* [to-be-updated] powerpc-mm-switch-device-dax-to-shared-tail-vmemmap-pages.patch removed from -mm tree
@ 2026-09-10 23:07 Andrew Morton
0 siblings, 0 replies; 4+ messages in thread
From: Andrew Morton @ 2026-09-10 23:07 UTC (permalink / raw)
To: mm-commits, songmuchun, akpm
The quilt patch titled
Subject: powerpc/mm: switch device DAX to shared tail vmemmap pages
has been removed from the -mm tree. Its filename was
powerpc-mm-switch-device-dax-to-shared-tail-vmemmap-pages.patch
This patch was dropped because an updated version will be issued
------------------------------------------------------
From: Muchun Song <songmuchun@bytedance.com>
Subject: powerpc/mm: switch device DAX to shared tail vmemmap pages
Date: Tue, 8 Sep 2026 11:03:32 +0800
The powerpc radix compound vmemmap population path still finds a reusable
tail page by walking the vmemmap page tables.
Switch it to the common vmemmap_shared_tail_page() helper instead, so it
can use the shared vmemmap page directly to simplify the code.
This removes the powerpc-specific tail-page lookup and its fallback path
and aligns the device DAX vmemmap optimization path with HugeTLB.
Link:
https://lore.kernel.org/20260908030335.96549-9-songmuchun@bytedance.com
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
Cc: David Hildenbrand <david@kernel.org>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Lorenzo Stoakes <ljs@kernel.org>
Cc: Madhavan Srinivasan <maddy@linux.ibm.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: Nicholas Piggin <npiggin@gmail.com>
Cc: Oscar Salvador <osalvador@suse.de>
Cc: Qi Zheng <qi.zheng@linux.dev>
Cc: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---
arch/powerpc/mm/book3s64/radix_pgtable.c | 80 ++-------------------
1 file changed, 9 insertions(+), 71 deletions(-)
--- a/arch/powerpc/mm/book3s64/radix_pgtable.c~powerpc-mm-switch-device-dax-to-shared-tail-vmemmap-pages
+++ a/arch/powerpc/mm/book3s64/radix_pgtable.c
@@ -19,6 +19,7 @@
#include <linux/string_helpers.h>
#include <linux/memory.h>
#include <linux/kfence.h>
+#include <linux/vmemmap-optimization.h>
#include <asm/pgalloc.h>
#include <asm/mmu_context.h>
@@ -1250,59 +1251,6 @@ static pte_t * __meminit radix__vmemmap_
return pte;
}
-static pte_t * __meminit vmemmap_compound_tail_page(unsigned long addr,
- unsigned long pfn_offset, int node)
-{
- pgd_t *pgd;
- p4d_t *p4d;
- pud_t *pud;
- pmd_t *pmd;
- pte_t *pte;
- unsigned long map_addr;
-
- /* the second vmemmap page which we use for duplication */
- map_addr = addr - pfn_offset * sizeof(struct page) + PAGE_SIZE;
- pgd = pgd_offset_k(map_addr);
- p4d = p4d_offset(pgd, map_addr);
- pud = vmemmap_pud_alloc(p4d, node, map_addr);
- if (!pud)
- return NULL;
- pmd = vmemmap_pmd_alloc(pud, node, map_addr);
- if (!pmd)
- return NULL;
- if (pmd_leaf(*pmd))
- /*
- * The second page is mapped as a hugepage due to a nearby request.
- * Force our mapping to page size without deduplication
- */
- return NULL;
- pte = vmemmap_pte_alloc(pmd, node, map_addr);
- if (!pte)
- return NULL;
- /*
- * Check if there exist a mapping to the left
- */
- if (pte_none(*pte)) {
- /*
- * Populate the head page vmemmap page.
