 2020-01-17 9:42 UTC (permalink / raw) To: David Rientjes Cc: Michal Hocko, Wei Yang, hannes, vdavydov.dev, akpm, kirill.shutemov, yang.shi, cgroups, linux-mm, linux-kernel, alexander.duyck, stable On 17.01.2020 12:32, David Rientjes wrote: > On Fri, 17 Jan 2020, Kirill Tkhai wrote: > >>>> 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? >>> >>> 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? >> >> 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? 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? ^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: [Patch v3] mm: thp: grab the lock before manipulation defer list 2020-01-17 9:42 ` Kirill Tkhai @ 2020-01-17 11:59 ` Michal Hocko 0 siblings, 0 replies; 15+ messages in thread From: Michal Hocko @ 2020-01-17 11:59 UTC (permalink / raw) To: Kirill Tkhai, g, kirill.shutemov Cc: David Rientjes, Wei Yang, hannes, vdavydov.dev, akpm, yang.shi, cgroups, linux-mm, linux-kernel, alexander.duyck, stable On Fri 17-01-20 12:42:05, Kirill Tkhai wrote: > On 17.01.2020 12:32, David Rientjes wrote: > > On Fri, 17 Jan 2020, Kirill Tkhai wrote: > > > >>>> 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? > >>> > >>> 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? > >> > >> 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? > > 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? -- Michal Hocko SUSE Labs ^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: [Patch v3] mm: thp: grab the lock before manipulation defer list 2020-01-17 9:10 ` Michal Hocko 2020-01-17 9:26 ` Kirill Tkhai @ 2020-01-17 9:31 ` David Rientjes 2020-01-17 15:38 ` Kirill A. Shutemov 1 sibling, 1 reply; 15+ messages in thread From: David Rientjes @ 2020-01-17 9:31 UTC (permalink / raw) To: Michal Hocko Cc: Kirill Tkhai, Wei Yang, hannes, vdavydov.dev, akpm, kirill.shutemov, yang.shi, cgroups, linux-mm, linux-kernel, alexander.duyck, stable On Fri, 17 Jan 2020, Michal Hocko wrote: > On Thu 16-01-20 14:01:59, David Rientjes wrote: > > On Thu, 16 Jan 2020, Kirill Tkhai wrote: > > > > > > diff --git a/mm/memcontrol.c b/mm/memcontrol.c > > > > index c5b5f74cfd4d..6450bbe394e2 100644 > > > > --- a/mm/memcontrol.c > > > > +++ b/mm/memcontrol.c > > > > @@ -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 > > > > > > 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: > > > > > > diff --git a/mm/memcontrol.c b/mm/memcontrol.c > > > index d4394ae4e5be..0be0136adaa6 100644 > > > --- a/mm/memcontrol.c > > > +++ b/mm/memcontrol.c > > > @@ -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), > > > > > > > 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? > > 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? > 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. ^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: [Patch v3] mm: thp: grab the lock before manipulation defer list 2020-01-17 9:31 ` David Rientjes @ 2020-01-17 15:38 ` Kirill A. Shutemov 2020-01-17 19:11 ` David Rientjes 2020-01-17 19:17 ` Yang Shi 0 siblings, 2 replies; 15+ messages in thread From: Kirill A. Shutemov @ 2020-01-17 15:38 UTC (permalink / raw) To: David Rientjes Cc: Michal Hocko, Kirill Tkhai, Wei Yang, hannes, vdavydov.dev, akpm, kirill.shutemov, yang.shi, cgroups, linux-mm, linux-kernel, alexander.duyck, stable On Fri, Jan 17, 2020 at 01:31:50AM -0800, David Rientjes wrote: > On Fri, 17 Jan 2020, Michal Hocko wrote: > > > On Thu 16-01-20 14:01:59, David Rientjes wrote: > > > On Thu, 16 Jan 2020, Kirill Tkhai wrote: > > > > > > > > diff --git a/mm/memcontrol.c b/mm/memcontrol.c > > > > > index c5b5f74cfd4d..6450bbe394e2 100644 > > > > > --- a/mm/memcontrol.c > > > > > +++ b/mm/memcontrol.c > > > > > @@ -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 > > > > > > > > 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: > > > > > > > > diff --git a/mm/memcontrol.c b/mm/memcontrol.c > > > > index d4394ae4e5be..0be0136adaa6 100644 > > > > --- a/mm/memcontrol.c > > > > +++ b/mm/memcontrol.c > > > > @@ -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), > > > > > > > > > > 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? > > > > 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? > > > > 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. 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 diff --git a/mm/memcontrol.c b/mm/memcontrol.c index c5b5f74cfd4d..e87ee4c10f6e 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -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; -- Kirill A. Shutemov ^ permalink raw reply related [flat|nested] 15+ messages in thread
