[PATCH v9 0/8] mm: optimize zone-device memmap initialization

[PATCH v9 0/8] mm: optimize zone-device memmap initialization

[Li Zhe]

memmap_init_zone_device() can take a noticeable amount of time when large
pmem namespaces are bound or rebound, because it initializes nearly
identical struct page descriptors one PFN at a time. This series reduces
that ZONE_DEVICE memmap initialization overhead by reusing prepared
struct page templates and, on x86, using memcpy_nontemporal() for the
template copy path.

The main target is large fsdax/devdax pmem configurations, where the
cost of initializing the memmap shows up directly in nd_pmem/dax_pmem
bind and rebind latency. This matters because the cost is paid in the
synchronous probe/bind path for large DAX/PMEM ZONE_DEVICE mappings.
Userspace workflows such as provisioning or reconfiguring
nd_pmem/dax_pmem namespaces, bringing hot-added PMEM-backed capacity
online, and recovering or rebinding a device after driver or device
changes all wait for this initialization to finish. Reducing this cost
will yield benefits as lower user-visible provisioning, hot-add,
recovery, and rebind latency for large DAX/PMEM devices.

Patches 1-3 are preparatory cleanups and helper extraction. Patches 4-5
add the template-copy path for head pages and compound tails. Patch 6
introduces memcpy_nontemporal(). Patch 7 switches the ZONE_DEVICE
template-copy path over to memcpy_nontemporal(). Patch 8 extends the x86
fixed-size memcpy_flushcache() inline cases used by the x86
memcpy_nontemporal() backend for struct page sized copies.

Architectures without a specialized memcpy_nontemporal() backend fall
back to memcpy(), so the generic template-copy optimization remains
available without arch-specific support. On x86, memcpy_nontemporal()
maps to the existing memcpy_flushcache() backend and can use the
fixed-size MOVNTI paths added by this series for struct page sized
copies.

memcpy_nontemporal() is only a copy primitive. It does not imply a drain
or a publication barrier. Callers that use it before a producer-consumer
or device-visible handoff must provide the required ordering. The
ZONE_DEVICE template-copy path uses it only while initializing struct
page metadata, so the copy primitive itself does not grow a separate
drain contract.

The numbers below measure the time spent in memmap_init_zone_device()
during driver bind/rebind. They are not measurements of the full
nd_pmem or dax_pmem bind/rebind operation.

Tested in a VM with a 100 GB fsdax namespace device configured with
map=dev and a 100 GB devdax namespace (align=2097152) on Intel Ice Lake
server.

Test procedure:
Rebind the nd_pmem and dax_pmem drivers 30 times and collect the memmap
initialization time from the pr_debug() output of
memmap_init_zone_device().

Base(v7.2-rc1):
  Average of nd_pmem rebinds: 244.28 ms
  Average of dax_pmem rebinds: 273.31 ms

With this series applied:
  Average of nd_pmem rebinds: 96.79 ms
  Average of dax_pmem rebinds: 119.04 ms

This reduces the average memmap initialization time measured during
rebind by about 60.4% for nd_pmem and 56.4% for dax_pmem.

As an additional x86_64 data point, I also ran a smaller set of
measurements on the same physical host with a 100 GB PMEM region created
via the memmap= kernel command line, configured as fsdax and devdax
namespaces with map=dev and 2 MiB alignment.

For brevity, the individual patches keep only the VM results rather than
including a second set of physical-host measurements throughout the
series. The physical-host numbers below are included only as
supplemental evidence that the same optimization also provides a similar
benefit on a non-virtualized system.

Test procedure:
Reconfigure the namespace mode, rebind the nd_pmem or dax_pmem driver
once, and collect the memmap initialization time from the pr_debug()
output of memmap_init_zone_device().

Base (v7.2-rc1):
  nd_pmem / fsdax: 179 ms
  dax_pmem / devdax: 264 ms

With this series applied:
  nd_pmem / fsdax: 82 ms
  dax_pmem / devdax: 113 ms

This reduces the measured memmap initialization time during rebind by
about 54.2% for nd_pmem and 57.2% for dax_pmem on that setup, which is
broadly consistent with the VM results above.

As another supplemental data point, I measured the test_hmm.ko module on
the same physical x86_64 host, using the test_hmm.ko setup from the
previous discussion that times ten 64 GB
memremap_pages()/memunmap_pages() iterations during module insertion[1].
By default, module insertion initializes two DEVICE_PRIVATE dmirror
devices, so two avg memremap values are reported; each value is the
average for one 64 GB chunk.

This is not the primary target workload of the series, but it exercises
the same large ZONE_DEVICE memmap initialization path and shows the same
direction of improvement.

Base (v7.2-rc1):
  avg memremap reported during module insertion: 116689362 ns, 116539263 ns

With this series applied:
  avg memremap reported during module insertion: 54607108 ns, 54458236 ns

This corresponds to about a 53.2% reduction based on the mean of the
reported values, which is again consistent with the pmem bind/rebind
results above.

I also tested the generic template-copy part on an arm64 QEMU virt VM
with 64 KB pages and a 100 GB ACPI NVDIMM sparse backend. This setup
does not use the x86 MOVNTI fast paths, so it exercises the
architecture-independent part of the optimization.

For devdax, 2 MiB alignment is rejected in this 64 KB page setup, so the
devdax namespace was tested with the supported default 512 MiB
alignment.

Base (v7.2-rc1):
  Average of rebinds for nd_pmem driver: 25.60 ms
  Average of rebinds for dax_pmem driver: 25.60 ms

With this series applied:
  Average of rebinds for nd_pmem driver: 11.07 ms
  Average of rebinds for dax_pmem driver: 13.20 ms

This reduces the average memmap initialization time measured during
rebind by about 56.8% for nd_pmem and 48.4% for dax_pmem on that arm64
VM setup. Since this arm64 setup does not use the x86 MOVNTI fast paths,
the result also suggests that the generic template-copy optimization can
benefit architectures without an architecture-specific
memcpy_nontemporal() backend.

Li Zhe (8):
  mm: fix stale ZONE_DEVICE refcount comment
  mm: factor zone-device page init helpers out of
    __init_zone_device_page
  mm: add a set_page_section_from_pfn() helper
  mm: add a template-based fast path for zone-device page init
  mm: extend the template fast path to zone-device compound tails
  string: introduce memcpy_nontemporal()
  mm: use memcpy_nontemporal() in zone-device template copies
  x86/string: extend memcpy_flushcache() fixed-size fastpaths

 arch/x86/include/asm/string_64.h |  68 +++++++++++++++-
 include/linux/mm.h               |  15 +++-
 include/linux/string.h           |  13 +++
 mm/mm_init.c                     | 132 +++++++++++++++++++++++++------
 4 files changed, 200 insertions(+), 28 deletions(-)

---
v8: https://lore.kernel.org/all/20260727123429.5673-1-lizhe.67@bytedance.com/
v7: https://lore.kernel.org/all/20260720120259.1545-1-lizhe.67@bytedance.com/
v6: https://lore.kernel.org/all/20260709112520.24857-1-lizhe.67@bytedance.com/
v5: https://lore.kernel.org/all/20260701090553.62691-1-lizhe.67@bytedance.com/
v4: https://lore.kernel.org/all/20260603080152.64728-1-lizhe.67@bytedance.com/
v3: https://lore.kernel.org/all/20260527033636.28231-1-lizhe.67@bytedance.com/
v2: https://lore.kernel.org/all/20260521040124.10608-1-lizhe.67@bytedance.com/
v1: https://lore.kernel.org/all/20260515082045.63029-1-lizhe.67@bytedance.com/