- * It can fall in different pmd, hence
- * vmemmap_populate_address()
- */
- pte = radix__vmemmap_populate_address(map_addr - PAGE_SIZE, node, NULL, NULL);
- if (!pte)
- return NULL;
- /*
- * Populate the tail pages vmemmap page
- */
- pte = radix__vmemmap_pte_populate(pmd, map_addr, node, NULL, NULL);
- if (!pte)
- return NULL;
- vmemmap_verify(pte, node, map_addr, map_addr + PAGE_SIZE);
- return pte;
- }
- return pte;
-}
-
int __meminit vmemmap_populate_compound_pages(unsigned long start_pfn,
unsigned long start,
unsigned long end, int node,
@@ -1320,6 +1268,12 @@ int __meminit vmemmap_populate_compound_
pud_t *pud;
pmd_t *pmd;
pte_t *pte;
+ struct page *tail_page;
+ unsigned int order = pfn_to_section_order(start_pfn);
+
+ tail_page = vmemmap_shared_tail_page(order, device_zone(node));
+ if (!tail_page)
+ return -ENOMEM;
for (addr = start; addr < end; addr = next) {
@@ -1349,10 +1303,9 @@ int __meminit vmemmap_populate_compound_
next = addr + PAGE_SIZE;
continue;
} else {
- unsigned long nr_pages = pgmap_vmemmap_nr(pgmap);
+ unsigned long nr_pages = 1UL << order;
unsigned long addr_pfn = page_to_pfn((struct page *)addr);
unsigned long pfn_offset = addr_pfn - ALIGN_DOWN(addr_pfn, nr_pages);
- pte_t *tail_page_pte;
/*
* if the address is aligned to huge page size it is the
@@ -1377,23 +1330,8 @@ int __meminit vmemmap_populate_compound_
next = addr + 2 * PAGE_SIZE;
continue;
}
- /*
- * get the 2nd mapping details
- * Also create it if that doesn't exist
- */
- tail_page_pte = vmemmap_compound_tail_page(addr, pfn_offset, node);
- if (!tail_page_pte) {
-
- pte = radix__vmemmap_pte_populate(pmd, addr, node, NULL, NULL);
- if (!pte)
- return -ENOMEM;
- vmemmap_verify(pte, node, addr, addr + PAGE_SIZE);
-
- next = addr + PAGE_SIZE;
- continue;
- }
- pte = radix__vmemmap_pte_populate(pmd, addr, node, NULL, pte_page(*tail_page_pte));
+ pte = radix__vmemmap_pte_populate(pmd, addr, node, NULL, tail_page);
if (!pte)
return -ENOMEM;
vmemmap_verify(pte, node, addr, addr + PAGE_SIZE);
_
Patches currently in -mm which might be from songmuchun@bytedance.com are
mm-sparse-relax-struct-mem_section-size-constraints.patch
mm-sparse-vmemmap-rename-hvo-order-macros.patch
mm-mm_init-skip-initializing-shared-vmemmap-tail-pages.patch
mm-sparse-vmemmap-initialize-shared-tail-vmemmap-pages-on-allocation.patch
mm-sparse-vmemmap-support-section-based-vmemmap-accounting.patch
mm-mm_init-factor-out-pfn_to_zone.patch
mm-sparse-vmemmap-move-helpers-ahead-of-future-callers.patch
mm-sparse-vmemmap-support-section-based-vmemmap-optimization.patch
mm-sparse-initialize-memory-sections-earlier.patch
mm-hugetlb-switch-hugetlb-to-section-based-vmemmap-optimization.patch
mm-sparse-vmemmap-remove-sparsemem_vmemmap_preinit-support.patch
mm-sparse-inline-usemap-allocation-into-sparse_init_nid.patch
mm-sparse-remove-section_map_size.patch
mm-hugetlb-remove-huge_bootmem_hvo.patch
mm-hugetlb-remove-huge_bootmem_cma.patch
mm-hugetlb-localize-struct-huge_bootmem_page.patch
mm-hugetlb-localize-huge_bootmem_zones_valid.patch
mm-sparse-vmemmap-drop-the-extra-tail-page-from-device-dax-reservation.patch
mm-sparse-vmemmap-drop-unused-section_nr_vmemmap_pages-arguments.patch
documentation-mm-update-dax-vmemmap-deduplication-docs.patch
^ permalink raw reply [flat|nested] 4+ messages in thread
* [to-be-updated] powerpc-mm-switch-device-dax-to-shared-tail-vmemmap-pages.patch removed from -mm tree
@ 2026-09-17 1:29 Andrew Morton
0 siblings, 0 replies; 4+ messages in thread
From: Andrew Morton @ 2026-09-17 1:29 UTC (permalink / raw)
To: mm-commits, songmuchun, akpm
The quilt patch titled
Subject: powerpc/mm: switch device DAX to shared tail vmemmap pages
has been removed from the -mm tree. Its filename was
powerpc-mm-switch-device-dax-to-shared-tail-vmemmap-pages.patch
This patch was dropped because an updated version will be issued
------------------------------------------------------
From: Muchun Song <songmuchun@bytedance.com>
Subject: powerpc/mm: switch device DAX to shared tail vmemmap pages
Date: Fri, 11 Sep 2026 13:02:25 +0800
The powerpc radix compound vmemmap population path still finds a reusable
tail page by walking the vmemmap page tables.