* Re: [Patch v3] mm: thp: grab the lock before manipulation defer list 2020-01-17 15:38 ` Kirill A. Shutemov @ 2020-01-17 19:11 ` David Rientjes 2020-01-17 19:17 ` Yang Shi 1 sibling, 0 replies; 15+ messages in thread From: David Rientjes @ 2020-01-17 19:11 UTC (permalink / raw) To: Kirill A. Shutemov Cc: Michal Hocko, Kirill Tkhai, Wei Yang, hannes, vdavydov.dev, akpm, kirill.shutemov, yang.shi, cgroups, linux-mm, linux-kernel, alexander.duyck, stable On Fri, 17 Jan 2020, Kirill A. Shutemov wrote: > > 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. > > 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 > > diff --git a/mm/memcontrol.c b/mm/memcontrol.c > index c5b5f74cfd4d..e87ee4c10f6e 100644 > --- a/mm/memcontrol.c > +++ b/mm/memcontrol.c > @@ -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; 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+. ^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: [Patch v3] mm: thp: grab the lock before manipulation defer list 2020-01-17 15:38 ` Kirill A. Shutemov 2020-01-17 19:11 ` David Rientjes @ 2020-01-17 19:17 ` Yang Shi 2020-01-17 22:18 ` Wei Yang 1 sibling, 1 reply; 15+ messages in thread From: Yang Shi @ 2020-01-17 19:17 UTC (permalink / raw) To: Kirill A. Shutemov, David Rientjes Cc: Michal Hocko, Kirill Tkhai, Wei Yang, hannes, vdavydov.dev, akpm, kirill.shutemov, cgroups, linux-mm, linux-kernel, alexander.duyck, stable On 1/17/20 7:38 AM, Kirill A. Shutemov wrote: > On Fri, Jan 17, 2020 at 01:31:50AM -0800, David Rientjes wrote: >> On Fri, 17 Jan 2020, Michal Hocko wrote: >> >>> On Thu 16-01-20 14:01:59, David Rientjes wrote: >>>> On Thu, 16 Jan 2020, Kirill Tkhai wrote: >>>> >>>>>> diff --git a/mm/memcontrol.c b/mm/memcontrol.c >>>>>> index c5b5f74cfd4d..6450bbe394e2 100644 >>>>>> --- a/mm/memcontrol.c >>>>>> +++ b/mm/memcontrol.c >>>>>> @@ -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 >>>>> 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: >>>>> >>>>> diff --git a/mm/memcontrol.c b/mm/memcontrol.c >>>>> index d4394ae4e5be..0be0136adaa6 100644 >>>>> --- a/mm/memcontrol.c >>>>> +++ b/mm/memcontrol.c >>>>> @@ -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), >>>>> >>>> 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? >>> 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? >>> >> 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. > Right. It's broken indeed. Hmm... Yes, definitely. I wasn't realized this at the first place. > > 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 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". > > diff --git a/mm/memcontrol.c b/mm/memcontrol.c > index c5b5f74cfd4d..e87ee4c10f6e 100644 > --- a/mm/memcontrol.c > +++ b/mm/memcontrol.c > @@ -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; ^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: [Patch v3] mm: thp: grab the lock before manipulation defer list 2020-01-17 19:17 ` Yang Shi @ 2020-01-17 22:18 ` Wei Yang 2020-01-17 22:57 ` Kirill A. Shutemov 0 siblings, 1 reply; 15+ messages in thread From: Wei Yang @ 2020-01-17 22:18 UTC (permalink / raw) To: Yang Shi Cc: Kirill A. Shutemov, David Rientjes, Michal Hocko, Kirill Tkhai, Wei Yang, hannes, vdavydov.dev, akpm, kirill.shutemov, cgroups, linux-mm, linux-kernel, alexander.duyck, stable On Fri, Jan 17, 2020 at 11:17:38AM -0800, Yang Shi wrote: > > >On 1/17/20 7:38 AM, Kirill A. Shutemov wrote: >> On Fri, Jan 17, 2020 at 01:31:50AM -0800, David Rientjes wrote: >> > On Fri, 17 Jan 2020, Michal Hocko wrote: >> > >> > > On Thu 16-01-20 14:01:59, David Rientjes wrote: >> > > > On Thu, 16 Jan 2020, Kirill Tkhai wrote: >> > > > >> > > > > > diff --git a/mm/memcontrol.c b/mm/memcontrol.c >> > > > > > index c5b5f74cfd4d..6450bbe394e2 100644 >> > > > > > --- a/mm/memcontrol.c >> > > > > > +++ b/mm/memcontrol.c >> > > > > > @@ -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 >> > > > > 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: >> > > > > >> > > > > diff --git a/mm/memcontrol.c b/mm/memcontrol.c >> > > > > index d4394ae4e5be..0be0136adaa6 100644 >> > > > > --- a/mm/memcontrol.c >> > > > > +++ b/mm/memcontrol.c >> > > > > @@ -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), >> > > > > >> > > > 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? >> > > 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? >> > > >> > 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. >> Right. It's broken indeed. > >Hmm... Yes, definitely. I wasn't realized this at the first place. > >> >> 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 > >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? >> >> diff --git a/mm/memcontrol.c b/mm/memcontrol.c >> index c5b5f74cfd4d..e87ee4c10f6e 100644 >> --- a/mm/memcontrol.c >> +++ b/mm/memcontrol.c >> @@ -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; -- Wei Yang Help you, Help me ^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: [Patch v3] mm: thp: grab the lock before manipulation defer list 2020-01-17 22:18 ` Wei Yang @ 2020-01-17 22:57 ` Kirill A. Shutemov 0 siblings, 0 replies; 15+ messages in thread From: Kirill A. Shutemov @ 2020-01-17 22:57 UTC (permalink / raw) To: Wei Yang Cc: Yang Shi, Kirill A. Shutemov, David Rientjes, Michal Hocko, Kirill Tkhai, hannes, vdavydov.dev, akpm, cgroups, linux-mm, linux-kernel, alexander.duyck, stable On Fri, Jan 17, 2020 at 10:18:59PM +0000, Wei Yang wrote: > On Fri, Jan 17, 2020 at 11:17:38AM -0800, Yang Shi wrote: > > > > > >On 1/17/20 7:38 AM, Kirill A. Shutemov wrote: > >> On Fri, Jan 17, 2020 at 01:31:50AM -0800, David Rientjes wrote: > >> > On Fri, 17 Jan 2020, Michal Hocko wrote: > >> > > >> > > On Thu 16-01-20 14:01:59, David Rientjes wrote: > >> > > > On Thu, 16 Jan 2020, Kirill Tkhai wrote: > >> > > > > >> > > > > > diff --git a/mm/memcontrol.c b/mm/memcontrol.c > >> > > > > > index c5b5f74cfd4d..6450bbe394e2 100644 > >> > > > > > --- a/mm/memcontrol.c > >> > > > > > +++ b/mm/memcontrol.c > >> > > > > > @@ -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 > >> > > > > 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: > >> > > > > > >> > > > > diff --git a/mm/memcontrol.c b/mm/memcontrol.c > >> > > > > index d4394ae4e5be..0be0136adaa6 100644 > >> > > > > --- a/mm/memcontrol.c > >> > > > > +++ b/mm/memcontrol.c > >> > > > > @@ -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), > >> > > > > > >> > > > 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? > >> > > 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? > >> > > > >> > 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. > >> Right. It's broken indeed. > > > >Hmm... Yes, definitely. I wasn't realized this at the first place. > > > >> > >> 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 > > > >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? Go ahead. With my Suggested-by. -- Kirill A. Shutemov ^ permalink raw reply [flat|nested] 15+ messages in thread
end of thread, other threads:[~2020-01-17 22:58 UTC | newest] Thread overview: 15+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2020-01-16 1:31 [Patch v3] mm: thp: grab the lock before manipulation defer list Wei Yang 2020-01-16 9:35 ` Kirill Tkhai 2020-01-16 22:01 ` David Rientjes 2020-01-17 0:47 ` Wei Yang 2020-01-17 9:10 ` Michal Hocko 2020-01-17 9:26 ` Kirill Tkhai 2020-01-17 9:32 ` David Rientjes 2020-01-17 9:42 ` Kirill Tkhai 2020-01-17 11:59 ` Michal Hocko 2020-01-17 9:31 ` David Rientjes 2020-01-17 15:38 ` Kirill A. Shutemov 2020-01-17 19:11 ` David Rientjes 2020-01-17 19:17 ` Yang Shi 2020-01-17 22:18 ` Wei Yang 2020-01-17 22:57 ` Kirill A. Shutemov
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).