Changelogs:

v8->v9:
- Fold the removal of the local non-template fallback into the relevant
  template-copy patches and drop the separate cleanup patch.
- Reorder the memcpy_nontemporal() use before the x86 fixed-size
  fastpath patch, so the x86 patch shows its incremental ZONE_DEVICE
  benefit directly.
- Rework the x86 fixed-size fastpath commit message to justify the
  struct page sized copies and report real ZONE_DEVICE initialization
  data instead of a standalone microbenchmark.
- Add arm64 QEMU nd_pmem map=dev and dax_pmem devdax measurements for
  the generic template-copy path.
- Refresh the per-patch performance data for patches 4, 5, 7, and 8.

For changelogs of earlier revisions, please refer to the v8 cover
letter.

-- 
2.20.1

[PATCH v9 1/8] mm: fix stale ZONE_DEVICE refcount comment

[Li Zhe]

The comment in __init_zone_device_page() still uses the old
MEMORY_TYPE_* names and implies that FS_DAX pages regain a
refcount of 1 in the free path. That no longer matches the code.

Update the comment to describe the current policy correctly:
MEMORY_DEVICE_GENERIC pages regain a refcount of 1 in the free path,
while the remaining ZONE_DEVICE types start from 0 here and raise the
count again when the allocator or driver hands the page out.

No functional change intended.

Signed-off-by: Li Zhe <lizhe.67@bytedance.com>
Reviewed-by: David Hildenbrand (Arm) <david@kernel.org>
Reviewed-by: Alistair Popple <apopple@nvidia.com>
Reviewed-by: Muchun Song <muchun.song@linux.dev>
---
 mm/mm_init.c | 10 +++-------
 1 file changed, 3 insertions(+), 7 deletions(-)

diff --git a/mm/mm_init.c b/mm/mm_init.c
index 0f64909e8d20..95808ab5cfdb 100644
--- a/mm/mm_init.c
+++ b/mm/mm_init.c
@@ -1030,13 +1030,9 @@ static void __ref __init_zone_device_page(struct page *page, unsigned long pfn,
 	page->zone_device_data = NULL;
 
 	/*
-	 * ZONE_DEVICE pages other than MEMORY_TYPE_GENERIC are released
-	 * directly to the driver page allocator which will set the page count
-	 * to 1 when allocating the page.
-	 *
-	 * MEMORY_TYPE_GENERIC and MEMORY_TYPE_FS_DAX pages automatically have
-	 * their refcount reset to one whenever they are freed (ie. after
-	 * their refcount drops to 0).
+	 * MEMORY_DEVICE_GENERIC pages regain a refcount of 1 in the free
+	 * path. The remaining ZONE_DEVICE types start from 0 here and raise
+	 * the count again when the allocator or driver hands the page out.
 	 */
 	switch (pgmap->type) {
 	case MEMORY_DEVICE_FS_DAX:
-- 
2.20.1

[PATCH v9 2/8] mm: factor zone-device page init helpers out of __init_zone_device_page

[Li Zhe]

memmap_init_zone_device() currently mixes refcount policy and core
ZONE_DEVICE page setup in a single helper.

Factor the refcount-reset predicate into pagemap_requires_refcount_reset(),
move the common page initialization into __zone_device_page_init(), and
wrap the existing slow path in zone_device_page_init_slow().

This keeps the slow-path behaviour unchanged and gives later patches
reusable helper boundaries.

No functional change intended.

Signed-off-by: Li Zhe <lizhe.67@bytedance.com>
Reviewed-by: Mike Rapoport (Microsoft) <rppt@kernel.org>
---
 mm/mm_init.c | 56 ++++++++++++++++++++++++++++++++++------------------
 1 file changed, 37 insertions(+), 19 deletions(-)

diff --git a/mm/mm_init.c b/mm/mm_init.c
index 95808ab5cfdb..a70acb7431a6 100644
--- a/mm/mm_init.c
+++ b/mm/mm_init.c
@@ -1005,11 +1005,37 @@ static void __init memmap_init(void)
 }
 
 #ifdef CONFIG_ZONE_DEVICE
-static void __ref __init_zone_device_page(struct page *page, unsigned long pfn,
+/*
+ * Return true when memmap_init_zone_device() must initialize the page
+ * refcount to 0. MEMORY_DEVICE_GENERIC pages regain a refcount of 1 in
+ * the free path, while the remaining ZONE_DEVICE types start from 0 here
+ * and raise the count again when the allocator or driver hands the page
+ * out.
+ */
+static inline bool pagemap_requires_refcount_reset(const struct dev_pagemap *pgmap)
+{
+	/*
+	 * MEMORY_DEVICE_GENERIC pages regain a refcount of 1 in the free
+	 * path. The remaining ZONE_DEVICE types start from 0 here and raise
+	 * the count again when the allocator or driver hands the page out.
+	 */
+	switch (pgmap->type) {
+	case MEMORY_DEVICE_FS_DAX:
+	case MEMORY_DEVICE_PRIVATE:
+	case MEMORY_DEVICE_COHERENT:
+	case MEMORY_DEVICE_PCI_P2PDMA:
+		return true;
+	case MEMORY_DEVICE_GENERIC:
+		return false;
+	}
+
+	return false;
+}
+
+static void __ref __zone_device_page_init(struct page *page, unsigned long pfn,
 					  unsigned long zone_idx, int nid,
 					  struct dev_pagemap *pgmap)
 {
-
 	__init_single_page(page, pfn, zone_idx, nid);
 
 	/*
@@ -1028,23 +1054,15 @@ static void __ref __init_zone_device_page(struct page *page, unsigned long pfn,
 	 */
 	page_folio(page)->pgmap = pgmap;
 	page->zone_device_data = NULL;
+}
 
-	/*
-	 * MEMORY_DEVICE_GENERIC pages regain a refcount of 1 in the free
-	 * path. The remaining ZONE_DEVICE types start from 0 here and raise
-	 * the count again when the allocator or driver hands the page out.
-	 */
-	switch (pgmap->type) {
-	case MEMORY_DEVICE_FS_DAX:
-	case MEMORY_DEVICE_PRIVATE:
-	case MEMORY_DEVICE_COHERENT:
-	case MEMORY_DEVICE_PCI_P2PDMA:
+static void __ref zone_device_page_init_slow(struct page *page,
+		unsigned long pfn, unsigned long zone_idx, int nid,
+		struct dev_pagemap *pgmap)
+{
+	__zone_device_page_init(page, pfn, zone_idx, nid, pgmap);
+	if (pagemap_requires_refcount_reset(pgmap))
 		set_page_count(page, 0);
-		break;
-
-	case MEMORY_DEVICE_GENERIC:
-		break;
-	}
 }
 
 /*
@@ -1090,7 +1108,7 @@ 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);
+		zone_device_page_init_slow(page, pfn, zone_idx, nid, pgmap);
 		prep_compound_tail(page, head, order);
 		set_page_count(page, 0);
 	}
@@ -1126,7 +1144,7 @@ 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);
+		zone_device_page_init_slow(page, pfn, zone_idx, nid, pgmap);
 
 		if (IS_ALIGNED(pfn, PAGES_PER_SECTION))
 			cond_resched();
-- 
2.20.1

[PATCH v9 3/8] mm: add a set_page_section_from_pfn() helper

[Li Zhe]

Callers that want to update section bits from a PFN currently need to
open-code:

	set_page_section(page, pfn_to_section_nr(pfn));

and guard that sequence with #ifdef SECTION_IN_PAGE_FLAGS.