Switch it to the common vmemmap_shared_tail_page() helper instead, so it
can use the shared vmemmap page directly to simplify the code.
This removes the powerpc-specific tail-page lookup and its fallback path
and aligns the device DAX vmemmap optimization path with HugeTLB.
Link: https://lore.kernel.org/20260911050228.58884-9-songmuchun@bytedance.com
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Cc: David Hildenbrand (Arm) <david@kernel.org>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Lorenzo Stoakes <ljs@kernel.org>
Cc: Madhavan Srinivasan <maddy@linux.ibm.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: Nicholas Piggin <npiggin@gmail.com>
Cc: Oscar Salvador <osalvador@suse.de>
Cc: Qi Zheng <qi.zheng@linux.dev>
Cc: Randy Dunlap <rdunlap@infradead.org>
---
arch/powerpc/mm/book3s64/radix_pgtable.c | 80 ++-------------------
1 file changed, 9 insertions(+), 71 deletions(-)
--- a/arch/powerpc/mm/book3s64/radix_pgtable.c~powerpc-mm-switch-device-dax-to-shared-tail-vmemmap-pages
+++ a/arch/powerpc/mm/book3s64/radix_pgtable.c
@@ -19,6 +19,7 @@
#include <linux/string_helpers.h>
#include <linux/memory.h>
#include <linux/kfence.h>
+#include <linux/vmemmap-optimization.h>
#include <asm/pgalloc.h>
#include <asm/mmu_context.h>
@@ -1250,59 +1251,6 @@ static pte_t * __meminit radix__vmemmap_
return pte;
}
-static pte_t * __meminit vmemmap_compound_tail_page(unsigned long addr,
- unsigned long pfn_offset, int node)
-{
- pgd_t *pgd;
- p4d_t *p4d;
- pud_t *pud;
- pmd_t *pmd;
- pte_t *pte;
- unsigned long map_addr;
-
- /* the second vmemmap page which we use for duplication */
- map_addr = addr - pfn_offset * sizeof(struct page) + PAGE_SIZE;
- pgd = pgd_offset_k(map_addr);
- p4d = p4d_offset(pgd, map_addr);
- pud = vmemmap_pud_alloc(p4d, node, map_addr);
- if (!pud)
- return NULL;
- pmd = vmemmap_pmd_alloc(pud, node, map_addr);
- if (!pmd)
- return NULL;
- if (pmd_leaf(*pmd))
- /*
- * The second page is mapped as a hugepage due to a nearby request.
- * Force our mapping to page size without deduplication
- */
- return NULL;
- pte = vmemmap_pte_alloc(pmd, node, map_addr);
- if (!pte)
- return NULL;
- /*
- * Check if there exist a mapping to the left
- */
- if (pte_none(*pte)) {
- /*
- * Populate the head page vmemmap page.
