The patch titled
Subject: mm: pass nid to reserve_bootmem_region()
has been added to the -mm mm-unstable branch. Its filename is
mm-pass-nid-to-reserve_bootmem_region.patch
This patch will shortly appear at
https://git.kernel.org/pub/scm/linux/kernel/git/akpm/25-new.git/tree/patches/mm-pass-nid-to-reserve_bootmem_region.patch
This patch will later appear in the mm-unstable branch at
git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
Before you just go and hit "reply", please:
a) Consider who else should be cc'ed
b) Prefer to cc a suitable mailing list as well
c) Ideally: find the original patch on the mailing list and do a
reply-to-all to that, adding suitable additional cc's
*** Remember to use Documentation/process/submit-checklist.rst when testing your code ***
The -mm tree is included into linux-next via the mm-everything
branch at git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
and is updated there every 2-3 working days
------------------------------------------------------
From: Yajun Deng <yajun.deng@xxxxxxxxx>
Subject: mm: pass nid to reserve_bootmem_region()
Date: Mon, 19 Jun 2023 10:34:06 +0800
early_pfn_to_nid() is called frequently in init_reserved_page(), it
returns the node id of the PFN. These PFN are probably from the same
memory region, they have the same node id. It's not necessary to call
early_pfn_to_nid() for each PFN.
Pass nid to reserve_bootmem_region() and drop the call to
early_pfn_to_nid() in init_reserved_page(). Also, set nid on all reserved
pages before doing this, as some reserved memory regions may not be set
nid.
The most beneficial function is memmap_init_reserved_pages() if
CONFIG_DEFERRED_STRUCT_PAGE_INIT is enabled.
The following data was tested on an x86 machine with 190GB of RAM.
before:
memmap_init_reserved_pages() 67ms
after:
memmap_init_reserved_pages() 20ms
Link: https://lkml.kernel.org/r/20230619023406.424298-1-yajun.deng@xxxxxxxxx
Signed-off-by: Yajun Deng <yajun.deng@xxxxxxxxx>
Reviewed-by: Mike Rapoport (IBM) <rppt@xxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
include/linux/mm.h | 3 ++-
mm/memblock.c | 31 +++++++++++++++++++++----------
mm/mm_init.c | 30 +++++++++++++++++-------------
3 files changed, 40 insertions(+), 24 deletions(-)
--- a/include/linux/mm.h~mm-pass-nid-to-reserve_bootmem_region
+++ a/include/linux/mm.h
@@ -2940,7 +2940,8 @@ extern unsigned long free_reserved_area(
extern void adjust_managed_page_count(struct page *page, long count);
-extern void reserve_bootmem_region(phys_addr_t start, phys_addr_t end);
+extern void reserve_bootmem_region(phys_addr_t start,
+ phys_addr_t end, int nid);
/* Free the reserved page into the buddy system, so it gets managed. */
static inline void free_reserved_page(struct page *page)
--- a/mm/memblock.c~mm-pass-nid-to-reserve_bootmem_region
+++ a/mm/memblock.c
@@ -2082,19 +2082,30 @@ static void __init memmap_init_reserved_
{
struct memblock_region *region;
phys_addr_t start, end;
- u64 i;
+ int nid;
+
+ /*
+ * set nid on all reserved pages and also treat struct
+ * pages for the NOMAP regions as PageReserved
+ */
+ for_each_mem_region(region) {
+ nid = memblock_get_region_node(region);
+ start = region->base;
+ end = start + region->size;
+
+ if (memblock_is_nomap(region))
+ reserve_bootmem_region(start, end, nid);
+
+ memblock_set_node(start, end, &memblock.reserved, nid);