Add set_page_section_from_pfn() to wrap that update in one place. When
section bits are stored in page flags, the helper derives the section
number from the PFN and updates the page flags. Otherwise keep it as a
no-op so callers can use one helper without open-coding
SECTION_IN_PAGE_FLAGS.

Convert set_page_links() to use the new helper so later ZONE_DEVICE
fast-path patches can also update section bits without open-coding
SECTION_IN_PAGE_FLAGS at each callsite.

This keeps the PFN-to-section translation local to the configurations
that actually store section bits in struct page flags, and avoids
exposing that detail to generic callers.

No functional change intended.

Signed-off-by: Li Zhe <lizhe.67@bytedance.com>
Reviewed-by: Mike Rapoport (Microsoft) <rppt@kernel.org>
Acked-by: Muchun Song <muchun.song@linux.dev>
Reviewed-by: Balbir Singh <balbirs@nvidia.com>
---
 include/linux/mm.h | 15 ++++++++++++---
 1 file changed, 12 insertions(+), 3 deletions(-)

diff --git a/include/linux/mm.h b/include/linux/mm.h
index 485df9c2dbdd..43343bfce493 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -2541,11 +2541,22 @@ static inline void set_page_section(struct page *page, unsigned long section)
 	page->flags.f |= (section & SECTIONS_MASK) << SECTIONS_PGSHIFT;
 }
 
+static inline void set_page_section_from_pfn(struct page *page,
+		unsigned long pfn)
+{
+	set_page_section(page, pfn_to_section_nr(pfn));
+}
+
 static inline unsigned long memdesc_section(memdesc_flags_t mdf)
 {
 	return (mdf.f >> SECTIONS_PGSHIFT) & SECTIONS_MASK;
 }
 #else /* !SECTION_IN_PAGE_FLAGS */
+static inline void set_page_section_from_pfn(struct page *page,
+		unsigned long pfn)
+{
+}
+
 static inline unsigned long memdesc_section(memdesc_flags_t mdf)
 {
 	return 0;
@@ -2768,9 +2779,7 @@ static inline void set_page_links(struct page *page, enum zone_type zone,
 {
 	set_page_zone(page, zone);
 	set_page_node(page, node);
-#ifdef SECTION_IN_PAGE_FLAGS
-	set_page_section(page, pfn_to_section_nr(pfn));
-#endif
+	set_page_section_from_pfn(page, pfn);
 }
 
 /**
-- 
2.20.1

[PATCH v9 4/8] mm: add a template-based fast path for zone-device page init

[Li Zhe]

memmap_init_zone_device() repeats nearly identical head-page
initialization for each PFN. Prepare one reusable ZONE_DEVICE head-page
template through the existing slow path, refresh the PFN-dependent
fields in that template before each copy, and memcpy it into each
destination page.

Use the template path unconditionally, as suggested by Muchun. The
page_ref_set tracepoint is primarily a debugging aid, while this code is
still initializing struct pages before they are handed out. From the
perspective of users of those pages, the initialization-time refcount
transitions are not part of the observable page lifetime.

This means page_ref_set will no longer observe every initialization-time
refcount assignment for copied ZONE_DEVICE head pages. The impact is
controlled because the final initialized struct page state is unchanged,
and keeping a separate non-template path only for this local tracepoint
observability would add complexity to the common path.

This patch accelerates head-page initialization. The pfns_per_compound
== 1 case gets the full benefit here, compound tails are handled in the
next patch.

Tested in a VM with a 100 GB fsdax namespace device configured with
map=dev on Intel Ice Lake server. This test exercises the nd_pmem rebind
path (pfns_per_compound == 1).

Test procedure:
Rebind the nd_pmem driver 30 times and collect the memmap initialization
time from the pr_debug() output of memmap_init_zone_device().

Base(v7.2-rc1):
  Average of rebinds for nd_pmem driver: 244.28 ms

With this patch and its prerequisites applied:
  Average of rebinds for nd_pmem driver: 215.55 ms

This reduces the average memmap initialization time measured during rebind
from 244.28 ms to 215.55 ms, or about 11%.

Suggested-by: Muchun Song <muchun.song@linux.dev>
Suggested-by: Mike Rapoport (Microsoft) <rppt@kernel.org>
Signed-off-by: Li Zhe <lizhe.67@bytedance.com>
---
 mm/mm_init.c | 47 ++++++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 46 insertions(+), 1 deletion(-)

diff --git a/mm/mm_init.c b/mm/mm_init.c
index a70acb7431a6..56a36a71ba89 100644
--- a/mm/mm_init.c
+++ b/mm/mm_init.c
@@ -1065,6 +1065,35 @@ static void __ref zone_device_page_init_slow(struct page *page,
 		set_page_count(page, 0);
 }
 
+/*
+ * 'template' is a reusable page prototype rather than a strictly immutable
+ * object. Most ZONE_DEVICE fields stay constant across the pages covered by
+ * the current template, but section bits and page->virtual may still depend
+ * on the PFN. Refresh those PFN-dependent fields in the template before
+ * copying it into @page.
+ */
+static inline void zone_device_page_update_template(struct page *template,
+		unsigned long pfn)
+{
+	set_page_section_from_pfn(template, pfn);
+#ifdef WANT_PAGE_VIRTUAL
+	if (!is_highmem_idx(ZONE_DEVICE))
+		set_page_address(template, __va(pfn << PAGE_SHIFT));
+#endif
+}
+
+static void zone_device_page_init_from_template(struct page *page,
+		unsigned long pfn, struct page *template)
+{
+	/*
+	 * 'template' carries the invariant portion of a ZONE_DEVICE struct
+	 * page. Update the PFN-dependent fields in place before copying it
+	 * to the destination page.
+	 */
+	zone_device_page_update_template(template, pfn);
+	memcpy(page, template, sizeof(*page));
+}
+
 /*
  * With compound page geometry and when struct pages are stored in ram most
  * tail pages are reused. Consequently, the amount of unique struct pages to
@@ -1127,6 +1156,7 @@ void __ref memmap_init_zone_device(struct zone *zone,
 	unsigned long zone_idx = zone_idx(zone);
 	unsigned long start = jiffies;
 	int nid = pgdat->node_id;
+	struct page template;
 
 	if (WARN_ON_ONCE(!pgmap || zone_idx != ZONE_DEVICE))
 		return;
@@ -1144,7 +1174,22 @@ 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);
 
-		zone_device_page_init_slow(page, pfn, zone_idx, nid, pgmap);
+		if (pfn == start_pfn) {
+			/*
+			 * Seed the reusable head-page template from the
+			 * first real struct page. This initializes the
+			 * first page through the existing slow path and
+			 * then reuses that final state as the template
+			 * for subsequent pages.
+			 */
+			zone_device_page_init_slow(page, pfn, zone_idx,
+						   nid, pgmap);
+			/* init template page */
+			memcpy(&template, page, sizeof(*page));
+		} else {
+			zone_device_page_init_from_template(page, pfn,
+							    &template);
+		}
 
 		if (IS_ALIGNED(pfn, PAGES_PER_SECTION))
 			cond_resched();
-- 
2.20.1

[PATCH v9 5/8] mm: extend the template fast path to zone-device compound tails

[Li Zhe]

The template fast path from the previous patch only accelerates head
pages. Compound tails in memmap_init_compound() still go through the
zone_device_page_init_slow() one by one.