- * It can fall in different pmd, hence
- * vmemmap_populate_address()
- */
- pte = radix__vmemmap_populate_address(map_addr - PAGE_SIZE, node, NULL, NULL);
- if (!pte)
- return NULL;
- /*
- * Populate the tail pages vmemmap page
- */
- pte = radix__vmemmap_pte_populate(pmd, map_addr, node, NULL, NULL);
- if (!pte)
- return NULL;
- vmemmap_verify(pte, node, map_addr, map_addr + PAGE_SIZE);
- return pte;
- }
- return pte;
-}
-
int __meminit vmemmap_populate_compound_pages(unsigned long start_pfn,
unsigned long start,
unsigned long end, int node,
@@ -1320,6 +1268,12 @@ int __meminit vmemmap_populate_compound_
pud_t *pud;
pmd_t *pmd;
pte_t *pte;
+ struct page *tail_page;
+ unsigned int order = pfn_to_section_compound_order(start_pfn);
+
+ tail_page = vmemmap_shared_tail_page(order, device_zone(node));
+ if (!tail_page)
+ return -ENOMEM;
for (addr = start; addr < end; addr = next) {
@@ -1349,10 +1303,9 @@ int __meminit vmemmap_populate_compound_
next = addr + PAGE_SIZE;
continue;
} else {
- unsigned long nr_pages = pgmap_vmemmap_nr(pgmap);
+ unsigned long nr_pages = 1UL << order;
unsigned long addr_pfn = page_to_pfn((struct page *)addr);
unsigned long pfn_offset = addr_pfn - ALIGN_DOWN(addr_pfn, nr_pages);
- pte_t *tail_page_pte;
/*
* if the address is aligned to huge page size it is the
@@ -1377,23 +1330,8 @@ int __meminit vmemmap_populate_compound_
next = addr + 2 * PAGE_SIZE;
continue;
}
- /*
- * get the 2nd mapping details
- * Also create it if that doesn't exist
- */
- tail_page_pte = vmemmap_compound_tail_page(addr, pfn_offset, node);
- if (!tail_page_pte) {
-
- pte = radix__vmemmap_pte_populate(pmd, addr, node, NULL, NULL);
- if (!pte)
- return -ENOMEM;
- vmemmap_verify(pte, node, addr, addr + PAGE_SIZE);
-
- next = addr + PAGE_SIZE;
- continue;
- }
- pte = radix__vmemmap_pte_populate(pmd, addr, node, NULL, pte_page(*tail_page_pte));
+ pte = radix__vmemmap_pte_populate(pmd, addr, node, NULL, tail_page);
if (!pte)
return -ENOMEM;
vmemmap_verify(pte, node, addr, addr + PAGE_SIZE);
_
Patches currently in -mm which might be from songmuchun@bytedance.com are
mm-sparse-relax-struct-mem_section-size-constraints.patch
mm-sparse-vmemmap-rename-hvo-order-macros.patch
mm-mm_init-skip-initializing-shared-vmemmap-tail-pages.patch
mm-sparse-vmemmap-initialize-shared-tail-vmemmap-pages-on-allocation.patch
mm-sparse-vmemmap-support-section-based-vmemmap-accounting.patch
mm-mm_init-factor-out-pfn_to_zone.patch
mm-sparse-vmemmap-move-helpers-ahead-of-future-callers.patch
mm-sparse-vmemmap-support-section-based-vmemmap-optimization.patch
mm-sparse-initialize-memory-sections-earlier.patch
mm-hugetlb-switch-hugetlb-to-section-based-vmemmap-optimization.patch
mm-sparse-vmemmap-remove-sparsemem_vmemmap_preinit-support.patch
mm-sparse-inline-usemap-allocation-into-sparse_init_nid.patch
mm-sparse-remove-section_map_size.patch
mm-hugetlb-remove-huge_bootmem_hvo.patch
mm-hugetlb-remove-huge_bootmem_cma.patch
mm-hugetlb-localize-struct-huge_bootmem_page.patch
mm-hugetlb-localize-huge_bootmem_zones_valid.patch
mm-sparse-vmemmap-drop-the-extra-tail-page-from-device-dax-reservation.patch
mm-sparse-vmemmap-drop-unused-section_nr_vmemmap_pages-arguments.patch
documentation-mm-update-dax-vmemmap-deduplication-docs.patch
^ permalink raw reply [flat|nested] 4+ messages in thread
* [to-be-updated] powerpc-mm-switch-device-dax-to-shared-tail-vmemmap-pages.patch removed from -mm tree
@ 2026-09-27 5:43 Andrew Morton
0 siblings, 0 replies; 4+ messages in thread
From: Andrew Morton @ 2026-09-27 5:43 UTC (permalink / raw)
To: mm-commits, songmuchun, akpm
The quilt patch titled
Subject: powerpc/mm: switch device DAX to shared tail vmemmap pages
has been removed from the -mm tree. Its filename was
powerpc-mm-switch-device-dax-to-shared-tail-vmemmap-pages.patch
This patch was dropped because an updated version will be issued
------------------------------------------------------
From: Muchun Song <songmuchun@bytedance.com>
Subject: powerpc/mm: switch device DAX to shared tail vmemmap pages
Date: Wed, 16 Sep 2026 14:43:38 +0800
The powerpc radix compound vmemmap population path still finds a reusable
tail page by walking the vmemmap page tables.
