Re: [stable-6.6.y] mm: khugepaged refuses to freeze

["David Hildenbrand (Arm)" <david@kernel.org>]

On 2/6/26 06:12, Baolin Wang wrote:
> 
> 
> On 2/6/26 12:31 PM, Sergey Senozhatsky wrote:
>> On (26/02/06 12:38), Sergey Senozhatsky wrote:
>> [..]
>>>
>>> Right, I thought about it but wasn't sure.  Could the inner loop (e.g.
>>> collapse_file() in this particular case) loop long enough to fail 
>>> suspend
>>> w/o ever giving the outer loop (khugepaged_do_scan()) a chance to 
>>> freeze?
> 
> Yes, that’s possible. However, if we add a try_to_freeze() check in the 
> inner loop, we need to consider various scenarios (such as anonymous 
> folio swap-in and other potential cases?), which feels too hacky to me.
> 
>> For inner loops I wondered if cond_resched() could be an indicator of
>> where try_to_freeze() should be placed.  Those cond_resched() calls
>> are there for a reason, after all.   E.g. something like:
>>
>> ---
>>
>> diff --git a/mm/khugepaged.c b/mm/khugepaged.c
>> index fa6a018b20a8..cee08466a069 100644
>> --- a/mm/khugepaged.c
>> +++ b/mm/khugepaged.c
>> @@ -2431,6 +2431,9 @@ static unsigned int 
>> khugepaged_scan_mm_slot(unsigned int pages, enum scan_result
>>           unsigned long hstart, hend;
>>           cond_resched();
>> +        if (try_to_freeze())
>> +            break;
>> +
>>           if (unlikely(hpage_collapse_test_exit_or_disable(mm))) {
>>               progress++;
>>               break;
>> @@ -2453,6 +2456,9 @@ static unsigned int 
>> khugepaged_scan_mm_slot(unsigned int pages, enum scan_result
>>               bool mmap_locked = true;
>>               cond_resched();
>> +            if (try_to_freeze())
>> +                goto breakouterloop;
>> +
>>               if (unlikely(hpage_collapse_test_exit_or_disable(mm)))
>>                   goto breakouterloop;
> 
> This looks better than the previous version. Let’s also wait to see if 
> others have any better suggestions.

What prevents other callpaths (faults, read(), write(), etc) from 
similarly triggering swapin?

I recall that there is a notifier when the system is preparing to sleep 
(pm notifier or something). Could we simply hook into that to tell 
khugepaged to suspend+resume?

Essentially, making hpage_collapse_test_exit_or_disable() break our for us.

-- 
Cheers,

David