Build separate head and tail templates and reuse one prepared tail
template across the tail pages in a compound range. Head pages preserve
the existing refcount policy, while compound tails always start with a
refcount of 0 after prep_compound_tail().

This extends the template-copy fast path to pfns_per_compound > 1
without changing the existing zone_device_page_init_slow() helper.
Tail-page PFN-dependent fields are refreshed in the reusable tail
template before each copy.

Do not keep a separate non-template fallback for compound tails either.
This follows the same simplification suggested by Muchun: these pages
are still under memmap initialization, and the initialization-time
refcount updates are not part of the observable lifetime of pages handed
out later.

The impact is controlled for the same reason as for head pages. The
first tail page still seeds the reusable tail template through the
normal tail initialization sequence, and the copied tail pages have the
same final initialized state except for the PFN-dependent fields
refreshed before each copy.

Tested in a VM with a 100 GB devdax namespace (align=2097152) on Intel
Ice Lake server. This test exercises the dax_pmem rebind path and
measures memmap initialization latency.

Test procedure:
Unbind and rebind the dax_pmem driver 30 times, collect memmap
initialization time from the pr_debug() output of memmap_init_zone_device().

Base(v7.2-rc1):
  Average of rebinds for dax_pmem driver: 273.31 ms

With this patch and its prerequisites applied:
  Average of rebinds for dax_pmem driver: 244.37 ms

This reduces the average memmap initialization time measured during rebind
from 273.31 ms to 244.37 ms, or about 10.6%.

Suggested-by: Muchun Song <muchun.song@linux.dev>
Suggested-by: Mike Rapoport (Microsoft) <rppt@kernel.org>
Signed-off-by: Li Zhe <lizhe.67@bytedance.com>
---
 mm/mm_init.c | 31 ++++++++++++++++++++++++++++---
 1 file changed, 28 insertions(+), 3 deletions(-)

diff --git a/mm/mm_init.c b/mm/mm_init.c
index 56a36a71ba89..9691fa2a060d 100644
--- a/mm/mm_init.c
+++ b/mm/mm_init.c
@@ -1065,6 +1065,16 @@ static void __ref zone_device_page_init_slow(struct page *page,
 		set_page_count(page, 0);
 }
 
+static inline void zone_device_tail_page_init(struct page *page,
+		unsigned long pfn, unsigned long zone_idx, int nid,
+		struct dev_pagemap *pgmap, const struct page *head,
+		unsigned int order)
+{
+	zone_device_page_init_slow(page, pfn, zone_idx, nid, pgmap);
+	prep_compound_tail(page, head, order);
+	set_page_count(page, 0);
+}
+
 /*
  * 'template' is a reusable page prototype rather than a strictly immutable
  * object. Most ZONE_DEVICE fields stay constant across the pages covered by
@@ -1126,6 +1136,7 @@ static void __ref memmap_init_compound(struct page *head,
 {
 	unsigned long pfn, end_pfn = head_pfn + nr_pages;
 	unsigned int order = pgmap->vmemmap_shift;
+	struct page template;
 
 	/*
 	 * We have to initialize the pages, including setting up page links.
@@ -1134,12 +1145,26 @@ static void __ref memmap_init_compound(struct page *head,
 	 * the pages in the same go.
 	 */
 	__SetPageHead(head);
+
 	for (pfn = head_pfn + 1; pfn < end_pfn; pfn++) {
 		struct page *page = pfn_to_page(pfn);
 
-		zone_device_page_init_slow(page, pfn, zone_idx, nid, pgmap);
-		prep_compound_tail(page, head, order);
-		set_page_count(page, 0);
+		if (pfn == head_pfn + 1) {
+			/*
+			 * All tails of the same compound page share the
+			 * state established by prep_compound_tail(). Reuse
+			 * one tail template for the whole range and
+			 * refresh only the PFN-dependent fields in that
+			 * template before each copy.
+			 */
+			zone_device_tail_page_init(page, pfn, zone_idx, nid,
+						   pgmap, head, order);
+			/* init template page */
+			memcpy(&template, page, sizeof(*page));
+		} else {
+			zone_device_page_init_from_template(page, pfn,
+							    &template);
+		}
 	}
 	prep_compound_head(head, order);
 }
-- 
2.20.1

[PATCH v9 6/8] string: introduce memcpy_nontemporal()

[Li Zhe]

Introduce memcpy_nontemporal() for write-once copy sites that want a
named non-temporal copy primitive.

On x86_64, override the helper in arch/x86/include/asm/string_64.h using
the usual self-macro pattern, next to the existing memcpy_flushcache()
backend that memcpy_nontemporal() wraps.

include/linux/string.h provides the generic memcpy_nontemporal()
fallback as

  #define memcpy_nontemporal(dst, src, len) \
	((void)memcpy(dst, src, len))

instead of an inline wrapper, so architectures without a specialized
backend keep the usual memcpy() FORTIFY coverage when the compiler can
still see object sizes at the original call site. It also makes the
memcpy_nontemporal() API uniformly void, matching memcpy_flushcache()
and the x86 backend, so callers cannot accidentally depend on a return
value on fallback architectures.

memcpy_nontemporal() is only a copy primitive. It does not imply a drain
or a publication barrier. Callers that use it before a producer-consumer
or device-visible handoff must provide the required ordering at that
handoff point.

The immediate user is the ZONE_DEVICE template-copy path. It populates
struct page descriptors in a write-once pattern, so a regular cached
memcpy() can incur avoidable write-allocate traffic and cache pollution
for data with little near-term reuse.

