Re: [PATCH v3] mm/memory hotplug/unplug: Optimize zone contiguous check when changing pfn range

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

On 4/9/26 16:40, Mike Rapoport wrote:
> On Wed, Apr 08, 2026 at 09:36:14AM +0200, David Hildenbrand (Arm) wrote:
>> On 4/8/26 05:16, Yuan Liu wrote:
>>> When move_pfn_range_to_zone() or remove_pfn_range_from_zone() updates a
>>> zone, set_zone_contiguous() rescans the entire zone pageblock-by-pageblock
>>> to rebuild zone->contiguous. For large zones this is a significant cost
>>> during memory hotplug and hot-unplug.
>>>
>>> Add a new zone member pages_with_online_memmap that tracks the number of
>>> pages within the zone span that have an online memory map (including present
>>> pages and memory holes whose memory map has been initialized). When
>>> spanned_pages == pages_with_online_memmap the zone is contiguous and
>>> pfn_to_page() can be called on any PFN in the zone span without further
>>> pfn_valid() checks.
>>>
>>> Only pages that fall within the current zone span are accounted towards
>>> pages_with_online_memmap. A "too small" value is safe, it merely prevents
>>> detecting a contiguous zone.
>>>
>>> The following test cases of memory hotplug for a VM [1], tested in the
>>> environment [2], show that this optimization can significantly reduce the
>>> memory hotplug time [3].
>>>
>>> +----------------+------+---------------+--------------+----------------+
>>> |                | Size | Time (before) | Time (after) | Time Reduction |
>>> |                +------+---------------+--------------+----------------+
>>> | Plug Memory    | 256G |      10s      |      3s      |       70%      |
>>> |                +------+---------------+--------------+----------------+
>>> |                | 512G |      36s      |      7s      |       81%      |
>>> +----------------+------+---------------+--------------+----------------+
>>>
>>> +----------------+------+---------------+--------------+----------------+
>>> |                | Size | Time (before) | Time (after) | Time Reduction |
>>> |                +------+---------------+--------------+----------------+
>>> | Unplug Memory  | 256G |      11s      |      4s      |       64%      |
>>> |                +------+---------------+--------------+----------------+
>>> |                | 512G |      36s      |      9s      |       75%      |
>>> +----------------+------+---------------+--------------+----------------+
>>>
>>> [1] Qemu commands to hotplug 256G/512G memory for a VM:
>>>     object_add memory-backend-ram,id=hotmem0,size=256G/512G,share=on
>>>     device_add virtio-mem-pci,id=vmem1,memdev=hotmem0,bus=port1
>>>     qom-set vmem1 requested-size 256G/512G (Plug Memory)
>>>     qom-set vmem1 requested-size 0G (Unplug Memory)
>>>
>>> [2] Hardware     : Intel Icelake server
>>>     Guest Kernel : v7.0-rc4
>>>     Qemu         : v9.0.0
>>>
>>>     Launch VM    :
>>>     qemu-system-x86_64 -accel kvm -cpu host \
>>>     -drive file=./Centos10_cloud.qcow2,format=qcow2,if=virtio \
>>>     -drive file=./seed.img,format=raw,if=virtio \
>>>     -smp 3,cores=3,threads=1,sockets=1,maxcpus=3 \
>>>     -m 2G,slots=10,maxmem=2052472M \
>>>     -device pcie-root-port,id=port1,bus=pcie.0,slot=1,multifunction=on \
>>>     -device pcie-root-port,id=port2,bus=pcie.0,slot=2 \
>>>     -nographic -machine q35 \
>>>     -nic user,hostfwd=tcp::3000-:22
>>>
>>>     Guest kernel auto-onlines newly added memory blocks:
>>>     echo online > /sys/devices/system/memory/auto_online_blocks
>>>
>>> [3] The time from typing the QEMU commands in [1] to when the output of
>>>     'grep MemTotal /proc/meminfo' on Guest reflects that all hotplugged
>>>     memory is recognized.
>>>
>>> Reported-by: Nanhai Zou <nanhai.zou@intel.com>
