The patch titled
Subject: mm/memremap_pages: initialize all ZONE_DEVICE pages to start at refcount 0
has been added to the -mm mm-unstable branch. Its filename is
mm-memremap_pages-initialize-all-zone_device-pages-to-start-at-refcount-0.patch
This patch will shortly appear at
https://git.kernel.org/pub/scm/linux/kernel/git/akpm/25-new.git/tree/patches/mm-memremap_pages-initialize-all-zone_device-pages-to-start-at-refcount-0.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: Dan Williams <dan.j.williams@xxxxxxxxx>
Subject: mm/memremap_pages: initialize all ZONE_DEVICE pages to start at refcount 0
Date: Fri, 14 Oct 2022 16:59:12 -0700
The initial memremap_pages() implementation inherited the
__init_single_page() default of pages starting life with an elevated
reference count. This originally allowed for the page->pgmap pointer to
alias with the storage for page->lru since a page was only allowed to be
on an lru list when its reference count was zero.
Since then, 'struct page' definition cleanups have arranged for dedicated
space for the ZONE_DEVICE page metadata, the
MEMORY_DEVICE_{PRIVATE,COHERENT} work has arranged for the 1 -> 0
page->_refcount transition to route the page to free_zone_device_page()
and not the core-mm page-free, and MEMORY_DEVICE_{PRIVATE,COHERENT} now
arranges for its ZONE_DEVICE pages to start at _refcount 0. With those
cleanups in place and with filesystem-dax and device-dax now converted to
take and drop references at map and truncate time, it is possible to start
MEMORY_DEVICE_FS_DAX and MEMORY_DEVICE_GENERIC reference counts at 0 as
well.
This conversion also unifies all @pgmap accounting to be relative to
pgmap_request_folio() and the paired folio_put() calls for those requested
folios. This allows pgmap_release_folios() to be simplified to just a
folio_put() helper.
Link: https://lkml.kernel.org/r/166579195218.2236710.8731183545033177929.stgit@xxxxxxxxxxxxxxxxxxxxxxxxx
Signed-off-by: Dan Williams <dan.j.williams@xxxxxxxxx>
Cc: Matthew Wilcox <willy@xxxxxxxxxxxxx>
Cc: Jan Kara <jack@xxxxxxx>
Cc: "Darrick J. Wong" <djwong@xxxxxxxxxx>
Cc: Christoph Hellwig <hch@xxxxxx>
Cc: John Hubbard <jhubbard@xxxxxxxxxx>
Cc: Alistair Popple <apopple@xxxxxxxxxx>
Cc: Jason Gunthorpe <jgg@xxxxxxxxxx>
Cc: Alex Deucher <alexander.deucher@xxxxxxx>
Cc: Ben Skeggs <bskeggs@xxxxxxxxxx>
Cc: "Christian König" <christian.koenig@xxxxxxx>
Cc: Daniel Vetter <daniel@xxxxxxxx>
Cc: Dave Chinner <david@xxxxxxxxxxxxx>
Cc: David Airlie <airlied@xxxxxxxx>
Cc: Felix Kuehling <Felix.Kuehling@xxxxxxx>
Cc: Jerome Glisse <jglisse@xxxxxxxxxx>
Cc: Karol Herbst <kherbst@xxxxxxxxxx>
Cc: kernel test robot <lkp@xxxxxxxxx>
Cc: Lyude Paul <lyude@xxxxxxxxxx>
Cc: "Pan, Xinhui" <Xinhui.Pan@xxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
drivers/dax/mapping.c | 2 +-
include/linux/dax.h | 2 +-
include/linux/memremap.h | 6 ++----
mm/memremap.c | 36 ++++++++++++++++--------------------
mm/page_alloc.c | 9 +--------
5 files changed, 21 insertions(+), 34 deletions(-)
--- a/drivers/dax/mapping.c~mm-memremap_pages-initialize-all-zone_device-pages-to-start-at-refcount-0
+++ a/drivers/dax/mapping.c
@@ -691,7 +691,7 @@ static struct page *dax_zap_pages(struct
dax_for_each_folio(entry, folio, i) {
if (zap)
- pgmap_release_folios(folio_pgmap(folio), folio, 1);
+ pgmap_release_folios(folio, 1);
if (!ret && !dax_folio_idle(folio))
ret = folio_page(folio, 0);
}
--- a/include/linux/dax.h~mm-memremap_pages-initialize-all-zone_device-pages-to-start-at-refcount-0
+++ a/include/linux/dax.h
@@ -235,7 +235,7 @@ static inline void dax_unlock_mapping_en
*/
static inline bool dax_page_idle(struct page *page)
{
- return page_ref_count(page) == 1;
+ return page_ref_count(page) == 0;
}
static inline bool dax_folio_idle(struct folio *folio)
--- a/include/linux/memremap.h~mm-memremap_pages-initialize-all-zone_device-pages-to-start-at-refcount-0
+++ a/include/linux/memremap.h
@@ -195,8 +195,7 @@ struct dev_pagemap *get_dev_pagemap(unsi
struct dev_pagemap *pgmap);
bool pgmap_request_folios(struct dev_pagemap *pgmap, struct folio *folio,
int nr_folios);
-void pgmap_release_folios(struct dev_pagemap *pgmap, struct folio *folio,
- int nr_folios);
+void pgmap_release_folios(struct folio *folio, int nr_folios);
bool pgmap_pfn_valid(struct dev_pagemap *pgmap, unsigned long pfn);
unsigned long vmem_altmap_offset(struct vmem_altmap *altmap);
@@ -238,8 +237,7 @@ static inline bool pgmap_request_folios(
return false;
}
-static inline void pgmap_release_folios(struct dev_pagemap *pgmap,
- struct folio *folio, int nr_folios)
+static inline void pgmap_release_folios(struct folio *folio, int nr_folios)
{
}
--- a/mm/memremap.c~mm-memremap_pages-initialize-all-zone_device-pages-to-start-at-refcount-0
+++ a/mm/memremap.c
@@ -469,8 +469,10 @@ EXPORT_SYMBOL_GPL(get_dev_pagemap);
void free_zone_device_page(struct page *page)
{
- if (WARN_ON_ONCE(!page->pgmap->ops || !page->pgmap->ops->page_free))
- return;
+ struct dev_pagemap *pgmap = page->pgmap;
+
+ /* wake filesystem 'break dax layouts' waiters */
+ wake_up_var(page);
mem_cgroup_uncharge(page_folio(page));
@@ -505,17 +507,9 @@ void free_zone_device_page(struct page *
* to clear page->mapping.