Signed-off-by: Li Zhe <lizhe.67@bytedance.com>
---
 arch/x86/include/asm/string_64.h | 12 ++++++++++++
 include/linux/string.h           | 13 +++++++++++++
 2 files changed, 25 insertions(+)

diff --git a/arch/x86/include/asm/string_64.h b/arch/x86/include/asm/string_64.h
index 4635616863f5..21ae515ae35a 100644
--- a/arch/x86/include/asm/string_64.h
+++ b/arch/x86/include/asm/string_64.h
@@ -100,6 +100,18 @@ static __always_inline void memcpy_flushcache(void *dst, const void *src, size_t
 	}
 	__memcpy_flushcache(dst, src, cnt);
 }
+
+#define memcpy_nontemporal memcpy_nontemporal
+/*
+ * Reuse the existing x86 flushcache backend as the non-temporal copy
+ * primitive.
+ */
+static __always_inline void memcpy_nontemporal(void *dst, const void *src,
+		size_t cnt)
+{
+	memcpy_flushcache(dst, src, cnt);
+}
+
 #endif
 
 #endif /* __KERNEL__ */
diff --git a/include/linux/string.h b/include/linux/string.h
index 5702daca4326..6cb5cdd01158 100644
--- a/include/linux/string.h
+++ b/include/linux/string.h
@@ -278,6 +278,19 @@ static inline void memcpy_flushcache(void *dst, const void *src, size_t cnt)
 }
 #endif
 
+#ifndef memcpy_nontemporal
+/*
+ * memcpy_nontemporal() requests a non-temporal copy when the
+ * architecture has a suitable backend. Architectures without a
+ * specialized backend fall back to memcpy(). Keep this as a
+ * function-like macro so the compiler can still see the original
+ * memcpy() call site and preserve the usual FORTIFY coverage when
+ * object sizes remain visible there, while keeping the API void.
+ */
+#define memcpy_nontemporal(dst, src, len) \
+	((void)memcpy(dst, src, len))
+#endif
+
 void *memchr_inv(const void *s, int c, size_t n);
 char *strreplace(char *str, char old, char new);
 
-- 
2.20.1

[PATCH v9 7/8] mm: use memcpy_nontemporal() in zone-device template copies

[Li Zhe]

The template fast path currently uses memcpy() for the actual struct
page copy. Switch zone_device_page_init_from_template() to
memcpy_nontemporal().

ZONE_DEVICE memmap initialization is largely write-once: each struct
page is populated once, and most destination cachelines are not expected
to be reused immediately afterwards. On x86, a regular cached memcpy()
can therefore incur write-allocate traffic by pulling destination
cachelines into the cache before writeback, and can populate the cache
with data that has little near-term reuse. Using memcpy_nontemporal()
lets this path request nontemporal stores for that copy pattern, which
can reduce cache pollution and avoid part of the associated
write-allocate overhead, while architectures without a specialized
backend still fall back to memcpy().

Do not add a KASAN/KMSAN-specific fallback around this call site. As
Muchun pointed out, special KASAN handling for memcpy_flushcache() or
memcpy_nontemporal(), if needed, belongs in the low-level helper rather
than in this ZONE_DEVICE caller.

No separate drain is added here. memcpy_nontemporal() is used only as
the copy primitive while memmap_init_zone_device() is still initializing
the struct page array. The ordinary stores that follow in this path,
such as compound-page setup, are part of the same CPU's initialization
sequence; they are not used as a publication store that tells another CPU
or device to consume data written by the non-temporal copy.

Therefore this call site does not need a helper-level drain for
correctness. Callers that use memcpy_nontemporal() as part of a
producer-consumer or device-visible handoff must add the required
ordering themselves.

Tested in a VM with a 100 GB fsdax namespace device configured with
map=dev and a 100 GB devdax namespace (align=2097152) on Intel Ice Lake
server.

Test procedure:
Rebind the nd_pmem and dax_pmem driver 30 times and collect the memmap
initialization time from the pr_debug() output of
memmap_init_zone_device().

Base(v7.2-rc1):
  Average of rebinds for nd_pmem driver: 244.28 ms
  Average of rebinds for dax_pmem driver: 273.31 ms

With this patch and its prerequisites applied:
  Average of rebinds for nd_pmem driver: 150.83 ms
  Average of rebinds for dax_pmem driver: 153.55 ms

This reduces the average memmap initialization time measured during rebind
by about 38.3% for nd_pmem and 43.8% for dax_pmem.

Signed-off-by: Li Zhe <lizhe.67@bytedance.com>
---
 mm/mm_init.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/mm/mm_init.c b/mm/mm_init.c
index 9691fa2a060d..bb2007806a28 100644
--- a/mm/mm_init.c
+++ b/mm/mm_init.c
@@ -1101,7 +1101,7 @@ static void zone_device_page_init_from_template(struct page *page,
 	 * to the destination page.
 	 */
 	zone_device_page_update_template(template, pfn);
-	memcpy(page, template, sizeof(*page));
+	memcpy_nontemporal(page, template, sizeof(*page));
 }
 
 /*
-- 
2.20.1

[PATCH v9 8/8] x86/string: extend memcpy_flushcache() fixed-size fastpaths

[Li Zhe]

The x86 memcpy_nontemporal() helper maps to memcpy_flushcache(), and the
ZONE_DEVICE template-copy path uses it to copy one struct page at a
time.

The relevant copy size is sizeof(struct page). On x86_64, the base struct
page layout is 64 bytes. Adding either the KMSAN metadata pointers or an
out-of-flags last_cpupid field can make it 80 bytes after alignment, and
enabling both can make it 96 bytes.

memcpy_flushcache() currently only has inline fixed-size cases for 4, 8,
and 16 bytes. As a result, these constant-sized struct page copies fall
through to __memcpy_flushcache() even though the compiler knows the copy
size at the call site.

Add fixed-size MOVNTI cases up to 96 bytes so the ZONE_DEVICE
template-copy path can keep these struct page copies in the inline
memcpy_flushcache() path.

This matters for ZONE_DEVICE memmap initialization because the copy
happens once per initialized struct page. For a 100 GB fsdax namespace
with map=dev, this is about 25 million struct page copies during nd_pmem
binding or rebinding.

Tested in a VM with a 100 GB fsdax namespace device configured with
map=dev and a 100 GB devdax namespace (align=2097152) on Intel Ice Lake
server.

Test procedure:
Rebind the nd_pmem and dax_pmem drivers 30 times and collect the memmap
initialization time from the pr_debug() output of
memmap_init_zone_device().

With memcpy_nontemporal() used by the ZONE_DEVICE template-copy path:
  Average of rebinds for nd_pmem driver: 150.83 ms
  Average of rebinds for dax_pmem driver: 153.55 ms

With this x86 fixed-size fastpath patch applied:
  Average of rebinds for nd_pmem driver: 96.79 ms
  Average of rebinds for dax_pmem driver: 119.04 ms

This further reduces the average memmap initialization time measured
during rebind by about 35.8% for nd_pmem and 22.5% for dax_pmem.