Switch it to the common vmemmap_shared_tail_page() helper instead, so it
can use the shared vmemmap page directly to simplify the code.
This removes the powerpc-specific tail-page lookup and its fallback path
and aligns the device DAX vmemmap optimization path with HugeTLB.
Link: https://lore.kernel.org/20260916064341.1825793-9-songmuchun@bytedance.com
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Cc: David Hildenbrand (Arm) <david@kernel.org>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Lorenzo Stoakes <ljs@kernel.org>
Cc: Madhavan Srinivasan <maddy@linux.ibm.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Mike Rapoport (Microsoft) <rppt@kernel.org>
Cc: Nicholas Piggin <npiggin@gmail.com>
Cc: Oscar Salvador <osalvador@suse.de>
Cc: Qi Zheng <qi.zheng@linux.dev>
Cc: Randy Dunlap <rdunlap@infradead.org>
---
arch/powerpc/mm/book3s64/radix_pgtable.c | 80 ++-------------------
1 file changed, 9 insertions(+), 71 deletions(-)
--- a/arch/powerpc/mm/book3s64/radix_pgtable.c~powerpc-mm-switch-device-dax-to-shared-tail-vmemmap-pages
+++ a/arch/powerpc/mm/book3s64/radix_pgtable.c
@@ -19,6 +19,7 @@
#include <linux/string_helpers.h>
#include <linux/memory.h>
#include <linux/kfence.h>
+#include <linux/vmemmap-optimization.h>
#include <asm/pgalloc.h>
#include <asm/mmu_context.h>
@@ -1250,59 +1251,6 @@ static pte_t * __meminit radix__vmemmap_
return pte;
}
-static pte_t * __meminit vmemmap_compound_tail_page(unsigned long addr,
- unsigned long pfn_offset, int node)
-{
- pgd_t *pgd;
- p4d_t *p4d;
- pud_t *pud;
- pmd_t *pmd;
- pte_t *pte;
- unsigned long map_addr;
-
- /* the second vmemmap page which we use for duplication */
- map_addr = addr - pfn_offset * sizeof(struct page) + PAGE_SIZE;
- pgd = pgd_offset_k(map_addr);
- p4d = p4d_offset(pgd, map_addr);
- pud = vmemmap_pud_alloc(p4d, node, map_addr);
- if (!pud)
- return NULL;
- pmd = vmemmap_pmd_alloc(pud, node, map_addr);
- if (!pmd)
- return NULL;
- if (pmd_leaf(*pmd))
- /*
- * The second page is mapped as a hugepage due to a nearby request.
- * Force our mapping to page size without deduplication
- */
- return NULL;
- pte = vmemmap_pte_alloc(pmd, node, map_addr);
- if (!pte)
- return NULL;
- /*
- * Check if there exist a mapping to the left
- */
- if (pte_none(*pte)) {
- /*
- * Populate the head page vmemmap page.
