[PATCH v2 0/2] mm: Don't set and reset page count in MEMINIT_EARLY

[PATCH v2 0/2] mm: Don't set and reset page count in MEMINIT_EARLY

[Yajun Deng]

__init_single_page would set page count and __free_pages_core would
reset it. A lot of pages don't need to do this when in MEMINIT_EARLY
context. It's unnecessary and time-consuming.

The 1st patch is pass page count and reserved to __init_single_page.
It's in preparation for the 2nd patch, it didn't change anything.

The 2nd patch only set page count for the reserved region, not all
of the region.

Yajun Deng (2):
  mm: pass page count and reserved to __init_single_page
  mm: Init page count in reserve_bootmem_region when MEMINIT_EARLY

 mm/hugetlb.c    |  2 +-
 mm/internal.h   |  8 +++++++-
 mm/mm_init.c    | 45 ++++++++++++++++++++++++++++-----------------
 mm/page_alloc.c | 20 ++++++++++++--------
 4 files changed, 48 insertions(+), 27 deletions(-)

-- 
2.25.1

[PATCH v2 1/2] mm: pass page count and reserved to __init_single_page

[Yajun Deng]

When we init a single page, we need to mark this page reserved if it
does. And some pages may not need to set page count, such as compound
pages.

Introduce INIT_PAGE_COUNT and INIT_PAGE_RESERVED, let the caller
decide if it needs to set page count and mark page reserved or not.

Signed-off-by: Yajun Deng <yajun.deng@linux.dev>
---
 mm/hugetlb.c  |  2 +-
 mm/internal.h |  8 +++++++-
 mm/mm_init.c  | 31 +++++++++++++++++--------------
 3 files changed, 25 insertions(+), 16 deletions(-)

diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index e2123d1bb4a2..b63635abe5c3 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -3196,7 +3196,7 @@ static void __init hugetlb_folio_init_tail_vmemmap(struct folio *folio,
 	for (pfn = head_pfn + start_page_number; pfn < end_pfn; pfn++) {
 		struct page *page = pfn_to_page(pfn);
 
-		__init_single_page(page, pfn, zone, nid);
+		__init_single_page(page, pfn, zone, nid, INIT_PAGE_COUNT);
 		prep_compound_tail((struct page *)folio, pfn - head_pfn);
 		ret = page_ref_freeze(page, 1);
 		VM_BUG_ON(!ret);
diff --git a/mm/internal.h b/mm/internal.h
index 7a961d12b088..e9366cce461c 100644
--- a/mm/internal.h
+++ b/mm/internal.h
@@ -1209,8 +1209,14 @@ struct vma_prepare {
 	struct vm_area_struct *remove2;
 };
 
+enum {
+	INIT_PAGE_COUNT    = (1 << 0),
+	INIT_PAGE_RESERVED = (1 << 1),
+};
+
 void __meminit __init_single_page(struct page *page, unsigned long pfn,
-				unsigned long zone, int nid);
+				  unsigned long zone, int nid,
+				  unsigned int flags);
 
 /* shrinker related functions */
 unsigned long shrink_slab(gfp_t gfp_mask, int nid, struct mem_cgroup *memcg,
diff --git a/mm/mm_init.c b/mm/mm_init.c
index 06a72c223bce..61df37133331 100644
--- a/mm/mm_init.c
+++ b/mm/mm_init.c
@@ -557,11 +557,11 @@ static void __init find_zone_movable_pfns_for_nodes(void)
 }
 
 void __meminit __init_single_page(struct page *page, unsigned long pfn,
-				unsigned long zone, int nid)
+				  unsigned long zone, int nid,
+				  unsigned int flags)
 {
 	mm_zero_struct_page(page);
 	set_page_links(page, zone, nid, pfn);
-	init_page_count(page);
 	page_mapcount_reset(page);
 	page_cpupid_reset_last(page);
 	page_kasan_tag_reset(page);
@@ -572,6 +572,10 @@ void __meminit __init_single_page(struct page *page, unsigned long pfn,
 	if (!is_highmem_idx(zone))
 		set_page_address(page, __va(pfn << PAGE_SHIFT));
 #endif
+	if (flags & INIT_PAGE_COUNT)
+		init_page_count(page);
+	if (flags & INIT_PAGE_RESERVED)
+		__SetPageReserved(page);
 }
 
 #ifdef CONFIG_NUMA
@@ -714,7 +718,7 @@ static void __meminit init_reserved_page(unsigned long pfn, int nid)
 		if (zone_spans_pfn(zone, pfn))
 			break;
 	}
-	__init_single_page(pfn_to_page(pfn), pfn, zid, nid);
+	__init_single_page(pfn_to_page(pfn), pfn, zid, nid, INIT_PAGE_COUNT);
 }
 #else
 static inline void pgdat_set_deferred_range(pg_data_t *pgdat) {}
@@ -821,8 +825,8 @@ static void __init init_unavailable_range(unsigned long spfn,
 			pfn = pageblock_end_pfn(pfn) - 1;
 			continue;
 		}
-		__init_single_page(pfn_to_page(pfn), pfn, zone, node);
-		__SetPageReserved(pfn_to_page(pfn));
+		__init_single_page(pfn_to_page(pfn), pfn, zone, node,
+				   INIT_PAGE_COUNT | INIT_PAGE_RESERVED);
 		pgcnt++;
 	}
 