Signed-off-by: Li Zhe <lizhe.67@bytedance.com>
---
 arch/x86/include/asm/string_64.h | 56 +++++++++++++++++++++++++++++++-
 1 file changed, 55 insertions(+), 1 deletion(-)

diff --git a/arch/x86/include/asm/string_64.h b/arch/x86/include/asm/string_64.h
index 21ae515ae35a..bc6a9f34b346 100644
--- a/arch/x86/include/asm/string_64.h
+++ b/arch/x86/include/asm/string_64.h
@@ -82,7 +82,35 @@ int strcmp(const char *cs, const char *ct);
 #ifdef CONFIG_ARCH_HAS_UACCESS_FLUSHCACHE
 #define __HAVE_ARCH_MEMCPY_FLUSHCACHE 1
 void __memcpy_flushcache(void *dst, const void *src, size_t cnt);
-static __always_inline void memcpy_flushcache(void *dst, const void *src, size_t cnt)
+
+static __always_inline void movnti_8(void *dst, const void *src)
+{
+	asm volatile("movntiq %1, %0"
+		     : "=m"(*(u64 *)dst)
+		     : "r"(*(const u64 *)src)
+		     : "memory");
+}
+
+static __always_inline void movnti_16(void *dst, const void *src)
+{
+	movnti_8(dst, src);
+	movnti_8(dst + 8, src + 8);
+}
+
+static __always_inline void movnti_32(void *dst, const void *src)
+{
+	movnti_16(dst, src);
+	movnti_16(dst + 16, src + 16);
+}
+
+static __always_inline void movnti_64(void *dst, const void *src)
+{
+	movnti_32(dst, src);
+	movnti_32(dst + 32, src + 32);
+}
+
+static __always_inline void memcpy_flushcache(void *dst, const void *src,
+					      size_t cnt)
 {
 	if (__builtin_constant_p(cnt)) {
 		switch (cnt) {
@@ -96,8 +124,34 @@ static __always_inline void memcpy_flushcache(void *dst, const void *src, size_t
 				asm ("movntiq %1, %0" : "=m"(*(u64 *)dst) : "r"(*(u64 *)src));
 				asm ("movntiq %1, %0" : "=m"(*(u64 *)(dst + 8)) : "r"(*(u64 *)(src + 8)));
 				return;
+			/*
+			 * The relevant fixed-size copies here are the
+			 * x86_64 struct page sizes: 64, 80, and 96 bytes.
+			 * Keep 32-byte and 48-byte copies inline as well
+			 * instead of sending those nearby fixed-size
+			 * cases back to __memcpy_flushcache().
+			 */
+			case 32:
+				movnti_32(dst, src);
+				return;
+			case 48:
+				movnti_32(dst, src);
+				movnti_16(dst + 32, src + 32);
+				return;
+			case 64:
+				movnti_64(dst, src);
+				return;
+			case 80:
+				movnti_64(dst, src);
+				movnti_16(dst + 64, src + 64);
+				return;
+			case 96:
+				movnti_64(dst, src);
+				movnti_32(dst + 64, src + 64);
+				return;
 		}
 	}
+
 	__memcpy_flushcache(dst, src, cnt);
 }
 
-- 
2.20.1

Re: [PATCH v9 4/8] mm: add a template-based fast path for zone-device page init

[Muchun Song]

> On Aug 3, 2026, at 15:09, Li Zhe <lizhe.67@bytedance.com> wrote:
> 
> memmap_init_zone_device() repeats nearly identical head-page
> initialization for each PFN. Prepare one reusable ZONE_DEVICE head-page
> template through the existing slow path, refresh the PFN-dependent
> fields in that template before each copy, and memcpy it into each
> destination page.
> 
> Use the template path unconditionally, as suggested by Muchun. The
> page_ref_set tracepoint is primarily a debugging aid, while this code is
> still initializing struct pages before they are handed out. From the
> perspective of users of those pages, the initialization-time refcount
> transitions are not part of the observable page lifetime.
> 
> This means page_ref_set will no longer observe every initialization-time
> refcount assignment for copied ZONE_DEVICE head pages. The impact is
> controlled because the final initialized struct page state is unchanged,
> and keeping a separate non-template path only for this local tracepoint
> observability would add complexity to the common path.
> 
> This patch accelerates head-page initialization. The pfns_per_compound
> == 1 case gets the full benefit here, compound tails are handled in the
> next patch.
> 
> Tested in a VM with a 100 GB fsdax namespace device configured with
> map=dev on Intel Ice Lake server. This test exercises the nd_pmem rebind
> path (pfns_per_compound == 1).
> 
> Test procedure:
> Rebind the nd_pmem driver 30 times and collect the memmap initialization
> time from the pr_debug() output of memmap_init_zone_device().
> 
> Base(v7.2-rc1):
>  Average of rebinds for nd_pmem driver: 244.28 ms
> 
> With this patch and its prerequisites applied:
>  Average of rebinds for nd_pmem driver: 215.55 ms
> 
> This reduces the average memmap initialization time measured during rebind
> from 244.28 ms to 215.55 ms, or about 11%.
> 
> Suggested-by: Muchun Song <muchun.song@linux.dev>

I personally feel you might have used the wrong trailer here, because it
implies that the overall design is based on someone's suggestion. But for me,
I think what you actually meant is that in a previous version, I made a
suggestion about simplifying the code, rather than dictating the overall
direction of this entire patch series. So I feel I shouldn't be credited here.
This information is already enough just appearing in the change log.

Muchun,
Thanks.

Re: [PATCH v9 0/8] mm: optimize zone-device memmap initialization

[Andrew Morton]

On Mon,  3 Aug 2026 15:09:21 +0800 "Li Zhe" <lizhe.67@bytedance.com> wrote:

> Test procedure:
> Rebind the nd_pmem and dax_pmem drivers 30 times and collect the memmap
> initialization time from the pr_debug() output of
> memmap_init_zone_device().
> 
> Base(v7.2-rc1):
>   Average of nd_pmem rebinds: 244.28 ms
>   Average of dax_pmem rebinds: 273.31 ms
> 
> With this series applied:
>   Average of nd_pmem rebinds: 96.79 ms
>   Average of dax_pmem rebinds: 119.04 ms
> 
> This reduces the average memmap initialization time measured during
> rebind by about 60.4% for nd_pmem and 56.4% for dax_pmem.

Well that's nice.

Review is partial and it's late in the cycle so I'd prefer to schedule
this for the next -rc cycle.

AI review might have found a couple of things, please check?
	https://sashiko.dev/#/patchset/20260803070929.86075-1-lizhe.67@bytedance.com

Re: [PATCH v9 8/8] x86/string: extend memcpy_flushcache() fixed-size fastpaths

[Borislav Petkov]

On Mon, Aug 03, 2026 at 03:09:29PM +0800, Li Zhe wrote:
> Tested in a VM with a 100 GB fsdax namespace device configured with
> map=dev and a 100 GB devdax namespace (align=2097152) on Intel Ice Lake
> server.
> 
> Test procedure:
> Rebind the nd_pmem and dax_pmem drivers 30 times and collect the memmap
> initialization time from the pr_debug() output of
> memmap_init_zone_device().
> 
> With memcpy_nontemporal() used by the ZONE_DEVICE template-copy path:
>   Average of rebinds for nd_pmem driver: 150.83 ms
>   Average of rebinds for dax_pmem driver: 153.55 ms
> 
> With this x86 fixed-size fastpath patch applied:
>   Average of rebinds for nd_pmem driver: 96.79 ms
>   Average of rebinds for dax_pmem driver: 119.04 ms
> 
> This further reduces the average memmap initialization time measured
> during rebind by about 35.8% for nd_pmem and 22.5% for dax_pmem.

Testing methodology doesn't usually belong in the commit message but under the
"---" lines below but this is important enough to keep it here.

> Signed-off-by: Li Zhe <lizhe.67@bytedance.com>
> ---
>  arch/x86/include/asm/string_64.h | 56 +++++++++++++++++++++++++++++++-
>  1 file changed, 55 insertions(+), 1 deletion(-)

A quick and untested cleanup ontop with the potential for a bunch more
unification. But later.