- * It can fall in different pmd, hence
- * vmemmap_populate_address()
- */
- pte = radix__vmemmap_populate_address(map_addr - PAGE_SIZE, node, NULL, NULL);
- if (!pte)
- return NULL;
- /*
- * Populate the tail pages vmemmap page
- */
- pte = radix__vmemmap_pte_populate(pmd, map_addr, node, NULL, NULL);
- if (!pte)
- return NULL;
- vmemmap_verify(pte, node, map_addr, map_addr + PAGE_SIZE);
- return pte;
- }
- return pte;
-}
-
int __meminit vmemmap_populate_compound_pages(unsigned long start_pfn,
unsigned long start,
unsigned long end, int node,
@@ -1320,6 +1268,12 @@ int __meminit vmemmap_populate_compound_
pud_t *pud;
pmd_t *pmd;
pte_t *pte;
+ struct page *tail_page;
+ unsigned int order = pfn_to_section_compound_order(start_pfn);
+
+ tail_page = vmemmap_shared_tail_page(order, device_zone(node));
+ if (!tail_page)
+ return -ENOMEM;
for (addr = start; addr < end; addr = next) {
@@ -1349,10 +1303,9 @@ int __meminit vmemmap_populate_compound_
next = addr + PAGE_SIZE;
continue;
} else {
- unsigned long nr_pages = pgmap_vmemmap_nr(pgmap);
+ unsigned long nr_pages = 1UL << order;
unsigned long addr_pfn = page_to_pfn((struct page *)addr);
unsigned long pfn_offset = addr_pfn - ALIGN_DOWN(addr_pfn, nr_pages);
- pte_t *tail_page_pte;
/*
* if the address is aligned to huge page size it is the
@@ -1377,23 +1330,8 @@ int __meminit vmemmap_populate_compound_
next = addr + 2 * PAGE_SIZE;
continue;
}
- /*
- * get the 2nd mapping details
- * Also create it if that doesn't exist
- */
- tail_page_pte = vmemmap_compound_tail_page(addr, pfn_offset, node);
- if (!tail_page_pte) {
-
- pte = radix__vmemmap_pte_populate(pmd, addr, node, NULL, NULL);
- if (!pte)
- return -ENOMEM;
- vmemmap_verify(pte, node, addr, addr + PAGE_SIZE);
-
- next = addr + PAGE_SIZE;
- continue;
- }
- pte = radix__vmemmap_pte_populate(pmd, addr, node, NULL, pte_page(*tail_page_pte));
+ pte = radix__vmemmap_pte_populate(pmd, addr, node, NULL, tail_page);
if (!pte)
return -ENOMEM;
vmemmap_verify(pte, node, addr, addr + PAGE_SIZE);
_
Patches currently in -mm which might be from songmuchun@bytedance.com are
mm-sparse-relax-struct-mem_section-size-constraints.patch
mm-sparse-vmemmap-rename-hvo-order-macros.patch
mm-mm_init-skip-initializing-shared-vmemmap-tail-pages.patch
mm-sparse-vmemmap-initialize-shared-tail-vmemmap-pages-on-allocation.patch
mm-sparse-vmemmap-support-section-based-vmemmap-accounting.patch
mm-mm_init-factor-out-pfn_to_zone.patch
mm-sparse-vmemmap-move-helpers-ahead-of-future-callers.patch
mm-sparse-vmemmap-support-section-based-vmemmap-optimization.patch
mm-sparse-initialize-memory-sections-earlier.patch
mm-hugetlb-switch-hugetlb-to-section-based-vmemmap-optimization.patch
mm-sparse-vmemmap-remove-sparsemem_vmemmap_preinit-support.patch
mm-sparse-inline-usemap-allocation-into-sparse_init_nid.patch
mm-sparse-remove-section_map_size.patch
mm-hugetlb-remove-huge_bootmem_hvo.patch
mm-hugetlb-remove-huge_bootmem_cma.patch
mm-hugetlb-localize-struct-huge_bootmem_page.patch
mm-hugetlb-localize-huge_bootmem_zones_valid.patch
mm-sparse-vmemmap-drop-the-extra-tail-page-from-device-dax-reservation.patch
mm-sparse-vmemmap-drop-unused-section_nr_vmemmap_pages-arguments.patch
documentation-mm-update-dax-vmemmap-deduplication-docs.patch
^ permalink raw reply [flat|nested] 4+ messages in thread
* [to-be-updated] powerpc-mm-switch-device-dax-to-shared-tail-vmemmap-pages.patch removed from -mm tree
@ 2026-09-30 18:40 Andrew Morton
0 siblings, 0 replies; 4+ messages in thread
From: Andrew Morton @ 2026-09-30 18:40 UTC (permalink / raw)
To: mm-commits, muchun.song, akpm
The quilt patch titled
Subject: powerpc/mm: switch device DAX to shared tail vmemmap pages
has been removed from the -mm tree. Its filename was
powerpc-mm-switch-device-dax-to-shared-tail-vmemmap-pages.patch
This patch was dropped because an updated version will be issued
------------------------------------------------------
From: Muchun Song <muchun.song@linux.dev>
Subject: powerpc/mm: switch device DAX to shared tail vmemmap pages
Date: Sun, 27 Sep 2026 10:54:38 +0800
The powerpc radix compound vmemmap population path still finds a reusable
tail page by walking the vmemmap page tables.
