[PATCH v7 0/2] Optimize zone->contiguous update and issue fix

[PATCH v7 0/2] Optimize zone->contiguous update and issue fix

[Tianyou Li]

This series contains 2 patches, the first one fix an issue when check
the zone->contiguous during zone grows, the second one add a fast path
to check the zone->contiguous.  The issue fixed by the first patch can
be found in the original code path without the second patch.

Yuan Liu (1):
  mm/memory hotplug: fix zone->contiguous always false when hotplug

Tianyou Li (1):
  mm/memory hotplug/unplug: Optimize zone->contiguous update when
    changes pfn range


Changes History
===============
v2 changes:
   Add check_zone_contiguous_fast function to check zone contiguity for
   new  memory PFN ranges.

v3 changes:
   Add zone contiguity check for empty zones.

v4 changes:
   1. Improve coding style.
   2. Add fast path for zone contiguity check in memory unplugged cases,
      and update test results.
   3. Refactor set_zone_contiguous: the new set_zone_contiguous updates
      zone contiguity based on the fast path results.

v5 changes:
   1. Improve coding style.
   2. Fix a issue in which zone->contiguous was always false when adding
      new memory, leveraging the fast path optimization.

v6 changes:
   1. Improve coding style.
   2. Add comments.

v7 changes:
   1. Rebased to 6.19-rc1
   2. Reorder the patches so that the fix will be the first in the series. 

Tianyou Li (2):
  mm/memory hotplug: fix zone->contiguous always false when hotplug
  mm/memory hotplug/unplug: Optimize zone->contiguous update when
    changes pfn range

 mm/internal.h       |  8 ++++-
 mm/memory_hotplug.c | 75 +++++++++++++++++++++++++++++++++++++++++++--
 mm/mm_init.c        | 12 ++++++--
 3 files changed, 89 insertions(+), 6 deletions(-)

-- 
2.47.1

[PATCH v7 1/2] mm/memory hotplug: fix zone->contiguous always false when hotplug

[Tianyou Li]

Function set_zone_contiguous used __pageblock_pfn_to_page to
check the whole pageblock is in the same zone. One assumption is
the memory section must online, otherwise the __pageblock_pfn_to_page
will return NULL, then the set_zone_contiguous will be false.
When move_pfn_range_to_zone invoked set_zone_contiguous, since the
memory section did not online, the return value will always be false.

To fix this issue, we removed the set_zone_contiguous from the
move_pfn_range_to_zone, and place it after memory section onlined.

Function remove_pfn_range_from_zone did not have this issue because
memory section remains online at the time set_zone_contiguous invoked.

Reviewed-by: Nanhai Zou <nanhai.zou@intel.com>
Signed-off-by: Yuan Liu <yuan1.liu@intel.com>
Signed-off-by: Tianyou Li <tianyou.li@intel.com>
---
 mm/memory_hotplug.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c
index a63ec679d861..ce6caf8674a5 100644
--- a/mm/memory_hotplug.c
+++ b/mm/memory_hotplug.c
@@ -782,8 +782,6 @@ void move_pfn_range_to_zone(struct zone *zone, unsigned long start_pfn,
 	memmap_init_range(nr_pages, nid, zone_idx(zone), start_pfn, 0,
 			 MEMINIT_HOTPLUG, altmap, migratetype,
 			 isolate_pageblock);
-
-	set_zone_contiguous(zone);
 }
 
 struct auto_movable_stats {
@@ -1115,6 +1113,7 @@ int mhp_init_memmap_on_memory(unsigned long pfn, unsigned long nr_pages,
 	if (nr_pages >= PAGES_PER_SECTION)
 	        online_mem_sections(pfn, ALIGN_DOWN(end_pfn, PAGES_PER_SECTION));
 
+	set_zone_contiguous(zone);
 	return ret;
 }
 
@@ -1205,6 +1204,7 @@ int online_pages(unsigned long pfn, unsigned long nr_pages,
 	}
 
 	online_pages_range(pfn, nr_pages);
+	set_zone_contiguous(zone);
 	adjust_present_page_count(pfn_to_page(pfn), group, nr_pages);
 
 	if (node_arg.nid >= 0)
-- 
2.47.1

[PATCH v7 2/2] mm/memory hotplug/unplug: Optimize zone->contiguous update when changes pfn range

[Tianyou Li]

When invoke move_pfn_range_to_zone or remove_pfn_range_from_zone, it will
update the zone->contiguous by checking the new zone's pfn range from the
beginning to the end, regardless the previous state of the old zone. When
the zone's pfn range is large, the cost of traversing the pfn range to
update the zone->contiguous could be significant.

Add fast paths to quickly detect cases where zone is definitely not
contiguous without scanning the new zone. The cases are: when the new range
did not overlap with previous range, the contiguous should be false; if the
new range adjacent with the previous range, just need to check the new
range; if the new added pages could not fill the hole of previous zone, the
contiguous should be false.