Note that you're inconsistent about the "memory" clobber. The current code
doesn't have it, you're adding it to movnti_8(). I've added it now everywhere
to be on the safe side - it would be interesting to know whether you see any
perf difference with and without it in your use case.

Thx.

---
diff --git a/arch/x86/include/asm/string_64.h b/arch/x86/include/asm/string_64.h
index bc6a9f34b346..3a1e1f17431b 100644
--- a/arch/x86/include/asm/string_64.h
+++ b/arch/x86/include/asm/string_64.h
@@ -83,6 +83,14 @@ int strcmp(const char *cs, const char *ct);
 #define __HAVE_ARCH_MEMCPY_FLUSHCACHE 1
 void __memcpy_flushcache(void *dst, const void *src, size_t cnt);
 
+static __always_inline void movnti_4(void *dst, const void *src)
+{
+	asm volatile("movntil %1, %0"
+		     : "=m"(*(u32 *)dst)
+		     : "r"(*(u32 *)src)
+		     : "memory");
+}
+
 static __always_inline void movnti_8(void *dst, const void *src)
 {
 	asm volatile("movntiq %1, %0"
@@ -112,47 +120,26 @@ static __always_inline void movnti_64(void *dst, const void *src)
 static __always_inline void memcpy_flushcache(void *dst, const void *src,
 					      size_t cnt)
 {
-	if (__builtin_constant_p(cnt)) {
-		switch (cnt) {
-			case 4:
-				asm ("movntil %1, %0" : "=m"(*(u32 *)dst) : "r"(*(u32 *)src));
-				return;
-			case 8:
-				asm ("movntiq %1, %0" : "=m"(*(u64 *)dst) : "r"(*(u64 *)src));
-				return;
-			case 16:
-				asm ("movntiq %1, %0" : "=m"(*(u64 *)dst) : "r"(*(u64 *)src));
-				asm ("movntiq %1, %0" : "=m"(*(u64 *)(dst + 8)) : "r"(*(u64 *)(src + 8)));
-				return;
-			/*
-			 * The relevant fixed-size copies here are the
-			 * x86_64 struct page sizes: 64, 80, and 96 bytes.
-			 * Keep 32-byte and 48-byte copies inline as well
-			 * instead of sending those nearby fixed-size
-			 * cases back to __memcpy_flushcache().
-			 */
-			case 32:
-				movnti_32(dst, src);
-				return;
-			case 48:
-				movnti_32(dst, src);
-				movnti_16(dst + 32, src + 32);
-				return;
-			case 64:
-				movnti_64(dst, src);
-				return;
-			case 80:
-				movnti_64(dst, src);
-				movnti_16(dst + 64, src + 64);
-				return;
-			case 96:
-				movnti_64(dst, src);
-				movnti_32(dst + 64, src + 64);
-				return;
-		}
+	if (!__builtin_constant_p(cnt))
+		return __memcpy_flushcache(dst, src, cnt);
+
+	/*
+	 * The relevant fixed-size copies here are the x86_64 struct page sizes:
+	 * 64, 80, and 96 bytes. Keep 32-byte and 48-byte copies inline as well
+	 * instead of sending those nearby fixed-size cases back to
+	 * __memcpy_flushcache().
+	 */
+	switch (cnt) {
+	case 4:  movnti_4(dst, src); break;
+	case 8:	 movnti_8(dst, src); break;
+	case 16: movnti_16(dst, src); break;
+	case 32: movnti_32(dst, src); break;
+	case 48: movnti_32(dst, src); movnti_16(dst + 32, src + 32); break;
+	case 64: movnti_64(dst, src); break;
+	case 80: movnti_64(dst, src); movnti_16(dst + 64, src + 64); break;
+	case 96: movnti_64(dst, src); movnti_32(dst + 64, src + 64); break;
+	default: __memcpy_flushcache(dst, src, cnt); break;
 	}
-
-	__memcpy_flushcache(dst, src, cnt);
 }
 
 #define memcpy_nontemporal memcpy_nontemporal

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette

Re: [PATCH v9 0/8] mm: optimize zone-device memmap initialization

[Li Zhe]

On 8/4/26 5:40 AM, Andrew Morton wrote:
> On Mon,  3 Aug 2026 15:09:21 +0800 "Li Zhe" <lizhe.67@bytedance.com> wrote:
>
>> Test procedure:
>> Rebind the nd_pmem and dax_pmem drivers 30 times and collect the memmap
>> initialization time from the pr_debug() output of
>> memmap_init_zone_device().
>>
>> Base(v7.2-rc1):
>>    Average of nd_pmem rebinds: 244.28 ms
>>    Average of dax_pmem rebinds: 273.31 ms
>>
>> With this series applied:
>>    Average of nd_pmem rebinds: 96.79 ms
>>    Average of dax_pmem rebinds: 119.04 ms
>>
>> This reduces the average memmap initialization time measured during
>> rebind by about 60.4% for nd_pmem and 56.4% for dax_pmem.
> Well that's nice.
>
> Review is partial and it's late in the cycle so I'd prefer to schedule
> this for the next -rc cycle.
>
> AI review might have found a couple of things, please check?
> 	https://sashiko.dev/#/patchset/20260803070929.86075-1-lizhe.67@bytedance.com


Thanks.

I checked the Sashiko comments. The reported items are the same points
that were discussed in the previous v7/v8 reviews. v9 already folds the
related explanation into patch 7, so I do not plan any additional code
change for these items.

Thanks,
Zhe

Re: [PATCH v9 4/8] mm: add a template-based fast path for zone-device page init

[Li Zhe]

On 8/3/26 4:39 PM, Muchun Song wrote:
>
>> On Aug 3, 2026, at 15:09, Li Zhe <lizhe.67@bytedance.com> wrote:
>>
>> memmap_init_zone_device() repeats nearly identical head-page
>> initialization for each PFN. Prepare one reusable ZONE_DEVICE head-page
>> template through the existing slow path, refresh the PFN-dependent
>> fields in that template before each copy, and memcpy it into each
>> destination page.
>>
>> Use the template path unconditionally, as suggested by Muchun. The
>> page_ref_set tracepoint is primarily a debugging aid, while this code is
>> still initializing struct pages before they are handed out. From the
>> perspective of users of those pages, the initialization-time refcount
>> transitions are not part of the observable page lifetime.
>>
>> This means page_ref_set will no longer observe every initialization-time
>> refcount assignment for copied ZONE_DEVICE head pages. The impact is
>> controlled because the final initialized struct page state is unchanged,
>> and keeping a separate non-template path only for this local tracepoint
>> observability would add complexity to the common path.
>>
>> This patch accelerates head-page initialization. The pfns_per_compound
>> == 1 case gets the full benefit here, compound tails are handled in the
>> next patch.
>>
>> Tested in a VM with a 100 GB fsdax namespace device configured with
>> map=dev on Intel Ice Lake server. This test exercises the nd_pmem rebind
>> path (pfns_per_compound == 1).
>>
>> Test procedure:
>> Rebind the nd_pmem driver 30 times and collect the memmap initialization
>> time from the pr_debug() output of memmap_init_zone_device().
>>
>> Base(v7.2-rc1):
>>   Average of rebinds for nd_pmem driver: 244.28 ms
>>
>> With this patch and its prerequisites applied:
>>   Average of rebinds for nd_pmem driver: 215.55 ms
>>
>> This reduces the average memmap initialization time measured during rebind
>> from 244.28 ms to 215.55 ms, or about 11%.
>>
>> Suggested-by: Muchun Song <muchun.song@linux.dev>
> I personally feel you might have used the wrong trailer here, because it
> implies that the overall design is based on someone's suggestion. But for me,
> I think what you actually meant is that in a previous version, I made a
> suggestion about simplifying the code, rather than dictating the overall
> direction of this entire patch series. So I feel I shouldn't be credited here.
> This information is already enough just appearing in the change log.
>
> Muchun,
> Thanks.