Switch it to the common vmemmap_shared_tail_page() helper instead, so it
can use the shared vmemmap page directly to simplify the code.
This removes the powerpc-specific tail-page lookup and its fallback path
and aligns the device DAX vmemmap optimization path with HugeTLB.
Link: https://lore.kernel.org/20260927025441.741633-10-songmuchun@bytedance.com
Signed-off-by: Muchun Song <muchun.song@linux.dev>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Cc: David Hildenbrand <david@kernel.org>
Cc: Oscar Salvador <osalvador@suse.de>
Cc: Madhavan Srinivasan <maddy@linux.ibm.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Lorenzo Stoakes <ljs@kernel.org>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: Qi Zheng <qi.zheng@linux.dev>
Cc: Nicholas Piggin <npiggin@gmail.com>
Cc: Christophe Leroy <chleroy@kernel.org>
Cc: Randy Dunlap <rdunlap@infradead.org>
Cc: Lance Yang <lance.yang@linux.dev>
---
arch/powerpc/mm/book3s64/radix_pgtable.c | 80 ++-------------------
include/linux/vmemmap-optimization.h | 6 +
mm/sparse-vmemmap.c | 6 -
3 files changed, 15 insertions(+), 77 deletions(-)
--- a/arch/powerpc/mm/book3s64/radix_pgtable.c~powerpc-mm-switch-device-dax-to-shared-tail-vmemmap-pages
+++ a/arch/powerpc/mm/book3s64/radix_pgtable.c
@@ -19,6 +19,7 @@
#include <linux/string_helpers.h>
#include <linux/memory.h>
#include <linux/kfence.h>
+#include <linux/vmemmap-optimization.h>
#include <asm/pgalloc.h>
#include <asm/mmu_context.h>
@@ -1250,59 +1251,6 @@ static pte_t * __meminit radix__vmemmap_
return pte;
}
-static pte_t * __meminit vmemmap_compound_tail_page(unsigned long addr,
- unsigned long pfn_offset, int node)
-{
- pgd_t *pgd;
- p4d_t *p4d;
- pud_t *pud;
- pmd_t *pmd;
- pte_t *pte;
- unsigned long map_addr;
-
- /* the second vmemmap page which we use for duplication */
- map_addr = addr - pfn_offset * sizeof(struct page) + PAGE_SIZE;
- pgd = pgd_offset_k(map_addr);
- p4d = p4d_offset(pgd, map_addr);
- pud = vmemmap_pud_alloc(p4d, node, map_addr);
- if (!pud)
- return NULL;
- pmd = vmemmap_pmd_alloc(pud, node, map_addr);
- if (!pmd)
- return NULL;
- if (pmd_leaf(*pmd))
- /*
- * The second page is mapped as a hugepage due to a nearby request.
- * Force our mapping to page size without deduplication
- */
- return NULL;
- pte = vmemmap_pte_alloc(pmd, node, map_addr);
- if (!pte)
- return NULL;
- /*
- * Check if there exist a mapping to the left
- */
- if (pte_none(*pte)) {
- /*
- * Populate the head page vmemmap page.