@@ -884,7 +888,7 @@ void __meminit memmap_init_range(unsigned long size, int nid, unsigned long zone
 		}
 
 		page = pfn_to_page(pfn);
-		__init_single_page(page, pfn, zone, nid);
+		__init_single_page(page, pfn, zone, nid, INIT_PAGE_COUNT);
 		if (context == MEMINIT_HOTPLUG)
 			__SetPageReserved(page);
 
@@ -965,11 +969,9 @@ static void __init memmap_init(void)
 #ifdef CONFIG_ZONE_DEVICE
 static void __ref __init_zone_device_page(struct page *page, unsigned long pfn,
 					  unsigned long zone_idx, int nid,
-					  struct dev_pagemap *pgmap)
+					  struct dev_pagemap *pgmap,
+					  unsigned int flags)
 {
-
-	__init_single_page(page, pfn, zone_idx, nid);
-
 	/*
 	 * Mark page reserved as it will need to wait for onlining
 	 * phase for it to be fully associated with a zone.
@@ -977,7 +979,7 @@ static void __ref __init_zone_device_page(struct page *page, unsigned long pfn,
 	 * We can use the non-atomic __set_bit operation for setting
 	 * the flag as we are still initializing the pages.
 	 */
-	__SetPageReserved(page);
+	__init_single_page(page, pfn, zone_idx, nid, flags | INIT_PAGE_RESERVED);
 
 	/*
 	 * ZONE_DEVICE pages union ->lru with a ->pgmap back pointer
@@ -1041,7 +1043,7 @@ static void __ref memmap_init_compound(struct page *head,
 	for (pfn = head_pfn + 1; pfn < end_pfn; pfn++) {
 		struct page *page = pfn_to_page(pfn);
 
-		__init_zone_device_page(page, pfn, zone_idx, nid, pgmap);
+		__init_zone_device_page(page, pfn, zone_idx, nid, pgmap, 0);
 		prep_compound_tail(head, pfn - head_pfn);
 		set_page_count(page, 0);
 
@@ -1084,7 +1086,8 @@ void __ref memmap_init_zone_device(struct zone *zone,
 	for (pfn = start_pfn; pfn < end_pfn; pfn += pfns_per_compound) {
 		struct page *page = pfn_to_page(pfn);
 
-		__init_zone_device_page(page, pfn, zone_idx, nid, pgmap);
+		__init_zone_device_page(page, pfn, zone_idx, nid, pgmap,
+					INIT_PAGE_COUNT);
 
 		if (pfns_per_compound == 1)
 			continue;
@@ -2058,7 +2061,7 @@ static unsigned long  __init deferred_init_pages(struct zone *zone,
 		} else {
 			page++;
 		}
-		__init_single_page(page, pfn, zid, nid);
+		__init_single_page(page, pfn, zid, nid, INIT_PAGE_COUNT);
 		nr_pages++;
 	}
 	return (nr_pages);
-- 
2.25.1

[PATCH v2 2/2] mm: Init page count in reserve_bootmem_region when MEMINIT_EARLY

[Yajun Deng]

memmap_init_range() would set page count of all pages, but the free
pages count would be reset in __free_pages_core(). There are opposite
operations. It's unnecessary and time-consuming when it's MEMINIT_EARLY
context.

Init page count in reserve_bootmem_region when in MEMINIT_EARLY context,
and check the page count before reset it.

At the same time, the INIT_LIST_HEAD in reserve_bootmem_region isn't
need, as it already done in __init_single_page.

The following data was tested on an x86 machine with 190GB of RAM.

before:
free_low_memory_core_early()    341ms

after:
free_low_memory_core_early()    285ms

Signed-off-by: Yajun Deng <yajun.deng@linux.dev>
---
 mm/mm_init.c    | 18 +++++++++++++-----
 mm/page_alloc.c | 20 ++++++++++++--------
 2 files changed, 25 insertions(+), 13 deletions(-)

diff --git a/mm/mm_init.c b/mm/mm_init.c
index 61df37133331..64c00ebaf4ef 100644
--- a/mm/mm_init.c
+++ b/mm/mm_init.c
@@ -718,7 +718,7 @@ static void __meminit init_reserved_page(unsigned long pfn, int nid)
 		if (zone_spans_pfn(zone, pfn))
 			break;
 	}
-	__init_single_page(pfn_to_page(pfn), pfn, zid, nid, INIT_PAGE_COUNT);
+	__init_single_page(pfn_to_page(pfn), pfn, zid, nid, 0);
 }
 #else
 static inline void pgdat_set_deferred_range(pg_data_t *pgdat) {}
@@ -756,8 +756,8 @@ void __meminit reserve_bootmem_region(phys_addr_t start,
 
 			init_reserved_page(start_pfn, nid);
 