Thanks.

I will drop it and keep the simplification note only in the changelog.

Thanks,
Zhe

Re: [PATCH v9 8/8] x86/string: extend memcpy_flushcache() fixed-size fastpaths

[Li Zhe]

On 8/5/26 4:35 AM, Borislav Petkov wrote:
> On Mon, Aug 03, 2026 at 03:09:29PM +0800, Li Zhe wrote:
>> Tested in a VM with a 100 GB fsdax namespace device configured with
>> map=dev and a 100 GB devdax namespace (align=2097152) on Intel Ice Lake
>> server.
>>
>> Test procedure:
>> Rebind the nd_pmem and dax_pmem drivers 30 times and collect the memmap
>> initialization time from the pr_debug() output of
>> memmap_init_zone_device().
>>
>> With memcpy_nontemporal() used by the ZONE_DEVICE template-copy path:
>>    Average of rebinds for nd_pmem driver: 150.83 ms
>>    Average of rebinds for dax_pmem driver: 153.55 ms
>>
>> With this x86 fixed-size fastpath patch applied:
>>    Average of rebinds for nd_pmem driver: 96.79 ms
>>    Average of rebinds for dax_pmem driver: 119.04 ms
>>
>> This further reduces the average memmap initialization time measured
>> during rebind by about 35.8% for nd_pmem and 22.5% for dax_pmem.
> Testing methodology doesn't usually belong in the commit message but under the
> "---" lines below but this is important enough to keep it here.
>
>> Signed-off-by: Li Zhe <lizhe.67@bytedance.com>
>> ---
>>   arch/x86/include/asm/string_64.h | 56 +++++++++++++++++++++++++++++++-
>>   1 file changed, 55 insertions(+), 1 deletion(-)
> A quick and untested cleanup ontop with the potential for a bunch more
> unification. But later.
>
> Note that you're inconsistent about the "memory" clobber. The current code
> doesn't have it, you're adding it to movnti_8(). I've added it now everywhere
> to be on the safe side - it would be interesting to know whether you see any
> perf difference with and without it in your use case.
>
> Thx.


Thanks, I will fold the cleaned-up version into v10.

I also tested the memcpy_flushcache() fixed-size path with and without
the "memory" clobber. In this build, the generated assembly for this
path is the same in both cases.

I also ran the ZONE_DEVICE rebind test with both variants. The difference
was below 1% for both nd_pmem and dax_pmem, with no stable direction, so
I do not see a measurable performance difference from the "memory"
clobber in this use case.

Given that, I will keep the clobber for consistency and safety.

Thanks,
Zhe

>
> ---
> diff --git a/arch/x86/include/asm/string_64.h b/arch/x86/include/asm/string_64.h
> index bc6a9f34b346..3a1e1f17431b 100644
> --- a/arch/x86/include/asm/string_64.h
> +++ b/arch/x86/include/asm/string_64.h
> @@ -83,6 +83,14 @@ int strcmp(const char *cs, const char *ct);
>   #define __HAVE_ARCH_MEMCPY_FLUSHCACHE 1
>   void __memcpy_flushcache(void *dst, const void *src, size_t cnt);
>   
> +static __always_inline void movnti_4(void *dst, const void *src)
> +{
> +	asm volatile("movntil %1, %0"
> +		     : "=m"(*(u32 *)dst)
> +		     : "r"(*(u32 *)src)
> +		     : "memory");
> +}
> +
>   static __always_inline void movnti_8(void *dst, const void *src)
>   {
>   	asm volatile("movntiq %1, %0"
> @@ -112,47 +120,26 @@ static __always_inline void movnti_64(void *dst, const void *src)
>   static __always_inline void memcpy_flushcache(void *dst, const void *src,
>   					      size_t cnt)
>   {
> -	if (__builtin_constant_p(cnt)) {
> -		switch (cnt) {
> -			case 4:
> -				asm ("movntil %1, %0" : "=m"(*(u32 *)dst) : "r"(*(u32 *)src));
> -				return;
> -			case 8:
> -				asm ("movntiq %1, %0" : "=m"(*(u64 *)dst) : "r"(*(u64 *)src));
> -				return;
> -			case 16:
> -				asm ("movntiq %1, %0" : "=m"(*(u64 *)dst) : "r"(*(u64 *)src));
> -				asm ("movntiq %1, %0" : "=m"(*(u64 *)(dst + 8)) : "r"(*(u64 *)(src + 8)));
> -				return;
> -			/*
> -			 * The relevant fixed-size copies here are the
> -			 * x86_64 struct page sizes: 64, 80, and 96 bytes.
> -			 * Keep 32-byte and 48-byte copies inline as well
> -			 * instead of sending those nearby fixed-size
> -			 * cases back to __memcpy_flushcache().
> -			 */
> -			case 32:
> -				movnti_32(dst, src);
> -				return;
> -			case 48:
> -				movnti_32(dst, src);
> -				movnti_16(dst + 32, src + 32);
> -				return;
> -			case 64:
> -				movnti_64(dst, src);
> -				return;
> -			case 80:
> -				movnti_64(dst, src);
> -				movnti_16(dst + 64, src + 64);
> -				return;
> -			case 96:
> -				movnti_64(dst, src);
> -				movnti_32(dst + 64, src + 64);
> -				return;
> -		}
> +	if (!__builtin_constant_p(cnt))
> +		return __memcpy_flushcache(dst, src, cnt);
> +
> +	/*
> +	 * The relevant fixed-size copies here are the x86_64 struct page sizes:
> +	 * 64, 80, and 96 bytes. Keep 32-byte and 48-byte copies inline as well
> +	 * instead of sending those nearby fixed-size cases back to
> +	 * __memcpy_flushcache().
> +	 */
> +	switch (cnt) {
> +	case 4:  movnti_4(dst, src); break;
> +	case 8:	 movnti_8(dst, src); break;
> +	case 16: movnti_16(dst, src); break;
> +	case 32: movnti_32(dst, src); break;
> +	case 48: movnti_32(dst, src); movnti_16(dst + 32, src + 32); break;
> +	case 64: movnti_64(dst, src); break;
> +	case 80: movnti_64(dst, src); movnti_16(dst + 64, src + 64); break;
> +	case 96: movnti_64(dst, src); movnti_32(dst + 64, src + 64); break;
> +	default: __memcpy_flushcache(dst, src, cnt); break;
>   	}
> -
> -	__memcpy_flushcache(dst, src, cnt);
>   }
>   
>   #define memcpy_nontemporal memcpy_nontemporal