*/
page->mapping = NULL;
- page->pgmap->ops->page_free(page);
-
- if (page->pgmap->type != MEMORY_DEVICE_PRIVATE &&
- page->pgmap->type != MEMORY_DEVICE_COHERENT)
- /*
- * Reset the page count to 1 to prepare for handing out the page
- * again.
- */
- set_page_count(page, 1);
- else
- put_dev_pagemap(page->pgmap);
+ if (pgmap->ops && pgmap->ops->page_free)
+ pgmap->ops->page_free(page);
+ put_dev_pagemap(page->pgmap);
}
static bool folio_span_valid(struct dev_pagemap *pgmap, struct folio *folio,
@@ -576,17 +570,19 @@ bool pgmap_request_folios(struct dev_pag
}
EXPORT_SYMBOL_GPL(pgmap_request_folios);
-void pgmap_release_folios(struct dev_pagemap *pgmap, struct folio *folio, int nr_folios)
+/*
+ * A symmetric helper to undo the page references acquired by
+ * pgmap_request_folios(), but the caller can also just arrange
+ * folio_put() on all the folios it acquired previously for the same
+ * effect.
+ */
+void pgmap_release_folios(struct folio *folio, int nr_folios)
{
struct folio *iter;
int i;
- for (iter = folio, i = 0; i < nr_folios; iter = folio_next(iter), i++) {
- if (!put_devmap_managed_page(&iter->page))
- folio_put(iter);
- if (!folio_ref_count(iter))
- put_dev_pagemap(pgmap);
- }
+ for (iter = folio, i = 0; i < nr_folios; iter = folio_next(folio), i++)
+ folio_put(iter);
}
#ifdef CONFIG_FS_DAX
--- a/mm/page_alloc.c~mm-memremap_pages-initialize-all-zone_device-pages-to-start-at-refcount-0
+++ a/mm/page_alloc.c
@@ -6791,6 +6791,7 @@ static void __ref __init_zone_device_pag
{
__init_single_page(page, pfn, zone_idx, nid);
+ set_page_count(page, 0);
/*
* Mark page reserved as it will need to wait for onlining
@@ -6823,14 +6824,6 @@ static void __ref __init_zone_device_pag
set_pageblock_migratetype(page, MIGRATE_MOVABLE);
cond_resched();
}
-
- /*
- * ZONE_DEVICE pages are released directly to the driver page allocator
- * which will set the page count to 1 when allocating the page.
- */
- if (pgmap->type == MEMORY_DEVICE_PRIVATE ||
- pgmap->type == MEMORY_DEVICE_COHERENT)
- set_page_count(page, 0);
}
/*
_
Patches currently in -mm which might be from dan.j.williams@xxxxxxxxx are
fsdax-wait-on-page-not-page-_refcount.patch
fsdax-use-dax_page_idle-to-document-dax-busy-page-checking.patch
fsdax-include-unmapped-inodes-for-page-idle-detection.patch
fsdax-introduce-dax_zap_mappings.patch
fsdax-wait-for-pinned-pages-during-truncate_inode_pages_final.patch
fsdax-validate-dax-layouts-broken-before-truncate.patch
fsdax-hold-dax-lock-over-mapping-insertion.patch
fsdax-update-dax_insert_entry-calling-convention-to-return-an-error.patch
fsdax-rework-for_each_mapped_pfn-to-dax_for_each_folio.patch
fsdax-introduce-pgmap_request_folios.patch
fsdax-rework-dax_insert_entry-calling-convention.patch
fsdax-cleanup-dax_associate_entry.patch
devdax-minor-warning-fixups.patch
devdax-fix-sparse-lock-imbalance-warning.patch
libnvdimm-pmem-support-pmem-block-devices-without-dax.patch
devdax-move-address_space-helpers-to-the-dax-core.patch
devdax-sparse-fixes-for-xarray-locking.patch
devdax-sparse-fixes-for-vmfault_t-dax-entry-conversions.patch
devdax-sparse-fixes-for-vm_fault_t-in-tracepoints.patch
devdax-add-pud-support-to-the-dax-mapping-infrastructure.patch
devdax-use-dax_insert_entry-dax_delete_mapping_entry.patch
mm-memremap_pages-replace-zone_device_page_init-with-pgmap_request_folios.patch
mm-memremap_pages-initialize-all-zone_device-pages-to-start-at-refcount-0.patch
mm-meremap_pages-delete-put_devmap_managed_page_refs.patch
mm-gup-drop-dax-pgmap-accounting.patch