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      |      2s      |       80%      |
|                +------+---------------+--------------+----------------+
|                | 512G |      33s      |      6s      |       81%      |
+----------------+------+---------------+--------------+----------------+

+----------------+------+---------------+--------------+----------------+
|                | Size | Time (before) | Time (after) | Time Reduction |
|                +------+---------------+--------------+----------------+
| Unplug Memory  | 256G |      10s      |      2s      |       80%      |
|                +------+---------------+--------------+----------------+
|                | 512G |      34s      |      6s      |       82%      |
+----------------+------+---------------+--------------+----------------+

[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 : v6.18-rc2
    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>
Reviewed-by: Yuan Liu <yuan1.liu@intel.com>
Signed-off-by: Tianyou Li <tianyou.li@intel.com>
---
 mm/internal.h       |  8 ++++-
 mm/memory_hotplug.c | 75 +++++++++++++++++++++++++++++++++++++++++++--
 mm/mm_init.c        | 12 ++++++--
 3 files changed, 89 insertions(+), 6 deletions(-)

diff --git a/mm/internal.h b/mm/internal.h
index e430da900430..828aed5c2fef 100644
--- a/mm/internal.h
+++ b/mm/internal.h
@@ -730,7 +730,13 @@ static inline struct page *pageblock_pfn_to_page(unsigned long start_pfn,
 	return __pageblock_pfn_to_page(start_pfn, end_pfn, zone);
 }
 
-void set_zone_contiguous(struct zone *zone);
+enum zone_contig_state {
+	ZONE_CONTIG_YES,
+	ZONE_CONTIG_NO,
+	ZONE_CONTIG_MAYBE,
+};
+
+void set_zone_contiguous(struct zone *zone, enum zone_contig_state state);
 bool pfn_range_intersects_zones(int nid, unsigned long start_pfn,
 			   unsigned long nr_pages);
 
diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c
index ce6caf8674a5..f51293be12eb 100644
--- a/mm/memory_hotplug.c
+++ b/mm/memory_hotplug.c
@@ -544,6 +544,28 @@ static void update_pgdat_span(struct pglist_data *pgdat)
 	pgdat->node_spanned_pages = node_end_pfn - node_start_pfn;
 }
 
+static enum zone_contig_state zone_contig_state_after_shrinking(struct zone *zone,
+				unsigned long start_pfn, unsigned long nr_pages)
+{
+	const unsigned long end_pfn = start_pfn + nr_pages;
+
+	/*
+	 * If the removed pfn range inside the original zone span, the contiguous
+	 * property is surely false.
+	 */
+	if (start_pfn > zone->zone_start_pfn && end_pfn < zone_end_pfn(zone))
+		return ZONE_CONTIG_NO;
+
+	/* If the removed pfn range is at the beginning or end of the
+	 * original zone span, the contiguous property is preserved when
+	 * the original zone is contiguous.
+	 */
+	if (start_pfn == zone->zone_start_pfn || end_pfn == zone_end_pfn(zone))
+		return zone->contiguous ? ZONE_CONTIG_YES : ZONE_CONTIG_MAYBE;
+
+	return ZONE_CONTIG_MAYBE;
+}
+
 void remove_pfn_range_from_zone(struct zone *zone,
 				      unsigned long start_pfn,
 				      unsigned long nr_pages)
@@ -551,6 +573,7 @@ void remove_pfn_range_from_zone(struct zone *zone,
 	const unsigned long end_pfn = start_pfn + nr_pages;
 	struct pglist_data *pgdat = zone->zone_pgdat;
 	unsigned long pfn, cur_nr_pages;
+	enum zone_contig_state new_contiguous_state = ZONE_CONTIG_MAYBE;
 
 	/* Poison struct pages because they are now uninitialized again. */
 	for (pfn = start_pfn; pfn < end_pfn; pfn += cur_nr_pages) {
@@ -571,12 +594,14 @@ void remove_pfn_range_from_zone(struct zone *zone,
 	if (zone_is_zone_device(zone))
 		return;
 
+	new_contiguous_state = zone_contig_state_after_shrinking(zone, start_pfn,
+								 nr_pages);
 	clear_zone_contiguous(zone);
 
 	shrink_zone_span(zone, start_pfn, start_pfn + nr_pages);
 	update_pgdat_span(pgdat);
 
-	set_zone_contiguous(zone);
+	set_zone_contiguous(zone, new_contiguous_state);
 }
 
 /**
@@ -736,6 +761,39 @@ static inline void section_taint_zone_device(unsigned long pfn)
 }
 #endif
 