+ }
/* initialize struct pages for the reserved regions */
- for_each_reserved_mem_range(i, &start, &end)
- reserve_bootmem_region(start, end);
+ for_each_reserved_mem_region(region) {
+ nid = memblock_get_region_node(region);
+ start = region->base;
+ end = start + region->size;
- /* and also treat struct pages for the NOMAP regions as PageReserved */
- for_each_mem_region(region) {
- if (memblock_is_nomap(region)) {
- start = region->base;
- end = start + region->size;
- reserve_bootmem_region(start, end);
- }
+ reserve_bootmem_region(start, end, nid);
}
}
--- a/mm/mm_init.c~mm-pass-nid-to-reserve_bootmem_region
+++ a/mm/mm_init.c
@@ -646,10 +646,8 @@ static inline void pgdat_set_deferred_ra
}
/* Returns true if the struct page for the pfn is initialised */
-static inline bool __meminit early_page_initialised(unsigned long pfn)
+static inline bool __meminit early_page_initialised(unsigned long pfn, int nid)
{
- int nid = early_pfn_to_nid(pfn);
-
if (node_online(nid) && pfn >= NODE_DATA(nid)->first_deferred_pfn)
return false;
@@ -695,15 +693,14 @@ defer_init(int nid, unsigned long pfn, u
return false;
}
-static void __meminit init_reserved_page(unsigned long pfn)
+static void __meminit init_reserved_page(unsigned long pfn, int nid)
{
pg_data_t *pgdat;
- int nid, zid;
+ int zid;
- if (early_page_initialised(pfn))
+ if (early_page_initialised(pfn, nid))
return;
- nid = early_pfn_to_nid(pfn);
pgdat = NODE_DATA(nid);
for (zid = 0; zid < MAX_NR_ZONES; zid++) {
@@ -717,7 +714,7 @@ static void __meminit init_reserved_page
#else
static inline void pgdat_set_deferred_range(pg_data_t *pgdat) {}
-static inline bool early_page_initialised(unsigned long pfn)
+static inline bool early_page_initialised(unsigned long pfn, int nid)
{
return true;
}
@@ -727,7 +724,7 @@ static inline bool defer_init(int nid, u
return false;
}
-static inline void init_reserved_page(unsigned long pfn)
+static inline void init_reserved_page(unsigned long pfn, int nid)
{
}
#endif /* CONFIG_DEFERRED_STRUCT_PAGE_INIT */
@@ -738,7 +735,8 @@ static inline void init_reserved_page(un
* marks the pages PageReserved. The remaining valid pages are later
* sent to the buddy page allocator.
*/
-void __meminit reserve_bootmem_region(phys_addr_t start, phys_addr_t end)
+void __meminit reserve_bootmem_region(phys_addr_t start,
+ phys_addr_t end, int nid)
{
unsigned long start_pfn = PFN_DOWN(start);
unsigned long end_pfn = PFN_UP(end);
@@ -747,7 +745,7 @@ void __meminit reserve_bootmem_region(ph
if (pfn_valid(start_pfn)) {
struct page *page = pfn_to_page(start_pfn);
- init_reserved_page(start_pfn);
+ init_reserved_page(start_pfn, nid);
/* Avoid false-positive PageTail() */
INIT_LIST_HEAD(&page->lru);
@@ -2572,8 +2570,14 @@ void __init set_dma_reserve(unsigned lon
void __init memblock_free_pages(struct page *page, unsigned long pfn,
unsigned int order)
{
- if (!early_page_initialised(pfn))
- return;
+
+ if (IS_ENABLED(CONFIG_DEFERRED_STRUCT_PAGE_INIT)) {
+ int nid = early_pfn_to_nid(pfn);
+
+ if (!early_page_initialised(pfn, nid))
+ return;
+ }
+
if (!kmsan_memblock_free_pages(page, order)) {
/* KMSAN will take care of these pages. */
return;
_
Patches currently in -mm which might be from yajun.deng@xxxxxxxxx are
dma-contiguous-support-per-numa-cma-for-all-architectures.patch
mm-pass-nid-to-reserve_bootmem_region.patch