-			/* Avoid false-positive PageTail() */
-			INIT_LIST_HEAD(&page->lru);
+			/* Set page count for the reserve region */
+			init_page_count(page);
 
 			/*
 			 * no need for atomic set_bit because the struct
@@ -888,9 +888,17 @@ void __meminit memmap_init_range(unsigned long size, int nid, unsigned long zone
 		}
 
 		page = pfn_to_page(pfn);
-		__init_single_page(page, pfn, zone, nid, INIT_PAGE_COUNT);
-		if (context == MEMINIT_HOTPLUG)
+
+		/* If the context is MEMINIT_EARLY, we will set page count and
+		 * mark page reserved in reserve_bootmem_region, the free region
+		 * wouldn't have page count and we will check the pages count
+		 * in __free_pages_core.
+		 */
+		__init_single_page(page, pfn, zone, nid, 0);
+		if (context == MEMINIT_HOTPLUG) {
+			init_page_count(page);
 			__SetPageReserved(page);
+		}
 
 		/*
 		 * Usually, we want to mark the pageblock MIGRATE_MOVABLE,
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 06be8821d833..b868caabe8dc 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1285,18 +1285,22 @@ void __free_pages_core(struct page *page, unsigned int order)
 	unsigned int loop;
 
 	/*
-	 * When initializing the memmap, __init_single_page() sets the refcount
-	 * of all pages to 1 ("allocated"/"not free"). We have to set the
-	 * refcount of all involved pages to 0.
+	 * When initializing the memmap, memmap_init_range sets the refcount
+	 * of all pages to 1 ("reserved" and "free") in hotplug context. We
+	 * have to set the refcount of all involved pages to 0. Otherwise,
+	 * we don't do it, as reserve_bootmem_region only set the refcount on
+	 * reserve region ("reserved") in early context.
 	 */
-	prefetchw(p);
-	for (loop = 0; loop < (nr_pages - 1); loop++, p++) {
-		prefetchw(p + 1);
+	if (page_count(page)) {
+		prefetchw(p);
+		for (loop = 0; loop < (nr_pages - 1); loop++, p++) {
+			prefetchw(p + 1);
+			__ClearPageReserved(p);
+			set_page_count(p, 0);
+		}
 		__ClearPageReserved(p);
 		set_page_count(p, 0);
 	}
-	__ClearPageReserved(p);
-	set_page_count(p, 0);
 
 	atomic_long_add(nr_pages, &page_zone(page)->managed_pages);
 
-- 
2.25.1

Re: [PATCH v2 1/2] mm: pass page count and reserved to __init_single_page

[Andrew Morton]

On Mon, 25 Sep 2023 15:21:49 +0800 Yajun Deng <yajun.deng@linux.dev> wrote:

> When we init a single page, we need to mark this page reserved if it
> does. And some pages may not need to set page count, such as compound
> pages.
> 
> Introduce INIT_PAGE_COUNT and INIT_PAGE_RESERVED, let the caller
> decide if it needs to set page count and mark page reserved or not.
> 
> ...
>
> --- a/mm/hugetlb.c
> +++ b/mm/hugetlb.c
> @@ -3196,7 +3196,7 @@ static void __init hugetlb_folio_init_tail_vmemmap(struct folio *folio,
>  	for (pfn = head_pfn + start_page_number; pfn < end_pfn; pfn++) {
>  		struct page *page = pfn_to_page(pfn);
>  
> -		__init_single_page(page, pfn, zone, nid);
> +		__init_single_page(page, pfn, zone, nid, INIT_PAGE_COUNT);
>  		prep_compound_tail((struct page *)folio, pfn - head_pfn);
>  		ret = page_ref_freeze(page, 1);
>  		VM_BUG_ON(!ret);
> diff --git a/mm/internal.h b/mm/internal.h
> index 7a961d12b088..e9366cce461c 100644
> --- a/mm/internal.h
> +++ b/mm/internal.h
> @@ -1209,8 +1209,14 @@ struct vma_prepare {
>  	struct vm_area_struct *remove2;
>  };
>  
> +enum {
> +	INIT_PAGE_COUNT    = (1 << 0),
> +	INIT_PAGE_RESERVED = (1 << 1),
> +};

It would be neater to give this enum a name

>  void __meminit __init_single_page(struct page *page, unsigned long pfn,
> -				unsigned long zone, int nid);
> +				  unsigned long zone, int nid,
> +				  unsigned int flags);

Then use this enum for `flags'.