>>> Reported-by: Chen Zhang <zhangchen.kidd@jd.com>
>>> Tested-by: Yuan Liu <yuan1.liu@intel.com>
>>> Reviewed-by: Tim Chen <tim.c.chen@linux.intel.com>
>>> Reviewed-by: Qiuxu Zhuo <qiuxu.zhuo@intel.com>
>>> Reviewed-by: Yu C Chen <yu.c.chen@intel.com>
>>> Reviewed-by: Pan Deng <pan.deng@intel.com>
>>> Reviewed-by: Nanhai Zou <nanhai.zou@intel.com>
>>> Co-developed-by: Tianyou Li <tianyou.li@intel.com>
>>> Signed-off-by: Tianyou Li <tianyou.li@intel.com>
>>> Signed-off-by: Yuan Liu <yuan1.liu@intel.com>
>>> Acked-by: David Hildenbrand (Arm) <david@kernel.org>
>>> ---
>>
>> [...]
>>
>>> @@ -842,7 +842,7 @@ overlap_memmap_init(unsigned long zone, unsigned long *pfn)
>>>   *   zone/node above the hole except for the trailing pages in the last
>>>   *   section that will be appended to the zone/node below.
>>>   */
>>> -static void __init init_unavailable_range(unsigned long spfn,
>>> +static unsigned long __init init_unavailable_range(unsigned long spfn,
>>>  					  unsigned long epfn,
>>>  					  int zone, int node)
>>>  {
>>> @@ -858,6 +858,7 @@ static void __init init_unavailable_range(unsigned long spfn,
>>>  	if (pgcnt)
>>>  		pr_info("On node %d, zone %s: %lld pages in unavailable ranges\n",
>>>  			node, zone_names[zone], pgcnt);
>>> +	return pgcnt;
>>>  }
>>>  
>>>  /*
>>> @@ -956,9 +957,22 @@ static void __init memmap_init_zone_range(struct zone *zone,
>>>  	memmap_init_range(end_pfn - start_pfn, nid, zone_id, start_pfn,
>>>  			  zone_end_pfn, MEMINIT_EARLY, NULL, MIGRATE_MOVABLE,
>>>  			  false);
>>> +	zone->pages_with_online_memmap += end_pfn - start_pfn;
>>>  
>>> -	if (*hole_pfn < start_pfn)
>>> -		init_unavailable_range(*hole_pfn, start_pfn, zone_id, nid);
>>> +	if (*hole_pfn < start_pfn) {
>>> +		unsigned long pgcnt;
>>> +
>>> +		if (*hole_pfn < zone_start_pfn) {
>>> +			init_unavailable_range(*hole_pfn, zone_start_pfn,
>>> +					       zone_id, nid);
>>> +			pgcnt = init_unavailable_range(zone_start_pfn,
>>> +					start_pfn, zone_id, nid);
>>
>> Indentation of parameters.
>>
>>> +		} else {
>>> +			pgcnt = init_unavailable_range(*hole_pfn, start_pfn,
>>> +					zone_id, nid);
>>
>>
>> Same here.
>>
>>> +		}
>>> +		zone->pages_with_online_memmap += pgcnt;
>>> +	}
>>
>>
>> Maybe something like the following could make it nicer to read, just a
>> thought.
>>
>>
>> unsigned long hole_start_pfn = *hole_pfn;
>>
>> if (hole_start_pfn < zone_start_pfn) {
>> 	init_unavailable_range(hole_start_pfn, zone_start_pfn,
>> 			       zone_id, nid);
>> 	hole_start_pfn = zone_start_pfn;
>> }
>> pgcnt = init_unavailable_range(hole_start_pfn, start_pfn,
>> 			       zone_id, nid);
>>
> 
> Yeah, this looks better :)
> 
> sashiko had several comments
> https://sashiko.dev/#/patchset/20260408031615.1831922-1-yuan1.liu%40intel.com
> 
> I skipped the ones related to hotplug, but in the mm_init part the comment
> about zones that can have overlapping physical spans when mirrored
> kernelcore is enabled seems valid.

The set_zone_contiguous/clear_zone_contiguous can be ignored I think.

The comment about shrink_zone_span() is likely not realistic.
shrink_zone_span() would not shrink over boot holes.

Well, unless we have an odd case where the hole+memory starts in the
middle of a "PAGES_PER_SUBSECTION". That would already be problematic if
memory starts/ends in the middle of a PAGES_PER_SUBSECTION chunk. I
don't such a case exists.

We could improve shrink_zone_span() to let
find_smallest_section_pfn/find_biggest_section_pfn test the pfn_to_nid()
and page_zone() not on;y on the smallest/highest pfn, but also on the
highest/smallest PFN in a PAGES_PER_SUBSECTION chunk.

No need to test pfn_to_online_page() twice, as that is the same result
for all pages in a PAGES_PER_SUBSECTION chunk.

-- 
Cheers,

David