+static enum zone_contig_state zone_contig_state_after_growing(struct zone *zone,
+				unsigned long start_pfn, unsigned long nr_pages)
+{
+	const unsigned long end_pfn = start_pfn + nr_pages;
+
+	if (zone_is_empty(zone))
+		return ZONE_CONTIG_YES;
+
+	/*
+	 * If the moved pfn range does not intersect with the original zone spa
+	 * the contiguous property is surely false.
+	 */
+	if (end_pfn < zone->zone_start_pfn || start_pfn > zone_end_pfn(zone))
+		return ZONE_CONTIG_NO;
+
+	/*
+	 * If the moved pfn range is adjacent to the original zone span, given
+	 * the moved pfn range's contiguous property is always true, the zone's
+	 * contiguous property inherited from the original value.
+	 */
+	if (end_pfn == zone->zone_start_pfn || start_pfn == zone_end_pfn(zone))
+		return zone->contiguous ? ZONE_CONTIG_YES : ZONE_CONTIG_NO;
+
+	/*
+	 * If the original zone's hole larger than the moved pages in the range
+	 * the contiguous property is surely false.
+	 */
+	if (nr_pages < (zone->spanned_pages - zone->present_pages))
+		return ZONE_CONTIG_NO;
+
+	return ZONE_CONTIG_MAYBE;
+}
+
 /*
  * Associate the pfn range with the given zone, initializing the memmaps
  * and resizing the pgdat/zone data to span the added pages. After this
@@ -1090,11 +1148,20 @@ int mhp_init_memmap_on_memory(unsigned long pfn, unsigned long nr_pages,
 {
 	unsigned long end_pfn = pfn + nr_pages;
 	int ret, i;
+	enum zone_contig_state new_contiguous_state = ZONE_CONTIG_NO;
 
 	ret = kasan_add_zero_shadow(__va(PFN_PHYS(pfn)), PFN_PHYS(nr_pages));
 	if (ret)
 		return ret;
 
+	/*
+	 * If the allocated memmap pages are not in a full section, keep the
+	 * contiguous state as ZONE_CONTIG_NO.
+	 */
+	if (IS_ALIGNED(end_pfn, PAGES_PER_SECTION))
+		new_contiguous_state = zone_contig_state_after_growing(zone,
+								pfn, nr_pages);
+
 	move_pfn_range_to_zone(zone, pfn, nr_pages, NULL, MIGRATE_UNMOVABLE,
 			       false);
 
@@ -1113,7 +1180,7 @@ int mhp_init_memmap_on_memory(unsigned long pfn, unsigned long nr_pages,
 	if (nr_pages >= PAGES_PER_SECTION)
 	        online_mem_sections(pfn, ALIGN_DOWN(end_pfn, PAGES_PER_SECTION));
 
-	set_zone_contiguous(zone);
+	set_zone_contiguous(zone, new_contiguous_state);
 	return ret;
 }
 
@@ -1153,6 +1220,7 @@ int online_pages(unsigned long pfn, unsigned long nr_pages,
 	const int nid = zone_to_nid(zone);
 	int need_zonelists_rebuild = 0;
 	unsigned long flags;
+	enum zone_contig_state new_contiguous_state = ZONE_CONTIG_NO;
 	int ret;
 
 	/*
@@ -1166,6 +1234,7 @@ int online_pages(unsigned long pfn, unsigned long nr_pages,
 			 !IS_ALIGNED(pfn + nr_pages, PAGES_PER_SECTION)))
 		return -EINVAL;
 
+	new_contiguous_state = zone_contig_state_after_growing(zone, pfn, nr_pages);
 
 	/* associate pfn range with the zone */
 	move_pfn_range_to_zone(zone, pfn, nr_pages, NULL, MIGRATE_MOVABLE,
@@ -1204,7 +1273,7 @@ int online_pages(unsigned long pfn, unsigned long nr_pages,
 	}
 
 	online_pages_range(pfn, nr_pages);
-	set_zone_contiguous(zone);
+	set_zone_contiguous(zone, new_contiguous_state);
 	adjust_present_page_count(pfn_to_page(pfn), group, nr_pages);
 
 	if (node_arg.nid >= 0)
diff --git a/mm/mm_init.c b/mm/mm_init.c
index fc2a6f1e518f..0c41f1004847 100644
--- a/mm/mm_init.c
+++ b/mm/mm_init.c
@@ -2263,11 +2263,19 @@ void __init init_cma_pageblock(struct page *page)
 }
 #endif
 
-void set_zone_contiguous(struct zone *zone)
+void set_zone_contiguous(struct zone *zone, enum zone_contig_state state)
 {
 	unsigned long block_start_pfn = zone->zone_start_pfn;
 	unsigned long block_end_pfn;
 
+	if (state == ZONE_CONTIG_YES) {
+		zone->contiguous = true;
+		return;
+	}
+
+	if (state == ZONE_CONTIG_NO)
+		return;
+
 	block_end_pfn = pageblock_end_pfn(block_start_pfn);
 	for (; block_start_pfn < zone_end_pfn(zone);
 			block_start_pfn = block_end_pfn,
@@ -2348,7 +2356,7 @@ void __init page_alloc_init_late(void)
 		shuffle_free_memory(NODE_DATA(nid));
 
 	for_each_populated_zone(zone)
-		set_zone_contiguous(zone);
+		set_zone_contiguous(zone, ZONE_CONTIG_MAYBE);
 
 	/* Initialize page ext after all struct pages are initialized. */
 	if (deferred_struct_pages)
-- 
2.47.1