Re: [PATCH RFC] mm/page_alloc: Fix pcp->count race between drain_pages_zone() vs __rmqueue_pcplist()

["Zhijian Li (Fujitsu)" <lizhijian@fujitsu.com>]

On 18/07/2024 22:02, David Hildenbrand wrote:
> On 18.07.24 13:16, Vlastimil Babka (SUSE) wrote:
>> On 7/16/24 9:39 AM, Li Zhijian wrote:
>>> It's expected that no page should be left in pcp_list after calling
>>> zone_pcp_disable() in offline_pages(). Previously, it's observed that
>>> offline_pages() gets stuck [1] due to some pages remaining in pcp_list.
>>>
>>> Cause:
>>> There is a race condition between drain_pages_zone() and __rmqueue_pcplist()
>>> involving the pcp->count variable. See below scenario:
>>>
>>>           CPU0                              CPU1
>>>      ----------------                    ---------------
>>>                                        spin_lock(&pcp->lock);
>>>                                        __rmqueue_pcplist() {
>>> zone_pcp_disable() {
>>>                                          /* list is empty */
>>>                                          if (list_empty(list)) {
>>>                                            /* add pages to pcp_list */
>>>                                            alloced = rmqueue_bulk()
>>>    mutex_lock(&pcp_batch_high_lock)
>>>    ...
>>>    __drain_all_pages() {
>>>      drain_pages_zone() {
>>>        /* read pcp->count, it's 0 here */
>>>        count = READ_ONCE(pcp->count)
>>>        /* 0 means nothing to drain */
>>>                                            /* update pcp->count */
>>>                                            pcp->count += alloced << order;
>>>        ...
>>>                                        ...
>>>                                        spin_unlock(&pcp->lock);
>>>
>>> In this case, after calling zone_pcp_disable() though, there are still some
>>> pages in pcp_list. And these pages in pcp_list are neither movable nor
>>> isolated, offline_pages() gets stuck as a result.
>>>
>>> Solution:
>>> Expand the scope of the pcp->lock to also protect pcp->count in
>>> drain_pages_zone(), ensuring no pages are left in the pcp list.
>>>
>>> [1] https://lore.kernel.org/linux-mm/6a07125f-e720-404c-b2f9-e55f3f166e85@fujitsu.com/
>>>
>>> Cc: David Hildenbrand <david@redhat.com>
>>> Reported-by: Yao Xingtao <yaoxt.fnst@fujitsu.com>
>>> Signed-off-by: Li Zhijian <lizhijian@fujitsu.com>
>>> ---
>>>   mm/page_alloc.c | 7 ++++---
>>>   1 file changed, 4 insertions(+), 3 deletions(-)
>>>
>>> diff --git a/mm/page_alloc.c b/mm/page_alloc.c
>>> index 9ecf99190ea2..1780df31d5f5 100644
>>> --- a/mm/page_alloc.c
>>> +++ b/mm/page_alloc.c
>>> @@ -2323,16 +2323,17 @@ void drain_zone_pages(struct zone *zone, struct per_cpu_pages *pcp)
>>>   static void drain_pages_zone(unsigned int cpu, struct zone *zone)
>>>   {
>>>       struct per_cpu_pages *pcp = per_cpu_ptr(zone->per_cpu_pageset, cpu);
>>> -    int count = READ_ONCE(pcp->count);
>>> +    int count;
>>> +    spin_lock(&pcp->lock);
>>> +    count = pcp->count;
>>>       while (count) {
>>>           int to_drain = min(count, pcp->batch << CONFIG_PCP_BATCH_SCALE_MAX);
>>>           count -= to_drain;
>>> -        spin_lock(&pcp->lock);
>>>           free_pcppages_bulk(zone, to_drain, pcp, 0);
>>> -        spin_unlock(&pcp->lock);
>>>       }
>>> +    spin_unlock(&pcp->lock);
>>
>> This way seems to be partially going against the purpose of 55f77df7d715
>> ("mm: page_alloc: control latency caused by zone PCP draining") - the zone
>> lock hold time will still be limited by the batch, but not the pcp lock
>> time. It should still be possible to relock between the iterations? To
>> prevent the race I think the main part is determining pcp->count under the
>> lock, but release/retake should still be ok if the pcp->count is reread
>> after relocking.

Okay, I will try it.

> 
> Agreed, had the smame thing in mind when skimming over this patch.


> 
> @Li, with this patch the problems you have been seeing are fully resolved, correct?
> 

Yeah, It worked in my thousand test runs.

P.S, Previously, It occurred more than 5%.