Re: [PATCH v2] mm: compaction: skip memory hole rapidly when isolating migratable pages

["Huang, Ying" <ying.huang@intel.com>]

David Hildenbrand <david@redhat.com> writes:

> On 13.06.23 13:13, Baolin Wang wrote:
>> On 6/13/2023 5:56 PM, David Hildenbrand wrote:
>>> On 13.06.23 10:55, Baolin Wang wrote:
>>>> On some machines, the normal zone can have a large memory hole like
>>>> below memory layout, and we can see the range from 0x100000000 to
>>>> 0x1800000000 is a hole. So when isolating some migratable pages, the
>>>> scanner can meet the hole and it will take more time to skip the large
>>>> hole. From my measurement, I can see the isolation scanner will take
>>>> 80us ~ 100us to skip the large hole [0x100000000 - 0x1800000000].
>>>>
>>>> So adding a new helper to fast search next online memory section
>>>> to skip the large hole can help to find next suitable pageblock
>>>> efficiently. With this patch, I can see the large hole scanning only
>>>> takes < 1us.
>>>>
>>>> [    0.000000] Zone ranges:
>>>> [    0.000000]   DMA      [mem 0x0000000040000000-0x00000000ffffffff]
>>>> [    0.000000]   DMA32    empty
>>>> [    0.000000]   Normal   [mem 0x0000000100000000-0x0000001fa7ffffff]
>>>> [    0.000000] Movable zone start for each node
>>>> [    0.000000] Early memory node ranges
>>>> [    0.000000]   node   0: [mem 0x0000000040000000-0x0000000fffffffff]
>>>> [    0.000000]   node   0: [mem 0x0000001800000000-0x0000001fa3c7ffff]
>>>> [    0.000000]   node   0: [mem 0x0000001fa3c80000-0x0000001fa3ffffff]
>>>> [    0.000000]   node   0: [mem 0x0000001fa4000000-0x0000001fa402ffff]
>>>> [    0.000000]   node   0: [mem 0x0000001fa4030000-0x0000001fa40effff]
>>>> [    0.000000]   node   0: [mem 0x0000001fa40f0000-0x0000001fa73cffff]
>>>> [    0.000000]   node   0: [mem 0x0000001fa73d0000-0x0000001fa745ffff]
>>>> [    0.000000]   node   0: [mem 0x0000001fa7460000-0x0000001fa746ffff]
>>>> [    0.000000]   node   0: [mem 0x0000001fa7470000-0x0000001fa758ffff]
>>>> [    0.000000]   node   0: [mem 0x0000001fa7590000-0x0000001fa7ffffff]
>>>>
>>>> Signed-off-by: Baolin Wang <baolin.wang@linux.alibaba.com>
>>>> ---
>>>> Changes from v1:
>>>>    - Fix building errors if CONFIG_SPARSEMEM is not selected.
>>>>    - Use NR_MEM_SECTIONS instead of '-1' per Huang Ying.
>>>> ---
>>>>    include/linux/mmzone.h | 10 ++++++++++
>>>>    mm/compaction.c        | 30 +++++++++++++++++++++++++++++-
>>>>    2 files changed, 39 insertions(+), 1 deletion(-)
>>>>
>>>> diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
>>>> index 5a7ada0413da..5ff1fa2efe28 100644
>>>> --- a/include/linux/mmzone.h
>>>> +++ b/include/linux/mmzone.h
>>>> @@ -2000,6 +2000,16 @@ static inline unsigned long
>>>> next_present_section_nr(unsigned long section_nr)
>>>>        return -1;
>>>>    }
>>>> +static inline unsigned long next_online_section_nr(unsigned long
>>>> section_nr)
>>>> +{
>>>> +    while (++section_nr <= __highest_present_section_nr) {
>>>> +        if (online_section_nr(section_nr))
>>>> +            return section_nr;
>>>> +    }
>>>> +
>>>> +    return NR_MEM_SECTIONS;
>>>> +}
>>>> +
>>>>    /*
>>>>     * These are _only_ used during initialisation, therefore they
>>>>     * can use __initdata ...  They could have names to indicate
>>>> diff --git a/mm/compaction.c b/mm/compaction.c
>>>> index 3398ef3a55fe..c31ff6123891 100644
>>>> --- a/mm/compaction.c
>>>> +++ b/mm/compaction.c
>>>> @@ -229,6 +229,28 @@ static void reset_cached_positions(struct zone
>>>> *zone)
>>>>                    pageblock_start_pfn(zone_end_pfn(zone) - 1);
>>>>    }
>>>> +#ifdef CONFIG_SPARSEMEM
>>>> +static unsigned long skip_hole_pageblock(unsigned long start_pfn)
>>>> +{
>>>> +    unsigned long next_online_nr;
>>>> +    unsigned long start_nr = pfn_to_section_nr(start_pfn);
>>>> +
>>>> +    if (online_section_nr(start_nr))
>>>> +        return 0;
>>>> +
>>>> +    next_online_nr = next_online_section_nr(start_nr);
>>>> +    if (next_online_nr < NR_MEM_SECTIONS)
>>>> +        return section_nr_to_pfn(next_online_nr);
>>>> +
>>>
>>> I would simply inline next_online_section_nr and simplify (and add a
>>> comment):
>>>
>>> /*
>>>    * If the PFN falls into an offline section, return the start PFN of the
>>>    * next online section. If the PFN falls into an online section or if
>>>    * there is no next online section, return 0.
>>>    */
>>> static unsigned long skip_hole_pageblock(unsigned long start_pfn)
>>> {
>>>       unsigned long nr = pfn_to_section_nr(start_pfn);
>>>
>>>       if (online_section_nr(nr))
>>>           return 0;
>>>
>>>       while (++nr <= __highest_present_section_nr) {
>>>           if (online_section_nr(nr))
>>>               return section_nr_to_pfn(nr);
>>>       }
>>>       return 0
>>> }
>>>
>>> Easier, no?
>> Originally I want to add a common helper like
>> next_present_section_nr(),
>> which can be used by other users. But yes, your suggestion is easier,
>> and I am fine with that.
>> 
>>> And maybe just call that function "skip_offline_sections()" then?
>>> Because we're not operating on pageblocks.
>> OK. Thanks.
>> 
>
> Feel free to add to the simplified version
>
> Acked-by: David Hildenbrand <david@redhat.com>

With David's above comments addressed, feel free to add

Acked-by: "Huang, Ying" <ying.huang@intel.com>