- * It can fall in different pmd, hence
- * vmemmap_populate_address()
- */
- pte = radix__vmemmap_populate_address(map_addr - PAGE_SIZE, node, NULL, NULL);
- if (!pte)
- return NULL;
- /*
- * Populate the tail pages vmemmap page
- */
- pte = radix__vmemmap_pte_populate(pmd, map_addr, node, NULL, NULL);
- if (!pte)
- return NULL;
- vmemmap_verify(pte, node, map_addr, map_addr + PAGE_SIZE);
- return pte;
- }
- return pte;
-}
-
int __meminit vmemmap_populate_compound_pages(unsigned long start_pfn,
unsigned long start,
unsigned long end, int node,
@@ -1320,6 +1268,12 @@ int __meminit vmemmap_populate_compound_
pud_t *pud;
pmd_t *pmd;
pte_t *pte;
+ struct page *tail_page;
+ unsigned int order = pfn_to_section_compound_order(start_pfn);
+
+ tail_page = vmemmap_shared_tail_page(order, device_zone(node));
+ if (!tail_page)
+ return -ENOMEM;
for (addr = start; addr < end; addr = next) {
@@ -1349,10 +1303,9 @@ int __meminit vmemmap_populate_compound_
next = addr + PAGE_SIZE;
continue;
} else {
- unsigned long nr_pages = pgmap_vmemmap_nr(pgmap);
+ unsigned long nr_pages = 1UL << order;
unsigned long addr_pfn = page_to_pfn((struct page *)addr);
unsigned long pfn_offset = addr_pfn - ALIGN_DOWN(addr_pfn, nr_pages);
- pte_t *tail_page_pte;
/*
* if the address is aligned to huge page size it is the
@@ -1377,23 +1330,8 @@ int __meminit vmemmap_populate_compound_
next = addr + 2 * PAGE_SIZE;
continue;
}
- /*
- * get the 2nd mapping details
- * Also create it if that doesn't exist
- */
- tail_page_pte = vmemmap_compound_tail_page(addr, pfn_offset, node);
- if (!tail_page_pte) {
-
- pte = radix__vmemmap_pte_populate(pmd, addr, node, NULL, NULL);
- if (!pte)
- return -ENOMEM;
- vmemmap_verify(pte, node, addr, addr + PAGE_SIZE);
-
- next = addr + PAGE_SIZE;
- continue;
- }
- pte = radix__vmemmap_pte_populate(pmd, addr, node, NULL, pte_page(*tail_page_pte));
+ pte = radix__vmemmap_pte_populate(pmd, addr, node, NULL, tail_page);
if (!pte)
return -ENOMEM;
vmemmap_verify(pte, node, addr, addr + PAGE_SIZE);
--- a/include/linux/vmemmap-optimization.h~powerpc-mm-switch-device-dax-to-shared-tail-vmemmap-pages
+++ a/include/linux/vmemmap-optimization.h
@@ -83,6 +83,12 @@ static inline unsigned int pfn_to_sectio
{
return 0;
}
+
+static inline struct page *vmemmap_shared_tail_page(unsigned int order,
+ struct zone *zone)
+{
+ return NULL;
+}
#endif /* CONFIG_VMEMMAP_OPTIMIZATION */
static inline bool vmemmap_optimizable_pfn(unsigned long pfn)
--- a/mm/sparse-vmemmap.c~powerpc-mm-switch-device-dax-to-shared-tail-vmemmap-pages
+++ a/mm/sparse-vmemmap.c
@@ -236,12 +236,6 @@ struct page __ref *vmemmap_shared_tail_p
return page;
}
-#else
-static inline struct page *vmemmap_shared_tail_page(unsigned int order,
- struct zone *zone)
-{
- return NULL;
-}
#endif
static __meminit void *vmemmap_alloc_pte(unsigned long pfn, int node,
_
Patches currently in -mm which might be from muchun.song@linux.dev are
mm-sparse-vmemmap-drop-the-extra-tail-page-from-device-dax-reservation.patch
mm-sparse-vmemmap-drop-unused-section_nr_vmemmap_pages-arguments.patch
documentation-mm-update-dax-vmemmap-deduplication-docs.patch
mm-sparse-vmemmap-drop-vmemmap_populate_dax.patch
mm-sparse-vmemmap-support-device-dax-in-common-vmemmap-path.patch
mm-sparse-vmemmap-drop-device-dax-specific-population-path.patch
mm-sparse-vmemmap-remove-the-unused-ptpfn-argument.patch
mm-sparse-vmemmap-open-code-vmemmap_populate_address.patch
mm-mm_init-add-zone-mismatch-warning-during-page-init.patch
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2026-09-30 18:40 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-09-30 18:40 [to-be-updated] powerpc-mm-switch-device-dax-to-shared-tail-vmemmap-pages.patch removed from -mm tree Andrew Morton
-- strict thread matches above, loose matches on Subject: below --
2026-09-27 5:43 Andrew Morton
2026-09-17 1:29 Andrew Morton
2026-09-10 23:07 Andrew Morton
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox