* [PATCH v3 01/26] set_memory: add folio_{zap,restore}_direct_map helpers
2026-07-26 22:22 [PATCH v3 00/26] mm: Add ALLOC_UNMAPPED and AS_NO_DIRECT_MAP Brendan Jackman
@ 2026-07-26 22:22 ` Brendan Jackman
2026-07-27 10:33 ` Mike Rapoport
2026-07-26 22:22 ` [PATCH v3 02/26] mm/secretmem: make use of folio_{zap,restore}_direct_map Brendan Jackman
` (24 subsequent siblings)
25 siblings, 1 reply; 29+ messages in thread
From: Brendan Jackman @ 2026-07-26 22:22 UTC (permalink / raw)
To: Borislav Petkov, Dave Hansen, Peter Zijlstra, Andrew Morton,
David Hildenbrand, Vlastimil Babka, Mike Rapoport, Wei Xu,
Johannes Weiner, Zi Yan, Lorenzo Stoakes
Cc: linux-mm, linux-kernel, x86, rppt, Sumit Garg, Will Deacon,
rientjes, Kalyazin, Nikita, patrick.roy, Itazuri, Takahiro,
Andy Lutomirski, David Kaplan, Thomas Gleixner, Yosry Ahmed,
Patrick Bellasi, Reiji Watanabe, Sean Christopherson,
Brendan Jackman, Nikita Kalyazin
From: Nikita Kalyazin <nikita.kalyazin@linux.dev>
Let's provide folio_{zap,restore}_direct_map helpers as preparation for
supporting removal of the direct map for guest_memfd folios.
In folio_zap_direct_map(), flush TLB to make sure the data is not
accessible. On some architectures, there may be a double TLB flush
issued because set_direct_map_valid_noflush already performs a flush
internally.
The new helpers need to be accessible to KVM on architectures that
support guest_memfd (x86 and arm64).
Direct map removal gives guest_memfd the same protection that
memfd_secret does, such as hardening against Spectre-like attacks
through in-kernel gadgets.
Acked-by: David Hildenbrand (Arm) <david@kernel.org>
Signed-off-by: Nikita Kalyazin <nikita.kalyazin@linux.dev>
[Added comment, dropped modified set_direct_map API, added highmem check]
Signed-off-by: Brendan Jackman <jackmanb@google.com>
---
include/linux/set_memory.h | 13 +++++++++++++
mm/memory.c | 46 ++++++++++++++++++++++++++++++++++++++++++++++
2 files changed, 59 insertions(+)
diff --git a/include/linux/set_memory.h b/include/linux/set_memory.h
index 3030d9245f5ac..1bf2a15bca118 100644
--- a/include/linux/set_memory.h
+++ b/include/linux/set_memory.h
@@ -40,6 +40,15 @@ static inline int set_direct_map_valid_noflush(struct page *page,
return 0;
}
+static inline int folio_zap_direct_map(struct folio *folio)
+{
+ return 0;
+}
+
+static inline void folio_restore_direct_map(struct folio *folio)
+{
+}
+
static inline bool kernel_page_present(struct page *page)
{
return true;
@@ -56,6 +65,10 @@ static inline bool can_set_direct_map(void)
}
#define can_set_direct_map can_set_direct_map
#endif
+
+int folio_zap_direct_map(struct folio *folio);
+void folio_restore_direct_map(struct folio *folio);
+
#endif /* CONFIG_ARCH_HAS_SET_DIRECT_MAP */
#ifdef CONFIG_X86_64
diff --git a/mm/memory.c b/mm/memory.c
index a73af1fccb3d0..789c65a6d6a0e 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -78,6 +78,7 @@
#include <linux/sched/sysctl.h>
#include <linux/pgalloc.h>
#include <linux/uaccess.h>
+#include <linux/set_memory.h>
#include <trace/events/kmem.h>
@@ -7758,3 +7759,48 @@ void vma_pgtable_walk_end(struct vm_area_struct *vma)
if (is_vm_hugetlb_page(vma))
hugetlb_vma_unlock_read(vma);
}
+
+#ifdef CONFIG_ARCH_HAS_SET_DIRECT_MAP
+/**
+ * folio_zap_direct_map - remove a folio from the kernel direct map
+ * @folio: folio to remove from the direct map
+ *
+ * Removes the folio from the kernel direct map and flushes the TLB. This may
+ * require splitting huge pages in the direct map, which can fail due to memory
+ * allocation. So far, only order-0 folios are supported; this guarantees
+ * the unmap is either a complete success or a total failure.
+ *
+ * Return: 0 on success, or a negative error code on failure.
+ */
+int folio_zap_direct_map(struct folio *folio)
+{
+ struct page *page = folio_page(folio, 0);
+ unsigned long addr = (unsigned long)page_address(page);
+ int ret;
+
+ if (folio_test_large(folio) || folio_test_highmem(folio))
+ return -EINVAL;
+
+ ret = set_direct_map_valid_noflush(page, 1, false);
+ flush_tlb_kernel_range(addr, addr + folio_size(folio));
+
+ return ret;
+}
+EXPORT_SYMBOL_FOR_MODULES(folio_zap_direct_map, "kvm");
+
+/**
+ * folio_restore_direct_map - restore the kernel direct map entry for a folio
+ * @folio: folio whose direct map entry is to be restored
+ *
+ * This may only be called after a prior successful folio_zap_direct_map() on
+ * the same folio. Because the zap will have already split any huge pages in
+ * the direct map, restoration here only updates protection bits and cannot
+ * fail.
+ */
+void folio_restore_direct_map(struct folio *folio)
+{
+ WARN_ON_ONCE(set_direct_map_valid_noflush(folio_page(folio, 0),
+ folio_nr_pages(folio), true));
+}
+EXPORT_SYMBOL_FOR_MODULES(folio_restore_direct_map, "kvm");
+#endif /* CONFIG_ARCH_HAS_SET_DIRECT_MAP */
--
2.54.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* Re: [PATCH v3 01/26] set_memory: add folio_{zap,restore}_direct_map helpers
2026-07-26 22:22 ` [PATCH v3 01/26] set_memory: add folio_{zap,restore}_direct_map helpers Brendan Jackman
@ 2026-07-27 10:33 ` Mike Rapoport
0 siblings, 0 replies; 29+ messages in thread
From: Mike Rapoport @ 2026-07-27 10:33 UTC (permalink / raw)
To: Brendan Jackman
Cc: Borislav Petkov, Dave Hansen, Peter Zijlstra, Andrew Morton,
David Hildenbrand, Vlastimil Babka, Wei Xu, Johannes Weiner,
Zi Yan, Lorenzo Stoakes, linux-mm, linux-kernel, x86, Sumit Garg,
Will Deacon, rientjes, Kalyazin, Nikita, patrick.roy,
Itazuri, Takahiro, Andy Lutomirski, David Kaplan, Thomas Gleixner,
Yosry Ahmed, Patrick Bellasi, Reiji Watanabe, Sean Christopherson,
Nikita Kalyazin
Hi Brendan,
On Sun, Jul 26, 2026 at 10:22:34PM +0000, Brendan Jackman wrote:
> From: Nikita Kalyazin <nikita.kalyazin@linux.dev>
>
> Let's provide folio_{zap,restore}_direct_map helpers as preparation for
> supporting removal of the direct map for guest_memfd folios.
> In folio_zap_direct_map(), flush TLB to make sure the data is not
> accessible. On some architectures, there may be a double TLB flush
> issued because set_direct_map_valid_noflush already performs a flush
> internally.
>
> The new helpers need to be accessible to KVM on architectures that
> support guest_memfd (x86 and arm64).
>
> Direct map removal gives guest_memfd the same protection that
> memfd_secret does, such as hardening against Spectre-like attacks
> through in-kernel gadgets.
>
> Acked-by: David Hildenbrand (Arm) <david@kernel.org>
> Signed-off-by: Nikita Kalyazin <nikita.kalyazin@linux.dev>
> [Added comment, dropped modified set_direct_map API, added highmem check]
> Signed-off-by: Brendan Jackman <jackmanb@google.com>
> ---
> include/linux/set_memory.h | 13 +++++++++++++
> mm/memory.c | 46 ++++++++++++++++++++++++++++++++++++++++++++++
> 2 files changed, 59 insertions(+)
>
> diff --git a/include/linux/set_memory.h b/include/linux/set_memory.h
> index 3030d9245f5ac..1bf2a15bca118 100644
> --- a/include/linux/set_memory.h
> +++ b/include/linux/set_memory.h
> @@ -40,6 +40,15 @@ static inline int set_direct_map_valid_noflush(struct page *page,
> return 0;
> }
>
> +static inline int folio_zap_direct_map(struct folio *folio)
> +{
> + return 0;
> +}
> +
> +static inline void folio_restore_direct_map(struct folio *folio)
> +{
> +}
> +
> static inline bool kernel_page_present(struct page *page)
> {
> return true;
> @@ -56,6 +65,10 @@ static inline bool can_set_direct_map(void)
> }
> #define can_set_direct_map can_set_direct_map
> #endif
> +
> +int folio_zap_direct_map(struct folio *folio);
> +void folio_restore_direct_map(struct folio *folio);
> +
> #endif /* CONFIG_ARCH_HAS_SET_DIRECT_MAP */
>
> #ifdef CONFIG_X86_64
> diff --git a/mm/memory.c b/mm/memory.c
> index a73af1fccb3d0..789c65a6d6a0e 100644
> --- a/mm/memory.c
> +++ b/mm/memory.c
> @@ -78,6 +78,7 @@
> #include <linux/sched/sysctl.h>
> #include <linux/pgalloc.h>
> #include <linux/uaccess.h>
> +#include <linux/set_memory.h>
>
> #include <trace/events/kmem.h>
>
> @@ -7758,3 +7759,48 @@ void vma_pgtable_walk_end(struct vm_area_struct *vma)
> if (is_vm_hugetlb_page(vma))
> hugetlb_vma_unlock_read(vma);
> }
> +
> +#ifdef CONFIG_ARCH_HAS_SET_DIRECT_MAP
> +/**
> + * folio_zap_direct_map - remove a folio from the kernel direct map
> + * @folio: folio to remove from the direct map
> + *
> + * Removes the folio from the kernel direct map and flushes the TLB. This may
> + * require splitting huge pages in the direct map, which can fail due to memory
> + * allocation. So far, only order-0 folios are supported; this guarantees
> + * the unmap is either a complete success or a total failure.
> + *
> + * Return: 0 on success, or a negative error code on failure.
> + */
> +int folio_zap_direct_map(struct folio *folio)
> +{
> + struct page *page = folio_page(folio, 0);
> + unsigned long addr = (unsigned long)page_address(page);
> + int ret;
> +
> + if (folio_test_large(folio) || folio_test_highmem(folio))
> + return -EINVAL;
> +
> + ret = set_direct_map_valid_noflush(page, 1, false);
There was a discussion about slight differences in the semantics of
set_direct_map_valid() on x86 and on arm64 and that execmem should
apparently switch to set_direct_map_{invalid,default}.
Maybe for this series it would be better to add a patch that adds numpages
to set_direct_map_{invalid,default}_noflush and use
set_direct_map_default_noflush() here?
And maybe also pick another Nikita's patch [2] that makes set_direct_map_*
to take address?
[1] https://lore.kernel.org/all/DJ69RCVRBO0Y.3JCYSW50IC4RC@linux.dev/
[2] https://lore.kernel.org/all/20260410151746.61150-2-kalyazin@amazon.com/
> + flush_tlb_kernel_range(addr, addr + folio_size(folio));
> +
> + return ret;
> +}
> +EXPORT_SYMBOL_FOR_MODULES(folio_zap_direct_map, "kvm");
> +
> +/**
> + * folio_restore_direct_map - restore the kernel direct map entry for a folio
> + * @folio: folio whose direct map entry is to be restored
> + *
> + * This may only be called after a prior successful folio_zap_direct_map() on
> + * the same folio. Because the zap will have already split any huge pages in
> + * the direct map, restoration here only updates protection bits and cannot
> + * fail.
> + */
> +void folio_restore_direct_map(struct folio *folio)
> +{
> + WARN_ON_ONCE(set_direct_map_valid_noflush(folio_page(folio, 0),
> + folio_nr_pages(folio), true));
> +}
> +EXPORT_SYMBOL_FOR_MODULES(folio_restore_direct_map, "kvm");
> +#endif /* CONFIG_ARCH_HAS_SET_DIRECT_MAP */
>
> --
> 2.54.0
>
--
Sincerely yours,
Mike.
^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH v3 02/26] mm/secretmem: make use of folio_{zap,restore}_direct_map
2026-07-26 22:22 [PATCH v3 00/26] mm: Add ALLOC_UNMAPPED and AS_NO_DIRECT_MAP Brendan Jackman
2026-07-26 22:22 ` [PATCH v3 01/26] set_memory: add folio_{zap,restore}_direct_map helpers Brendan Jackman
@ 2026-07-26 22:22 ` Brendan Jackman
2026-07-27 10:40 ` Mike Rapoport
2026-07-26 22:22 ` [PATCH v3 03/26] mm: introduce AS_NO_DIRECT_MAP Brendan Jackman
` (23 subsequent siblings)
25 siblings, 1 reply; 29+ messages in thread
From: Brendan Jackman @ 2026-07-26 22:22 UTC (permalink / raw)
To: Borislav Petkov, Dave Hansen, Peter Zijlstra, Andrew Morton,
David Hildenbrand, Vlastimil Babka, Mike Rapoport, Wei Xu,
Johannes Weiner, Zi Yan, Lorenzo Stoakes
Cc: linux-mm, linux-kernel, x86, rppt, Sumit Garg, Will Deacon,
rientjes, Kalyazin, Nikita, patrick.roy, Itazuri, Takahiro,
Andy Lutomirski, David Kaplan, Thomas Gleixner, Yosry Ahmed,
Patrick Bellasi, Reiji Watanabe, Sean Christopherson,
Brendan Jackman, Nikita Kalyazin, Ackerley Tng
From: Nikita Kalyazin <nikita.kalyazin@linux.dev>
Replace set_direct_map_*_noflush with newly available
folio_zap_direct_map calls that take folio's address internally. A side
effect is even if filemap_add_folio fails, the TLB is still flushed,
which is not expected to be on the hot path.
Acked-by: David Hildenbrand (Arm) <david@kernel.org>
Reviewed-by: Ackerley Tng <ackerleytng@google.com>
Signed-off-by: Nikita Kalyazin <nikita.kalyazin@linux.dev>
Signed-off-by: Brendan Jackman <jackmanb@google.com>
---
mm/secretmem.c | 8 ++------
1 file changed, 2 insertions(+), 6 deletions(-)
diff --git a/mm/secretmem.c b/mm/secretmem.c
index d29865075b6ea..4a4934769f8ba 100644
--- a/mm/secretmem.c
+++ b/mm/secretmem.c
@@ -53,7 +53,6 @@ static vm_fault_t secretmem_fault(struct vm_fault *vmf)
struct inode *inode = file_inode(vmf->vma->vm_file);
pgoff_t offset = vmf->pgoff;
gfp_t gfp = vmf->gfp_mask;
- unsigned long addr;
struct folio *folio;
vm_fault_t ret;
int err;
@@ -72,7 +71,7 @@ static vm_fault_t secretmem_fault(struct vm_fault *vmf)
goto out;
}
- err = set_direct_map_invalid_noflush(folio_page(folio, 0));
+ err = folio_zap_direct_map(folio);
if (err) {
folio_put(folio);
ret = vmf_error(err);
@@ -87,7 +86,7 @@ static vm_fault_t secretmem_fault(struct vm_fault *vmf)
* already happened when we marked the page invalid
* which guarantees that this call won't fail
*/
- set_direct_map_default_noflush(folio_page(folio, 0));
+ folio_restore_direct_map(folio);
folio_put(folio);
if (err == -EEXIST)
goto retry;
@@ -95,9 +94,6 @@ static vm_fault_t secretmem_fault(struct vm_fault *vmf)
ret = vmf_error(err);
goto out;
}
-
- addr = (unsigned long)folio_address(folio);
- flush_tlb_kernel_range(addr, addr + PAGE_SIZE);
}
vmf->page = folio_file_page(folio, vmf->pgoff);
--
2.54.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* Re: [PATCH v3 02/26] mm/secretmem: make use of folio_{zap,restore}_direct_map
2026-07-26 22:22 ` [PATCH v3 02/26] mm/secretmem: make use of folio_{zap,restore}_direct_map Brendan Jackman
@ 2026-07-27 10:40 ` Mike Rapoport
0 siblings, 0 replies; 29+ messages in thread
From: Mike Rapoport @ 2026-07-27 10:40 UTC (permalink / raw)
To: Brendan Jackman
Cc: Borislav Petkov, Dave Hansen, Peter Zijlstra, Andrew Morton,
David Hildenbrand, Vlastimil Babka, Wei Xu, Johannes Weiner,
Zi Yan, Lorenzo Stoakes, linux-mm, linux-kernel, x86, Sumit Garg,
Will Deacon, rientjes, Kalyazin, Nikita, patrick.roy,
Itazuri, Takahiro, Andy Lutomirski, David Kaplan, Thomas Gleixner,
Yosry Ahmed, Patrick Bellasi, Reiji Watanabe, Sean Christopherson,
Nikita Kalyazin, Ackerley Tng
On Sun, Jul 26, 2026 at 10:22:35PM +0000, Brendan Jackman wrote:
> From: Nikita Kalyazin <nikita.kalyazin@linux.dev>
>
> Replace set_direct_map_*_noflush with newly available
> folio_zap_direct_map calls that take folio's address internally. A side
> effect is even if filemap_add_folio fails, the TLB is still flushed,
> which is not expected to be on the hot path.
>
> Acked-by: David Hildenbrand (Arm) <david@kernel.org>
> Reviewed-by: Ackerley Tng <ackerleytng@google.com>
> Signed-off-by: Nikita Kalyazin <nikita.kalyazin@linux.dev>
> Signed-off-by: Brendan Jackman <jackmanb@google.com>
Reviewed-by: Mike Rapoport (Microsoft) <rppt@kernel.org>
> ---
> mm/secretmem.c | 8 ++------
> 1 file changed, 2 insertions(+), 6 deletions(-)
--
Sincerely yours,
Mike.
^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH v3 03/26] mm: introduce AS_NO_DIRECT_MAP
2026-07-26 22:22 [PATCH v3 00/26] mm: Add ALLOC_UNMAPPED and AS_NO_DIRECT_MAP Brendan Jackman
2026-07-26 22:22 ` [PATCH v3 01/26] set_memory: add folio_{zap,restore}_direct_map helpers Brendan Jackman
2026-07-26 22:22 ` [PATCH v3 02/26] mm/secretmem: make use of folio_{zap,restore}_direct_map Brendan Jackman
@ 2026-07-26 22:22 ` Brendan Jackman
2026-07-26 22:22 ` [PATCH v3 04/26] x86/mm: split out preallocate_sub_pgd() Brendan Jackman
` (22 subsequent siblings)
25 siblings, 0 replies; 29+ messages in thread
From: Brendan Jackman @ 2026-07-26 22:22 UTC (permalink / raw)
To: Borislav Petkov, Dave Hansen, Peter Zijlstra, Andrew Morton,
David Hildenbrand, Vlastimil Babka, Mike Rapoport, Wei Xu,
Johannes Weiner, Zi Yan, Lorenzo Stoakes
Cc: linux-mm, linux-kernel, x86, rppt, Sumit Garg, Will Deacon,
rientjes, Kalyazin, Nikita, patrick.roy, Itazuri, Takahiro,
Andy Lutomirski, David Kaplan, Thomas Gleixner, Yosry Ahmed,
Patrick Bellasi, Reiji Watanabe, Sean Christopherson,
Brendan Jackman, Nikita Kalyazin
From: Patrick Roy <patrick.roy@linux.dev>
Add AS_NO_DIRECT_MAP for mappings where direct map entries of folios are
set to not present. Currently, mappings that match this description are
secretmem mappings (memfd_secret()). Later, some guest_memfd
configurations will also fall into this category.
Reject this new type of mappings in all locations that currently reject
secretmem mappings, on the assumption that if secretmem mappings are
rejected somewhere, it is precisely because of an inability to deal with
folios without direct map entries, and then make memfd_secret() use
AS_NO_DIRECT_MAP on its address_space to drop its special
vma_is_secretmem()/secretmem_mapping() checks.
Use a new flag instead of overloading AS_INACCESSIBLE (which is already
set by guest_memfd) because not all guest_memfd mappings will end up
being direct map removed (e.g. in pKVM setups, parts of guest_memfd that
can be mapped to userspace should also be GUP-able, and generally not
have restrictions on who can access it).
Signed-off-by: Patrick Roy <patrick.roy@linux.dev>
Signed-off-by: Nikita Kalyazin <nikita.kalyazin@linux.dev>
[Moved zapping to page cache; removed review tags]
Signed-off-by: Brendan Jackman <jackmanb@google.com>
---
include/linux/pagemap.h | 25 +++++++++++
include/linux/secretmem.h | 18 --------
include/linux/set_memory.h | 2 +
lib/buildid.c | 8 +++-
mm/filemap.c | 105 ++++++++++++++++++++++++++++++++++++---------
mm/gup.c | 9 ++--
mm/mlock.c | 2 +-
mm/secretmem.c | 46 +++-----------------
8 files changed, 128 insertions(+), 87 deletions(-)
diff --git a/include/linux/pagemap.h b/include/linux/pagemap.h
index 4e8b2b29f6d3e..011f6e34859cc 100644
--- a/include/linux/pagemap.h
+++ b/include/linux/pagemap.h
@@ -15,6 +15,7 @@
#include <linux/bitops.h>
#include <linux/hardirq.h> /* for in_interrupt() */
#include <linux/hugetlb_inline.h>
+#include <linux/set_memory.h>
struct folio_batch;
@@ -210,6 +211,7 @@ enum mapping_flags {
AS_WRITEBACK_MAY_DEADLOCK_ON_RECLAIM = 9,
AS_KERNEL_FILE = 10, /* mapping for a fake kernel file that shouldn't
account usage to user cgroups */
+ AS_NO_DIRECT_MAP = 11, /* Folios in the mapping are not in the direct map */
/* Bits 16-25 are used for FOLIO_ORDER */
AS_FOLIO_ORDER_BITS = 5,
AS_FOLIO_ORDER_MIN = 16,
@@ -345,6 +347,9 @@ static inline bool mapping_writeback_may_deadlock_on_reclaim(const struct addres
return test_bit(AS_WRITEBACK_MAY_DEADLOCK_ON_RECLAIM, &mapping->flags);
}
+static inline unsigned int
+mapping_max_folio_order(const struct address_space *mapping);
+
static inline gfp_t mapping_gfp_mask(const struct address_space *mapping)
{
return mapping->gfp_mask;
@@ -366,6 +371,24 @@ static inline void mapping_set_gfp_mask(struct address_space *m, gfp_t mask)
m->gfp_mask = mask;
}
+static inline void mapping_set_no_direct_map(struct address_space *mapping)
+{
+ WARN_ON(!can_set_direct_map());
+ /* folio_zap_direct_map() doesn't support large folios. */
+ WARN_ON(mapping_max_folio_order(mapping));
+ set_bit(AS_NO_DIRECT_MAP, &mapping->flags);
+}
+
+static inline bool mapping_no_direct_map(const struct address_space *mapping)
+{
+ return test_bit(AS_NO_DIRECT_MAP, &mapping->flags);
+}
+
+static inline bool vma_has_no_direct_map(const struct vm_area_struct *vma)
+{
+ return vma->vm_file && mapping_no_direct_map(vma->vm_file->f_mapping);
+}
+
/*
* There are some parts of the kernel which assume that PMD entries
* are exactly HPAGE_PMD_ORDER. Those should be fixed, but until then,
@@ -454,6 +477,8 @@ static inline void mapping_set_folio_min_order(struct address_space *mapping,
*/
static inline void mapping_set_large_folios(struct address_space *mapping)
{
+ /* folio_zap_direct_map() doesn't support large folios. */
+ WARN_ON(mapping_no_direct_map(mapping));
mapping_set_folio_order_range(mapping, 0, MAX_PAGECACHE_ORDER);
}
diff --git a/include/linux/secretmem.h b/include/linux/secretmem.h
index e918f96881f56..0ae1fb057b3d2 100644
--- a/include/linux/secretmem.h
+++ b/include/linux/secretmem.h
@@ -4,28 +4,10 @@
#ifdef CONFIG_SECRETMEM
-extern const struct address_space_operations secretmem_aops;
-
-static inline bool secretmem_mapping(struct address_space *mapping)
-{
- return mapping->a_ops == &secretmem_aops;
-}
-
-bool vma_is_secretmem(struct vm_area_struct *vma);
bool secretmem_active(void);
#else
-static inline bool vma_is_secretmem(struct vm_area_struct *vma)
-{
- return false;
-}
-
-static inline bool secretmem_mapping(struct address_space *mapping)
-{
- return false;
-}
-
static inline bool secretmem_active(void)
{
return false;
diff --git a/include/linux/set_memory.h b/include/linux/set_memory.h
index 1bf2a15bca118..0eff85b6d3980 100644
--- a/include/linux/set_memory.h
+++ b/include/linux/set_memory.h
@@ -53,6 +53,8 @@ static inline bool kernel_page_present(struct page *page)
{
return true;
}
+
+#define can_set_direct_map() false
#else /* CONFIG_ARCH_HAS_SET_DIRECT_MAP */
/*
* Some architectures, e.g. ARM64 can disable direct map modifications at
diff --git a/lib/buildid.c b/lib/buildid.c
index c4b7376406215..ba79bf28f7e69 100644
--- a/lib/buildid.c
+++ b/lib/buildid.c
@@ -47,6 +47,10 @@ static int freader_get_folio(struct freader *r, loff_t file_off)
freader_put_folio(r);
+ /* reject folios without direct map entries (e.g. from memfd_secret() or guest_memfd()) */
+ if (mapping_no_direct_map(r->file->f_mapping))
+ return -EFAULT;
+
/* only use page cache lookup - fail if not already cached */
r->folio = filemap_get_folio(r->file->f_mapping, file_off >> PAGE_SHIFT);
@@ -87,8 +91,8 @@ const void *freader_fetch(struct freader *r, loff_t file_off, size_t sz)
return r->data + file_off;
}
- /* reject secretmem folios created with memfd_secret() */
- if (secretmem_mapping(r->file->f_mapping)) {
+ /* reject folios without direct map entries (e.g. from memfd_secret() or guest_memfd()) */
+ if (mapping_no_direct_map(r->file->f_mapping)) {
r->err = -EFAULT;
return NULL;
}
diff --git a/mm/filemap.c b/mm/filemap.c
index 08cb1efe665ac..341c7c5b28bc7 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -14,6 +14,7 @@
#include <linux/compiler.h>
#include <linux/dax.h>
#include <linux/fs.h>
+#include <linux/set_memory.h>
#include <linux/sched/signal.h>
#include <linux/uaccess.h>
#include <linux/capability.h>
@@ -239,6 +240,54 @@ static void filemap_free_folio(const struct address_space *mapping,
folio_put_refs(folio, folio_nr_pages(folio));
}
+#ifdef CONFIG_ARCH_HAS_SET_DIRECT_MAP
+static inline int prep_add_unmapped_folio(struct address_space *mapping,
+ struct folio *folio)
+{
+ if (!mapping_no_direct_map(mapping))
+ return 0;
+
+ return folio_zap_direct_map(folio);
+}
+
+static inline void prep_remove_unmapped_folio(struct address_space *mapping,
+ struct folio *folio)
+{
+ if (!mapping_no_direct_map(mapping))
+ return;
+
+ folio_restore_direct_map(folio);
+}
+
+static inline void prep_remove_unmapped_batch(struct address_space *mapping,
+ struct folio_batch *fbatch)
+{
+ if (!mapping_no_direct_map(mapping))
+ return;
+
+ for (int i = 0; i < folio_batch_count(fbatch); i++)
+ folio_restore_direct_map(fbatch->folios[i]);
+}
+#else
+static inline int prep_add_unmapped_folio(struct address_space *mapping, struct folio *folio)
+{
+ VM_WARN_ON(mapping_no_direct_map(mapping));
+ return 0;
+}
+
+static inline void prep_remove_unmapped_folio(struct address_space *mapping,
+ struct folio *folio)
+{
+ VM_WARN_ON(mapping_no_direct_map(mapping));
+}
+
+static inline void prep_remove_unmapped_batch(struct address_space *mapping,
+ struct folio_batch *fbatch)
+{
+ VM_WARN_ON(mapping_no_direct_map(mapping));
+}
+#endif
+
/**
* filemap_remove_folio - Remove folio from page cache.
* @folio: The folio.
@@ -260,6 +309,8 @@ void filemap_remove_folio(struct folio *folio)
inode_lru_list_add(mapping->host);
spin_unlock(&mapping->host->i_lock);
+ prep_remove_unmapped_folio(mapping, folio);
+
filemap_free_folio(mapping, folio);
}
@@ -339,6 +390,8 @@ void delete_from_page_cache_batch(struct address_space *mapping,
inode_lru_list_add(mapping->host);
spin_unlock(&mapping->host->i_lock);
+ prep_remove_unmapped_batch(mapping, fbatch);
+
for (i = 0; i < folio_batch_count(fbatch); i++)
filemap_free_folio(mapping, fbatch->folios[i]);
}
@@ -963,28 +1016,38 @@ int filemap_add_folio(struct address_space *mapping, struct folio *folio,
return ret;
__folio_set_locked(folio);
+
+ ret = prep_add_unmapped_folio(mapping, folio);
+ if (unlikely(ret))
+ goto err;
+
ret = __filemap_add_folio(mapping, folio, index, gfp, &shadow);
- if (unlikely(ret)) {
- mem_cgroup_uncharge(folio);
- __folio_clear_locked(folio);
- } else {
- /*
- * The folio might have been evicted from cache only
- * recently, in which case it should be activated like
- * any other repeatedly accessed folio.
- * The exception is folios getting rewritten; evicting other
- * data from the working set, only to cache data that will
- * get overwritten with something else, is a waste of memory.
- */
- WARN_ON_ONCE(folio_test_active(folio));
- if (!(gfp & __GFP_WRITE) && shadow)
- workingset_refault(folio, shadow);
- folio_add_lru(folio);
- if (kernel_file)
- mod_node_page_state(folio_pgdat(folio),
- NR_KERNEL_FILE_PAGES,
- folio_nr_pages(folio));
- }
+ if (unlikely(ret))
+ goto err_restore;
+
+ /*
+ * The folio might have been evicted from cache only
+ * recently, in which case it should be activated like
+ * any other repeatedly accessed folio.
+ * The exception is folios getting rewritten; evicting other
+ * data from the working set, only to cache data that will
+ * get overwritten with something else, is a waste of memory.
+ */
+ WARN_ON_ONCE(folio_test_active(folio));
+ if (!(gfp & __GFP_WRITE) && shadow)
+ workingset_refault(folio, shadow);
+ folio_add_lru(folio);
+ if (kernel_file)
+ mod_node_page_state(folio_pgdat(folio),
+ NR_KERNEL_FILE_PAGES,
+ folio_nr_pages(folio));
+ return ret;
+
+err_restore:
+ prep_remove_unmapped_folio(mapping, folio);
+err:
+ mem_cgroup_uncharge(folio);
+ __folio_clear_locked(folio);
return ret;
}
EXPORT_SYMBOL_GPL(filemap_add_folio);
diff --git a/mm/gup.c b/mm/gup.c
index 1d32e9a3dc79c..5297f0f84101d 100644
--- a/mm/gup.c
+++ b/mm/gup.c
@@ -11,7 +11,6 @@
#include <linux/rmap.h>
#include <linux/swap.h>
#include <linux/swapops.h>
-#include <linux/secretmem.h>
#include <linux/sched/signal.h>
#include <linux/rwsem.h>
@@ -1216,7 +1215,7 @@ static int check_vma_flags(struct vm_area_struct *vma, unsigned long gup_flags)
if ((gup_flags & FOLL_SPLIT_PMD) && is_vm_hugetlb_page(vma))
return -EOPNOTSUPP;
- if (vma_is_secretmem(vma))
+ if (vma_has_no_direct_map(vma))
return -EFAULT;
if (write) {
@@ -2731,7 +2730,7 @@ EXPORT_SYMBOL(get_user_pages_unlocked);
* This call assumes the caller has pinned the folio, that the lowest page table
* level still points to this folio, and that interrupts have been disabled.
*
- * GUP-fast must reject all secretmem folios.
+ * GUP-fast must reject all folios without direct map entries (such as secretmem).
*
* Writing to pinned file-backed dirty tracked folios is inherently problematic
* (see comment describing the writable_file_mapping_allowed() function). We
@@ -2769,7 +2768,7 @@ static bool gup_fast_folio_allowed(struct folio *folio, unsigned int flags)
if (WARN_ON_ONCE(folio_test_slab(folio)))
return false;
- /* hugetlb neither requires dirty-tracking nor can be secretmem. */
+ /* hugetlb neither requires dirty-tracking nor can be without direct map. */
if (folio_test_hugetlb(folio))
return true;
@@ -2812,7 +2811,7 @@ static bool gup_fast_folio_allowed(struct folio *folio, unsigned int flags)
* At this point, we know the mapping is non-null and points to an
* address_space object.
*/
- if (check_secretmem && secretmem_mapping(mapping))
+ if (mapping_no_direct_map(mapping))
return false;
/* The only remaining allowed file system is shmem. */
return !reject_file_backed || shmem_mapping(mapping);
diff --git a/mm/mlock.c b/mm/mlock.c
index efa6716e4dfbd..045b6779440b1 100644
--- a/mm/mlock.c
+++ b/mm/mlock.c
@@ -474,7 +474,7 @@ static int mlock_fixup(struct vma_iterator *vmi, struct vm_area_struct *vma,
int ret = 0;
if (vma_flags_same_pair(&old_vma_flags, new_vma_flags) ||
- vma_is_secretmem(vma) || !vma_supports_mlock(vma)) {
+ vma_has_no_direct_map(vma) || !vma_supports_mlock(vma)) {
/*
* Don't set VMA_LOCKED_BIT or VMA_LOCKONFAULT_BIT and don't
* count. For secretmem, don't allow the memory to be unlocked.
diff --git a/mm/secretmem.c b/mm/secretmem.c
index 4a4934769f8ba..c043c53687d95 100644
--- a/mm/secretmem.c
+++ b/mm/secretmem.c
@@ -52,49 +52,20 @@ static vm_fault_t secretmem_fault(struct vm_fault *vmf)
struct address_space *mapping = vmf->vma->vm_file->f_mapping;
struct inode *inode = file_inode(vmf->vma->vm_file);
pgoff_t offset = vmf->pgoff;
- gfp_t gfp = vmf->gfp_mask;
struct folio *folio;
vm_fault_t ret;
- int err;
if (((loff_t)vmf->pgoff << PAGE_SHIFT) >= i_size_read(inode))
return vmf_error(-EINVAL);
filemap_invalidate_lock_shared(mapping);
-retry:
- folio = filemap_lock_folio(mapping, offset);
+ folio = filemap_grab_folio(mapping, offset);
if (IS_ERR(folio)) {
- folio = folio_alloc(gfp | __GFP_ZERO, 0);
- if (!folio) {
- ret = VM_FAULT_OOM;
- goto out;
- }
-
- err = folio_zap_direct_map(folio);
- if (err) {
- folio_put(folio);
- ret = vmf_error(err);
- goto out;
- }
-
- __folio_mark_uptodate(folio);
- err = filemap_add_folio(mapping, folio, offset, gfp);
- if (unlikely(err)) {
- /*
- * If a split of large page was required, it
- * already happened when we marked the page invalid
- * which guarantees that this call won't fail
- */
- folio_restore_direct_map(folio);
- folio_put(folio);
- if (err == -EEXIST)
- goto retry;
-
- ret = vmf_error(err);
- goto out;
- }
+ ret = vmf_error(PTR_ERR(folio));
+ goto out;
}
+ folio_mark_uptodate(folio);
vmf->page = folio_file_page(folio, vmf->pgoff);
ret = VM_FAULT_LOCKED;
@@ -129,11 +100,6 @@ static int secretmem_mmap_prepare(struct vm_area_desc *desc)
return 0;
}
-bool vma_is_secretmem(struct vm_area_struct *vma)
-{
- return vma->vm_ops == &secretmem_vm_ops;
-}
-
static const struct file_operations secretmem_fops = {
.release = secretmem_release,
.mmap_prepare = secretmem_mmap_prepare,
@@ -147,11 +113,10 @@ static int secretmem_migrate_folio(struct address_space *mapping,
static void secretmem_free_folio(struct folio *folio)
{
- set_direct_map_default_noflush(folio_page(folio, 0));
folio_zero_segment(folio, 0, folio_size(folio));
}
-const struct address_space_operations secretmem_aops = {
+static const struct address_space_operations secretmem_aops = {
.dirty_folio = noop_dirty_folio,
.free_folio = secretmem_free_folio,
.migrate_folio = secretmem_migrate_folio,
@@ -200,6 +165,7 @@ static struct file *secretmem_file_create(unsigned long flags)
mapping_set_gfp_mask(inode->i_mapping, GFP_USER);
mapping_set_unevictable(inode->i_mapping);
+ mapping_set_no_direct_map(inode->i_mapping);
inode->i_op = &secretmem_iops;
inode->i_mapping->a_ops = &secretmem_aops;
--
2.54.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v3 04/26] x86/mm: split out preallocate_sub_pgd()
2026-07-26 22:22 [PATCH v3 00/26] mm: Add ALLOC_UNMAPPED and AS_NO_DIRECT_MAP Brendan Jackman
` (2 preceding siblings ...)
2026-07-26 22:22 ` [PATCH v3 03/26] mm: introduce AS_NO_DIRECT_MAP Brendan Jackman
@ 2026-07-26 22:22 ` Brendan Jackman
2026-07-26 22:22 ` [PATCH v3 05/26] x86: move PAE PMD preallocation defines to header Brendan Jackman
` (21 subsequent siblings)
25 siblings, 0 replies; 29+ messages in thread
From: Brendan Jackman @ 2026-07-26 22:22 UTC (permalink / raw)
To: Borislav Petkov, Dave Hansen, Peter Zijlstra, Andrew Morton,
David Hildenbrand, Vlastimil Babka, Mike Rapoport, Wei Xu,
Johannes Weiner, Zi Yan, Lorenzo Stoakes
Cc: linux-mm, linux-kernel, x86, rppt, Sumit Garg, Will Deacon,
rientjes, Kalyazin, Nikita, patrick.roy, Itazuri, Takahiro,
Andy Lutomirski, David Kaplan, Thomas Gleixner, Yosry Ahmed,
Patrick Bellasi, Reiji Watanabe, Sean Christopherson,
Brendan Jackman
This code will be needed elsewhere in a following patch. Split out the
trivial code move for easy review.
As a side effect, change the logging slightly: instead of directly
reporting the level of the failure in panic(), show a generic panic
message, will be preceded by a separate warn that reports the level of
the failure. This is a simple way to have this helper suit the needs of
its new user as well as the existing one.
Other than logging, no functional change intended.
Signed-off-by: Brendan Jackman <jackmanb@google.com>
---
arch/x86/include/asm/pgalloc.h | 3 +++
arch/x86/mm/init_64.c | 44 +++++++-----------------------------------
arch/x86/mm/pgtable.c | 38 ++++++++++++++++++++++++++++++++++++
3 files changed, 48 insertions(+), 37 deletions(-)
diff --git a/arch/x86/include/asm/pgalloc.h b/arch/x86/include/asm/pgalloc.h
index c88691b15f3c6..2aba6cfabf495 100644
--- a/arch/x86/include/asm/pgalloc.h
+++ b/arch/x86/include/asm/pgalloc.h
@@ -2,6 +2,7 @@
#ifndef _ASM_X86_PGALLOC_H
#define _ASM_X86_PGALLOC_H
+#include <linux/printk.h>
#include <linux/threads.h>
#include <linux/mm.h> /* for struct page */
#include <linux/pagemap.h>
@@ -128,6 +129,8 @@ static inline void __pud_free_tlb(struct mmu_gather *tlb, pud_t *pud,
___pud_free_tlb(tlb, pud);
}
+extern int preallocate_sub_pgd(struct mm_struct *mm, unsigned long addr);
+
#if CONFIG_PGTABLE_LEVELS > 4
static inline void pgd_populate(struct mm_struct *mm, pgd_t *pgd, p4d_t *p4d)
{
diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
index ab4c5a02326f7..ac6688c70872e 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -1293,46 +1293,16 @@ static struct kcore_list kcore_vsyscall;
static void __init preallocate_vmalloc_pages(void)
{
unsigned long addr;
- const char *lvl;
for (addr = VMALLOC_START; addr <= VMEMORY_END; addr = ALIGN(addr + 1, PGDIR_SIZE)) {
- pgd_t *pgd = pgd_offset_k(addr);
- p4d_t *p4d;
- pud_t *pud;
-
- lvl = "p4d";
- p4d = p4d_alloc(&init_mm, pgd, addr);
- if (!p4d)
- goto failed;
-
- if (pgtable_l5_enabled())
- continue;
-
- /*
- * The goal here is to allocate all possibly required
- * hardware page tables pointed to by the top hardware
- * level.
- *
- * On 4-level systems, the P4D layer is folded away and
- * the above code does no preallocation. Below, go down
- * to the pud _software_ level to ensure the second
- * hardware level is allocated on 4-level systems too.
- */
- lvl = "pud";
- pud = pud_alloc(&init_mm, p4d, addr);
- if (!pud)
- goto failed;
+ if (preallocate_sub_pgd(&init_mm, addr)) {
+ /*
+ * The pages have to be there now or they will be
+ * missing in process page-tables later.
+ */
+ panic("Failed to pre-allocate pagetables for vmalloc area\n");
+ }
}
-
- return;
-
-failed:
-
- /*
- * The pages have to be there now or they will be missing in
- * process page-tables later.
- */
- panic("Failed to pre-allocate %s pages for vmalloc area\n", lvl);
}
void __init arch_mm_preinit(void)
diff --git a/arch/x86/mm/pgtable.c b/arch/x86/mm/pgtable.c
index f32facdb30354..fdd3709509946 100644
--- a/arch/x86/mm/pgtable.c
+++ b/arch/x86/mm/pgtable.c
@@ -833,3 +833,41 @@ void arch_check_zapped_pud(struct vm_area_struct *vma, pud_t pud)
/* See note in arch_check_zapped_pte() */
VM_WARN_ON_ONCE(!(vma->vm_flags & VM_SHADOW_STACK) && pud_shstk(pud));
}
+
+#if CONFIG_PGTABLE_LEVELS > 3
+/*
+ * Allocate all possibly required hardware page tables pointed to ths
+ * top hardware level. In other words, allocate a p4d on 5-level or a
+ * pud on 4-level.
+ */
+int preallocate_sub_pgd(struct mm_struct *mm, unsigned long addr)
+{
+ const char *lvl;
+ p4d_t *p4d;
+ pud_t *pud;
+
+ lvl = "p4d";
+ p4d = p4d_alloc(mm, pgd_offset_pgd(mm->pgd, addr), addr);
+ if (!p4d)
+ goto failed;
+
+ if (pgtable_l5_enabled())
+ return 0;
+
+ /*
+ * On 4-level systems, the P4D layer is folded away and
+ * the above code does no preallocation. Below, go down
+ * to the pud _software_ level to ensure the second
+ * hardware level is allocated on 4-level systems too.
+ */
+ lvl = "pud";
+ pud = pud_alloc(mm, p4d, addr);
+ if (!pud)
+ goto failed;
+ return 0;
+
+failed:
+ pr_warn_ratelimited("Failed to preallocate %s\n", lvl);
+ return -ENOMEM;
+}
+#endif
--
2.54.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v3 05/26] x86: move PAE PMD preallocation defines to header
2026-07-26 22:22 [PATCH v3 00/26] mm: Add ALLOC_UNMAPPED and AS_NO_DIRECT_MAP Brendan Jackman
` (3 preceding siblings ...)
2026-07-26 22:22 ` [PATCH v3 04/26] x86/mm: split out preallocate_sub_pgd() Brendan Jackman
@ 2026-07-26 22:22 ` Brendan Jackman
2026-07-26 22:22 ` [PATCH v3 06/26] x86/tlb: Expose some flush function declarations to modules Brendan Jackman
` (20 subsequent siblings)
25 siblings, 0 replies; 29+ messages in thread
From: Brendan Jackman @ 2026-07-26 22:22 UTC (permalink / raw)
To: Borislav Petkov, Dave Hansen, Peter Zijlstra, Andrew Morton,
David Hildenbrand, Vlastimil Babka, Mike Rapoport, Wei Xu,
Johannes Weiner, Zi Yan, Lorenzo Stoakes
Cc: linux-mm, linux-kernel, x86, rppt, Sumit Garg, Will Deacon,
rientjes, Kalyazin, Nikita, patrick.roy, Itazuri, Takahiro,
Andy Lutomirski, David Kaplan, Thomas Gleixner, Yosry Ahmed,
Patrick Bellasi, Reiji Watanabe, Sean Christopherson,
Brendan Jackman
In a subsequent patch these defines will need to be referenced from a new
file, move them as a separate patch for easy review.
The comments have style violations (personal pronouns etc), do not fix
them as this is just code movement.
No functional change intended.
Signed-off-by: Brendan Jackman <jackmanb@google.com>
---
arch/x86/include/asm/page.h | 32 ++++++++++++++++++++++++++++++++
arch/x86/mm/pgtable.c | 29 -----------------------------
2 files changed, 32 insertions(+), 29 deletions(-)
diff --git a/arch/x86/include/asm/page.h b/arch/x86/include/asm/page.h
index 416dc88e35c15..4de4715c3b40f 100644
--- a/arch/x86/include/asm/page.h
+++ b/arch/x86/include/asm/page.h
@@ -78,6 +78,38 @@ static __always_inline u64 __is_canonical_address(u64 vaddr, u8 vaddr_bits)
return __canonical_address(vaddr, vaddr_bits) == vaddr;
}
+#ifdef CONFIG_X86_PAE
+
+/*
+ * In PAE mode, we need to do a cr3 reload (=tlb flush) when
+ * updating the top-level pagetable entries to guarantee the
+ * processor notices the update. Since this is expensive, and
+ * all 4 top-level entries are used almost immediately in a
+ * new process's life, we just pre-populate them here.
+ */
+#define PREALLOCATED_PMDS PTRS_PER_PGD
+/*
+ * "USER_PMDS" are the PMDs for the user copy of the page tables when
+ * PTI is enabled. They do not exist when PTI is disabled. Note that
+ * this is distinct from the user _portion_ of the kernel page tables
+ * which always exists.
+ *
+ * We allocate separate PMDs for the kernel part of the user page-table
+ * when PTI is enabled. We need them to map the per-process LDT into the
+ * user-space page-table.
+ */
+#define PREALLOCATED_USER_PMDS (boot_cpu_has(X86_FEATURE_PTI) ? KERNEL_PGD_PTRS : 0)
+#define MAX_PREALLOCATED_USER_PMDS KERNEL_PGD_PTRS
+
+#else /* !CONFIG_X86_PAE */
+
+/* No need to prepopulate any pagetable entries in non-PAE modes. */
+#define PREALLOCATED_PMDS 0
+#define PREALLOCATED_USER_PMDS 0
+#define MAX_PREALLOCATED_USER_PMDS 0
+
+#endif /* CONFIG_X86_PAE */
+
#endif /* __ASSEMBLER__ */
#include <asm-generic/memory_model.h>
diff --git a/arch/x86/mm/pgtable.c b/arch/x86/mm/pgtable.c
index fdd3709509946..8f3505b1eefec 100644
--- a/arch/x86/mm/pgtable.c
+++ b/arch/x86/mm/pgtable.c
@@ -100,29 +100,6 @@ static void pgd_dtor(pgd_t *pgd)
}
#ifdef CONFIG_X86_PAE
-/*
- * In PAE mode, we need to do a cr3 reload (=tlb flush) when
- * updating the top-level pagetable entries to guarantee the
- * processor notices the update. Since this is expensive, and
- * all 4 top-level entries are used almost immediately in a
- * new process's life, we just pre-populate them here.
- */
-#define PREALLOCATED_PMDS PTRS_PER_PGD
-
-/*
- * "USER_PMDS" are the PMDs for the user copy of the page tables when
- * PTI is enabled. They do not exist when PTI is disabled. Note that
- * this is distinct from the user _portion_ of the kernel page tables
- * which always exists.
- *
- * We allocate separate PMDs for the kernel part of the user page-table
- * when PTI is enabled. We need them to map the per-process LDT into the
- * user-space page-table.
- */
-#define PREALLOCATED_USER_PMDS (boot_cpu_has(X86_FEATURE_PTI) ? \
- KERNEL_PGD_PTRS : 0)
-#define MAX_PREALLOCATED_USER_PMDS KERNEL_PGD_PTRS
-
void pud_populate(struct mm_struct *mm, pud_t *pudp, pmd_t *pmd)
{
paravirt_alloc_pmd(mm, __pa(pmd) >> PAGE_SHIFT);
@@ -139,12 +116,6 @@ void pud_populate(struct mm_struct *mm, pud_t *pudp, pmd_t *pmd)
*/
flush_tlb_mm(mm);
}
-#else /* !CONFIG_X86_PAE */
-
-/* No need to prepopulate any pagetable entries in non-PAE modes. */
-#define PREALLOCATED_PMDS 0
-#define PREALLOCATED_USER_PMDS 0
-#define MAX_PREALLOCATED_USER_PMDS 0
#endif /* CONFIG_X86_PAE */
static void free_pmds(struct mm_struct *mm, pmd_t *pmds[], int count)
--
2.54.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v3 06/26] x86/tlb: Expose some flush function declarations to modules
2026-07-26 22:22 [PATCH v3 00/26] mm: Add ALLOC_UNMAPPED and AS_NO_DIRECT_MAP Brendan Jackman
` (4 preceding siblings ...)
2026-07-26 22:22 ` [PATCH v3 05/26] x86: move PAE PMD preallocation defines to header Brendan Jackman
@ 2026-07-26 22:22 ` Brendan Jackman
2026-07-26 22:22 ` [PATCH v3 07/26] x86/mm: introduce mm-local region Brendan Jackman
` (19 subsequent siblings)
25 siblings, 0 replies; 29+ messages in thread
From: Brendan Jackman @ 2026-07-26 22:22 UTC (permalink / raw)
To: Borislav Petkov, Dave Hansen, Peter Zijlstra, Andrew Morton,
David Hildenbrand, Vlastimil Babka, Mike Rapoport, Wei Xu,
Johannes Weiner, Zi Yan, Lorenzo Stoakes
Cc: linux-mm, linux-kernel, x86, rppt, Sumit Garg, Will Deacon,
rientjes, Kalyazin, Nikita, patrick.roy, Itazuri, Takahiro,
Andy Lutomirski, David Kaplan, Thomas Gleixner, Yosry Ahmed,
Patrick Bellasi, Reiji Watanabe, Sean Christopherson,
Brendan Jackman
In commit bfe3d8f6313d ("x86/tlb: Restrict access to tlbstate") some
low-level logic (the important detail here is flush_tlb_info) was hidden
from modules, along with functions associated with that data.
Later, the set of functions defined here changed and there are now a
bunch of flush_tlb_*() functions that do not depend on x86 internals
like flush_tlb_info.
This leads to some build fragility: KVM (which can be a module) cares
about TLB flushing and includes {linux->asm}/mmu_context.h which
includes asm/tlb.h and asm/tlbflush.h. This x86 TLB code expects these
helpers to be defined (e.g. tlb_flush() calls flush_tlb_mm_range()).
Modules probably shouldn't call these helpers - luckily this is already
enforced by the lack of EXPORT_SYMBOL(). Therefore keep things simple
and just expose the declarations anyway to prevent build failures.
Signed-off-by: Brendan Jackman <jackmanb@google.com>
---
arch/x86/include/asm/tlbflush.h | 43 +++++++++++++++++++++--------------------
1 file changed, 22 insertions(+), 21 deletions(-)
diff --git a/arch/x86/include/asm/tlbflush.h b/arch/x86/include/asm/tlbflush.h
index 0545fe75c3fa1..56cff03b65aa2 100644
--- a/arch/x86/include/asm/tlbflush.h
+++ b/arch/x86/include/asm/tlbflush.h
@@ -251,7 +251,6 @@ struct flush_tlb_info {
u8 trim_cpumask;
};
-void flush_tlb_local(void);
void flush_tlb_one_user(unsigned long addr);
void flush_tlb_one_kernel(unsigned long addr);
void flush_tlb_multi(const struct cpumask *cpumask,
@@ -325,26 +324,6 @@ static inline void mm_clear_asid_transition(struct mm_struct *mm) { }
static inline bool mm_in_asid_transition(struct mm_struct *mm) { return false; }
#endif /* CONFIG_BROADCAST_TLB_FLUSH */
-#define flush_tlb_mm(mm) \
- flush_tlb_mm_range(mm, 0UL, TLB_FLUSH_ALL, 0UL, true)
-
-#define flush_tlb_range(vma, start, end) \
- flush_tlb_mm_range((vma)->vm_mm, start, end, \
- ((vma)->vm_flags & VM_HUGETLB) \
- ? huge_page_shift(hstate_vma(vma)) \
- : PAGE_SHIFT, true)
-
-extern void flush_tlb_all(void);
-extern void flush_tlb_mm_range(struct mm_struct *mm, unsigned long start,
- unsigned long end, unsigned int stride_shift,
- bool freed_tables);
-extern void flush_tlb_kernel_range(unsigned long start, unsigned long end);
-
-static inline void flush_tlb_page(struct vm_area_struct *vma, unsigned long a)
-{
- flush_tlb_mm_range(vma->vm_mm, a, a + PAGE_SIZE, PAGE_SHIFT, false);
-}
-
static inline bool arch_tlbbatch_should_defer(struct mm_struct *mm)
{
bool should_defer = false;
@@ -513,4 +492,26 @@ static inline void __native_tlb_flush_global(unsigned long cr4)
native_write_cr4(cr4 ^ X86_CR4_PGE);
native_write_cr4(cr4);
}
+
+#define flush_tlb_mm(mm) \
+ flush_tlb_mm_range(mm, 0UL, TLB_FLUSH_ALL, 0UL, true)
+
+#define flush_tlb_range(vma, start, end) \
+ flush_tlb_mm_range((vma)->vm_mm, start, end, \
+ ((vma)->vm_flags & VM_HUGETLB) \
+ ? huge_page_shift(hstate_vma(vma)) \
+ : PAGE_SHIFT, true)
+
+void flush_tlb_local(void);
+extern void flush_tlb_all(void);
+extern void flush_tlb_mm_range(struct mm_struct *mm, unsigned long start,
+ unsigned long end, unsigned int stride_shift,
+ bool freed_tables);
+extern void flush_tlb_kernel_range(unsigned long start, unsigned long end);
+
+static inline void flush_tlb_page(struct vm_area_struct *vma, unsigned long a)
+{
+ flush_tlb_mm_range(vma->vm_mm, a, a + PAGE_SIZE, PAGE_SHIFT, false);
+}
+
#endif /* _ASM_X86_TLBFLUSH_H */
--
2.54.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v3 07/26] x86/mm: introduce mm-local region
2026-07-26 22:22 [PATCH v3 00/26] mm: Add ALLOC_UNMAPPED and AS_NO_DIRECT_MAP Brendan Jackman
` (5 preceding siblings ...)
2026-07-26 22:22 ` [PATCH v3 06/26] x86/tlb: Expose some flush function declarations to modules Brendan Jackman
@ 2026-07-26 22:22 ` Brendan Jackman
2026-07-26 22:22 ` [PATCH v3 08/26] x86/mm: move LDT remap into " Brendan Jackman
` (18 subsequent siblings)
25 siblings, 0 replies; 29+ messages in thread
From: Brendan Jackman @ 2026-07-26 22:22 UTC (permalink / raw)
To: Borislav Petkov, Dave Hansen, Peter Zijlstra, Andrew Morton,
David Hildenbrand, Vlastimil Babka, Mike Rapoport, Wei Xu,
Johannes Weiner, Zi Yan, Lorenzo Stoakes
Cc: linux-mm, linux-kernel, x86, rppt, Sumit Garg, Will Deacon,
rientjes, Kalyazin, Nikita, patrick.roy, Itazuri, Takahiro,
Andy Lutomirski, David Kaplan, Thomas Gleixner, Yosry Ahmed,
Patrick Bellasi, Reiji Watanabe, Sean Christopherson,
Brendan Jackman
Various security features benefit from having process-local address
mappings within the kernel. Examples include no-direct-map guest_memfd
[2] and significant optimizations for ASI [1].
With the currently envisaged usecases, there will be many situations
where almost no processes have any need for the mm-local region.
Therefore, avoid its overhead (memory cost of pagetables, alloc/free
overhead during fork/exit) for processes that don't use it by requiring
its users to explicitly initialize it via the new mm_local_* API.
As pointed out by Andy in [0], x86 already has a PGD entry that is local
to the mm, which is used for the LDT. In a subsequent patch, the LDT
remap will be unified with the general mm-local region, but to help
keep the patch to a manageable size, first just introduce the mm-local
region.
On 64-bit, give the mm-local region a whole PGD. On 32-bit, just give it
one PMD. No investigation has been done into whether it's feasible to
expand the region on 32-bit. Most likely there is no strong usecase for
that anyway.
In order to combine the need for an on-demand mm initialisation, with
the desire to transparently handle propagating mappings to userspace
under KPTI, the user and kernel pagetables are shared at the highest
level possible. For PAE that means the PTE table is shared and for
64-bit the P4D/PUD. This is implemented by pre-allocating the first
shared table when the mm-local region is first initialised.
[0] https://lore.kernel.org/linux-mm/CALCETrXHbS9VXfZ80kOjiTrreM2EbapYeGp68mvJPbosUtorYA@mail.gmail.com/
[1] https://linuxasi.dev/
[2] https://lore.kernel.org/all/20250924151101.2225820-1-patrick.roy@campus.lmu.de
Signed-off-by: Brendan Jackman <jackmanb@google.com>
---
arch/x86/Kconfig | 2 +
arch/x86/include/asm/mmu_context.h | 118 +++++++++++++++++++++++++++++++-
arch/x86/include/asm/pgtable_32_areas.h | 9 ++-
arch/x86/include/asm/pgtable_64_types.h | 12 +++-
arch/x86/mm/pgtable.c | 3 +
include/linux/mm.h | 13 ++++
include/linux/mm_types.h | 2 +
kernel/fork.c | 1 +
mm/Kconfig | 7 ++
9 files changed, 161 insertions(+), 6 deletions(-)
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index fb298e2191792..3efab3524a6cf 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -132,6 +132,8 @@ config X86
select ARCH_SUPPORTS_LTO_CLANG
select ARCH_SUPPORTS_LTO_CLANG_THIN
select ARCH_SUPPORTS_RT
+ # LDT remap temporarily clashes with mm-local region, can't have both.
+ select ARCH_SUPPORTS_MM_LOCAL_REGION if X86_64 || X86_PAE && !MODIFY_LDT_SYSCALL
select ARCH_USE_BUILTIN_BSWAP
select ARCH_USE_CMPXCHG_LOCKREF
select ARCH_USE_MEMTEST
diff --git a/arch/x86/include/asm/mmu_context.h b/arch/x86/include/asm/mmu_context.h
index ef5b507de34e2..3d4f54673014f 100644
--- a/arch/x86/include/asm/mmu_context.h
+++ b/arch/x86/include/asm/mmu_context.h
@@ -8,8 +8,10 @@
#include <trace/events/tlb.h>
+#include <asm/tlb.h>
#include <asm/tlbflush.h>
#include <asm/paravirt.h>
+#include <asm/pgalloc.h>
#include <asm/debugreg.h>
#include <asm/gsseg.h>
#include <asm/desc.h>
@@ -223,10 +225,124 @@ static inline int arch_dup_mmap(struct mm_struct *oldmm, struct mm_struct *mm)
return ldt_dup_context(oldmm, mm);
}
+#ifdef CONFIG_MM_LOCAL_REGION
+static inline void mm_local_region_free(struct mm_struct *mm)
+{
+ if (!mm_local_region_used(mm))
+ return;
+
+ struct mmu_gather tlb;
+ unsigned long start = MM_LOCAL_BASE_ADDR;
+ unsigned long end = MM_LOCAL_END_ADDR;
+
+ /*
+ * Although free_pgd_range() is intended for freeing user
+ * page-tables, it also works out for kernel mappings on x86.
+ * Use tlb_gather_mmu_fullmm() to avoid confusing the
+ * range-tracking logic in __tlb_adjust_range().
+ */
+ tlb_gather_mmu_fullmm(&tlb, mm);
+ free_pgd_range(&tlb, start, end, start, end);
+ tlb_finish_mmu(&tlb);
+
+ mm_flags_clear(MMF_LOCAL_REGION_USED, mm);
+}
+
+#if defined(CONFIG_MITIGATION_PAGE_TABLE_ISOLATION) && defined(CONFIG_X86_PAE)
+static inline pmd_t *pgd_to_pmd_walk(pgd_t *pgd, unsigned long va)
+{
+ p4d_t *p4d;
+ pud_t *pud;
+
+ if (pgd->pgd == 0)
+ return NULL;
+
+ p4d = p4d_offset(pgd, va);
+ if (p4d_none(*p4d))
+ return NULL;
+
+ pud = pud_offset(p4d, va);
+ if (pud_none(*pud))
+ return NULL;
+
+ return pmd_offset(pud, va);
+}
+
+static inline int mm_local_map_to_user(struct mm_struct *mm)
+{
+ BUILD_BUG_ON(!PREALLOCATED_PMDS);
+ pgd_t *k_pgd = pgd_offset(mm, MM_LOCAL_BASE_ADDR);
+ pgd_t *u_pgd = kernel_to_user_pgdp(k_pgd);
+ pmd_t *k_pmd, *u_pmd;
+ int err;
+
+ k_pmd = pgd_to_pmd_walk(k_pgd, MM_LOCAL_BASE_ADDR);
+ u_pmd = pgd_to_pmd_walk(u_pgd, MM_LOCAL_BASE_ADDR);
+
+ BUILD_BUG_ON(MM_LOCAL_END_ADDR - MM_LOCAL_BASE_ADDR > PMD_SIZE);
+
+ /* Preallocate the PTE table so it can be shared. */
+ err = pte_alloc(mm, k_pmd);
+ if (err)
+ return err;
+
+ /* Point the userspace PMD at the same PTE as the kernel PMD. */
+ set_pmd(u_pmd, *k_pmd);
+ return 0;
+}
+#elif defined(CONFIG_MITIGATION_PAGE_TABLE_ISOLATION)
+static inline int mm_local_map_to_user(struct mm_struct *mm)
+{
+ pgd_t *pgd;
+ int err;
+
+ err = preallocate_sub_pgd(mm, MM_LOCAL_BASE_ADDR);
+ if (err)
+ return err;
+
+ pgd = pgd_offset(mm, MM_LOCAL_BASE_ADDR);
+ set_pgd(kernel_to_user_pgdp(pgd), *pgd);
+ return 0;
+}
+#else
+static inline int mm_local_map_to_user(struct mm_struct *mm)
+{
+ WARN_ONCE(1, "mm_local_map_to_user() not implemented");
+ return -EINVAL;
+}
+#endif
+
+/*
+ * Do initial setup of the mm-local region. Call from process context.
+ *
+ * Under PTI, userspace shares the pagetables for the mm-local region with the
+ * kernel (if you map stuff here, it's immediately mapped into userspace too).
+ * LDT remap. It's assuming nothing gets mapped in here that needs to be
+ * protected from Meltdown-type attacks from the current process.
+ */
+static inline int mm_local_region_init(struct mm_struct *mm)
+{
+ int err;
+
+ if (boot_cpu_has(X86_FEATURE_PTI)) {
+ err = mm_local_map_to_user(mm);
+ if (err)
+ return err;
+ }
+
+ mm_flags_set(MMF_LOCAL_REGION_USED, mm);
+
+ return 0;
+}
+
+#else
+static inline void mm_local_region_free(struct mm_struct *mm) { }
+#endif /* CONFIG_MM_LOCAL_REGION */
+
static inline void arch_exit_mmap(struct mm_struct *mm)
{
paravirt_arch_exit_mmap(mm);
- ldt_arch_exit_mmap(mm);
+ mm_local_region_free(mm);
}
#ifdef CONFIG_X86_64
diff --git a/arch/x86/include/asm/pgtable_32_areas.h b/arch/x86/include/asm/pgtable_32_areas.h
index 921148b429676..7fccb887f8b33 100644
--- a/arch/x86/include/asm/pgtable_32_areas.h
+++ b/arch/x86/include/asm/pgtable_32_areas.h
@@ -30,9 +30,14 @@ extern bool __vmalloc_start_set; /* set once high_memory is set */
#define CPU_ENTRY_AREA_BASE \
((FIXADDR_TOT_START - PAGE_SIZE*(CPU_ENTRY_AREA_PAGES+1)) & PMD_MASK)
-#define LDT_BASE_ADDR \
- ((CPU_ENTRY_AREA_BASE - PAGE_SIZE) & PMD_MASK)
+/*
+ * On 32-bit the mm-local region is currently completely consumed by the LDT
+ * remap.
+ */
+#define MM_LOCAL_BASE_ADDR ((CPU_ENTRY_AREA_BASE - PAGE_SIZE) & PMD_MASK)
+#define MM_LOCAL_END_ADDR (MM_LOCAL_BASE_ADDR + PMD_SIZE)
+#define LDT_BASE_ADDR MM_LOCAL_BASE_ADDR
#define LDT_END_ADDR (LDT_BASE_ADDR + PMD_SIZE)
#define PKMAP_BASE \
diff --git a/arch/x86/include/asm/pgtable_64_types.h b/arch/x86/include/asm/pgtable_64_types.h
index 7eb61ef6a185f..1181565966405 100644
--- a/arch/x86/include/asm/pgtable_64_types.h
+++ b/arch/x86/include/asm/pgtable_64_types.h
@@ -5,8 +5,11 @@
#include <asm/sparsemem.h>
#ifndef __ASSEMBLER__
+#include <linux/build_bug.h>
#include <linux/types.h>
#include <asm/kaslr.h>
+#include <asm/page_types.h>
+#include <uapi/asm/ldt.h>
/*
* These are used to make use of C type-checking..
@@ -100,9 +103,12 @@ extern unsigned int ptrs_per_p4d;
#define GUARD_HOLE_BASE_ADDR (GUARD_HOLE_PGD_ENTRY << PGDIR_SHIFT)
#define GUARD_HOLE_END_ADDR (GUARD_HOLE_BASE_ADDR + GUARD_HOLE_SIZE)
-#define LDT_PGD_ENTRY -240UL
-#define LDT_BASE_ADDR (LDT_PGD_ENTRY << PGDIR_SHIFT)
-#define LDT_END_ADDR (LDT_BASE_ADDR + PGDIR_SIZE)
+#define MM_LOCAL_PGD_ENTRY -240UL
+#define MM_LOCAL_BASE_ADDR (MM_LOCAL_PGD_ENTRY << PGDIR_SHIFT)
+#define MM_LOCAL_END_ADDR ((MM_LOCAL_PGD_ENTRY + 1) << PGDIR_SHIFT)
+
+#define LDT_BASE_ADDR MM_LOCAL_BASE_ADDR
+#define LDT_END_ADDR (LDT_BASE_ADDR + PMD_SIZE)
#define __VMALLOC_BASE_L4 0xffffc90000000000UL
#define __VMALLOC_BASE_L5 0xffa0000000000000UL
diff --git a/arch/x86/mm/pgtable.c b/arch/x86/mm/pgtable.c
index 8f3505b1eefec..cc768ea8c19c8 100644
--- a/arch/x86/mm/pgtable.c
+++ b/arch/x86/mm/pgtable.c
@@ -335,6 +335,9 @@ pgd_t *pgd_alloc(struct mm_struct *mm)
void pgd_free(struct mm_struct *mm, pgd_t *pgd)
{
+ /* Should be cleaned up in mmap exit path. */
+ VM_WARN_ON_ONCE(mm_local_region_used(mm));
+
pgd_mop_up_pmds(mm, pgd);
pgd_dtor(pgd);
paravirt_pgd_free(mm, pgd);
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 7fabe6c66b4b7..28022c96edcb5 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -989,6 +989,19 @@ static inline void mm_flags_clear_all(struct mm_struct *mm)
bitmap_zero(ACCESS_PRIVATE(&mm->flags, __mm_flags), NUM_MM_FLAG_BITS);
}
+#ifdef CONFIG_MM_LOCAL_REGION
+static inline bool mm_local_region_used(struct mm_struct *mm)
+{
+ return mm_flags_test(MMF_LOCAL_REGION_USED, mm);
+}
+#else
+static inline bool mm_local_region_used(struct mm_struct *mm)
+{
+ VM_WARN_ON_ONCE(mm_flags_test(MMF_LOCAL_REGION_USED, mm));
+ return false;
+}
+#endif
+
extern const struct vm_operations_struct vma_dummy_vm_ops;
static inline void vma_init(struct vm_area_struct *vma, struct mm_struct *mm)
diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
index b5d4cd3b067bf..d39fddf57edc8 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -1997,6 +1997,8 @@ enum {
#define MMF_TOPDOWN 31 /* mm searches top down by default */
#define MMF_TOPDOWN_MASK BIT(MMF_TOPDOWN)
+#define MMF_LOCAL_REGION_USED 32
+
#define MMF_INIT_LEGACY_MASK (MMF_DUMP_FILTER_MASK |\
MMF_DISABLE_THP_MASK | MMF_HAS_MDWE_MASK |\
MMF_VM_MERGE_ANY_MASK | MMF_TOPDOWN_MASK)
diff --git a/kernel/fork.c b/kernel/fork.c
index f0e2e131a9a5a..4fb23ea33b7da 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -1154,6 +1154,7 @@ static struct mm_struct *mm_init(struct mm_struct *mm, struct task_struct *p)
fail_nocontext:
mm_free_id(mm);
fail_noid:
+ WARN_ON_ONCE(mm_local_region_used(mm));
mm_free_pgd(mm);
fail_mm_init:
free_mm(mm);
diff --git a/mm/Kconfig b/mm/Kconfig
index c52ab6afcb159..cb531c1436f77 100644
--- a/mm/Kconfig
+++ b/mm/Kconfig
@@ -1500,6 +1500,13 @@ config LAZY_MMU_MODE_KUNIT_TEST
If unsure, say N.
+config ARCH_SUPPORTS_MM_LOCAL_REGION
+ bool
+
+config MM_LOCAL_REGION
+ bool
+ depends on ARCH_SUPPORTS_MM_LOCAL_REGION
+
source "mm/damon/Kconfig"
endmenu
--
2.54.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v3 08/26] x86/mm: move LDT remap into mm-local region
2026-07-26 22:22 [PATCH v3 00/26] mm: Add ALLOC_UNMAPPED and AS_NO_DIRECT_MAP Brendan Jackman
` (6 preceding siblings ...)
2026-07-26 22:22 ` [PATCH v3 07/26] x86/mm: introduce mm-local region Brendan Jackman
@ 2026-07-26 22:22 ` Brendan Jackman
2026-07-26 22:22 ` [PATCH v3 09/26] mm: Create flags arg for __apply_to_page_range() Brendan Jackman
` (17 subsequent siblings)
25 siblings, 0 replies; 29+ messages in thread
From: Brendan Jackman @ 2026-07-26 22:22 UTC (permalink / raw)
To: Borislav Petkov, Dave Hansen, Peter Zijlstra, Andrew Morton,
David Hildenbrand, Vlastimil Babka, Mike Rapoport, Wei Xu,
Johannes Weiner, Zi Yan, Lorenzo Stoakes
Cc: linux-mm, linux-kernel, x86, rppt, Sumit Garg, Will Deacon,
rientjes, Kalyazin, Nikita, patrick.roy, Itazuri, Takahiro,
Andy Lutomirski, David Kaplan, Thomas Gleixner, Yosry Ahmed,
Patrick Bellasi, Reiji Watanabe, Sean Christopherson,
Brendan Jackman
Now that x86 processes have a general mm-local region, the LDT-specific
management of the higher-level pagetables can mostly be replaced by just
using the generic mm-local API.
Drop all management of pagetable allocation and freeing; that is now
handled automatically by virtue of the pagetables being in the mm-local
region.
Drop explicit logic to map LDTs into the user pagetables under PTI; that
also happens automatically for this region.
Unify the sanity-checking logic between x86_64 and PAE: use the generic
set_memory.c mechanism to walk pagetables. This means the
sanity-checking is slightly more relaxed, since lookup_address_in_pgd()
is more flexible than pgd_to_pmd_walk(), but this seems to be worth it
for the simplified code. It means that ldt.c doesn't have to know about
the exact structure of the mm-local region's pagetables.
Signed-off-by: Brendan Jackman <jackmanb@google.com>
---
Documentation/arch/x86/x86_64/mm.rst | 4 +-
arch/x86/Kconfig | 4 +-
arch/x86/include/asm/mmu_context.h | 2 -
arch/x86/kernel/ldt.c | 124 +++++------------------------------
4 files changed, 20 insertions(+), 114 deletions(-)
diff --git a/Documentation/arch/x86/x86_64/mm.rst b/Documentation/arch/x86/x86_64/mm.rst
index a6cf05d51bd8c..fa2bb7bab6a42 100644
--- a/Documentation/arch/x86/x86_64/mm.rst
+++ b/Documentation/arch/x86/x86_64/mm.rst
@@ -53,7 +53,7 @@ Complete virtual memory map with 4-level page tables
____________________________________________________________|___________________________________________________________
| | | |
ffff800000000000 | -128 TB | ffff87ffffffffff | 8 TB | ... guard hole, also reserved for hypervisor
- ffff880000000000 | -120 TB | ffff887fffffffff | 0.5 TB | LDT remap for PTI
+ ffff880000000000 | -120 TB | ffff887fffffffff | 0.5 TB | MM-local kernel data. Includes LDT remap for PTI
ffff888000000000 | -119.5 TB | ffffc87fffffffff | 64 TB | direct mapping of all physical memory (page_offset_base)
ffffc88000000000 | -55.5 TB | ffffc8ffffffffff | 0.5 TB | ... unused hole
ffffc90000000000 | -55 TB | ffffe8ffffffffff | 32 TB | vmalloc/ioremap space (vmalloc_base)
@@ -123,7 +123,7 @@ Complete virtual memory map with 5-level page tables
____________________________________________________________|___________________________________________________________
| | | |
ff00000000000000 | -64 PB | ff0fffffffffffff | 4 PB | ... guard hole, also reserved for hypervisor
- ff10000000000000 | -60 PB | ff10ffffffffffff | 0.25 PB | LDT remap for PTI
+ ff10000000000000 | -60 PB | ff10ffffffffffff | 0.25 PB | MM-local kernel data. Includes LDT remap for PTI
ff11000000000000 | -59.75 PB | ff90ffffffffffff | 32 PB | direct mapping of all physical memory (page_offset_base)
ff91000000000000 | -27.75 PB | ff9fffffffffffff | 3.75 PB | ... unused hole
ffa0000000000000 | -24 PB | ffd1ffffffffffff | 12.5 PB | vmalloc/ioremap space (vmalloc_base)
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 3efab3524a6cf..33c1282bfbf93 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -132,8 +132,7 @@ config X86
select ARCH_SUPPORTS_LTO_CLANG
select ARCH_SUPPORTS_LTO_CLANG_THIN
select ARCH_SUPPORTS_RT
- # LDT remap temporarily clashes with mm-local region, can't have both.
- select ARCH_SUPPORTS_MM_LOCAL_REGION if X86_64 || X86_PAE && !MODIFY_LDT_SYSCALL
+ select ARCH_SUPPORTS_MM_LOCAL_REGION if X86_64 || X86_PAE
select ARCH_USE_BUILTIN_BSWAP
select ARCH_USE_CMPXCHG_LOCKREF
select ARCH_USE_MEMTEST
@@ -2280,6 +2279,7 @@ config CMDLINE_OVERRIDE
config MODIFY_LDT_SYSCALL
bool "Enable the LDT (local descriptor table)" if EXPERT
default y
+ select MM_LOCAL_REGION if MITIGATION_PAGE_TABLE_ISOLATION || X86_PAE
help
Linux can allow user programs to install a per-process x86
Local Descriptor Table (LDT) using the modify_ldt(2) system
diff --git a/arch/x86/include/asm/mmu_context.h b/arch/x86/include/asm/mmu_context.h
index 3d4f54673014f..09b8d8e6a56ea 100644
--- a/arch/x86/include/asm/mmu_context.h
+++ b/arch/x86/include/asm/mmu_context.h
@@ -61,7 +61,6 @@ static inline void init_new_context_ldt(struct mm_struct *mm)
}
int ldt_dup_context(struct mm_struct *oldmm, struct mm_struct *mm);
void destroy_context_ldt(struct mm_struct *mm);
-void ldt_arch_exit_mmap(struct mm_struct *mm);
#else /* CONFIG_MODIFY_LDT_SYSCALL */
static inline void init_new_context_ldt(struct mm_struct *mm) { }
static inline int ldt_dup_context(struct mm_struct *oldmm,
@@ -70,7 +69,6 @@ static inline int ldt_dup_context(struct mm_struct *oldmm,
return 0;
}
static inline void destroy_context_ldt(struct mm_struct *mm) { }
-static inline void ldt_arch_exit_mmap(struct mm_struct *mm) { }
#endif
#ifdef CONFIG_MODIFY_LDT_SYSCALL
diff --git a/arch/x86/kernel/ldt.c b/arch/x86/kernel/ldt.c
index 40c5bf97dd5cc..685664c1ee770 100644
--- a/arch/x86/kernel/ldt.c
+++ b/arch/x86/kernel/ldt.c
@@ -186,10 +186,16 @@ static struct ldt_struct *alloc_ldt_struct(unsigned int num_entries)
#ifdef CONFIG_MITIGATION_PAGE_TABLE_ISOLATION
-static void do_sanity_check(struct mm_struct *mm,
- bool had_kernel_mapping,
- bool had_user_mapping)
+static void sanity_check_ldt_mapping(struct mm_struct *mm)
{
+ pgd_t *k_pgd = pgd_offset(mm, LDT_BASE_ADDR);
+ pgd_t *u_pgd = kernel_to_user_pgdp(k_pgd);
+ unsigned int k_level, u_level;
+ bool had_kernel_mapping, had_user_mapping;
+
+ had_kernel_mapping = lookup_address_in_pgd(k_pgd, LDT_BASE_ADDR, &k_level);
+ had_user_mapping = lookup_address_in_pgd(u_pgd, LDT_BASE_ADDR, &u_level);
+
if (mm->context.ldt) {
/*
* We already had an LDT. The top-level entry should already
@@ -210,76 +216,6 @@ static void do_sanity_check(struct mm_struct *mm,
}
}
-#ifdef CONFIG_X86_PAE
-
-static pmd_t *pgd_to_pmd_walk(pgd_t *pgd, unsigned long va)
-{
- p4d_t *p4d;
- pud_t *pud;
-
- if (pgd->pgd == 0)
- return NULL;
-
- p4d = p4d_offset(pgd, va);
- if (p4d_none(*p4d))
- return NULL;
-
- pud = pud_offset(p4d, va);
- if (pud_none(*pud))
- return NULL;
-
- return pmd_offset(pud, va);
-}
-
-static void map_ldt_struct_to_user(struct mm_struct *mm)
-{
- pgd_t *k_pgd = pgd_offset(mm, LDT_BASE_ADDR);
- pgd_t *u_pgd = kernel_to_user_pgdp(k_pgd);
- pmd_t *k_pmd, *u_pmd;
-
- k_pmd = pgd_to_pmd_walk(k_pgd, LDT_BASE_ADDR);
- u_pmd = pgd_to_pmd_walk(u_pgd, LDT_BASE_ADDR);
-
- if (boot_cpu_has(X86_FEATURE_PTI) && !mm->context.ldt)
- set_pmd(u_pmd, *k_pmd);
-}
-
-static void sanity_check_ldt_mapping(struct mm_struct *mm)
-{
- pgd_t *k_pgd = pgd_offset(mm, LDT_BASE_ADDR);
- pgd_t *u_pgd = kernel_to_user_pgdp(k_pgd);
- bool had_kernel, had_user;
- pmd_t *k_pmd, *u_pmd;
-
- k_pmd = pgd_to_pmd_walk(k_pgd, LDT_BASE_ADDR);
- u_pmd = pgd_to_pmd_walk(u_pgd, LDT_BASE_ADDR);
- had_kernel = (k_pmd->pmd != 0);
- had_user = (u_pmd->pmd != 0);
-
- do_sanity_check(mm, had_kernel, had_user);
-}
-
-#else /* !CONFIG_X86_PAE */
-
-static void map_ldt_struct_to_user(struct mm_struct *mm)
-{
- pgd_t *pgd = pgd_offset(mm, LDT_BASE_ADDR);
-
- if (boot_cpu_has(X86_FEATURE_PTI) && !mm->context.ldt)
- set_pgd(kernel_to_user_pgdp(pgd), *pgd);
-}
-
-static void sanity_check_ldt_mapping(struct mm_struct *mm)
-{
- pgd_t *pgd = pgd_offset(mm, LDT_BASE_ADDR);
- bool had_kernel = (pgd->pgd != 0);
- bool had_user = (kernel_to_user_pgdp(pgd)->pgd != 0);
-
- do_sanity_check(mm, had_kernel, had_user);
-}
-
-#endif /* CONFIG_X86_PAE */
-
/*
* If PTI is enabled, this maps the LDT into the kernelmode and
* usermode tables for the given mm.
@@ -290,7 +226,7 @@ map_ldt_struct(struct mm_struct *mm, struct ldt_struct *ldt, int slot)
unsigned long va;
bool is_vmalloc;
spinlock_t *ptl;
- int i, nr_pages;
+ int i, nr_pages, err;
if (!boot_cpu_has(X86_FEATURE_PTI))
return 0;
@@ -304,6 +240,10 @@ map_ldt_struct(struct mm_struct *mm, struct ldt_struct *ldt, int slot)
/* Check if the current mappings are sane */
sanity_check_ldt_mapping(mm);
+ err = mm_local_region_init(mm);
+ if (err)
+ return err;
+
is_vmalloc = is_vmalloc_addr(ldt->entries);
nr_pages = DIV_ROUND_UP(ldt->nr_entries * LDT_ENTRY_SIZE, PAGE_SIZE);
@@ -339,9 +279,6 @@ map_ldt_struct(struct mm_struct *mm, struct ldt_struct *ldt, int slot)
pte_unmap_unlock(ptep, ptl);
}
- /* Propagate LDT mapping to the user page-table */
- map_ldt_struct_to_user(mm);
-
ldt->slot = slot;
return 0;
}
@@ -390,28 +327,6 @@ static void unmap_ldt_struct(struct mm_struct *mm, struct ldt_struct *ldt)
}
#endif /* CONFIG_MITIGATION_PAGE_TABLE_ISOLATION */
-static void free_ldt_pgtables(struct mm_struct *mm)
-{
-#ifdef CONFIG_MITIGATION_PAGE_TABLE_ISOLATION
- struct mmu_gather tlb;
- unsigned long start = LDT_BASE_ADDR;
- unsigned long end = LDT_END_ADDR;
-
- if (!boot_cpu_has(X86_FEATURE_PTI))
- return;
-
- /*
- * Although free_pgd_range() is intended for freeing user
- * page-tables, it also works out for kernel mappings on x86.
- * We use tlb_gather_mmu_fullmm() to avoid confusing the
- * range-tracking logic in __tlb_adjust_range().
- */
- tlb_gather_mmu_fullmm(&tlb, mm);
- free_pgd_range(&tlb, start, end, start, end);
- tlb_finish_mmu(&tlb);
-#endif
-}
-
/* After calling this, the LDT is immutable. */
static void finalize_ldt_struct(struct ldt_struct *ldt)
{
@@ -472,7 +387,6 @@ int ldt_dup_context(struct mm_struct *old_mm, struct mm_struct *mm)
retval = map_ldt_struct(mm, new_ldt, 0);
if (retval) {
- free_ldt_pgtables(mm);
free_ldt_struct(new_ldt);
goto out_unlock;
}
@@ -494,11 +408,6 @@ void destroy_context_ldt(struct mm_struct *mm)
mm->context.ldt = NULL;
}
-void ldt_arch_exit_mmap(struct mm_struct *mm)
-{
- free_ldt_pgtables(mm);
-}
-
static int read_ldt(void __user *ptr, unsigned long bytecount)
{
struct mm_struct *mm = current->mm;
@@ -645,10 +554,9 @@ static int write_ldt(void __user *ptr, unsigned long bytecount, int oldmode)
/*
* This only can fail for the first LDT setup. If an LDT is
* already installed then the PTE page is already
- * populated. Mop up a half populated page table.
+ * populated.
*/
- if (!WARN_ON_ONCE(old_ldt))
- free_ldt_pgtables(mm);
+ WARN_ON_ONCE(old_ldt);
free_ldt_struct(new_ldt);
goto out_unlock;
}
--
2.54.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v3 09/26] mm: Create flags arg for __apply_to_page_range()
2026-07-26 22:22 [PATCH v3 00/26] mm: Add ALLOC_UNMAPPED and AS_NO_DIRECT_MAP Brendan Jackman
` (7 preceding siblings ...)
2026-07-26 22:22 ` [PATCH v3 08/26] x86/mm: move LDT remap into " Brendan Jackman
@ 2026-07-26 22:22 ` Brendan Jackman
2026-07-26 22:22 ` [PATCH v3 10/26] mm: Add more flags " Brendan Jackman
` (16 subsequent siblings)
25 siblings, 0 replies; 29+ messages in thread
From: Brendan Jackman @ 2026-07-26 22:22 UTC (permalink / raw)
To: Borislav Petkov, Dave Hansen, Peter Zijlstra, Andrew Morton,
David Hildenbrand, Vlastimil Babka, Mike Rapoport, Wei Xu,
Johannes Weiner, Zi Yan, Lorenzo Stoakes
Cc: linux-mm, linux-kernel, x86, rppt, Sumit Garg, Will Deacon,
rientjes, Kalyazin, Nikita, patrick.roy, Itazuri, Takahiro,
Andy Lutomirski, David Kaplan, Thomas Gleixner, Yosry Ahmed,
Patrick Bellasi, Reiji Watanabe, Sean Christopherson,
Brendan Jackman
Preparatory patch, no functional change intended.
To prepare for making this function more generic, convert the boolean
"create" arg into a flags arg with a single flag that has the same
meaning.
Signed-off-by: Brendan Jackman <jackmanb@google.com>
---
mm/internal.h | 10 ++++++++++
mm/memory.c | 29 +++++++++++++++++------------
2 files changed, 27 insertions(+), 12 deletions(-)
diff --git a/mm/internal.h b/mm/internal.h
index f47f06c555481..395331a12d62d 100644
--- a/mm/internal.h
+++ b/mm/internal.h
@@ -1660,4 +1660,14 @@ static inline bool can_spin_trylock(void)
return true;
}
+/*
+ * Create a mapping if it doesn't exist. (Otherwise, skip regions with no
+ * existing mapping, and return an error for regions with no leaf pagetable).
+ */
+#define PGRANGE_CREATE (1 << 0)
+
+int __apply_to_page_range(struct mm_struct *mm, unsigned long addr,
+ unsigned long size, pte_fn_t fn,
+ void *data, unsigned int flags);
+
#endif /* __MM_INTERNAL_H */
diff --git a/mm/memory.c b/mm/memory.c
index 789c65a6d6a0e..c4defefea1574 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -3438,9 +3438,10 @@ EXPORT_SYMBOL(vm_iomap_memory);
static int apply_to_pte_range(struct mm_struct *mm, pmd_t *pmd,
unsigned long addr, unsigned long end,
- pte_fn_t fn, void *data, bool create,
+ pte_fn_t fn, void *data, unsigned int flags,
pgtbl_mod_mask *mask)
{
+ bool create = flags & PGRANGE_CREATE;
pte_t *pte, *mapped_pte;
int err = 0;
spinlock_t *ptl;
@@ -3481,10 +3482,11 @@ static int apply_to_pte_range(struct mm_struct *mm, pmd_t *pmd,
static int apply_to_pmd_range(struct mm_struct *mm, pud_t *pud,
unsigned long addr, unsigned long end,
- pte_fn_t fn, void *data, bool create,
+ pte_fn_t fn, void *data, unsigned int flags,
pgtbl_mod_mask *mask)
{
pmd_t *pmd;
+ bool create = flags & PGRANGE_CREATE;
unsigned long next;
int err = 0;
@@ -3509,7 +3511,7 @@ static int apply_to_pmd_range(struct mm_struct *mm, pud_t *pud,
pmd_clear_bad(pmd);
}
err = apply_to_pte_range(mm, pmd, addr, next,
- fn, data, create, mask);
+ fn, data, flags, mask);
if (err)
break;
} while (pmd++, addr = next, addr != end);
@@ -3519,10 +3521,11 @@ static int apply_to_pmd_range(struct mm_struct *mm, pud_t *pud,
static int apply_to_pud_range(struct mm_struct *mm, p4d_t *p4d,
unsigned long addr, unsigned long end,
- pte_fn_t fn, void *data, bool create,
+ pte_fn_t fn, void *data, unsigned int flags,
pgtbl_mod_mask *mask)
{
pud_t *pud;
+ bool create = flags & PGRANGE_CREATE;
unsigned long next;
int err = 0;
@@ -3555,10 +3558,11 @@ static int apply_to_pud_range(struct mm_struct *mm, p4d_t *p4d,
static int apply_to_p4d_range(struct mm_struct *mm, pgd_t *pgd,
unsigned long addr, unsigned long end,
- pte_fn_t fn, void *data, bool create,
+ pte_fn_t fn, void *data, unsigned int flags,
pgtbl_mod_mask *mask)
{
p4d_t *p4d;
+ bool create = flags & PGRANGE_CREATE;
unsigned long next;
int err = 0;
@@ -3581,7 +3585,7 @@ static int apply_to_p4d_range(struct mm_struct *mm, pgd_t *pgd,
p4d_clear_bad(p4d);
}
err = apply_to_pud_range(mm, p4d, addr, next,
- fn, data, create, mask);
+ fn, data, flags, mask);
if (err)
break;
} while (p4d++, addr = next, addr != end);
@@ -3589,11 +3593,12 @@ static int apply_to_p4d_range(struct mm_struct *mm, pgd_t *pgd,
return err;
}
-static int __apply_to_page_range(struct mm_struct *mm, unsigned long addr,
- unsigned long size, pte_fn_t fn,
- void *data, bool create)
+int __apply_to_page_range(struct mm_struct *mm, unsigned long addr,
+ unsigned long size, pte_fn_t fn,
+ void *data, unsigned int flags)
{
pgd_t *pgd;
+ bool create = flags & PGRANGE_CREATE;
unsigned long start = addr, next;
unsigned long end = addr + size;
pgtbl_mod_mask mask = 0;
@@ -3617,7 +3622,7 @@ static int __apply_to_page_range(struct mm_struct *mm, unsigned long addr,
pgd_clear_bad(pgd);
}
err = apply_to_p4d_range(mm, pgd, addr, next,
- fn, data, create, &mask);
+ fn, data, flags, &mask);
if (err)
break;
} while (pgd++, addr = next, addr != end);
@@ -3635,7 +3640,7 @@ static int __apply_to_page_range(struct mm_struct *mm, unsigned long addr,
int apply_to_page_range(struct mm_struct *mm, unsigned long addr,
unsigned long size, pte_fn_t fn, void *data)
{
- return __apply_to_page_range(mm, addr, size, fn, data, true);
+ return __apply_to_page_range(mm, addr, size, fn, data, PGRANGE_CREATE);
}
EXPORT_SYMBOL_GPL(apply_to_page_range);
@@ -3649,7 +3654,7 @@ EXPORT_SYMBOL_GPL(apply_to_page_range);
int apply_to_existing_page_range(struct mm_struct *mm, unsigned long addr,
unsigned long size, pte_fn_t fn, void *data)
{
- return __apply_to_page_range(mm, addr, size, fn, data, false);
+ return __apply_to_page_range(mm, addr, size, fn, data, 0);
}
/*
--
2.54.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v3 10/26] mm: Add more flags for __apply_to_page_range()
2026-07-26 22:22 [PATCH v3 00/26] mm: Add ALLOC_UNMAPPED and AS_NO_DIRECT_MAP Brendan Jackman
` (8 preceding siblings ...)
2026-07-26 22:22 ` [PATCH v3 09/26] mm: Create flags arg for __apply_to_page_range() Brendan Jackman
@ 2026-07-26 22:22 ` Brendan Jackman
2026-07-26 22:22 ` [PATCH v3 11/26] x86/mm: introduce the mermap Brendan Jackman
` (15 subsequent siblings)
25 siblings, 0 replies; 29+ messages in thread
From: Brendan Jackman @ 2026-07-26 22:22 UTC (permalink / raw)
To: Borislav Petkov, Dave Hansen, Peter Zijlstra, Andrew Morton,
David Hildenbrand, Vlastimil Babka, Mike Rapoport, Wei Xu,
Johannes Weiner, Zi Yan, Lorenzo Stoakes
Cc: linux-mm, linux-kernel, x86, rppt, Sumit Garg, Will Deacon,
rientjes, Kalyazin, Nikita, patrick.roy, Itazuri, Takahiro,
Andy Lutomirski, David Kaplan, Thomas Gleixner, Yosry Ahmed,
Patrick Bellasi, Reiji Watanabe, Sean Christopherson,
Brendan Jackman
Add two flags to make this API more generic:
1. Separate "create" into two levels - one to allow creating new
mappings without allocating pagetables, and one for the current
behaviour that allows both of these.
2. Create a new flag to report that the caller has taken care of
synchronization and no locks are required.
Both of these will serve to allow calling this API from restricted
contexts where allocation and pagetable locking are not possible.
Signed-off-by: Brendan Jackman <jackmanb@google.com>
---
mm/internal.h | 26 +++++++++++++++++++++++++-
mm/memory.c | 59 ++++++++++++++++++++++++++++++++++-------------------------
2 files changed, 59 insertions(+), 26 deletions(-)
diff --git a/mm/internal.h b/mm/internal.h
index 395331a12d62d..5a237d9c5fa96 100644
--- a/mm/internal.h
+++ b/mm/internal.h
@@ -1662,9 +1662,33 @@ static inline bool can_spin_trylock(void)
/*
* Create a mapping if it doesn't exist. (Otherwise, skip regions with no
- * existing mapping, and return an error for regions with no leaf pagetable).
+ * existing mapping). This doesn't allow allocating, most users will want
+ * PGRANGE_ALLOC.
+ *
+ * Do not test this bit directly as it is implied by PGRANGE_ALLOC, use
+ * pgrange_create() instead.
*/
#define PGRANGE_CREATE (1 << 0)
+/*
+ * Allocate a pagetable if one is missing. (Otherwise, return an error for
+ * regions with no leaf pagetable). Also implies PGRANGE_CREATE.
+ *
+ * Note that __apply_to_page_range() assumes that pagetables for the area are
+ * already initialised down to PMD level, so this only affects PTEs in practice.
+ */
+#define PGRANGE_ALLOC (1 << 1)
+/*
+ * Do not take any locks. This means the caller has taken care of
+ * synchronisation. This is incompatible with PGRANGE_ALLOC and also with
+ * mm=&init_mm.
+ */
+#define PGRANGE_NOLOCK (1 << 2)
+
+
+static inline bool pgrange_create(unsigned int flags)
+{
+ return flags & (PGRANGE_CREATE | PGRANGE_ALLOC);
+}
int __apply_to_page_range(struct mm_struct *mm, unsigned long addr,
unsigned long size, pte_fn_t fn,
diff --git a/mm/memory.c b/mm/memory.c
index c4defefea1574..d00508db1021e 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -3441,30 +3441,35 @@ static int apply_to_pte_range(struct mm_struct *mm, pmd_t *pmd,
pte_fn_t fn, void *data, unsigned int flags,
pgtbl_mod_mask *mask)
{
- bool create = flags & PGRANGE_CREATE;
pte_t *pte, *mapped_pte;
int err = 0;
spinlock_t *ptl;
- if (create) {
+ if (flags & PGRANGE_ALLOC) {
+ VM_WARN_ON(flags & PGRANGE_NOLOCK);
+
mapped_pte = pte = (mm == &init_mm) ?
pte_alloc_kernel_track(pmd, addr, mask) :
pte_alloc_map_lock(mm, pmd, addr, &ptl);
if (!pte)
return -ENOMEM;
} else {
- mapped_pte = pte = (mm == &init_mm) ?
- pte_offset_kernel(pmd, addr) :
- pte_offset_map_lock(mm, pmd, addr, &ptl);
+ if (mm == &init_mm)
+ pte = pte_offset_kernel(pmd, addr);
+ else if (flags & PGRANGE_NOLOCK)
+ pte = pte_offset_map(pmd, addr);
+ else
+ pte = pte_offset_map_lock(mm, pmd, addr, &ptl);
if (!pte)
return -EINVAL;
+ mapped_pte = pte;
}
lazy_mmu_mode_enable();
if (fn) {
do {
- if (create || !pte_none(ptep_get(pte))) {
+ if (pgrange_create(flags) || !pte_none(ptep_get(pte))) {
err = fn(pte, addr, data);
if (err)
break;
@@ -3475,8 +3480,13 @@ static int apply_to_pte_range(struct mm_struct *mm, pmd_t *pmd,
lazy_mmu_mode_disable();
- if (mm != &init_mm)
- pte_unmap_unlock(mapped_pte, ptl);
+ if (mm != &init_mm) {
+ if (flags & PGRANGE_NOLOCK)
+ pte_unmap(mapped_pte);
+ else
+ pte_unmap_unlock(mapped_pte, ptl);
+ }
+
return err;
}
@@ -3486,13 +3496,12 @@ static int apply_to_pmd_range(struct mm_struct *mm, pud_t *pud,
pgtbl_mod_mask *mask)
{
pmd_t *pmd;
- bool create = flags & PGRANGE_CREATE;
unsigned long next;
int err = 0;
BUG_ON(pud_leaf(*pud));
- if (create) {
+ if (flags & PGRANGE_ALLOC) {
pmd = pmd_alloc_track(mm, pud, addr, mask);
if (!pmd)
return -ENOMEM;
@@ -3501,12 +3510,12 @@ static int apply_to_pmd_range(struct mm_struct *mm, pud_t *pud,
}
do {
next = pmd_addr_end(addr, end);
- if (pmd_none(*pmd) && !create)
+ if (pmd_none(*pmd) && !pgrange_create(flags))
continue;
if (WARN_ON_ONCE(pmd_leaf(*pmd)))
return -EINVAL;
if (!pmd_none(*pmd) && WARN_ON_ONCE(pmd_bad(*pmd))) {
- if (!create)
+ if (!pgrange_create(flags))
continue;
pmd_clear_bad(pmd);
}
@@ -3525,11 +3534,10 @@ static int apply_to_pud_range(struct mm_struct *mm, p4d_t *p4d,
pgtbl_mod_mask *mask)
{
pud_t *pud;
- bool create = flags & PGRANGE_CREATE;
unsigned long next;
int err = 0;
- if (create) {
+ if (flags & PGRANGE_ALLOC) {
pud = pud_alloc_track(mm, p4d, addr, mask);
if (!pud)
return -ENOMEM;
@@ -3538,17 +3546,17 @@ static int apply_to_pud_range(struct mm_struct *mm, p4d_t *p4d,
}
do {
next = pud_addr_end(addr, end);
- if (pud_none(*pud) && !create)
+ if (pud_none(*pud) && !pgrange_create(flags))
continue;
if (WARN_ON_ONCE(pud_leaf(*pud)))
return -EINVAL;
if (!pud_none(*pud) && WARN_ON_ONCE(pud_bad(*pud))) {
- if (!create)
+ if (!pgrange_create(flags))
continue;
pud_clear_bad(pud);
}
err = apply_to_pmd_range(mm, pud, addr, next,
- fn, data, create, mask);
+ fn, data, flags, mask);
if (err)
break;
} while (pud++, addr = next, addr != end);
@@ -3562,11 +3570,10 @@ static int apply_to_p4d_range(struct mm_struct *mm, pgd_t *pgd,
pgtbl_mod_mask *mask)
{
p4d_t *p4d;
- bool create = flags & PGRANGE_CREATE;
unsigned long next;
int err = 0;
- if (create) {
+ if (flags & PGRANGE_ALLOC) {
p4d = p4d_alloc_track(mm, pgd, addr, mask);
if (!p4d)
return -ENOMEM;
@@ -3575,12 +3582,12 @@ static int apply_to_p4d_range(struct mm_struct *mm, pgd_t *pgd,
}
do {
next = p4d_addr_end(addr, end);
- if (p4d_none(*p4d) && !create)
+ if (p4d_none(*p4d) && !pgrange_create(flags))
continue;
if (WARN_ON_ONCE(p4d_leaf(*p4d)))
return -EINVAL;
if (!p4d_none(*p4d) && WARN_ON_ONCE(p4d_bad(*p4d))) {
- if (!create)
+ if (!pgrange_create(flags))
continue;
p4d_clear_bad(p4d);
}
@@ -3598,7 +3605,6 @@ int __apply_to_page_range(struct mm_struct *mm, unsigned long addr,
void *data, unsigned int flags)
{
pgd_t *pgd;
- bool create = flags & PGRANGE_CREATE;
unsigned long start = addr, next;
unsigned long end = addr + size;
pgtbl_mod_mask mask = 0;
@@ -3606,18 +3612,21 @@ int __apply_to_page_range(struct mm_struct *mm, unsigned long addr,
if (WARN_ON(addr >= end))
return -EINVAL;
+ if (WARN_ON(flags & PGRANGE_NOLOCK &&
+ (mm == &init_mm || flags & PGRANGE_ALLOC)))
+ return -EINVAL;
pgd = pgd_offset(mm, addr);
do {
next = pgd_addr_end(addr, end);
- if (pgd_none(*pgd) && !create)
+ if (pgd_none(*pgd) && !pgrange_create(flags))
continue;
if (WARN_ON_ONCE(pgd_leaf(*pgd))) {
err = -EINVAL;
break;
}
if (!pgd_none(*pgd) && WARN_ON_ONCE(pgd_bad(*pgd))) {
- if (!create)
+ if (!pgrange_create(flags))
continue;
pgd_clear_bad(pgd);
}
@@ -3640,7 +3649,7 @@ int __apply_to_page_range(struct mm_struct *mm, unsigned long addr,
int apply_to_page_range(struct mm_struct *mm, unsigned long addr,
unsigned long size, pte_fn_t fn, void *data)
{
- return __apply_to_page_range(mm, addr, size, fn, data, PGRANGE_CREATE);
+ return __apply_to_page_range(mm, addr, size, fn, data, PGRANGE_ALLOC);
}
EXPORT_SYMBOL_GPL(apply_to_page_range);
--
2.54.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v3 11/26] x86/mm: introduce the mermap
2026-07-26 22:22 [PATCH v3 00/26] mm: Add ALLOC_UNMAPPED and AS_NO_DIRECT_MAP Brendan Jackman
` (9 preceding siblings ...)
2026-07-26 22:22 ` [PATCH v3 10/26] mm: Add more flags " Brendan Jackman
@ 2026-07-26 22:22 ` Brendan Jackman
2026-07-26 22:22 ` [PATCH v3 12/26] mm: KUnit tests for " Brendan Jackman
` (14 subsequent siblings)
25 siblings, 0 replies; 29+ messages in thread
From: Brendan Jackman @ 2026-07-26 22:22 UTC (permalink / raw)
To: Borislav Petkov, Dave Hansen, Peter Zijlstra, Andrew Morton,
David Hildenbrand, Vlastimil Babka, Mike Rapoport, Wei Xu,
Johannes Weiner, Zi Yan, Lorenzo Stoakes
Cc: linux-mm, linux-kernel, x86, rppt, Sumit Garg, Will Deacon,
rientjes, Kalyazin, Nikita, patrick.roy, Itazuri, Takahiro,
Andy Lutomirski, David Kaplan, Thomas Gleixner, Yosry Ahmed,
Patrick Bellasi, Reiji Watanabe, Sean Christopherson,
Brendan Jackman
The mermap provides a fast way to create ephemeral mm-local mappings of
physical pages. The purpose of this is to access pages that have been
removed from the direct map. Potential use cases are:
1. For zeroing direct-map-nonpresent pages (added in a later patch).
2. For populating guest_memfd pages that are protected by the
GUEST_MEMFD_NO_DIRECT_MAP feature [0].
3. For efficient access of pages protected by Address Space Isolation
[1].
[0] https://lore.kernel.org/all/20250924151101.2225820-1-patrick.roy@campus.lmu.de/
[1] https://linuxasi.dev
The details of this mechanism are described in the API comments. However
the key idea is to use CPU-local virtual regions to avoid a need for
synchronizing. On x86, this can also be used to prevent TLB shootdowns.
Because the virtual region is CPU-local, allocating from the mermap
disables migration. The caller is forbidden to use the returned value
from any other context, and migration is re-enabled when it's freed.
One might notice that mermap_get() bears a strong similarity to
kmap_local_page(). The most important differences between mermap_get()
and kmap_local_page() are:
1. mermap_get() allows mapping variable sizes while kmap_local_page()
specifically maps a single order-0 page.
2. As a consequence of 1 (combined with the need for mermap_get() to be
an extremely simple allocator), mermap_get() should be expected to
fail, while kmap_local_page() is guaranteed to work up to a certain
degree of nesting.
3. While the mappings provided by kmap_local_page() are _logically_
local to the calling context (it's a bug for software to access them
from elsewhere), they are _physically_ installed into the shared
kernel pagetables. This means their locality doesn't provide any
protection from hardware attacks. In contrast, the mermap is
physically local to the creating mm, taking advantage of the new
mm-local kernel address region.
So that the mermap is available even in contexts where failure is not
tolerable there is also a _reserved() variant, which is fixed at
allocating a single base page. This is useful, for example, for zeroing
unmapped pages, where handling failure would be extremely inconvenient.
The _reserved() variant is simply implemented by leaving one base-page
space unavailable for non-_reserved allocations, and requiring an atomic
context.
Note for Sashiko: Yes, the data mapped by the mermap is exposed to
Meltdown-style attacks by the current process. This is completely
intentional. Data is only supposed to be mapped there that the current
process is allowed to read anyway.
Signed-off-by: Brendan Jackman <jackmanb@google.com>
---
arch/x86/Kconfig | 1 +
arch/x86/include/asm/mermap.h | 23 +++
arch/x86/include/asm/pgtable_64_types.h | 8 +-
arch/x86/include/asm/pgtable_types.h | 2 +
include/linux/mermap.h | 63 ++++++
include/linux/mermap_types.h | 41 ++++
include/linux/mm_types.h | 4 +
kernel/fork.c | 5 +
mm/Kconfig | 9 +
mm/Makefile | 1 +
mm/mermap.c | 338 ++++++++++++++++++++++++++++++++
11 files changed, 494 insertions(+), 1 deletion(-)
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 33c1282bfbf93..6b4d81a280d3b 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -37,6 +37,7 @@ config X86_64
select ZONE_DMA32
select EXECMEM if DYNAMIC_FTRACE
select ACPI_MRRM if ACPI
+ select ARCH_SUPPORTS_MERMAP
config FORCE_DYNAMIC_FTRACE
def_bool y
diff --git a/arch/x86/include/asm/mermap.h b/arch/x86/include/asm/mermap.h
new file mode 100644
index 0000000000000..9d7614716b718
--- /dev/null
+++ b/arch/x86/include/asm/mermap.h
@@ -0,0 +1,23 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _ASM_X86_MERMAP_H
+#define _ASM_X86_MERMAP_H
+
+#include <asm/tlbflush.h>
+
+static inline void arch_mermap_flush_tlb(void)
+{
+ /*
+ * No shootdown allowed, IRQs may be off. Luckily other CPUs are not
+ * allowed to access our region so the stale mappings are harmless, as
+ * long as they still point to data belonging to this process.
+ */
+ __flush_tlb_all();
+}
+
+static inline bool arch_mermap_pgprot_allowed(pgprot_t prot)
+{
+ /* Mermap is mm-local so global mappings would be a bug. */
+ return !(pgprot_val(prot) & _PAGE_GLOBAL);
+}
+
+#endif /* _ASM_X86_MERMAP_H */
diff --git a/arch/x86/include/asm/pgtable_64_types.h b/arch/x86/include/asm/pgtable_64_types.h
index 1181565966405..fb6c3daacfeb8 100644
--- a/arch/x86/include/asm/pgtable_64_types.h
+++ b/arch/x86/include/asm/pgtable_64_types.h
@@ -105,11 +105,17 @@ extern unsigned int ptrs_per_p4d;
#define MM_LOCAL_PGD_ENTRY -240UL
#define MM_LOCAL_BASE_ADDR (MM_LOCAL_PGD_ENTRY << PGDIR_SHIFT)
-#define MM_LOCAL_END_ADDR ((MM_LOCAL_PGD_ENTRY + 1) << PGDIR_SHIFT)
+#define MM_LOCAL_START_ADDR ((MM_LOCAL_PGD_ENTRY) << PGDIR_SHIFT)
+#define MM_LOCAL_END_ADDR (MM_LOCAL_START_ADDR + (1UL << PGDIR_SHIFT))
#define LDT_BASE_ADDR MM_LOCAL_BASE_ADDR
#define LDT_END_ADDR (LDT_BASE_ADDR + PMD_SIZE)
+#define MERMAP_BASE_ADDR LDT_END_ADDR
+#define MERMAP_CPU_REGION_SIZE PMD_SIZE
+#define MERMAP_SIZE (MERMAP_CPU_REGION_SIZE * NR_CPUS)
+#define MERMAP_END_ADDR (MERMAP_BASE_ADDR + (NR_CPUS * MERMAP_CPU_REGION_SIZE))
+
#define __VMALLOC_BASE_L4 0xffffc90000000000UL
#define __VMALLOC_BASE_L5 0xffa0000000000000UL
diff --git a/arch/x86/include/asm/pgtable_types.h b/arch/x86/include/asm/pgtable_types.h
index af08d98be9309..f397e4311cf66 100644
--- a/arch/x86/include/asm/pgtable_types.h
+++ b/arch/x86/include/asm/pgtable_types.h
@@ -223,6 +223,7 @@ enum page_cache_mode {
#define __PAGE_KERNEL_RO (__PP| 0| 0|___A|__NX| 0| 0|___G)
#define __PAGE_KERNEL_ROX (__PP| 0| 0|___A| 0| 0| 0|___G)
#define __PAGE_KERNEL (__PP|__RW| 0|___A|__NX|___D| 0|___G)
+#define __PAGE_KERNEL_NOGLOBAL (__PP|__RW| 0|___A|__NX|___D| 0| 0)
#define __PAGE_KERNEL_EXEC (__PP|__RW| 0|___A| 0|___D| 0|___G)
#define __PAGE_KERNEL_NOCACHE (__PP|__RW| 0|___A|__NX|___D| 0|___G| __NC)
#define __PAGE_KERNEL_VVAR (__PP| 0|_USR|___A|__NX| 0| 0|___G)
@@ -245,6 +246,7 @@ enum page_cache_mode {
#define __pgprot_mask(x) __pgprot((x) & __default_kernel_pte_mask)
#define PAGE_KERNEL __pgprot_mask(__PAGE_KERNEL | _ENC)
+#define PAGE_KERNEL_NOGLOBAL __pgprot_mask(__PAGE_KERNEL_NOGLOBAL | _ENC)
#define PAGE_KERNEL_NOENC __pgprot_mask(__PAGE_KERNEL | 0)
#define PAGE_KERNEL_RO __pgprot_mask(__PAGE_KERNEL_RO | _ENC)
#define PAGE_KERNEL_EXEC __pgprot_mask(__PAGE_KERNEL_EXEC | _ENC)
diff --git a/include/linux/mermap.h b/include/linux/mermap.h
new file mode 100644
index 0000000000000..5457dcb8c9789
--- /dev/null
+++ b/include/linux/mermap.h
@@ -0,0 +1,63 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _LINUX_MERMAP_H
+#define _LINUX_MERMAP_H
+
+#include <linux/mermap_types.h>
+#include <linux/mm.h>
+
+#ifdef CONFIG_MERMAP
+
+#include <asm/mermap.h>
+
+int mermap_mm_prepare(struct mm_struct *mm);
+void mermap_mm_init(struct mm_struct *mm);
+void mermap_mm_teardown(struct mm_struct *mm);
+
+/* Can the mermap be called from this context? */
+static inline bool mermap_ready(void)
+{
+ return in_task() && current->mm && current->mm->mermap.cpu;
+}
+
+struct mermap_alloc *mermap_get(struct page *page, unsigned long size, pgprot_t prot);
+void *mermap_get_reserved(struct page *page, pgprot_t prot);
+void mermap_put(struct mermap_alloc *alloc);
+
+static inline void *mermap_addr(struct mermap_alloc *alloc)
+{
+ return (void *)alloc->base;
+}
+
+/*
+ * arch_mermap_flush_tlb() is called before a part of the local CPU's mermap
+ * region is remapped to a new address. No other CPU is allowed to _access_ that
+ * region, but the region was mapped there.
+ *
+ * This may be called with IRQs off.
+ *
+ * On arm64, this will need to be a broadcast TLB flush. Although the other CPUs
+ * are forbidden to access the region, they can leak the data that was mapped
+ * there via CPU exploits. Violating break-before-make would mean the data
+ * available to these CPU exploits is unpredictable.
+ */
+extern void arch_mermap_flush_tlb(void);
+extern bool arch_mermap_pgprot_allowed(pgprot_t prot);
+
+#if IS_ENABLED(CONFIG_KUNIT)
+struct mermap_alloc *__mermap_get(struct mm_struct *mm, struct page *page,
+ unsigned long size, pgprot_t prot, bool use_reserve);
+void __mermap_put(struct mm_struct *mm, struct mermap_alloc *alloc);
+unsigned long mermap_cpu_base(int cpu);
+unsigned long mermap_cpu_end(int cpu);
+#endif
+
+#else /* CONFIG_MERMAP */
+
+static inline int mermap_mm_prepare(struct mm_struct *mm) { return 0; }
+static inline void mermap_mm_init(struct mm_struct *mm) { }
+static inline void mermap_mm_teardown(struct mm_struct *mm) { }
+static inline bool mermap_ready(void) { return false; }
+
+#endif /* CONFIG_MERMAP */
+
+#endif /* _LINUX_MERMAP_H */
diff --git a/include/linux/mermap_types.h b/include/linux/mermap_types.h
new file mode 100644
index 0000000000000..c1c83b223c28d
--- /dev/null
+++ b/include/linux/mermap_types.h
@@ -0,0 +1,41 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _LINUX_MERMAP_TYPES_H
+#define _LINUX_MERMAP_TYPES_H
+
+#include <linux/mutex.h>
+#include <linux/percpu.h>
+#include <linux/types.h>
+
+#ifdef CONFIG_MERMAP
+
+/* Tracks an individual allocation in the mermap. */
+struct mermap_alloc {
+ /* Currently allocated. */
+ bool in_use;
+ /* Requires flush before reallocating. */
+ bool need_flush;
+ unsigned long base;
+ /* Non-inclusive. */
+ unsigned long end;
+};
+
+struct mermap_cpu {
+ /* Next address immediately available for alloc (no TLB flush needed). */
+ unsigned long next_addr;
+ struct mermap_alloc normal_allocs[3];
+ struct mermap_alloc reserve_alloc;
+};
+
+struct mermap {
+ struct mutex init_lock;
+ struct mermap_cpu __percpu *cpu;
+};
+
+#else /* CONFIG_MERMAP */
+
+struct mermap {};
+
+#endif /* CONFIG_MERMAP */
+
+#endif /* _LINUX_MERMAP_TYPES_H */
+
diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
index d39fddf57edc8..bb80d60cf3498 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -7,6 +7,7 @@
#include <linux/auxvec.h>
#include <linux/kref.h>
#include <linux/list.h>
+#include <linux/mermap_types.h>
#include <linux/spinlock.h>
#include <linux/rbtree.h>
#include <linux/maple_tree.h>
@@ -35,6 +36,7 @@
struct address_space;
struct futex_private_hash;
struct mem_cgroup;
+struct mermap;
typedef struct {
unsigned long f;
@@ -1211,6 +1213,8 @@ struct mm_struct {
atomic_t membarrier_state;
#endif
+ struct mermap mermap;
+
/**
* @mm_users: The number of users including userspace.
*
diff --git a/kernel/fork.c b/kernel/fork.c
index 4fb23ea33b7da..7c2050e76d1bc 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -13,6 +13,7 @@
*/
#include <linux/anon_inodes.h>
+#include <linux/mermap.h>
#include <linux/slab.h>
#include <linux/sched/autogroup.h>
#include <linux/sched/mm.h>
@@ -1143,6 +1144,9 @@ static struct mm_struct *mm_init(struct mm_struct *mm, struct task_struct *p)
goto fail_pcpu;
lru_gen_init_mm(mm);
+
+ mermap_mm_init(mm);
+
return mm;
fail_pcpu:
@@ -1186,6 +1190,7 @@ static inline void __mmput(struct mm_struct *mm)
ksm_exit(mm);
khugepaged_exit(mm); /* must run before exit_mmap */
exit_mmap(mm);
+ mermap_mm_teardown(mm);
mm_put_huge_zero_folio(mm);
set_mm_exe_file(mm, NULL);
if (!list_empty(&mm->mmlist)) {
diff --git a/mm/Kconfig b/mm/Kconfig
index cb531c1436f77..bf8c4c6264c73 100644
--- a/mm/Kconfig
+++ b/mm/Kconfig
@@ -1507,6 +1507,15 @@ config MM_LOCAL_REGION
bool
depends on ARCH_SUPPORTS_MM_LOCAL_REGION
+config ARCH_SUPPORTS_MERMAP
+ bool
+ select ARCH_SUPPORTS_MM_LOCAL_REGION
+
+config MERMAP
+ bool
+ depends on ARCH_SUPPORTS_MERMAP
+ select MM_LOCAL_REGION
+
source "mm/damon/Kconfig"
endmenu
diff --git a/mm/Makefile b/mm/Makefile
index ab37ef428d98d..9cf282c154104 100644
--- a/mm/Makefile
+++ b/mm/Makefile
@@ -147,3 +147,4 @@ obj-$(CONFIG_EXECMEM) += execmem.o
obj-$(CONFIG_TMPFS_QUOTA) += shmem_quota.o
obj-$(CONFIG_LAZY_MMU_MODE_KUNIT_TEST) += tests/lazy_mmu_mode_kunit.o
obj-$(CONFIG_MEM_ALLOC_PROFILING) += alloc_tag.o
+obj-$(CONFIG_MERMAP) += mermap.o
diff --git a/mm/mermap.c b/mm/mermap.c
new file mode 100644
index 0000000000000..2bead38eadfe8
--- /dev/null
+++ b/mm/mermap.c
@@ -0,0 +1,338 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/io.h>
+#include <linux/error-injection.h>
+#include <linux/mermap.h>
+#include <linux/mm.h>
+#include <linux/mmu_context.h>
+#include <linux/mutex.h>
+#include <linux/pagemap.h>
+#include <linux/pgtable.h>
+#include <linux/sched.h>
+
+#include <kunit/visibility.h>
+
+#include "internal.h"
+
+static inline int set_unmapped_pte(pte_t *ptep, unsigned long addr, void *data)
+{
+ set_pte(ptep, __pte(0));
+ return 0;
+}
+
+VISIBLE_IF_KUNIT void __mermap_put(struct mm_struct *mm, struct mermap_alloc *alloc)
+{
+ unsigned long size = PAGE_ALIGN(alloc->end - alloc->base);
+
+ __apply_to_page_range(mm, alloc->base, size, set_unmapped_pte,
+ NULL, PGRANGE_CREATE | PGRANGE_NOLOCK);
+
+ WRITE_ONCE(alloc->in_use, false);
+}
+EXPORT_SYMBOL_IF_KUNIT(__mermap_put);
+
+/* Return a region allocated by mermap_get(). */
+void mermap_put(struct mermap_alloc *alloc)
+{
+ cant_migrate();
+ if (WARN_ON_ONCE(!alloc->in_use))
+ return;
+ __mermap_put(current->mm, alloc);
+}
+EXPORT_SYMBOL(mermap_put);
+
+VISIBLE_IF_KUNIT inline unsigned long mermap_cpu_base(int cpu)
+{
+ return MERMAP_BASE_ADDR + (cpu * MERMAP_CPU_REGION_SIZE);
+
+}
+EXPORT_SYMBOL_IF_KUNIT(mermap_cpu_base);
+
+/* Non-inclusive */
+VISIBLE_IF_KUNIT inline unsigned long mermap_cpu_end(int cpu)
+{
+ return MERMAP_BASE_ADDR + ((cpu + 1) * MERMAP_CPU_REGION_SIZE);
+
+}
+EXPORT_SYMBOL_IF_KUNIT(mermap_cpu_end);
+
+static inline void mermap_flush_tlb(int cpu, struct mermap_cpu *mc)
+{
+#if IS_ENABLED(CONFIG_MERMAP_KUNIT_TEST)
+ mc->tlb_flushes++;
+#endif
+ arch_mermap_flush_tlb();
+}
+
+/* Call with preemption disabled if use_reserve, else with migration disabled. */
+static inline struct mermap_alloc *mermap_alloc(struct mm_struct *mm,
+ unsigned long size, bool use_reserve)
+{
+ int cpu = raw_smp_processor_id();
+ struct mermap_cpu *mc = this_cpu_ptr(mm->mermap.cpu);
+ unsigned long cpu_end = mermap_cpu_end(cpu);
+ struct mermap_alloc *alloc = NULL;
+
+ /*
+ * This is an extremely stupid allocator, there can only ever be a small
+ * number of allocations so everything just works on linear search.
+ *
+ * Allocations are "in order", i.e. if the whole region is free it
+ * allocates from the beginning. If there are any existing allocations
+ * it allocates from right after the last (highest address) one. Any
+ * free space before that goes unused.
+ *
+ * Once an allocation has been freed, the space it occupied must be flushed
+ * from the TLB before it can be reused.
+ *
+ * Visual example of how this is supposed to behave (A for allocated, T for
+ * TLB-flush-pending):
+ *
+ * _______________ Start with everything free.
+ * AaaA___________ Allocate something.
+ * TttT___________ Free it. (Region needs a TLB flush now).
+ * TttTAaaaaaaaA__ Allocate something else.
+ * TttTAaaaaaaaAAA Allocate the remaining space.
+ * TttTTtttttttTAA Free the allocation before last.
+ * ^^^^^^^^^^^^^ This could all be reused now but for simplicity it
+ * isn't. Another allocation at this point will fail.
+ * TttTTtttttttTTT Free the last allocation.
+ * _______________ Next time we allocate, first flush the TLB.
+ * AA_____________ Now we're back at the beginning.
+ */
+
+ /* Keep one page for mermap_get_reserved(). */
+ if (use_reserve) {
+ if (WARN_ON_ONCE(size != PAGE_SIZE))
+ return NULL;
+ lockdep_assert_preemption_disabled();
+ } else {
+ cpu_end -= PAGE_SIZE;
+ }
+
+ if (WARN_ON_ONCE(!in_task()))
+ return NULL;
+ guard(preempt)();
+
+ /* Out of already-available space? */
+ if (mc->next_addr + size > cpu_end) {
+ unsigned long new_next = mermap_cpu_base(cpu);
+
+ /* Would we have space after a TLB flush? */
+ for (int i = 0; i < ARRAY_SIZE(mc->normal_allocs); i++) {
+ struct mermap_alloc *alloc = &mc->normal_allocs[i];
+
+ /*
+ * The space between the uppermost allocated alloc->end
+ * (or the base of the CPU's region if there are no
+ * current allocations) and mc->next_addr has been
+ * unmapped in the pagetables, but not flushed from the
+ * TLB. Set new_next to point to the beginning of that
+ * space.
+ */
+ if (READ_ONCE(alloc->in_use))
+ new_next = max(new_next, alloc->end);
+ }
+ if (size > cpu_end - new_next)
+ return NULL;
+
+ mermap_flush_tlb(cpu, mc);
+ mc->next_addr = new_next;
+ }
+
+ /*
+ * Find an alloc-tracking structure to use. Keep one for
+ * mermap_get_reserved() - that should never be contended since it can
+ * only be allocated with preemption off.
+ */
+ if (WARN_ON_ONCE(mc->reserve_alloc.in_use))
+ return NULL;
+ if (use_reserve) {
+ alloc = &mc->reserve_alloc;
+ } else {
+ for (int i = 0; i < ARRAY_SIZE(mc->normal_allocs); i++) {
+ if (!READ_ONCE(mc->normal_allocs[i].in_use)) {
+ alloc = &mc->normal_allocs[i];
+ break;
+ }
+ }
+ if (!alloc)
+ return NULL;
+ }
+ alloc->in_use = true;
+ alloc->base = mc->next_addr;
+ alloc->end = alloc->base + size;
+ mc->next_addr += size;
+
+ return alloc;
+}
+
+struct set_pte_ctx {
+ pgprot_t prot;
+ unsigned long next_pfn;
+};
+
+static inline int do_set_pte(pte_t *pte, unsigned long addr, void *data)
+{
+ struct set_pte_ctx *ctx = data;
+
+ set_pte(pte, pfn_pte(ctx->next_pfn, ctx->prot));
+ ctx->next_pfn++;
+
+ return 0;
+}
+
+VISIBLE_IF_KUNIT struct mermap_alloc *
+__mermap_get(struct mm_struct *mm, struct page *page,
+ unsigned long size, pgprot_t prot, bool use_reserve)
+{
+ struct mermap_alloc *alloc = NULL;
+ struct set_pte_ctx ctx;
+ int err;
+
+ if (size > MERMAP_CPU_REGION_SIZE || WARN_ON_ONCE(!mm || !mm->mermap.cpu))
+ return NULL;
+ if (WARN_ON_ONCE(!arch_mermap_pgprot_allowed(prot)))
+ return NULL;
+
+ size = PAGE_ALIGN(size);
+
+ alloc = mermap_alloc(mm, size, use_reserve);
+ if (!alloc)
+ return NULL;
+
+ ctx.prot = prot;
+ ctx.next_pfn = page_to_pfn(page);
+ err = __apply_to_page_range(mm, alloc->base, size, do_set_pte, &ctx,
+ PGRANGE_CREATE | PGRANGE_NOLOCK);
+ if (err) {
+ WRITE_ONCE(alloc->in_use, false);
+ return NULL;
+ }
+
+ return alloc;
+}
+EXPORT_SYMBOL_IF_KUNIT(__mermap_get);
+
+/*
+ * Allocate a region of virtual memory, and map the page into it. This tries
+ * pretty hard to be fast but doesn't try very hard at all to actually succeed.
+ *
+ * Must be called with migration disabled, and it must stay disabled until you
+ * call mermap_put().
+ *
+ * The returned region is physically local to the current mm. It is _logically_
+ * local to the current CPU but this is not enforced by hardware so it can be
+ * exploited to mitigate CPU vulns. This means the caller must not map memory
+ * here that doesn't belong to the current process. The caller must also perform
+ * a full TLB flush of the region before freeing the pages that have been mapped
+ * here.
+ *
+ * This may only be called from process context, and the caller must arrange to
+ * first call mermap_mm_prepare(). (It would be possible to support this in IRQ,
+ * but it seems unlikely there's a valid usecase given the TLB flushing
+ * requirements).
+ *
+ * This is guaranteed not to allocate.
+ *
+ * Use mermap_addr() to get the actual address of the mapped region.
+ */
+struct mermap_alloc *mermap_get(struct page *page, unsigned long size, pgprot_t prot)
+{
+ struct mermap_alloc *alloc;
+
+ cant_migrate();
+ alloc = __mermap_get(current->mm, page, size, prot, false);
+ if (alloc) {
+ /* Leaving migration disabled is intentional. */
+ return alloc;
+ }
+ return NULL;
+}
+EXPORT_SYMBOL(mermap_get);
+ALLOW_ERROR_INJECTION(mermap_get, NULL);
+
+/*
+ * Allocate a single PAGE_SIZE page via mermap_get(), requiring preemption to be
+ * off until it is freed. This always succeeds.
+ */
+void *mermap_get_reserved(struct page *page, pgprot_t prot)
+{
+ lockdep_assert_preemption_disabled();
+ return __mermap_get(current->mm, page, PAGE_SIZE, prot, true);
+}
+EXPORT_SYMBOL(mermap_get_reserved);
+
+/*
+ * Internal - do unconditional (cheap) setup that's done for every mm. This
+ * doesn't actually prepare the mermap for use until someone calls
+ * mermap_mm_prepare().
+ */
+void mermap_mm_init(struct mm_struct *mm)
+{
+ mutex_init(&mm->mermap.init_lock);
+ mm->mermap.cpu = NULL;
+}
+
+/*
+ * Set up the mermap for this mm. The caller doesn't need to call
+ * mermap_mm_teardown(), that's take care of by the normal mm teardown
+ * mechanism. This is idempotent and thread-safe.
+ */
+int mermap_mm_prepare(struct mm_struct *mm)
+{
+ int err = 0;
+ int cpu;
+
+ guard(mutex)(&mm->mermap.init_lock);
+
+ /* Already done? */
+ if (likely(mm->mermap.cpu))
+ return 0;
+
+ mm->mermap.cpu = alloc_percpu_gfp(struct mermap_cpu,
+ GFP_KERNEL_ACCOUNT | __GFP_ZERO);
+ if (!mm->mermap.cpu)
+ return -ENOMEM;
+
+ /* So we can use this from the page allocator, preallocate pagetables. */
+ mm_flags_set(MMF_LOCAL_REGION_USED, mm);
+ for_each_possible_cpu(cpu) {
+ unsigned long base = mermap_cpu_base(cpu);
+
+ /* Note this pointlessly iterates over PTEs to initialise. */
+ err = apply_to_page_range(mm, base, MERMAP_CPU_REGION_SIZE,
+ set_unmapped_pte, NULL);
+ if (err) {
+ /*
+ * Clear .cpu now to inform mermap_ready(). Any partial
+ * page tables get cleared up by mm teardown.
+ */
+ free_percpu(mm->mermap.cpu);
+ mm->mermap.cpu = NULL;
+ break;
+ }
+ per_cpu_ptr(mm->mermap.cpu, cpu)->next_addr = base;
+ }
+
+ return err;
+}
+EXPORT_SYMBOL_GPL(mermap_mm_prepare);
+
+/* Clean up mermap stuff on mm teardown. */
+void mermap_mm_teardown(struct mm_struct *mm)
+{
+ int cpu;
+
+ if (!mm->mermap.cpu)
+ return;
+
+ for_each_possible_cpu(cpu) {
+ struct mermap_cpu *mc = per_cpu_ptr(mm->mermap.cpu, cpu);
+
+ for (int i = 0; i < ARRAY_SIZE(mc->normal_allocs); i++)
+ WARN_ON_ONCE(mc->normal_allocs[i].in_use);
+ WARN_ON_ONCE(mc->reserve_alloc.in_use);
+ }
+
+ free_percpu(mm->mermap.cpu);
+}
--
2.54.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v3 12/26] mm: KUnit tests for the mermap
2026-07-26 22:22 [PATCH v3 00/26] mm: Add ALLOC_UNMAPPED and AS_NO_DIRECT_MAP Brendan Jackman
` (10 preceding siblings ...)
2026-07-26 22:22 ` [PATCH v3 11/26] x86/mm: introduce the mermap Brendan Jackman
@ 2026-07-26 22:22 ` Brendan Jackman
2026-07-26 22:22 ` [PATCH v3 13/26] mm: introduce freetype_t Brendan Jackman
` (13 subsequent siblings)
25 siblings, 0 replies; 29+ messages in thread
From: Brendan Jackman @ 2026-07-26 22:22 UTC (permalink / raw)
To: Borislav Petkov, Dave Hansen, Peter Zijlstra, Andrew Morton,
David Hildenbrand, Vlastimil Babka, Mike Rapoport, Wei Xu,
Johannes Weiner, Zi Yan, Lorenzo Stoakes
Cc: linux-mm, linux-kernel, x86, rppt, Sumit Garg, Will Deacon,
rientjes, Kalyazin, Nikita, patrick.roy, Itazuri, Takahiro,
Andy Lutomirski, David Kaplan, Thomas Gleixner, Yosry Ahmed,
Patrick Bellasi, Reiji Watanabe, Sean Christopherson,
Brendan Jackman
Some simple smoke-tests for the mermap. Mainly aiming to test:
1. That there aren't any silly off-by-ones.
2. That the pagetables are not completely broken.
3. That the TLB appears to get flushed basically when expected.
This last point requires a bit of ifdeffery to detect when the flushing
has been performed.
Signed-off-by: Brendan Jackman <jackmanb@google.com>
---
include/linux/mermap_types.h | 3 +
mm/Kconfig | 11 ++
mm/Makefile | 1 +
mm/tests/mermap_kunit.c | 258 +++++++++++++++++++++++++++++++++++++++++++
4 files changed, 273 insertions(+)
diff --git a/include/linux/mermap_types.h b/include/linux/mermap_types.h
index c1c83b223c28d..13110fcb4c387 100644
--- a/include/linux/mermap_types.h
+++ b/include/linux/mermap_types.h
@@ -24,6 +24,9 @@ struct mermap_cpu {
unsigned long next_addr;
struct mermap_alloc normal_allocs[3];
struct mermap_alloc reserve_alloc;
+#if IS_ENABLED(CONFIG_MERMAP_KUNIT_TEST)
+ u64 tlb_flushes;
+#endif
};
struct mermap {
diff --git a/mm/Kconfig b/mm/Kconfig
index bf8c4c6264c73..adf71788971a2 100644
--- a/mm/Kconfig
+++ b/mm/Kconfig
@@ -1516,6 +1516,17 @@ config MERMAP
depends on ARCH_SUPPORTS_MERMAP
select MM_LOCAL_REGION
+config MERMAP_KUNIT_TEST
+ tristate "KUnit tests for the mermap" if !KUNIT_ALL_TESTS
+ depends on ARCH_SUPPORTS_MERMAP
+ depends on KUNIT
+ depends on MERMAP
+ default KUNIT_ALL_TESTS
+ help
+ KUnit test for the mermap.
+
+ If unsure, say N.
+
source "mm/damon/Kconfig"
endmenu
diff --git a/mm/Makefile b/mm/Makefile
index 9cf282c154104..3507cb7ed6a9a 100644
--- a/mm/Makefile
+++ b/mm/Makefile
@@ -148,3 +148,4 @@ obj-$(CONFIG_TMPFS_QUOTA) += shmem_quota.o
obj-$(CONFIG_LAZY_MMU_MODE_KUNIT_TEST) += tests/lazy_mmu_mode_kunit.o
obj-$(CONFIG_MEM_ALLOC_PROFILING) += alloc_tag.o
obj-$(CONFIG_MERMAP) += mermap.o
+obj-$(CONFIG_MERMAP_KUNIT_TEST) += tests/mermap_kunit.o
diff --git a/mm/tests/mermap_kunit.c b/mm/tests/mermap_kunit.c
new file mode 100644
index 0000000000000..737df7d84a340
--- /dev/null
+++ b/mm/tests/mermap_kunit.c
@@ -0,0 +1,258 @@
+// SPDX-License-Identifier: GPL-2.0-only
+#include <linux/cacheflush.h>
+#include <linux/kthread.h>
+#include <linux/mermap.h>
+#include <linux/pgtable.h>
+
+#include <kunit/test.h>
+
+#define NR_NORMAL_ALLOCS ARRAY_SIZE(((struct mm_struct *)NULL)->mermap.cpu->normal_allocs)
+
+KUNIT_DEFINE_ACTION_WRAPPER(__free_page_wrapper, __free_page, struct page *);
+
+static inline struct page *alloc_page_wrapper(struct kunit *test, gfp_t gfp)
+{
+ struct page *page = alloc_page(gfp);
+
+ KUNIT_ASSERT_NOT_NULL(test, page);
+ KUNIT_ASSERT_EQ(test, kunit_add_action_or_reset(test, __free_page_wrapper, page), 0);
+ return page;
+}
+
+KUNIT_DEFINE_ACTION_WRAPPER(mmput_wrapper, mmput, struct mm_struct *);
+
+static inline struct mm_struct *mm_alloc_wrapper(struct kunit *test)
+{
+ struct mm_struct *mm = mm_alloc();
+
+ KUNIT_ASSERT_NOT_NULL(test, mm);
+ KUNIT_ASSERT_EQ(test, kunit_add_action_or_reset(test, mmput_wrapper, mm), 0);
+ return mm;
+}
+
+static inline struct mm_struct *get_mm(struct kunit *test)
+{
+ struct mm_struct *mm = mm_alloc_wrapper(test);
+
+ KUNIT_ASSERT_EQ(test, mermap_mm_prepare(mm), 0);
+ return mm;
+}
+
+struct __mermap_put_args {
+ struct mm_struct *mm;
+ struct mermap_alloc *alloc;
+ unsigned long size;
+};
+
+static inline void __mermap_put_wrapper(void *ctx)
+{
+ struct __mermap_put_args *args = (struct __mermap_put_args *)ctx;
+
+ __mermap_put(args->mm, args->alloc);
+ mmdrop(args->mm);
+}
+
+/* Call __mermap_get() with use_reserve=false, deal with cleanup. */
+static inline struct __mermap_put_args *
+__mermap_get_wrapper(struct kunit *test, struct mm_struct *mm,
+ struct page *page, unsigned long size, pgprot_t prot)
+{
+ struct __mermap_put_args *args =
+ kunit_kmalloc(test, sizeof(struct __mermap_put_args), GFP_KERNEL);
+
+ KUNIT_ASSERT_NOT_NULL(test, args);
+ /*
+ * Grab reference to mm_struct/PGD to allow mermap_put() from the
+ * cleanup kthread.
+ */
+ mmgrab(mm);
+ args->mm = mm;
+ args->alloc = __mermap_get(mm, page, size, prot, false);
+ args->size = size;
+
+ if (args->alloc) {
+ int err = kunit_add_action_or_reset(test, __mermap_put_wrapper, args);
+
+ KUNIT_ASSERT_EQ(test, err, 0);
+ }
+
+ return args;
+}
+
+/* Do the cleanup from __mermap_get_wrapper, now. */
+static inline void __mermap_put_early(struct kunit *test, struct __mermap_put_args *args)
+{
+ kunit_release_action(test, __mermap_put_wrapper, args);
+}
+
+static void test_basic_alloc(struct kunit *test)
+{
+ struct page *page = alloc_page_wrapper(test, GFP_KERNEL);
+ struct mm_struct *mm = get_mm(test);
+ struct __mermap_put_args *args;
+
+ guard(migrate)();
+
+ args = __mermap_get_wrapper(test, mm, page, PAGE_SIZE, PAGE_KERNEL);
+ KUNIT_ASSERT_NOT_NULL(test, args->alloc);
+}
+
+/* Dumb check for off-by-ones. */
+static void test_size(struct kunit *test)
+{
+ struct page *page = alloc_page_wrapper(test, GFP_KERNEL);
+ struct __mermap_put_args *full, *large, *small, *fail;
+ struct mm_struct *mm = get_mm(test);
+ unsigned long region_size, large_size;
+ struct mermap_alloc *alloc;
+ int cpu;
+
+ guard(migrate)();
+ cpu = raw_smp_processor_id();
+ region_size = mermap_cpu_end(cpu) - mermap_cpu_base(cpu) - PAGE_SIZE;
+ large_size = region_size - PAGE_SIZE;
+
+ /* Allocate whole region at once. */
+ full = __mermap_get_wrapper(test, mm, page, region_size, PAGE_KERNEL);
+ KUNIT_ASSERT_NOT_NULL(test, full->alloc);
+ __mermap_put_early(test, full);
+
+ /* Allocate larger than region size. */
+ fail = __mermap_get_wrapper(test, mm, page, region_size + PAGE_SIZE, PAGE_KERNEL);
+ KUNIT_ASSERT_NULL(test, fail->alloc);
+
+ /* Tiptoe up to the edge then past it. */
+ large = __mermap_get_wrapper(test, mm, page, large_size, PAGE_KERNEL);
+ KUNIT_ASSERT_NOT_NULL(test, large->alloc);
+ small = __mermap_get_wrapper(test, mm, page, PAGE_SIZE, PAGE_KERNEL);
+ KUNIT_ASSERT_NOT_NULL(test, small->alloc);
+ fail = __mermap_get_wrapper(test, mm, page, PAGE_SIZE, PAGE_KERNEL);
+ KUNIT_ASSERT_NULL(test, fail->alloc);
+
+ /* Can still allocate the reserved page. */
+ local_irq_disable();
+ alloc = __mermap_get(mm, page, PAGE_SIZE, PAGE_KERNEL, true);
+ local_irq_enable();
+ KUNIT_ASSERT_NOT_NULL(test, alloc);
+ __mermap_put(mm, alloc);
+}
+
+static void test_multiple_allocs(struct kunit *test)
+{
+ struct __mermap_put_args *argss[NR_NORMAL_ALLOCS] = { };
+ struct page *pages[NR_NORMAL_ALLOCS + 1];
+ struct mermap_alloc *reserved_alloc;
+ struct mm_struct *mm = get_mm(test);
+ int magic = 0xE4A4;
+
+ for (int i = 0; i < ARRAY_SIZE(pages); i++) {
+ pages[i] = alloc_page_wrapper(test, GFP_KERNEL);
+ WRITE_ONCE(*(int *)page_to_virt(pages[i]), magic + i);
+ }
+
+ guard(migrate)();
+
+ for (int i = 0; i < ARRAY_SIZE(argss); i++) {
+ unsigned long base = mermap_cpu_base(raw_smp_processor_id());
+ unsigned long end = mermap_cpu_end(raw_smp_processor_id());
+ unsigned long addr;
+
+ argss[i] = __mermap_get_wrapper(test, mm, pages[i], PAGE_SIZE, PAGE_KERNEL);
+ KUNIT_ASSERT_NOT_NULL_MSG(test, argss[i]->alloc, "alloc %d failed", i);
+
+ addr = (unsigned long) mermap_addr(argss[i]->alloc);
+ KUNIT_EXPECT_GE_MSG(test, addr, base, "alloc %d out of range", i);
+ KUNIT_EXPECT_LT_MSG(test, addr, end, "alloc %d out of range", i);
+ }
+
+ /*
+ * Read through the mappings to try and detect if they point to the
+ * pages we wrote earlier.
+ */
+ kthread_use_mm(mm);
+ for (int i = 0; i < ARRAY_SIZE(pages) - 1; i++) {
+ int *ptr = (int *)mermap_addr(argss[i]->alloc);
+
+ KUNIT_EXPECT_EQ(test, *ptr, magic + i);
+ }
+
+ /* Run out of alloc structures, only reserved allocs should succeed now. */
+ KUNIT_EXPECT_NULL(test, __mermap_get(mm, pages[NR_NORMAL_ALLOCS],
+ PAGE_SIZE, PAGE_KERNEL, false));
+ preempt_disable();
+ reserved_alloc = __mermap_get(mm, pages[NR_NORMAL_ALLOCS],
+ PAGE_SIZE, PAGE_KERNEL, true);
+ KUNIT_EXPECT_NOT_NULL(test, reserved_alloc);
+ /* Also check if this mapping seems correct*/
+ if (reserved_alloc) {
+ int *ptr = (int *)mermap_addr(reserved_alloc);
+
+ KUNIT_EXPECT_EQ(test, *ptr, magic + NR_NORMAL_ALLOCS);
+
+ mermap_put(reserved_alloc);
+ }
+ preempt_enable();
+
+ kthread_unuse_mm(mm);
+}
+
+static void test_tlb_flushed(struct kunit *test)
+{
+ struct page *page = alloc_page_wrapper(test, GFP_KERNEL);
+ struct mm_struct *mm = get_mm(test);
+ unsigned long addr, prev_addr = 0;
+ /* Avoid running for ever in failure case. */
+ unsigned long max_iters = 1000000;
+ struct mermap_cpu *mc;
+
+ guard(migrate)();
+ mc = this_cpu_ptr(mm->mermap.cpu);
+
+ /*
+ * Allocate until we see an address less than what we had before - assume
+ * that means a reuse.
+ */
+ for (int i = 0; i < max_iters; i++) {
+ struct mermap_alloc *alloc;
+
+ /*
+ * Obviously flushing the TLB already is not wrong per se, but
+ * it's unexpected and probably means there's some bug.
+ * Use ASSERT to avoid spamming the log in the failure case.
+ */
+ KUNIT_ASSERT_EQ_MSG(test, mc->tlb_flushes, 0,
+ "unexpected flush before alloc %d", i);
+
+ alloc = __mermap_get(mm, page, PAGE_SIZE, PAGE_KERNEL, false);
+ KUNIT_ASSERT_NOT_NULL_MSG(test, alloc, "alloc %d failed", i);
+
+ addr = (unsigned long)mermap_addr(alloc);
+ __mermap_put(mm, alloc);
+ if (addr < prev_addr)
+ break;
+
+ prev_addr = addr;
+ cond_resched();
+ }
+ KUNIT_ASSERT_TRUE_MSG(test, addr < prev_addr, "no address reuse");
+ /* Again, more than one flush isn't wrong per se, but probably a bug. */
+ KUNIT_ASSERT_EQ(test, mc->tlb_flushes, 1);
+}
+
+static struct kunit_case mermap_test_cases[] = {
+ KUNIT_CASE(test_basic_alloc),
+ KUNIT_CASE(test_size),
+ KUNIT_CASE(test_multiple_allocs),
+ KUNIT_CASE(test_tlb_flushed),
+ {}
+};
+
+static struct kunit_suite mermap_test_suite = {
+ .name = "mermap",
+ .test_cases = mermap_test_cases,
+};
+kunit_test_suite(mermap_test_suite);
+
+MODULE_DESCRIPTION("Mermap unit tests");
+MODULE_LICENSE("GPL");
+MODULE_IMPORT_NS("EXPORTED_FOR_KUNIT_TESTING");
--
2.54.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v3 13/26] mm: introduce freetype_t
2026-07-26 22:22 [PATCH v3 00/26] mm: Add ALLOC_UNMAPPED and AS_NO_DIRECT_MAP Brendan Jackman
` (11 preceding siblings ...)
2026-07-26 22:22 ` [PATCH v3 12/26] mm: KUnit tests for " Brendan Jackman
@ 2026-07-26 22:22 ` Brendan Jackman
2026-07-26 22:22 ` [PATCH v3 14/26] mm: move migratetype definitions to freetype.h Brendan Jackman
` (12 subsequent siblings)
25 siblings, 0 replies; 29+ messages in thread
From: Brendan Jackman @ 2026-07-26 22:22 UTC (permalink / raw)
To: Borislav Petkov, Dave Hansen, Peter Zijlstra, Andrew Morton,
David Hildenbrand, Vlastimil Babka, Mike Rapoport, Wei Xu,
Johannes Weiner, Zi Yan, Lorenzo Stoakes
Cc: linux-mm, linux-kernel, x86, rppt, Sumit Garg, Will Deacon,
rientjes, Kalyazin, Nikita, patrick.roy, Itazuri, Takahiro,
Andy Lutomirski, David Kaplan, Thomas Gleixner, Yosry Ahmed,
Patrick Bellasi, Reiji Watanabe, Sean Christopherson,
Brendan Jackman
This is preparation for teaching the page allocator to break up free
pages according to properties that have nothing to do with mobility. For
example it can be used to allocate pages that are non-present in the
physmap, or pages that are sensitive in ASI.
For these usecases, certain allocator behaviours are desirable:
- A "pool" of pages with the given property is usually available, so
that pages can be provided with the correct sensitivity without
zeroing/TLB flushing.
- Pages are physically grouped by the property, so that large
allocations rarely have to alter the pagetables due to ASI.
- The properties can be forced to vary only at a certain fixed address
granularity, so that the pagetables can all be pre-allocated. This is
desirable because the page allocator will be changing mappings:
pre-allocation is a straightforward way to avoid recursive allocations
(of pagetables).
It seems that the existing infrastructure for grouping pages by
mobility, i.e. pageblocks and migratetypes, serves this purpose pretty
nicely. However, overloading migratetype itself for this purpose looks
like a road to maintenance hell. In particular, as soon as such
properties become orthogonal to migratetypes, it would start to require
"doubling" the migratetypes.
Therefore, introduce a new higher-level concept, called "freetype"
(because it is used to index "free"lists) that can encode extra
properties, orthogonally to mobility, via flags.
Since freetypes and migratetypes would be very easy to mix up, freetypes
are (at least for now) stored in a struct typedef similar to atomic_t.
This provides type-safety, but comes at the expense of being pretty
annoying to code with. For instance, freetype_t cannot be compared with
the == operator. Once this code matures, if the freetype/migratetype
distinction gets less confusing, it might be wise to drop this
struct and just use ints.
Because this will eventually be needed from pageblock-flags.h, put this
in its own header instead of directly in mmzone.h.
To try and reduce review pain for such a churny patch, first introduce
freetypes as nothing but an indirection over migratetypes. The helpers
concerned with the flags are defined, but only as stubs. Convert
everything over to using freetypes wherever they are needed to index
freelists, but maintain references to migratetypes in code that really
only cares specifically about mobility.
Leave some examples of code that still effectively assumes migratetypes
and freetypes are equivalent; for example the freetype flags are
completely ignored when indexing pcplists. Later patches will address
these cases incrementally.
Acked-by: Vlastimil Babka (SUSE) <vbabka@kernel.org>
Signed-off-by: Brendan Jackman <jackmanb@google.com>
---
MAINTAINERS | 1 +
include/linux/freetype.h | 45 +++++
include/linux/mmzone.h | 49 ++++-
mm/compaction.c | 36 ++--
mm/internal.h | 7 +-
mm/page_alloc.c | 473 ++++++++++++++++++++++++++++-------------------
mm/page_alloc.h | 26 ++-
mm/page_isolation.c | 2 +-
mm/page_owner.c | 7 +-
mm/page_reporting.c | 4 +-
mm/show_mem.c | 4 +-
11 files changed, 426 insertions(+), 228 deletions(-)
diff --git a/MAINTAINERS b/MAINTAINERS
index 15dd00c7ffec5..9087d2da9c397 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -16968,6 +16968,7 @@ W: http://www.linux-mm.org
T: git git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
F: Documentation/admin-guide/sysctl/vm.rst
F: include/linux/folio_batch.h
+F: include/linux/freetype.h
F: include/linux/gfp.h
F: include/linux/gfp_types.h
F: include/linux/highmem.h
diff --git a/include/linux/freetype.h b/include/linux/freetype.h
new file mode 100644
index 0000000000000..b56a029de4ea4
--- /dev/null
+++ b/include/linux/freetype.h
@@ -0,0 +1,45 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _LINUX_FREETYPE_H
+#define _LINUX_FREETYPE_H
+
+#include <linux/types.h>
+
+/*
+ * A freetype is the identifier for a page freelist. This consists of a
+ * migratetype, and other bits which encode orthogonal properties of memory.
+ */
+typedef struct {
+ int migratetype;
+} freetype_t;
+
+/*
+ * Return a dense linear index for freetypes.
+ */
+static inline int freetype_idx(freetype_t freetype)
+{
+ return freetype.migratetype;
+}
+
+static inline freetype_t freetype_from_idx(unsigned int idx)
+{
+ freetype_t freetype;
+
+ freetype.migratetype = idx;
+ return freetype;
+}
+
+/* No freetype flags actually exist yet. */
+#define NR_FREETYPE_IDXS MIGRATE_TYPES
+
+static inline unsigned int freetype_flags(freetype_t freetype)
+{
+ /* No flags supported yet. */
+ return 0;
+}
+
+static inline bool freetypes_equal(freetype_t a, freetype_t b)
+{
+ return a.migratetype == b.migratetype;
+}
+
+#endif /* _LINUX_FREETYPE_H */
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index a26c8b8552222..9fc77c4e70267 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -5,6 +5,7 @@
#ifndef __ASSEMBLER__
#ifndef __GENERATING_BOUNDS_H
+#include <linux/freetype.h>
#include <linux/spinlock.h>
#include <linux/list.h>
#include <linux/list_nulls.h>
@@ -179,24 +180,62 @@ static inline bool migratetype_is_mergeable(int mt)
#define for_each_free_list(list, zone, order) \
for (order = 0; order < NR_PAGE_ORDERS; order++) \
- for (unsigned int __type = 0; \
- __type < MIGRATE_TYPES && \
- (list = &(zone)->free_area[order].free_list[__type], 1); \
- __type++)
+ for (unsigned int __idx = 0; \
+ __idx < NR_FREETYPE_IDXS && \
+ (list = &(zone)->free_area[order].free_list[__idx], 1); \
+ __idx++)
+
+static inline freetype_t migrate_to_freetype(enum migratetype mt,
+ unsigned int flags)
+{
+ freetype_t freetype;
+
+ /* No flags supported yet. */
+ VM_WARN_ON_ONCE(flags);
+
+ freetype.migratetype = mt;
+ return freetype;
+}
+
+static inline enum migratetype free_to_migratetype(freetype_t freetype)
+{
+ return freetype.migratetype;
+}
+
+/* Convenience helper, return the freetype modified to have the migratetype. */
+static inline freetype_t freetype_with_migrate(freetype_t freetype,
+ enum migratetype migratetype)
+{
+ return migrate_to_freetype(migratetype, freetype_flags(freetype));
+}
extern int page_group_by_mobility_disabled;
+freetype_t get_pfnblock_freetype(const struct page *page, unsigned long pfn);
+
#define get_pageblock_migratetype(page) \
get_pfnblock_migratetype(page, page_to_pfn(page))
+#define get_pageblock_freetype(page) \
+ get_pfnblock_freetype(page, page_to_pfn(page))
+
#define folio_migratetype(folio) \
get_pageblock_migratetype(&folio->page)
struct free_area {
- struct list_head free_list[MIGRATE_TYPES];
+ struct list_head free_list[NR_FREETYPE_IDXS];
unsigned long nr_free;
};
+static inline
+struct list_head *free_area_list(struct free_area *area, freetype_t type)
+{
+ int idx = freetype_idx(type);
+
+ VM_WARN_ON(idx < 0);
+ return &area->free_list[idx];
+}
+
struct pglist_data;
#ifdef CONFIG_NUMA
diff --git a/mm/compaction.c b/mm/compaction.c
index 0568623d9384d..67b01af024e17 100644
--- a/mm/compaction.c
+++ b/mm/compaction.c
@@ -1376,7 +1376,8 @@ isolate_migratepages_range(struct compact_control *cc, unsigned long start_pfn,
static bool suitable_migration_source(struct compact_control *cc,
struct page *page)
{
- int block_mt;
+ freetype_t block_ft;
+ unsigned int block_mt;
if (pageblock_skip_persistent(page))
return false;
@@ -1389,7 +1390,8 @@ static bool suitable_migration_source(struct compact_control *cc,
if (!cc->direct_compaction)
return true;
- block_mt = get_pageblock_migratetype(page);
+ block_ft = get_pageblock_freetype(page);
+ block_mt = free_to_migratetype(block_ft);
/*
* CMA pages can only be taken by ALLOC_CMA requests. For anybody
@@ -1418,10 +1420,11 @@ static bool suitable_migration_source(struct compact_control *cc,
* (directly requested, or defrag_mode) is exempt as the allocator
* claims and converts these.
*/
- if (cc->migratetype == MIGRATE_MOVABLE || cc->order >= pageblock_order)
+ if (free_to_migratetype(cc->freetype) == MIGRATE_MOVABLE ||
+ cc->order >= pageblock_order)
return is_migrate_movable(block_mt);
else
- return block_mt == cc->migratetype;
+ return freetypes_equal(block_ft, cc->freetype);
}
/* Returns true if the page is within a block suitable for migration to */
@@ -2008,7 +2011,8 @@ static unsigned long fast_find_migrateblock(struct compact_control *cc)
* polluting movable blocks through fallbacks. Whole-block production
* is exempt as the allocator claims and converts these.
*/
- if (cc->direct_compaction && cc->migratetype != MIGRATE_MOVABLE &&
+ if (cc->direct_compaction &&
+ free_to_migratetype(cc->freetype) != MIGRATE_MOVABLE &&
cc->order < pageblock_order)
return pfn;
@@ -2280,7 +2284,7 @@ static bool should_proactive_compact_node(pg_data_t *pgdat)
static enum compact_result __compact_finished(struct compact_control *cc)
{
unsigned int order;
- const int migratetype = cc->migratetype;
+ const freetype_t freetype = cc->freetype;
int ret;
/* Compaction run completes if the migrate and free scanner meet */
@@ -2355,25 +2359,27 @@ static enum compact_result __compact_finished(struct compact_control *cc)
for (order = cc->order; order < NR_PAGE_ORDERS; order++) {
struct free_area *area = &cc->zone->free_area[order];
- /* Job done if page is free of the right migratetype */
- if (!free_area_empty(area, migratetype))
+ /* Job done if page is free of the right freetype */
+ if (!free_area_empty(area, freetype))
return COMPACT_SUCCESS;
#ifdef CONFIG_CMA
/* MIGRATE_MOVABLE can fallback on MIGRATE_CMA */
- if (migratetype == MIGRATE_MOVABLE &&
- !free_area_empty(area, MIGRATE_CMA))
+ if (free_to_migratetype(freetype) == MIGRATE_MOVABLE &&
+ !free_area_empty(area, freetype_with_migrate(cc->freetype,
+ MIGRATE_CMA)))
return COMPACT_SUCCESS;
#endif
/*
* Job done if allocation would steal freepages from
- * other migratetype buddy lists.
+ * other freetype buddy lists.
*/
- if (find_suitable_fallback(area, order, migratetype, true, NULL)
+ if (find_suitable_fallback(area, order, freetype, true, NULL)
== FALLBACK_FOUND)
/*
- * Movable pages are OK in any pageblock. If we are
- * stealing for a non-movable allocation, make sure
+ * Movable pages are OK in any pageblock of the right
+ * sensitivity. If we are stealing for a
+ * non-movable allocation, make sure
* we finish compacting the current pageblock first
* (which is assured by the above migrate_pfn align
* check) so it is as free as possible and we won't
@@ -2582,7 +2588,7 @@ compact_zone(struct compact_control *cc, struct capture_control *capc)
INIT_LIST_HEAD(&cc->freepages[order]);
INIT_LIST_HEAD(&cc->migratepages);
- cc->migratetype = gfp_migratetype(cc->gfp_mask);
+ cc->freetype = gfp_freetype(cc->gfp_mask);
if (!is_via_compact_memory(cc->order)) {
ret = compaction_suit_allocation_order(cc->zone, cc->order,
diff --git a/mm/internal.h b/mm/internal.h
index 5a237d9c5fa96..fada318d13541 100644
--- a/mm/internal.h
+++ b/mm/internal.h
@@ -10,6 +10,7 @@
#include <linux/fs.h>
#include <linux/khugepaged.h>
#include <linux/mm.h>
+#include <linux/mmzone.h>
#include <linux/mm_inline.h>
#include <linux/mmu_notifier.h>
#include <linux/pagemap.h>
@@ -730,7 +731,7 @@ int __meminit init_per_zone_wmark_min(void);
extern int __isolate_free_page(struct page *page, unsigned int order);
extern void __putback_isolated_page(struct page *page, unsigned int order,
- int mt);
+ freetype_t freetype);
/*
* This will have no effect, other than possibly generating a warning, if the
@@ -845,7 +846,7 @@ struct compact_control {
short search_order; /* order to start a fast search at */
const gfp_t gfp_mask; /* gfp mask of a direct compactor */
int order; /* order a direct compactor needs */
- int migratetype; /* migratetype of direct compactor */
+ freetype_t freetype; /* freetype of direct compactor */
const unsigned int alloc_flags; /* alloc flags of a direct compactor */
const int highest_zoneidx; /* zone index of a direct compactor */
enum migrate_mode mode; /* Async or sync migration mode */
@@ -870,7 +871,7 @@ struct compact_control {
*/
struct capture_control {
struct zone *zone;
- int migratetype;
+ freetype_t freetype;
/*
* Allocation request order. May differ from the compaction
* order: defrag_mode promotes sub-block allocations to
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 3c16075b12bc9..60aedd0e78b54 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -428,6 +428,46 @@ bool get_pfnblock_bit(const struct page *page, unsigned long pfn,
return test_bit(bitidx + pb_bit, bitmap_word);
}
+/**
+ * __get_pfnblock_freetype - Return the freetype of a pageblock, optionally
+ * ignoring the fact that it's currently isolated.
+ * @page: The page within the block of interest
+ * @pfn: The target page frame number
+ * @ignore_iso: If isolated, return the migratetype that the block had before
+ * isolation.
+ */
+static __always_inline freetype_t
+__get_pfnblock_freetype(const struct page *page, unsigned long pfn,
+ bool ignore_iso)
+{
+ unsigned long mask = PAGEBLOCK_MIGRATETYPE_MASK | PAGEBLOCK_ISO_MASK;
+ unsigned long flags;
+ enum migratetype mt;
+
+ flags = __get_pfnblock_flags_mask(page, pfn, mask);
+ mt = flags & PAGEBLOCK_MIGRATETYPE_MASK;
+
+#ifdef CONFIG_MEMORY_ISOLATION
+ if (!ignore_iso && flags & BIT(PB_migrate_isolate))
+ mt = MIGRATE_ISOLATE;
+#endif
+
+ return migrate_to_freetype(mt, 0);
+}
+
+/**
+ * get_pfnblock_freetype - Return the freetype of a pageblock
+ * @page: The page within the block of interest
+ * @pfn: The target page frame number
+ *
+ * Return: The freetype of the pageblock
+ */
+freetype_t get_pfnblock_freetype(const struct page *page, unsigned long pfn)
+{
+ return __get_pfnblock_freetype(page, pfn, false);
+}
+
+
/**
* get_pfnblock_migratetype - Return the migratetype of a pageblock
* @page: The page within the block of interest
@@ -441,16 +481,7 @@ bool get_pfnblock_bit(const struct page *page, unsigned long pfn,
enum migratetype
get_pfnblock_migratetype(const struct page *page, unsigned long pfn)
{
- unsigned long mask = PAGEBLOCK_MIGRATETYPE_MASK | PAGEBLOCK_ISO_MASK;
- unsigned long flags;
-
- flags = __get_pfnblock_flags_mask(page, pfn, mask);
-
-#ifdef CONFIG_MEMORY_ISOLATION
- if (flags & BIT(PB_migrate_isolate))
- return MIGRATE_ISOLATE;
-#endif
- return flags & PAGEBLOCK_MIGRATETYPE_MASK;
+ return free_to_migratetype(get_pfnblock_freetype(page, pfn));
}
/**
@@ -730,8 +761,11 @@ static inline struct capture_control *task_capc(struct zone *zone)
static inline bool
compaction_capture(struct capture_control *capc, struct page *page,
- int order, int migratetype)
+ int order, freetype_t freetype)
{
+ enum migratetype migratetype = free_to_migratetype(freetype);
+ enum migratetype capc_mt;
+
if (!capc || order != capc->order)
return false;
@@ -740,6 +774,8 @@ compaction_capture(struct capture_control *capc, struct page *page,
is_migrate_isolate(migratetype))
return false;
+ capc_mt = free_to_migratetype(capc->freetype);
+
/*
* Do not let lower order allocations pollute a movable pageblock
* unless compaction is also requesting movable pages.
@@ -748,12 +784,12 @@ compaction_capture(struct capture_control *capc, struct page *page,
* have trouble finding a high-order free page.
*/
if (order < pageblock_order && migratetype == MIGRATE_MOVABLE &&
- capc->migratetype != MIGRATE_MOVABLE)
+ capc_mt != MIGRATE_MOVABLE)
return false;
- if (migratetype != capc->migratetype)
+ if (migratetype != capc_mt)
trace_mm_page_alloc_extfrag(page, capc->order, order,
- capc->migratetype, migratetype);
+ capc_mt, migratetype);
capc->page = page;
return true;
@@ -767,7 +803,7 @@ static inline struct capture_control *task_capc(struct zone *zone)
static inline bool
compaction_capture(struct capture_control *capc, struct page *page,
- int order, int migratetype)
+ int order, freetype_t freetype)
{
return false;
}
@@ -792,23 +828,28 @@ static inline void account_freepages(struct zone *zone, int nr_pages,
/* Used for pages not on another list */
static inline void __add_to_free_list(struct page *page, struct zone *zone,
- unsigned int order, int migratetype,
+ unsigned int order, freetype_t freetype,
bool tail)
{
struct free_area *area = &zone->free_area[order];
int nr_pages = 1 << order;
- VM_WARN_ONCE(get_pageblock_migratetype(page) != migratetype,
- "page type is %d, passed migratetype is %d (nr=%d)\n",
- get_pageblock_migratetype(page), migratetype, nr_pages);
+ if (IS_ENABLED(CONFIG_DEBUG_VM)) {
+ freetype_t block_ft = get_pageblock_freetype(page);
+
+ VM_WARN_ONCE(!freetypes_equal(block_ft, freetype),
+ "page type is %d/%#x, passed type is %d/%#x (nr=%d)\n",
+ block_ft.migratetype, freetype_flags(block_ft),
+ freetype.migratetype, freetype_flags(freetype), nr_pages);
+ }
if (tail)
- list_add_tail(&page->buddy_list, &area->free_list[migratetype]);
+ list_add_tail(&page->buddy_list, free_area_list(area, freetype));
else
- list_add(&page->buddy_list, &area->free_list[migratetype]);
+ list_add(&page->buddy_list, free_area_list(area, freetype));
area->nr_free++;
- if (order >= pageblock_order && !is_migrate_isolate(migratetype))
+ if (order >= pageblock_order && !is_migrate_isolate(free_to_migratetype(freetype)))
__mod_zone_page_state(zone, NR_FREE_PAGES_BLOCKS, nr_pages);
}
@@ -818,17 +859,25 @@ static inline void __add_to_free_list(struct page *page, struct zone *zone,
* allocation again (e.g., optimization for memory onlining).
*/
static inline void move_to_free_list(struct page *page, struct zone *zone,
- unsigned int order, int old_mt, int new_mt)
+ unsigned int order,
+ freetype_t old_ft, freetype_t new_ft)
{
struct free_area *area = &zone->free_area[order];
+ int old_mt = free_to_migratetype(old_ft);
+ int new_mt = free_to_migratetype(new_ft);
int nr_pages = 1 << order;
/* Free page moving can fail, so it happens before the type update */
- VM_WARN_ONCE(get_pageblock_migratetype(page) != old_mt,
- "page type is %d, passed migratetype is %d (nr=%d)\n",
- get_pageblock_migratetype(page), old_mt, nr_pages);
+ if (IS_ENABLED(CONFIG_DEBUG_VM)) {
+ freetype_t block_ft = get_pageblock_freetype(page);
- list_move_tail(&page->buddy_list, &area->free_list[new_mt]);
+ VM_WARN_ONCE(!freetypes_equal(block_ft, old_ft),
+ "page type is %d/%#x, passed freetype is %d/%#x (nr=%d)\n",
+ block_ft.migratetype, freetype_flags(block_ft),
+ old_ft.migratetype, freetype_flags(old_ft), nr_pages);
+ }
+
+ list_move_tail(&page->buddy_list, free_area_list(area, new_ft));
account_freepages(zone, -nr_pages, old_mt);
account_freepages(zone, nr_pages, new_mt);
@@ -842,13 +891,18 @@ static inline void move_to_free_list(struct page *page, struct zone *zone,
}
static inline void __del_page_from_free_list(struct page *page, struct zone *zone,
- unsigned int order, int migratetype)
+ unsigned int order, freetype_t freetype)
{
int nr_pages = 1 << order;
- VM_WARN_ONCE(get_pageblock_migratetype(page) != migratetype,
- "page type is %d, passed migratetype is %d (nr=%d)\n",
- get_pageblock_migratetype(page), migratetype, nr_pages);
+ if (IS_ENABLED(CONFIG_DEBUG_VM)) {
+ freetype_t block_ft = get_pageblock_freetype(page);
+
+ VM_WARN_ONCE(!freetypes_equal(block_ft, freetype),
+ "page type is %d/%#x, passed freetype is %d/%#x (nr=%d)\n",
+ block_ft.migratetype, freetype_flags(block_ft),
+ freetype.migratetype, freetype_flags(freetype), nr_pages);
+ }
/* clear reported state and update reported page count */
if (page_reported(page))
@@ -859,21 +913,21 @@ static inline void __del_page_from_free_list(struct page *page, struct zone *zon
set_page_private(page, 0);
zone->free_area[order].nr_free--;
- if (order >= pageblock_order && !is_migrate_isolate(migratetype))
+ if (order >= pageblock_order && !is_migrate_isolate(free_to_migratetype(freetype)))
__mod_zone_page_state(zone, NR_FREE_PAGES_BLOCKS, -nr_pages);
}
static inline void del_page_from_free_list(struct page *page, struct zone *zone,
- unsigned int order, int migratetype)
+ unsigned int order, freetype_t freetype)
{
- __del_page_from_free_list(page, zone, order, migratetype);
- account_freepages(zone, -(1 << order), migratetype);
+ __del_page_from_free_list(page, zone, order, freetype);
+ account_freepages(zone, -(1 << order), free_to_migratetype(freetype));
}
static inline struct page *get_page_from_free_area(struct free_area *area,
- int migratetype)
+ freetype_t freetype)
{
- return list_first_entry_or_null(&area->free_list[migratetype],
+ return list_first_entry_or_null(free_area_list(area, freetype),
struct page, buddy_list);
}
@@ -940,9 +994,10 @@ static void change_pageblock_range(struct page *pageblock_page,
static inline void __free_one_page(struct page *page,
unsigned long pfn,
struct zone *zone, unsigned int order,
- int migratetype, fpi_t fpi_flags)
+ freetype_t freetype, fpi_t fpi_flags)
{
struct capture_control *capc = task_capc(zone);
+ int migratetype = free_to_migratetype(freetype);
unsigned long buddy_pfn = 0;
unsigned long combined_pfn;
struct page *buddy;
@@ -951,16 +1006,17 @@ static inline void __free_one_page(struct page *page,
VM_BUG_ON(!zone_is_initialized(zone));
VM_BUG_ON_PAGE(page->flags.f & PAGE_FLAGS_CHECK_AT_PREP, page);
- VM_BUG_ON(migratetype == -1);
+ VM_BUG_ON(freetype.migratetype == -1);
VM_BUG_ON_PAGE(pfn & ((1 << order) - 1), page);
VM_BUG_ON_PAGE(bad_range(zone, page), page);
account_freepages(zone, 1 << order, migratetype);
while (order < MAX_PAGE_ORDER) {
- int buddy_mt = migratetype;
+ freetype_t buddy_ft = freetype;
+ enum migratetype buddy_mt = free_to_migratetype(buddy_ft);
- if (compaction_capture(capc, page, order, migratetype)) {
+ if (compaction_capture(capc, page, order, freetype)) {
account_freepages(zone, -(1 << order), migratetype);
return;
}
@@ -976,7 +1032,8 @@ static inline void __free_one_page(struct page *page,
* pageblock isolation could cause incorrect freepage or CMA
* accounting or HIGHATOMIC accounting.
*/
- buddy_mt = get_pfnblock_migratetype(buddy, buddy_pfn);
+ buddy_ft = get_pfnblock_freetype(buddy, buddy_pfn);
+ buddy_mt = free_to_migratetype(buddy_ft);
if (migratetype != buddy_mt &&
(!migratetype_is_mergeable(migratetype) ||
@@ -991,7 +1048,7 @@ static inline void __free_one_page(struct page *page,
if (page_is_guard(buddy))
clear_page_guard(zone, buddy, order);
else
- __del_page_from_free_list(buddy, zone, order, buddy_mt);
+ __del_page_from_free_list(buddy, zone, order, buddy_ft);
if (unlikely(buddy_mt != migratetype)) {
/*
@@ -1018,7 +1075,7 @@ static inline void __free_one_page(struct page *page,
else
to_tail = buddy_merge_likely(pfn, buddy_pfn, page, order);
- __add_to_free_list(page, zone, order, migratetype, to_tail);
+ __add_to_free_list(page, zone, order, freetype, to_tail);
/* Notify page reporting subsystem of freed page */
if (!(fpi_flags & FPI_SKIP_REPORT_NOTIFY))
@@ -1501,19 +1558,20 @@ static void free_pcppages_bulk(struct zone *zone, int count,
nr_pages = 1 << order;
do {
unsigned long pfn;
- int mt;
+ freetype_t ft;
page = list_last_entry(list, struct page, pcp_list);
pfn = page_to_pfn(page);
- mt = get_pfnblock_migratetype(page, pfn);
+ ft = get_pfnblock_freetype(page, pfn);
/* must delete to avoid corrupting pcp list */
list_del(&page->pcp_list);
count -= nr_pages;
pcp->count -= nr_pages;
- __free_one_page(page, pfn, zone, order, mt, FPI_NONE);
- trace_mm_page_pcpu_drain(page, order, mt);
+ __free_one_page(page, pfn, zone, order, ft, FPI_NONE);
+ trace_mm_page_pcpu_drain(page, order,
+ free_to_migratetype(ft));
} while (count > 0 && !list_empty(list));
}
}
@@ -1532,9 +1590,9 @@ static void split_large_buddy(struct zone *zone, struct page *page,
order = pageblock_order;
do {
- int mt = get_pfnblock_migratetype(page, pfn);
+ freetype_t ft = get_pfnblock_freetype(page, pfn);
- __free_one_page(page, pfn, zone, order, mt, fpi);
+ __free_one_page(page, pfn, zone, order, ft, fpi);
pfn += 1 << order;
if (pfn == end)
break;
@@ -1712,7 +1770,7 @@ struct page *__pageblock_pfn_to_page(unsigned long start_pfn,
* -- nyc
*/
static inline unsigned int expand(struct zone *zone, struct page *page, int low,
- int high, int migratetype)
+ int high, freetype_t freetype)
{
unsigned int size = 1 << high;
unsigned int nr_added = 0;
@@ -1731,7 +1789,7 @@ static inline unsigned int expand(struct zone *zone, struct page *page, int low,
if (set_page_guard(zone, &page[size], high))
continue;
- __add_to_free_list(&page[size], zone, high, migratetype, false);
+ __add_to_free_list(&page[size], zone, high, freetype, false);
set_buddy_order(&page[size], high);
nr_added += size;
}
@@ -1741,13 +1799,13 @@ static inline unsigned int expand(struct zone *zone, struct page *page, int low,
static __always_inline void page_del_and_expand(struct zone *zone,
struct page *page, int low,
- int high, int migratetype)
+ int high, freetype_t freetype)
{
int nr_pages = 1 << high;
- __del_page_from_free_list(page, zone, high, migratetype);
- nr_pages -= expand(zone, page, low, high, migratetype);
- account_freepages(zone, -nr_pages, migratetype);
+ __del_page_from_free_list(page, zone, high, freetype);
+ nr_pages -= expand(zone, page, low, high, freetype);
+ account_freepages(zone, -nr_pages, free_to_migratetype(freetype));
}
static void check_new_page_bad(struct page *page)
@@ -1899,7 +1957,7 @@ static void prep_new_page(struct page *page, unsigned int order, gfp_t gfp_flags
*/
static __always_inline
struct page *__rmqueue_smallest(struct zone *zone, unsigned int order,
- int migratetype)
+ freetype_t freetype)
{
unsigned int current_order;
struct free_area *area;
@@ -1907,13 +1965,15 @@ struct page *__rmqueue_smallest(struct zone *zone, unsigned int order,
/* Find a page of the appropriate size in the preferred list */
for (current_order = order; current_order < NR_PAGE_ORDERS; ++current_order) {
+ enum migratetype migratetype = free_to_migratetype(freetype);
+
area = &(zone->free_area[current_order]);
- page = get_page_from_free_area(area, migratetype);
+ page = get_page_from_free_area(area, freetype);
if (!page)
continue;
page_del_and_expand(zone, page, order, current_order,
- migratetype);
+ freetype);
trace_mm_page_alloc_zone_locked(page, order, migratetype,
pcp_allowed_order(order) &&
migratetype < MIGRATE_PCPTYPES);
@@ -1938,13 +1998,18 @@ static int fallbacks[MIGRATE_PCPTYPES][MIGRATE_PCPTYPES - 1] = {
#ifdef CONFIG_CMA
static __always_inline struct page *__rmqueue_cma_fallback(struct zone *zone,
- unsigned int order)
+ unsigned int order, unsigned int ft_flags)
{
- return __rmqueue_smallest(zone, order, MIGRATE_CMA);
+ freetype_t freetype = migrate_to_freetype(MIGRATE_CMA, ft_flags);
+
+ return __rmqueue_smallest(zone, order, freetype);
}
#else
static inline struct page *__rmqueue_cma_fallback(struct zone *zone,
- unsigned int order) { return NULL; }
+ unsigned int order, unsigned int ft_flags)
+{
+ return NULL;
+}
#endif
/*
@@ -1952,7 +2017,7 @@ static inline struct page *__rmqueue_cma_fallback(struct zone *zone,
* change the block type.
*/
static int __move_freepages_block(struct zone *zone, unsigned long start_pfn,
- int old_mt, int new_mt)
+ freetype_t old_ft, freetype_t new_ft)
{
struct page *page;
unsigned long pfn, end_pfn;
@@ -1975,7 +2040,7 @@ static int __move_freepages_block(struct zone *zone, unsigned long start_pfn,
order = buddy_order(page);
- move_to_free_list(page, zone, order, old_mt, new_mt);
+ move_to_free_list(page, zone, order, old_ft, new_ft);
pfn += 1 << order;
pages_moved += 1 << order;
@@ -2035,7 +2100,7 @@ static bool prep_move_freepages_block(struct zone *zone, struct page *page,
}
static int move_freepages_block(struct zone *zone, struct page *page,
- int old_mt, int new_mt)
+ freetype_t old_ft, freetype_t new_ft)
{
unsigned long start_pfn;
int res;
@@ -2043,8 +2108,11 @@ static int move_freepages_block(struct zone *zone, struct page *page,
if (!prep_move_freepages_block(zone, page, &start_pfn, NULL, NULL))
return -1;
- res = __move_freepages_block(zone, start_pfn, old_mt, new_mt);
- set_pageblock_migratetype(pfn_to_page(start_pfn), new_mt);
+ WARN_ON(freetype_flags(old_ft) != freetype_flags(new_ft));
+
+ res = __move_freepages_block(zone, start_pfn, old_ft, new_ft);
+ set_pageblock_migratetype(pfn_to_page(start_pfn),
+ free_to_migratetype(new_ft));
return res;
@@ -2112,8 +2180,7 @@ static bool __move_freepages_block_isolate(struct zone *zone,
struct page *page, bool isolate)
{
unsigned long start_pfn, buddy_pfn;
- int from_mt;
- int to_mt;
+ freetype_t from_ft, to_ft;
struct page *buddy;
if (isolate == get_pageblock_isolate(page)) {
@@ -2136,7 +2203,7 @@ static bool __move_freepages_block_isolate(struct zone *zone,
int order = buddy_order(buddy);
del_page_from_free_list(buddy, zone, order,
- get_pfnblock_migratetype(buddy, buddy_pfn));
+ get_pfnblock_freetype(buddy, buddy_pfn));
toggle_pageblock_isolate(page, isolate);
split_large_buddy(zone, buddy, buddy_pfn, order, FPI_NONE);
return true;
@@ -2145,16 +2212,14 @@ static bool __move_freepages_block_isolate(struct zone *zone,
move:
/* Use PAGEBLOCK_MIGRATETYPE_MASK to get non-isolate migratetype */
if (isolate) {
- from_mt = __get_pfnblock_flags_mask(page, page_to_pfn(page),
- PAGEBLOCK_MIGRATETYPE_MASK);
- to_mt = MIGRATE_ISOLATE;
+ from_ft = __get_pfnblock_freetype(page, page_to_pfn(page), true);
+ to_ft = freetype_with_migrate(from_ft, MIGRATE_ISOLATE);
} else {
- from_mt = MIGRATE_ISOLATE;
- to_mt = __get_pfnblock_flags_mask(page, page_to_pfn(page),
- PAGEBLOCK_MIGRATETYPE_MASK);
+ to_ft = __get_pfnblock_freetype(page, page_to_pfn(page), true);
+ from_ft = freetype_with_migrate(to_ft, MIGRATE_ISOLATE);
}
- __move_freepages_block(zone, start_pfn, from_mt, to_mt);
+ __move_freepages_block(zone, start_pfn, from_ft, to_ft);
toggle_pageblock_isolate(pfn_to_page(start_pfn), isolate);
return true;
@@ -2316,15 +2381,16 @@ static bool should_try_claim_block(unsigned int order, int start_mt)
/*
* Check whether there is a suitable fallback freepage with requested order.
- * If claimable is true, this function returns fallback_mt only if
+ * If claimable is true, this function returns a fallback only if
* we would do this whole-block claiming. This would help to reduce
* fragmentation due to mixed migratetype pages in one pageblock.
*/
enum fallback_result
find_suitable_fallback(struct free_area *area, unsigned int order,
- int migratetype, bool claimable, int *mt_out)
+ freetype_t freetype, bool claimable, freetype_t *ft_out)
{
int i;
+ enum migratetype migratetype = free_to_migratetype(freetype);
if (claimable && !should_try_claim_block(order, migratetype))
return FALLBACK_NOCLAIM;
@@ -2334,10 +2400,15 @@ find_suitable_fallback(struct free_area *area, unsigned int order,
for (i = 0; i < MIGRATE_PCPTYPES - 1 ; i++) {
int fallback_mt = fallbacks[migratetype][i];
+ /*
+ * Fallback to different migratetypes, but currently always with
+ * the same freetype flags.
+ */
+ freetype_t fallback_ft = freetype_with_migrate(freetype, fallback_mt);
- if (!free_area_empty(area, fallback_mt)) {
- if (mt_out)
- *mt_out = fallback_mt;
+ if (!free_area_empty(area, fallback_ft)) {
+ if (ft_out)
+ *ft_out = fallback_ft;
return FALLBACK_FOUND;
}
}
@@ -2354,10 +2425,12 @@ find_suitable_fallback(struct free_area *area, unsigned int order,
*/
static struct page *
try_to_claim_block(struct zone *zone, struct page *page,
- int current_order, int order, int start_type,
- int block_type, unsigned int alloc_flags)
+ int current_order, int order, freetype_t start_type,
+ freetype_t block_type, unsigned int alloc_flags)
{
int free_pages, movable_pages, alike_pages;
+ int block_mt = free_to_migratetype(block_type);
+ int start_mt = free_to_migratetype(start_type);
unsigned long start_pfn;
/* Take ownership for orders >= pageblock_order */
@@ -2365,9 +2438,9 @@ try_to_claim_block(struct zone *zone, struct page *page,
unsigned int nr_added;
del_page_from_free_list(page, zone, current_order, block_type);
- change_pageblock_range(page, current_order, start_type);
+ change_pageblock_range(page, current_order, start_mt);
nr_added = expand(zone, page, order, current_order, start_type);
- account_freepages(zone, nr_added, start_type);
+ account_freepages(zone, nr_added, start_mt);
return page;
}
@@ -2389,7 +2462,7 @@ try_to_claim_block(struct zone *zone, struct page *page,
* For movable allocation, it's the number of movable pages which
* we just obtained. For other types it's a bit more tricky.
*/
- if (start_type == MIGRATE_MOVABLE) {
+ if (start_mt == MIGRATE_MOVABLE) {
alike_pages = movable_pages;
} else {
/*
@@ -2399,7 +2472,7 @@ try_to_claim_block(struct zone *zone, struct page *page,
* vice versa, be conservative since we can't distinguish the
* exact migratetype of non-movable pages.
*/
- if (block_type == MIGRATE_MOVABLE)
+ if (block_mt == MIGRATE_MOVABLE)
alike_pages = pageblock_nr_pages
- (free_pages + movable_pages);
else
@@ -2412,7 +2485,7 @@ try_to_claim_block(struct zone *zone, struct page *page,
if (free_pages + alike_pages >= (1 << (pageblock_order-1)) ||
page_group_by_mobility_disabled) {
__move_freepages_block(zone, start_pfn, block_type, start_type);
- set_pageblock_migratetype(pfn_to_page(start_pfn), start_type);
+ set_pageblock_migratetype(pfn_to_page(start_pfn), start_mt);
return __rmqueue_smallest(zone, order, start_type);
}
@@ -2428,14 +2501,13 @@ try_to_claim_block(struct zone *zone, struct page *page,
* condition simpler.
*/
static __always_inline struct page *
-__rmqueue_claim(struct zone *zone, int order, int start_migratetype,
+__rmqueue_claim(struct zone *zone, int order, freetype_t start_freetype,
unsigned int alloc_flags)
{
struct free_area *area;
int current_order;
int min_order = order;
struct page *page;
- int fallback_mt;
/*
* Do not steal pages from freelists belonging to other pageblocks
@@ -2452,11 +2524,13 @@ __rmqueue_claim(struct zone *zone, int order, int start_migratetype,
*/
for (current_order = MAX_PAGE_ORDER; current_order >= min_order;
--current_order) {
+ int start_mt = free_to_migratetype(start_freetype);
enum fallback_result result;
+ freetype_t fallback_ft;
area = &(zone->free_area[current_order]);
- result = find_suitable_fallback(area, current_order,
- start_migratetype, true, &fallback_mt);
+ result = find_suitable_fallback(area, current_order, start_freetype,
+ true, &fallback_ft);
if (result == FALLBACK_EMPTY)
continue;
@@ -2464,13 +2538,13 @@ __rmqueue_claim(struct zone *zone, int order, int start_migratetype,
if (result == FALLBACK_NOCLAIM)
break;
- page = get_page_from_free_area(area, fallback_mt);
+ page = get_page_from_free_area(area, fallback_ft);
page = try_to_claim_block(zone, page, current_order, order,
- start_migratetype, fallback_mt,
+ start_freetype, fallback_ft,
alloc_flags);
if (page) {
trace_mm_page_alloc_extfrag(page, order, current_order,
- start_migratetype, fallback_mt);
+ start_mt, free_to_migratetype(fallback_ft));
return page;
}
}
@@ -2483,26 +2557,27 @@ __rmqueue_claim(struct zone *zone, int order, int start_migratetype,
* the block as its current migratetype, potentially causing fragmentation.
*/
static __always_inline struct page *
-__rmqueue_steal(struct zone *zone, int order, int start_migratetype)
+__rmqueue_steal(struct zone *zone, int order, freetype_t start_freetype)
{
struct free_area *area;
int current_order;
struct page *page;
- int fallback_mt;
for (current_order = order; current_order < NR_PAGE_ORDERS; current_order++) {
enum fallback_result result;
+ freetype_t fallback_ft;
area = &(zone->free_area[current_order]);
- result = find_suitable_fallback(area, current_order, start_migratetype,
- false, &fallback_mt);
+ result = find_suitable_fallback(area, current_order, start_freetype,
+ false, &fallback_ft);
if (result == FALLBACK_EMPTY)
continue;
- page = get_page_from_free_area(area, fallback_mt);
- page_del_and_expand(zone, page, order, current_order, fallback_mt);
+ page = get_page_from_free_area(area, fallback_ft);
+ page_del_and_expand(zone, page, order, current_order, fallback_ft);
trace_mm_page_alloc_extfrag(page, order, current_order,
- start_migratetype, fallback_mt);
+ free_to_migratetype(start_freetype),
+ free_to_migratetype(fallback_ft));
return page;
}
@@ -2521,7 +2596,7 @@ enum rmqueue_mode {
* Call me with the zone->lock already held.
*/
static __always_inline struct page *
-__rmqueue(struct zone *zone, unsigned int order, int migratetype,
+__rmqueue(struct zone *zone, unsigned int order, freetype_t freetype,
unsigned int alloc_flags, enum rmqueue_mode *mode)
{
struct page *page;
@@ -2535,7 +2610,8 @@ __rmqueue(struct zone *zone, unsigned int order, int migratetype,
if (alloc_flags & ALLOC_CMA &&
zone_page_state(zone, NR_FREE_CMA_PAGES) >
zone_page_state(zone, NR_FREE_PAGES) / 2) {
- page = __rmqueue_cma_fallback(zone, order);
+ page = __rmqueue_cma_fallback(zone, order,
+ freetype_flags(freetype));
if (page)
return page;
}
@@ -2552,13 +2628,14 @@ __rmqueue(struct zone *zone, unsigned int order, int migratetype,
*/
switch (*mode) {
case RMQUEUE_NORMAL:
- page = __rmqueue_smallest(zone, order, migratetype);
+ page = __rmqueue_smallest(zone, order, freetype);
if (page)
return page;
fallthrough;
case RMQUEUE_CMA:
if (alloc_flags & ALLOC_CMA) {
- page = __rmqueue_cma_fallback(zone, order);
+ page = __rmqueue_cma_fallback(zone, order,
+ freetype_flags(freetype));
if (page) {
*mode = RMQUEUE_CMA;
return page;
@@ -2566,7 +2643,7 @@ __rmqueue(struct zone *zone, unsigned int order, int migratetype,
}
fallthrough;
case RMQUEUE_CLAIM:
- page = __rmqueue_claim(zone, order, migratetype, alloc_flags);
+ page = __rmqueue_claim(zone, order, freetype, alloc_flags);
if (page) {
/* Replenished preferred freelist, back to normal mode. */
*mode = RMQUEUE_NORMAL;
@@ -2575,7 +2652,7 @@ __rmqueue(struct zone *zone, unsigned int order, int migratetype,
fallthrough;
case RMQUEUE_STEAL:
if (!(alloc_flags & ALLOC_NOFRAGMENT)) {
- page = __rmqueue_steal(zone, order, migratetype);
+ page = __rmqueue_steal(zone, order, freetype);
if (page) {
*mode = RMQUEUE_STEAL;
return page;
@@ -2592,7 +2669,7 @@ __rmqueue(struct zone *zone, unsigned int order, int migratetype,
*/
static int rmqueue_bulk(struct zone *zone, unsigned int order,
unsigned long count, struct list_head *list,
- int migratetype, unsigned int alloc_flags)
+ freetype_t freetype, unsigned int alloc_flags)
{
enum rmqueue_mode rmqm = RMQUEUE_NORMAL;
unsigned long flags;
@@ -2605,7 +2682,7 @@ static int rmqueue_bulk(struct zone *zone, unsigned int order,
spin_lock_irqsave(&zone->lock, flags);
}
for (i = 0; i < count; ++i) {
- struct page *page = __rmqueue(zone, order, migratetype,
+ struct page *page = __rmqueue(zone, order, freetype,
alloc_flags, &rmqm);
if (unlikely(page == NULL))
break;
@@ -2900,7 +2977,7 @@ static int nr_pcp_high(struct per_cpu_pages *pcp, struct zone *zone,
* reacquired. Return true if pcp is locked, false otherwise.
*/
static bool free_frozen_page_commit(struct zone *zone,
- struct per_cpu_pages *pcp, struct page *page, int migratetype,
+ struct per_cpu_pages *pcp, struct page *page, freetype_t freetype,
unsigned int order, fpi_t fpi_flags)
{
int high, batch;
@@ -2917,7 +2994,7 @@ static bool free_frozen_page_commit(struct zone *zone,
*/
pcp->alloc_factor >>= 1;
__count_vm_events(PGFREE, 1 << order);
- pindex = order_to_pindex(migratetype, order);
+ pindex = order_to_pindex(free_to_migratetype(freetype), order);
list_add(&page->pcp_list, &pcp->lists[pindex]);
pcp->count += 1 << order;
@@ -3011,6 +3088,7 @@ static void __free_frozen_pages(struct page *page, unsigned int order,
struct zone *zone;
unsigned long pfn = page_to_pfn(page);
int migratetype;
+ freetype_t freetype;
if (!pcp_allowed_order(order)) {
__free_pages_ok(page, order, fpi_flags);
@@ -3028,13 +3106,14 @@ static void __free_frozen_pages(struct page *page, unsigned int order,
* excessively into the page allocator
*/
zone = page_zone(page);
- migratetype = get_pfnblock_migratetype(page, pfn);
+ freetype = get_pfnblock_freetype(page, pfn);
+ migratetype = free_to_migratetype(freetype);
if (unlikely(migratetype >= MIGRATE_PCPTYPES)) {
if (unlikely(is_migrate_isolate(migratetype))) {
free_one_page(zone, page, pfn, order, fpi_flags);
return;
}
- migratetype = MIGRATE_MOVABLE;
+ freetype = freetype_with_migrate(freetype, MIGRATE_MOVABLE);
}
if (unlikely((fpi_flags & FPI_NOLOCK) && !can_spin_trylock())) {
@@ -3043,7 +3122,7 @@ static void __free_frozen_pages(struct page *page, unsigned int order,
}
pcp = pcp_spin_trylock(zone->per_cpu_pageset);
if (pcp) {
- if (!free_frozen_page_commit(zone, pcp, page, migratetype,
+ if (!free_frozen_page_commit(zone, pcp, page, freetype,
order, fpi_flags))
return;
pcp_spin_unlock(pcp);
@@ -3100,10 +3179,12 @@ void free_unref_folios(struct folio_batch *folios)
struct zone *zone = folio_zone(folio);
unsigned long pfn = folio_pfn(folio);
unsigned int order = (unsigned long)folio->private;
+ freetype_t freetype;
int migratetype;
folio->private = NULL;
- migratetype = get_pfnblock_migratetype(&folio->page, pfn);
+ freetype = get_pfnblock_freetype(&folio->page, pfn);
+ migratetype = free_to_migratetype(freetype);
/* Different zone requires a different pcp lock */
if (zone != locked_zone ||
@@ -3142,11 +3223,12 @@ void free_unref_folios(struct folio_batch *folios)
* to the MIGRATE_MOVABLE pcp list.
*/
if (unlikely(migratetype >= MIGRATE_PCPTYPES))
- migratetype = MIGRATE_MOVABLE;
+ freetype = freetype_with_migrate(freetype,
+ MIGRATE_MOVABLE);
trace_mm_page_free_batched(&folio->page);
if (!free_frozen_page_commit(zone, pcp, &folio->page,
- migratetype, order, FPI_NONE)) {
+ freetype, order, FPI_NONE)) {
pcp = NULL;
locked_zone = NULL;
}
@@ -3190,9 +3272,9 @@ EXPORT_SYMBOL_GPL(split_page);
int __isolate_free_page(struct page *page, unsigned int order)
{
struct zone *zone = page_zone(page);
- int mt = get_pageblock_migratetype(page);
+ freetype_t ft = get_pageblock_freetype(page);
- if (!is_migrate_isolate(mt)) {
+ if (!is_migrate_isolate(free_to_migratetype(ft))) {
unsigned long watermark;
/*
* Obey watermarks as if the page was being allocated. We can
@@ -3205,7 +3287,7 @@ int __isolate_free_page(struct page *page, unsigned int order)
return 0;
}
- del_page_from_free_list(page, zone, order, mt);
+ del_page_from_free_list(page, zone, order, ft);
/*
* Set the pageblock if the isolated page is at least half of a
@@ -3214,14 +3296,16 @@ int __isolate_free_page(struct page *page, unsigned int order)
if (order >= pageblock_order - 1) {
struct page *endpage = page + (1 << order) - 1;
for (; page < endpage; page += pageblock_nr_pages) {
- int mt = get_pageblock_migratetype(page);
+ freetype_t old_ft = get_pageblock_freetype(page);
+ freetype_t new_ft = freetype_with_migrate(old_ft,
+ MIGRATE_MOVABLE);
+
/*
* Only change normal pageblocks (i.e., they can merge
* with others)
*/
- if (migratetype_is_mergeable(mt))
- move_freepages_block(zone, page, mt,
- MIGRATE_MOVABLE);
+ if (migratetype_is_mergeable(free_to_migratetype(ft)))
+ move_freepages_block(zone, page, old_ft, new_ft);
}
}
@@ -3232,12 +3316,13 @@ int __isolate_free_page(struct page *page, unsigned int order)
* __putback_isolated_page - Return a now-isolated page back where we got it
* @page: Page that was isolated
* @order: Order of the isolated page
- * @mt: The page's pageblock's migratetype
+ * @freetype: The page's pageblock's freetype
*
* This function is meant to return a page pulled from the free lists via
* __isolate_free_page back to the free lists they were pulled from.
*/
-void __putback_isolated_page(struct page *page, unsigned int order, int mt)
+void __putback_isolated_page(struct page *page, unsigned int order,
+ freetype_t freetype)
{
struct zone *zone = page_zone(page);
@@ -3245,7 +3330,7 @@ void __putback_isolated_page(struct page *page, unsigned int order, int mt)
lockdep_assert_held(&zone->lock);
/* Return isolated page to tail of freelist. */
- __free_one_page(page, page_to_pfn(page), zone, order, mt,
+ __free_one_page(page, page_to_pfn(page), zone, order, freetype,
FPI_SKIP_REPORT_NOTIFY | FPI_TO_TAIL);
}
@@ -3278,10 +3363,12 @@ static inline void zone_statistics(struct zone *preferred_zone, struct zone *z,
static __always_inline
struct page *rmqueue_buddy(struct zone *preferred_zone, struct zone *zone,
unsigned int order, unsigned int alloc_flags,
- int migratetype)
+ freetype_t freetype)
{
struct page *page;
unsigned long flags;
+ freetype_t ft_high = freetype_with_migrate(freetype,
+ MIGRATE_HIGHATOMIC);
do {
page = NULL;
@@ -3292,11 +3379,11 @@ struct page *rmqueue_buddy(struct zone *preferred_zone, struct zone *zone,
spin_lock_irqsave(&zone->lock, flags);
}
if (alloc_flags & ALLOC_HIGHATOMIC)
- page = __rmqueue_smallest(zone, order, MIGRATE_HIGHATOMIC);
+ page = __rmqueue_smallest(zone, order, ft_high);
if (!page) {
enum rmqueue_mode rmqm = RMQUEUE_NORMAL;
- page = __rmqueue(zone, order, migratetype, alloc_flags, &rmqm);
+ page = __rmqueue(zone, order, freetype, alloc_flags, &rmqm);
/*
* If the allocation fails, allow OOM handling and
@@ -3305,7 +3392,7 @@ struct page *rmqueue_buddy(struct zone *preferred_zone, struct zone *zone,
* high-order atomic allocation in the future.
*/
if (!page && (alloc_flags & (ALLOC_OOM|ALLOC_NON_BLOCK)))
- page = __rmqueue_smallest(zone, order, MIGRATE_HIGHATOMIC);
+ page = __rmqueue_smallest(zone, order, ft_high);
if (!page) {
spin_unlock_irqrestore(&zone->lock, flags);
@@ -3382,7 +3469,7 @@ static int nr_pcp_alloc(struct per_cpu_pages *pcp, struct zone *zone, int order)
/* Remove page from the per-cpu list, caller must protect the list */
static inline
struct page *__rmqueue_pcplist(struct zone *zone, unsigned int order,
- int migratetype,
+ freetype_t freetype,
unsigned int alloc_flags,
struct per_cpu_pages *pcp,
struct list_head *list)
@@ -3410,7 +3497,7 @@ struct page *__rmqueue_pcplist(struct zone *zone, unsigned int order,
alloced = rmqueue_bulk(zone, order,
batch, list,
- migratetype, alloc_flags);
+ freetype, alloc_flags);
pcp->count += alloced << order;
if (unlikely(list_empty(list)))
@@ -3428,7 +3515,7 @@ struct page *__rmqueue_pcplist(struct zone *zone, unsigned int order,
/* Lock and remove page from the per-cpu list */
static struct page *rmqueue_pcplist(struct zone *preferred_zone,
struct zone *zone, unsigned int order,
- int migratetype, unsigned int alloc_flags)
+ freetype_t freetype, unsigned int alloc_flags)
{
struct per_cpu_pages *pcp;
struct list_head *list;
@@ -3445,8 +3532,8 @@ static struct page *rmqueue_pcplist(struct zone *preferred_zone,
* frees.
*/
pcp->free_count >>= 1;
- list = &pcp->lists[order_to_pindex(migratetype, order)];
- page = __rmqueue_pcplist(zone, order, migratetype, alloc_flags, pcp, list);
+ list = &pcp->lists[order_to_pindex(free_to_migratetype(freetype), order)];
+ page = __rmqueue_pcplist(zone, order, freetype, alloc_flags, pcp, list);
pcp_spin_unlock(pcp);
if (page) {
__count_zid_vm_events(PGALLOC, page_zonenum(page), 1 << order);
@@ -3471,19 +3558,19 @@ static inline
struct page *rmqueue(struct zone *preferred_zone,
struct zone *zone, unsigned int order,
gfp_t gfp_flags, unsigned int alloc_flags,
- int migratetype)
+ freetype_t freetype)
{
struct page *page;
if (likely(pcp_allowed_order(order))) {
page = rmqueue_pcplist(preferred_zone, zone, order,
- migratetype, alloc_flags);
+ freetype, alloc_flags);
if (likely(page))
goto out;
}
page = rmqueue_buddy(preferred_zone, zone, order, alloc_flags,
- migratetype);
+ freetype);
out:
/* Separate test+clear to avoid unnecessary atomics */
@@ -3505,7 +3592,7 @@ struct page *rmqueue(struct zone *preferred_zone,
static void reserve_highatomic_pageblock(struct page *page, int order,
struct zone *zone)
{
- int mt;
+ freetype_t ft, ft_high;
unsigned long max_managed;
/*
@@ -3527,13 +3614,14 @@ static void reserve_highatomic_pageblock(struct page *page, int order,
return;
/* Yoink! */
- mt = get_pageblock_migratetype(page);
+ ft = get_pageblock_freetype(page);
/* Only reserve normal pageblocks (i.e., they can merge with others) */
- if (!migratetype_is_mergeable(mt))
+ if (!migratetype_is_mergeable(free_to_migratetype(ft)))
return;
+ ft_high = freetype_with_migrate(ft, MIGRATE_HIGHATOMIC);
if (order < pageblock_order) {
- if (move_freepages_block(zone, page, mt, MIGRATE_HIGHATOMIC) == -1)
+ if (move_freepages_block(zone, page, ft, ft_high) == -1)
return;
zone->nr_reserved_highatomic += pageblock_nr_pages;
} else {
@@ -3574,9 +3662,11 @@ static bool unreserve_highatomic_pageblock(const struct alloc_context *ac,
guard(spinlock_irqsave)(&zone->lock);
for (order = 0; order < NR_PAGE_ORDERS; order++) {
struct free_area *area = &(zone->free_area[order]);
+ freetype_t ft_high = freetype_with_migrate(ac->freetype,
+ MIGRATE_HIGHATOMIC);
unsigned long size;
- page = get_page_from_free_area(area, MIGRATE_HIGHATOMIC);
+ page = get_page_from_free_area(area, ft_high);
if (!page)
continue;
@@ -3603,14 +3693,14 @@ static bool unreserve_highatomic_pageblock(const struct alloc_context *ac,
*/
if (order < pageblock_order)
ret = move_freepages_block(zone, page,
- MIGRATE_HIGHATOMIC,
- ac->migratetype);
+ ft_high,
+ ac->freetype);
else {
move_to_free_list(page, zone, order,
- MIGRATE_HIGHATOMIC,
- ac->migratetype);
+ ft_high,
+ ac->freetype);
change_pageblock_range(page, order,
- ac->migratetype);
+ free_to_migratetype(ac->freetype));
ret = 1;
}
/*
@@ -3707,25 +3797,26 @@ bool __zone_watermark_ok(struct zone *z, unsigned int order, unsigned long mark,
/* For a high-order request, check at least one suitable page is free */
for (o = order; o < NR_PAGE_ORDERS; o++) {
struct free_area *area = &z->free_area[o];
- int mt;
if (!area->nr_free)
continue;
- for (mt = 0; mt < MIGRATE_PCPTYPES; mt++) {
- if (!free_area_empty(area, mt))
- return true;
- }
+ for (int ft_idx = 0; ft_idx < NR_FREETYPE_IDXS; ft_idx++) {
+ freetype_t ft = freetype_from_idx(ft_idx);
+ int mt = free_to_migratetype(ft);
+ if (list_empty(&area->free_list[ft_idx]))
+ continue;
+
+ if (mt < MIGRATE_PCPTYPES)
+ return true;
#ifdef CONFIG_CMA
- if ((alloc_flags & ALLOC_CMA) &&
- !free_area_empty(area, MIGRATE_CMA)) {
- return true;
- }
+ if ((alloc_flags & ALLOC_CMA) && mt == MIGRATE_CMA)
+ return true;
#endif
- if ((alloc_flags & (ALLOC_HIGHATOMIC|ALLOC_OOM)) &&
- !free_area_empty(area, MIGRATE_HIGHATOMIC)) {
- return true;
+ if ((alloc_flags & (ALLOC_HIGHATOMIC|ALLOC_OOM)) &&
+ mt == MIGRATE_HIGHATOMIC)
+ return true;
}
}
return false;
@@ -3844,7 +3935,7 @@ alloc_flags_nofragment(struct zone *zone, gfp_t gfp_mask)
static inline unsigned int alloc_flags_cma(gfp_t gfp_mask)
{
#ifdef CONFIG_CMA
- if (gfp_migratetype(gfp_mask) == MIGRATE_MOVABLE)
+ if (free_to_migratetype(gfp_freetype(gfp_mask)) == MIGRATE_MOVABLE)
return ALLOC_CMA;
#endif
return ALLOC_DEFAULT;
@@ -3996,7 +4087,7 @@ get_page_from_freelist(gfp_t gfp_mask, unsigned int order, int alloc_flags,
try_this_zone:
page = rmqueue(zonelist_zone(ac->preferred_zoneref), zone, order,
- gfp_mask, alloc_flags, ac->migratetype);
+ gfp_mask, alloc_flags, ac->freetype);
if (page) {
prep_new_page(page, order, gfp_mask, alloc_flags);
@@ -4203,7 +4294,7 @@ __alloc_pages_direct_compact(gfp_t gfp_mask, unsigned int order,
unsigned int noreclaim_flag;
struct capture_control capc = {
.zone = NULL,
- .migratetype = ac->migratetype,
+ .freetype = ac->freetype,
.order = order,
.page = NULL,
};
@@ -4221,7 +4312,8 @@ __alloc_pages_direct_compact(gfp_t gfp_mask, unsigned int order,
* For those, promote the order to help make blocks, instead
* of spinning in reclaim alone unproductively.
*/
- if ((alloc_flags & ALLOC_NOFRAGMENT) && ac->migratetype != MIGRATE_MOVABLE)
+ if ((alloc_flags & ALLOC_NOFRAGMENT) &&
+ free_to_migratetype(ac->freetype) != MIGRATE_MOVABLE)
compact_order = max(order, pageblock_order);
if (!compact_order)
@@ -4529,7 +4621,8 @@ __alloc_pages_direct_reclaim(gfp_t gfp_mask, unsigned int order,
int reclaim_order = order;
/* Match the slowpath compaction promotion in __alloc_pages_direct_compact */
- if ((alloc_flags & ALLOC_NOFRAGMENT) && ac->migratetype != MIGRATE_MOVABLE)
+ if ((alloc_flags & ALLOC_NOFRAGMENT) &&
+ free_to_migratetype(ac->freetype) != MIGRATE_MOVABLE)
reclaim_order = max(order, pageblock_order);
psi_memstall_enter(&pflags);
@@ -4859,6 +4952,7 @@ __alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order,
bool compact_first = false;
bool can_retry_reserves = true;
unsigned long alloc_start_time = jiffies;
+ enum migratetype migratetype = free_to_migratetype(ac->freetype);
if (unlikely(nofail)) {
/*
@@ -4893,8 +4987,7 @@ __alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order,
* try prevent permanent fragmentation by migrating from blocks of the
* same migratetype.
*/
- if (can_compact && (costly_order || (order > 0 &&
- ac->migratetype != MIGRATE_MOVABLE))) {
+ if (can_compact && (costly_order || (order > 0 && migratetype != MIGRATE_MOVABLE))) {
compact_first = true;
compact_priority = INIT_COMPACT_PRIORITY;
}
@@ -5158,7 +5251,7 @@ static inline bool prepare_alloc_pages(gfp_t gfp_mask, unsigned int order,
ac->highest_zoneidx = gfp_zone(gfp_mask);
ac->zonelist = node_zonelist(preferred_nid, gfp_mask);
ac->nodemask = nodemask;
- ac->migratetype = gfp_migratetype(gfp_mask);
+ ac->freetype = gfp_freetype(gfp_mask);
if (cpusets_enabled()) {
*alloc_gfp |= __GFP_HARDWALL;
@@ -5319,7 +5412,7 @@ unsigned long alloc_pages_bulk_noprof(gfp_t gfp, int preferred_nid,
goto failed;
/* Attempt the batch allocation */
- pcp_list = &pcp->lists[order_to_pindex(ac.migratetype, 0)];
+ pcp_list = &pcp->lists[order_to_pindex(free_to_migratetype(ac.freetype), 0)];
while (nr_populated < nr_pages) {
/* Skip existing pages */
@@ -5328,7 +5421,7 @@ unsigned long alloc_pages_bulk_noprof(gfp_t gfp, int preferred_nid,
continue;
}
- page = __rmqueue_pcplist(zone, 0, ac.migratetype, alloc_flags,
+ page = __rmqueue_pcplist(zone, 0, ac.freetype, alloc_flags,
pcp, pcp_list);
if (unlikely(!page)) {
/* Try and allocate at least one page */
@@ -5518,7 +5611,8 @@ struct page *__alloc_frozen_pages_noprof(gfp_t gfp, unsigned int order,
page = NULL;
}
- trace_mm_page_alloc(page, order, alloc_gfp, ac.migratetype);
+ trace_mm_page_alloc(page, order, alloc_gfp,
+ free_to_migratetype(ac.freetype));
kmsan_alloc_page(page, order, alloc_gfp);
return page;
@@ -7792,6 +7886,8 @@ unsigned long __offline_isolated_pages(unsigned long start_pfn,
zone = page_zone(pfn_to_page(pfn));
guard(spinlock_irqsave)(&zone->lock);
while (pfn < end_pfn) {
+ freetype_t ft_iso = migrate_to_freetype(MIGRATE_ISOLATE, 0);
+
page = pfn_to_page(pfn);
/*
* The HWPoisoned page may be not in buddy system, and
@@ -7815,9 +7911,9 @@ unsigned long __offline_isolated_pages(unsigned long start_pfn,
BUG_ON(page_count(page));
BUG_ON(!PageBuddy(page));
- VM_WARN_ON(get_pageblock_migratetype(page) != MIGRATE_ISOLATE);
+ VM_WARN_ON(!freetypes_equal(get_pageblock_freetype(page), ft_iso));
order = buddy_order(page);
- del_page_from_free_list(page, zone, order, MIGRATE_ISOLATE);
+ del_page_from_free_list(page, zone, order, ft_iso);
pfn += (1 << order);
}
@@ -7847,11 +7943,11 @@ EXPORT_SYMBOL(is_free_buddy_page);
#ifdef CONFIG_MEMORY_FAILURE
static inline void add_to_free_list(struct page *page, struct zone *zone,
- unsigned int order, int migratetype,
+ unsigned int order, freetype_t freetype,
bool tail)
{
- __add_to_free_list(page, zone, order, migratetype, tail);
- account_freepages(zone, 1 << order, migratetype);
+ __add_to_free_list(page, zone, order, freetype, tail);
+ account_freepages(zone, 1 << order, free_to_migratetype(freetype));
}
/*
@@ -7860,7 +7956,7 @@ static inline void add_to_free_list(struct page *page, struct zone *zone,
*/
static void break_down_buddy_pages(struct zone *zone, struct page *page,
struct page *target, int low, int high,
- int migratetype)
+ freetype_t freetype)
{
unsigned long size = 1 << high;
struct page *current_buddy;
@@ -7879,7 +7975,7 @@ static void break_down_buddy_pages(struct zone *zone, struct page *page,
if (set_page_guard(zone, current_buddy, high))
continue;
- add_to_free_list(current_buddy, zone, high, migratetype, false);
+ add_to_free_list(current_buddy, zone, high, freetype, false);
set_buddy_order(current_buddy, high);
}
}
@@ -7900,13 +7996,13 @@ bool take_page_off_buddy(struct page *page)
if (PageBuddy(page_head) && page_order >= order) {
unsigned long pfn_head = page_to_pfn(page_head);
- int migratetype = get_pfnblock_migratetype(page_head,
- pfn_head);
+ freetype_t freetype = get_pfnblock_freetype(page_head,
+ pfn_head);
del_page_from_free_list(page_head, zone, page_order,
- migratetype);
+ freetype);
break_down_buddy_pages(zone, page_head, page, 0,
- page_order, migratetype);
+ page_order, freetype);
SetPageHWPoisonTakenOff(page);
return true;
}
@@ -7926,12 +8022,13 @@ bool put_page_back_buddy(struct page *page)
guard(spinlock_irqsave)(&zone->lock);
if (put_page_testzero(page)) {
unsigned long pfn = page_to_pfn(page);
- int migratetype = get_pfnblock_migratetype(page, pfn);
+ freetype_t freetype = get_pfnblock_freetype(page, pfn);
ClearPageHWPoisonTakenOff(page);
- __free_one_page(page, pfn, zone, 0, migratetype, FPI_NONE);
- if (TestClearPageHWPoison(page))
+ __free_one_page(page, pfn, zone, 0, freetype, FPI_NONE);
+ if (TestClearPageHWPoison(page)) {
return true;
+ }
}
return false;
diff --git a/mm/page_alloc.h b/mm/page_alloc.h
index 409be31898ae7..9928aa9012588 100644
--- a/mm/page_alloc.h
+++ b/mm/page_alloc.h
@@ -77,7 +77,7 @@ struct alloc_context {
struct zonelist *zonelist;
const nodemask_t *nodemask;
struct zoneref *preferred_zoneref;
- int migratetype;
+ freetype_t freetype;
/*
* highest_zoneidx represents highest usable zone index of
@@ -268,7 +268,7 @@ extern void zone_pcp_enable(struct zone *zone);
extern void zone_pcp_init(struct zone *zone);
enum fallback_result {
- /* Found suitable migratetype, *mt_out is valid. */
+ /* Found suitable fallback, *ft_out is valid. */
FALLBACK_FOUND,
/* No fallback found in requested order. */
FALLBACK_EMPTY,
@@ -277,19 +277,21 @@ enum fallback_result {
};
enum fallback_result
find_suitable_fallback(struct free_area *area, unsigned int order,
- int migratetype, bool claimable, int *mt_out);
+ freetype_t freetype, bool claimable, freetype_t *ft_out);
-static inline bool free_area_empty(struct free_area *area, int migratetype)
+static inline bool free_area_empty(struct free_area *area, freetype_t freetype)
{
- return list_empty(&area->free_list[migratetype]);
+ return list_empty(free_area_list(area, freetype));
}
/* Convert GFP flags to their corresponding migrate type */
#define GFP_MOVABLE_MASK (__GFP_RECLAIMABLE|__GFP_MOVABLE)
#define GFP_MOVABLE_SHIFT 3
-static inline int gfp_migratetype(const gfp_t gfp_flags)
+static inline freetype_t gfp_freetype(const gfp_t gfp_flags)
{
+ unsigned int migratetype;
+
VM_WARN_ON((gfp_flags & GFP_MOVABLE_MASK) == GFP_MOVABLE_MASK);
BUILD_BUG_ON((1UL << GFP_MOVABLE_SHIFT) != ___GFP_MOVABLE);
BUILD_BUG_ON((___GFP_MOVABLE >> GFP_MOVABLE_SHIFT) != MIGRATE_MOVABLE);
@@ -297,11 +299,15 @@ static inline int gfp_migratetype(const gfp_t gfp_flags)
BUILD_BUG_ON(((___GFP_MOVABLE | ___GFP_RECLAIMABLE) >>
GFP_MOVABLE_SHIFT) != MIGRATE_HIGHATOMIC);
- if (unlikely(page_group_by_mobility_disabled))
- return MIGRATE_UNMOVABLE;
+ if (unlikely(page_group_by_mobility_disabled)) {
+ migratetype = MIGRATE_UNMOVABLE;
+ } else {
+ /* Group based on mobility */
+ migratetype = (__force unsigned long)(gfp_flags & GFP_MOVABLE_MASK)
+ >> GFP_MOVABLE_SHIFT;
+ }
- /* Group based on mobility */
- return (__force unsigned long)(gfp_flags & GFP_MOVABLE_MASK) >> GFP_MOVABLE_SHIFT;
+ return migrate_to_freetype(migratetype, 0);
}
#undef GFP_MOVABLE_MASK
#undef GFP_MOVABLE_SHIFT
diff --git a/mm/page_isolation.c b/mm/page_isolation.c
index e5dfc7bf49446..f81186431d3c3 100644
--- a/mm/page_isolation.c
+++ b/mm/page_isolation.c
@@ -275,7 +275,7 @@ static void unset_migratetype_isolate(struct page *page)
WARN_ON_ONCE(!pageblock_unisolate_and_move_free_pages(zone, page));
} else {
clear_pageblock_isolate(page);
- __putback_isolated_page(page, order, get_pageblock_migratetype(page));
+ __putback_isolated_page(page, order, get_pageblock_freetype(page));
}
zone->nr_isolate_pageblock--;
}
diff --git a/mm/page_owner.c b/mm/page_owner.c
index fbbda7ba914ba..ed7fcc0d70f3f 100644
--- a/mm/page_owner.c
+++ b/mm/page_owner.c
@@ -526,7 +526,8 @@ void pagetypeinfo_showmixedcount_print(struct seq_file *m,
goto ext_put_continue;
page_owner = get_page_owner(page_ext);
- page_mt = gfp_migratetype(page_owner->gfp_mask);
+ page_mt = free_to_migratetype(
+ gfp_freetype(page_owner->gfp_mask));
if (pageblock_mt != page_mt) {
if (is_migrate_cma(pageblock_mt))
count[MIGRATE_MOVABLE]++;
@@ -625,7 +626,7 @@ print_page_owner(char __user *buf, size_t count, unsigned long pfn,
/* Print information relevant to grouping pages by mobility */
pageblock_mt = get_pageblock_migratetype(page);
- page_mt = gfp_migratetype(page_owner->gfp_mask);
+ page_mt = free_to_migratetype(gfp_freetype(page_owner->gfp_mask));
ret += scnprintf(kbuf + ret, count - ret,
"PFN 0x%lx type %s Block %lu type %s Flags %pGp\n",
pfn,
@@ -685,7 +686,7 @@ void __dump_page_owner(const struct page *page)
page_owner = get_page_owner(page_ext);
gfp_mask = page_owner->gfp_mask;
- mt = gfp_migratetype(gfp_mask);
+ mt = free_to_migratetype(gfp_freetype(gfp_mask));
if (!test_bit(PAGE_EXT_OWNER, &page_ext->flags)) {
pr_alert("page_owner info is not present (never set?)\n");
diff --git a/mm/page_reporting.c b/mm/page_reporting.c
index 1cce8729696ee..9c8a27ff6cdd1 100644
--- a/mm/page_reporting.c
+++ b/mm/page_reporting.c
@@ -115,10 +115,10 @@ page_reporting_drain(struct page_reporting_dev_info *prdev,
*/
do {
struct page *page = sg_page(sg);
- int mt = get_pageblock_migratetype(page);
+ freetype_t ft = get_pageblock_freetype(page);
unsigned int order = get_order(sg->length);
- __putback_isolated_page(page, order, mt);
+ __putback_isolated_page(page, order, ft);
/* If the pages were not reported due to error skip flagging */
if (!reported)
diff --git a/mm/show_mem.c b/mm/show_mem.c
index d1288b4c2b640..607360a84bfab 100644
--- a/mm/show_mem.c
+++ b/mm/show_mem.c
@@ -380,7 +380,9 @@ static void show_free_areas(unsigned int filter, const nodemask_t *nodemask,
types[order] = 0;
for (type = 0; type < MIGRATE_TYPES; type++) {
- if (!free_area_empty(area, type))
+ freetype_t ft = migrate_to_freetype(type, 0);
+
+ if (!free_area_empty(area, ft))
types[order] |= 1 << type;
}
}
--
2.54.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v3 14/26] mm: move migratetype definitions to freetype.h
2026-07-26 22:22 [PATCH v3 00/26] mm: Add ALLOC_UNMAPPED and AS_NO_DIRECT_MAP Brendan Jackman
` (12 preceding siblings ...)
2026-07-26 22:22 ` [PATCH v3 13/26] mm: introduce freetype_t Brendan Jackman
@ 2026-07-26 22:22 ` Brendan Jackman
2026-07-26 22:22 ` [PATCH v3 15/26] mm/page_alloc: add support for freetypes with no freelist Brendan Jackman
` (11 subsequent siblings)
25 siblings, 0 replies; 29+ messages in thread
From: Brendan Jackman @ 2026-07-26 22:22 UTC (permalink / raw)
To: Borislav Petkov, Dave Hansen, Peter Zijlstra, Andrew Morton,
David Hildenbrand, Vlastimil Babka, Mike Rapoport, Wei Xu,
Johannes Weiner, Zi Yan, Lorenzo Stoakes
Cc: linux-mm, linux-kernel, x86, rppt, Sumit Garg, Will Deacon,
rientjes, Kalyazin, Nikita, patrick.roy, Itazuri, Takahiro,
Andy Lutomirski, David Kaplan, Thomas Gleixner, Yosry Ahmed,
Patrick Bellasi, Reiji Watanabe, Sean Christopherson,
Brendan Jackman
Since migratetypes are a sub-element of freetype, move the pure
definitions into the new freetype.h.
This will enable referring to these raw types from pageblock-flags.h.
Acked-by: Vlastimil Babka (SUSE) <vbabka@kernel.org>
Signed-off-by: Brendan Jackman <jackmanb@google.com>
---
include/linux/freetype.h | 84 ++++++++++++++++++++++++++++++++++++++++++++++++
include/linux/mmzone.h | 73 -----------------------------------------
2 files changed, 84 insertions(+), 73 deletions(-)
diff --git a/include/linux/freetype.h b/include/linux/freetype.h
index b56a029de4ea4..b16dcd58e184c 100644
--- a/include/linux/freetype.h
+++ b/include/linux/freetype.h
@@ -3,6 +3,66 @@
#define _LINUX_FREETYPE_H
#include <linux/types.h>
+#include <linux/mmdebug.h>
+
+/*
+ * A migratetype is the part of a freetype that encodes the mobility
+ * requirements for the allocations the freelist is intended to serve.
+ *
+ * It's also currently overloaded to encode page isolation state.
+ */
+enum migratetype {
+ MIGRATE_UNMOVABLE,
+ MIGRATE_MOVABLE,
+ MIGRATE_RECLAIMABLE,
+ MIGRATE_PCPTYPES, /* the number of types on the pcp lists */
+ MIGRATE_HIGHATOMIC = MIGRATE_PCPTYPES,
+#ifdef CONFIG_CMA
+ /*
+ * MIGRATE_CMA migration type is designed to mimic the way
+ * ZONE_MOVABLE works. Only movable pages can be allocated
+ * from MIGRATE_CMA pageblocks and page allocator never
+ * implicitly change migration type of MIGRATE_CMA pageblock.
+ *
+ * The way to use it is to change migratetype of a range of
+ * pageblocks to MIGRATE_CMA which can be done by
+ * __free_pageblock_cma() function.
+ */
+ MIGRATE_CMA,
+ __MIGRATE_TYPE_END = MIGRATE_CMA,
+#else
+ __MIGRATE_TYPE_END = MIGRATE_HIGHATOMIC,
+#endif
+#ifdef CONFIG_MEMORY_ISOLATION
+ MIGRATE_ISOLATE, /* can't allocate from here */
+#endif
+ MIGRATE_TYPES
+};
+
+/* In mm/page_alloc.c; keep in sync also with show_migration_types() there */
+extern const char * const migratetype_names[MIGRATE_TYPES];
+
+#ifdef CONFIG_CMA
+# define is_migrate_cma(migratetype) unlikely((migratetype) == MIGRATE_CMA)
+#else
+# define is_migrate_cma(migratetype) false
+#endif
+
+static inline bool is_migrate_movable(int mt)
+{
+ return is_migrate_cma(mt) || mt == MIGRATE_MOVABLE;
+}
+
+/*
+ * Check whether a migratetype can be merged with another migratetype.
+ *
+ * It is only mergeable when it can fall back to other migratetypes for
+ * allocation. See fallbacks[MIGRATE_TYPES][3] in page_alloc.c.
+ */
+static inline bool migratetype_is_mergeable(int mt)
+{
+ return mt < MIGRATE_PCPTYPES;
+}
/*
* A freetype is the identifier for a page freelist. This consists of a
@@ -42,4 +102,28 @@ static inline bool freetypes_equal(freetype_t a, freetype_t b)
return a.migratetype == b.migratetype;
}
+static inline freetype_t migrate_to_freetype(enum migratetype mt,
+ unsigned int flags)
+{
+ freetype_t freetype;
+
+ /* No flags supported yet. */
+ VM_WARN_ON_ONCE(flags);
+
+ freetype.migratetype = mt;
+ return freetype;
+}
+
+static inline enum migratetype free_to_migratetype(freetype_t freetype)
+{
+ return freetype.migratetype;
+}
+
+/* Convenience helper, return the freetype modified to have the migratetype. */
+static inline freetype_t freetype_with_migrate(freetype_t freetype,
+ enum migratetype migratetype)
+{
+ return migrate_to_freetype(migratetype, freetype_flags(freetype));
+}
+
#endif /* _LINUX_FREETYPE_H */
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 9fc77c4e70267..78455285dda7c 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -116,39 +116,7 @@
#define __NR_VMEMMAP_TAILS (MAX_FOLIO_ORDER - VMEMMAP_TAIL_MIN_ORDER + 1)
#define NR_VMEMMAP_TAILS (__NR_VMEMMAP_TAILS > 0 ? __NR_VMEMMAP_TAILS : 0)
-enum migratetype {
- MIGRATE_UNMOVABLE,
- MIGRATE_MOVABLE,
- MIGRATE_RECLAIMABLE,
- MIGRATE_PCPTYPES, /* the number of types on the pcp lists */
- MIGRATE_HIGHATOMIC = MIGRATE_PCPTYPES,
#ifdef CONFIG_CMA
- /*
- * MIGRATE_CMA migration type is designed to mimic the way
- * ZONE_MOVABLE works. Only movable pages can be allocated
- * from MIGRATE_CMA pageblocks and page allocator never
- * implicitly change migration type of MIGRATE_CMA pageblock.
- *
- * The way to use it is to change migratetype of a range of
- * pageblocks to MIGRATE_CMA which can be done by
- * __free_pageblock_cma() function.
- */
- MIGRATE_CMA,
- __MIGRATE_TYPE_END = MIGRATE_CMA,
-#else
- __MIGRATE_TYPE_END = MIGRATE_HIGHATOMIC,
-#endif
-#ifdef CONFIG_MEMORY_ISOLATION
- MIGRATE_ISOLATE, /* can't allocate from here */
-#endif
- MIGRATE_TYPES
-};
-
-/* In mm/page_alloc.c; keep in sync also with show_migration_types() there */
-extern const char * const migratetype_names[MIGRATE_TYPES];
-
-#ifdef CONFIG_CMA
-# define is_migrate_cma(migratetype) unlikely((migratetype) == MIGRATE_CMA)
# define is_migrate_cma_page(_page) (get_pageblock_migratetype(_page) == MIGRATE_CMA)
/*
* __dump_folio() in mm/debug.c passes a folio pointer to on-stack struct folio,
@@ -157,27 +125,10 @@ extern const char * const migratetype_names[MIGRATE_TYPES];
# define is_migrate_cma_folio(folio, pfn) \
(get_pfnblock_migratetype(&folio->page, pfn) == MIGRATE_CMA)
#else
-# define is_migrate_cma(migratetype) false
# define is_migrate_cma_page(_page) false
# define is_migrate_cma_folio(folio, pfn) false
#endif
-static inline bool is_migrate_movable(int mt)
-{
- return is_migrate_cma(mt) || mt == MIGRATE_MOVABLE;
-}
-
-/*
- * Check whether a migratetype can be merged with another migratetype.
- *
- * It is only mergeable when it can fall back to other migratetypes for
- * allocation. See fallbacks[MIGRATE_TYPES][3] in page_alloc.c.
- */
-static inline bool migratetype_is_mergeable(int mt)
-{
- return mt < MIGRATE_PCPTYPES;
-}
-
#define for_each_free_list(list, zone, order) \
for (order = 0; order < NR_PAGE_ORDERS; order++) \
for (unsigned int __idx = 0; \
@@ -185,30 +136,6 @@ static inline bool migratetype_is_mergeable(int mt)
(list = &(zone)->free_area[order].free_list[__idx], 1); \
__idx++)
-static inline freetype_t migrate_to_freetype(enum migratetype mt,
- unsigned int flags)
-{
- freetype_t freetype;
-
- /* No flags supported yet. */
- VM_WARN_ON_ONCE(flags);
-
- freetype.migratetype = mt;
- return freetype;
-}
-
-static inline enum migratetype free_to_migratetype(freetype_t freetype)
-{
- return freetype.migratetype;
-}
-
-/* Convenience helper, return the freetype modified to have the migratetype. */
-static inline freetype_t freetype_with_migrate(freetype_t freetype,
- enum migratetype migratetype)
-{
- return migrate_to_freetype(migratetype, freetype_flags(freetype));
-}
-
extern int page_group_by_mobility_disabled;
freetype_t get_pfnblock_freetype(const struct page *page, unsigned long pfn);
--
2.54.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v3 15/26] mm/page_alloc: add support for freetypes with no freelist
2026-07-26 22:22 [PATCH v3 00/26] mm: Add ALLOC_UNMAPPED and AS_NO_DIRECT_MAP Brendan Jackman
` (13 preceding siblings ...)
2026-07-26 22:22 ` [PATCH v3 14/26] mm: move migratetype definitions to freetype.h Brendan Jackman
@ 2026-07-26 22:22 ` Brendan Jackman
2026-07-26 22:22 ` [PATCH v3 16/26] mm: add definitions for allocating unmapped pages Brendan Jackman
` (10 subsequent siblings)
25 siblings, 0 replies; 29+ messages in thread
From: Brendan Jackman @ 2026-07-26 22:22 UTC (permalink / raw)
To: Borislav Petkov, Dave Hansen, Peter Zijlstra, Andrew Morton,
David Hildenbrand, Vlastimil Babka, Mike Rapoport, Wei Xu,
Johannes Weiner, Zi Yan, Lorenzo Stoakes
Cc: linux-mm, linux-kernel, x86, rppt, Sumit Garg, Will Deacon,
rientjes, Kalyazin, Nikita, patrick.roy, Itazuri, Takahiro,
Andy Lutomirski, David Kaplan, Thomas Gleixner, Yosry Ahmed,
Patrick Bellasi, Reiji Watanabe, Sean Christopherson,
Brendan Jackman
Freetypes are currently just a wrapper for migratetypes, but in a later
patch they will gain flags. The theoretically possible space of
freetypes will thus grow, but many freetypes will not actually be used
in practice.
To avoid bloating structures that contain arrays of freelists, mark
freetype_idx() as possibly returning -1, and add handling of this case
to code paths that will look up freelists by freetype (unless those
paths "know" that they're operating on a freetype that has a
valid freetype_idx).
It never actually returns -1 yet, this is just introduced as a separate
patch to make review easier.
Signed-off-by: Brendan Jackman <jackmanb@google.com>
---
include/linux/freetype.h | 3 ++-
mm/page_alloc.c | 13 +++++++++++--
mm/page_isolation.c | 6 ++++++
3 files changed, 19 insertions(+), 3 deletions(-)
diff --git a/include/linux/freetype.h b/include/linux/freetype.h
index b16dcd58e184c..3b0d44023b6a1 100644
--- a/include/linux/freetype.h
+++ b/include/linux/freetype.h
@@ -73,7 +73,8 @@ typedef struct {
} freetype_t;
/*
- * Return a dense linear index for freetypes.
+ * Return a dense linear index for freetypes that have lists in the free area.
+ * Return -1 for other freetypes.
*/
static inline int freetype_idx(freetype_t freetype)
{
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 60aedd0e78b54..4965019dc2286 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1963,6 +1963,9 @@ struct page *__rmqueue_smallest(struct zone *zone, unsigned int order,
struct free_area *area;
struct page *page;
+ if (freetype_idx(freetype) < 0)
+ return NULL;
+
/* Find a page of the appropriate size in the preferred list */
for (current_order = order; current_order < NR_PAGE_ORDERS; ++current_order) {
enum migratetype migratetype = free_to_migratetype(freetype);
@@ -2406,6 +2409,9 @@ find_suitable_fallback(struct free_area *area, unsigned int order,
*/
freetype_t fallback_ft = freetype_with_migrate(freetype, fallback_mt);
+ if (freetype_idx(fallback_ft) < 0)
+ continue;
+
if (!free_area_empty(area, fallback_ft)) {
if (ft_out)
*ft_out = fallback_ft;
@@ -3642,6 +3648,8 @@ static void reserve_highatomic_pageblock(struct page *page, int order,
static bool unreserve_highatomic_pageblock(const struct alloc_context *ac,
bool force)
{
+ freetype_t ft_high = freetype_with_migrate(ac->freetype,
+ MIGRATE_HIGHATOMIC);
struct zonelist *zonelist = ac->zonelist;
struct zoneref *z;
struct zone *zone;
@@ -3649,6 +3657,9 @@ static bool unreserve_highatomic_pageblock(const struct alloc_context *ac,
int order;
int ret;
+ if (freetype_idx(ft_high) < 0)
+ return false;
+
for_each_zone_zonelist_nodemask(zone, z, zonelist, ac->highest_zoneidx,
ac->nodemask) {
/*
@@ -3662,8 +3673,6 @@ static bool unreserve_highatomic_pageblock(const struct alloc_context *ac,
guard(spinlock_irqsave)(&zone->lock);
for (order = 0; order < NR_PAGE_ORDERS; order++) {
struct free_area *area = &(zone->free_area[order]);
- freetype_t ft_high = freetype_with_migrate(ac->freetype,
- MIGRATE_HIGHATOMIC);
unsigned long size;
page = get_page_from_free_area(area, ft_high);
diff --git a/mm/page_isolation.c b/mm/page_isolation.c
index f81186431d3c3..e6bd227fbf1db 100644
--- a/mm/page_isolation.c
+++ b/mm/page_isolation.c
@@ -176,6 +176,7 @@ static int set_migratetype_isolate(struct page *page, enum pb_isolate_mode mode,
struct zone *zone = page_zone(page);
struct page *unmovable;
unsigned long check_unmovable_start, check_unmovable_end;
+ freetype_t ft_isolate;
if (PageUnaccepted(page))
accept_page(page);
@@ -189,6 +190,11 @@ static int set_migratetype_isolate(struct page *page, enum pb_isolate_mode mode,
if (is_migrate_isolate_page(page))
return -EBUSY;
+ /* Do not isolate unsupported freetypes. */
+ ft_isolate = freetype_with_migrate(get_pageblock_freetype(page), MIGRATE_ISOLATE);
+ if (freetype_idx(ft_isolate) < 0)
+ return -EINVAL;
+
/*
* FIXME: Now, memory hotplug doesn't call shrink_slab() by
* itself. We just check MOVABLE pages.
--
2.54.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v3 16/26] mm: add definitions for allocating unmapped pages
2026-07-26 22:22 [PATCH v3 00/26] mm: Add ALLOC_UNMAPPED and AS_NO_DIRECT_MAP Brendan Jackman
` (14 preceding siblings ...)
2026-07-26 22:22 ` [PATCH v3 15/26] mm/page_alloc: add support for freetypes with no freelist Brendan Jackman
@ 2026-07-26 22:22 ` Brendan Jackman
2026-07-26 22:22 ` [PATCH v3 17/26] mm: encode freetype flags in pageblock flags Brendan Jackman
` (9 subsequent siblings)
25 siblings, 0 replies; 29+ messages in thread
From: Brendan Jackman @ 2026-07-26 22:22 UTC (permalink / raw)
To: Borislav Petkov, Dave Hansen, Peter Zijlstra, Andrew Morton,
David Hildenbrand, Vlastimil Babka, Mike Rapoport, Wei Xu,
Johannes Weiner, Zi Yan, Lorenzo Stoakes
Cc: linux-mm, linux-kernel, x86, rppt, Sumit Garg, Will Deacon,
rientjes, Kalyazin, Nikita, patrick.roy, Itazuri, Takahiro,
Andy Lutomirski, David Kaplan, Thomas Gleixner, Yosry Ahmed,
Patrick Bellasi, Reiji Watanabe, Sean Christopherson,
Brendan Jackman
Create ALLOC_UNMAPPED, which requests pages that are not present in the
direct map. Since this feature has a cost (e.g. more freelists), it's
behind a kconfig. Unlike other conditionally-defined alloc flags, it
doesn't fall back to being 0. This prevents building code that uses
ALLOC_UNMAPPED but doesn't depend on the necessary kconfig, since that
would lead to invisible security issues.
Create a freetype flag to record that pages on the freelists with this
flag are unmapped. This is currently only needed for MIGRATE_UNMOVABLE
pages, so the freetype encoding remains trivial.
Also create the corresponding pageblock flag to record the same thing.
To keep patches from being too overwhelming, the actual implementation
is added separately, this is just types, Kconfig boilerplate, etc.
Acked-by: Vlastimil Babka (SUSE) <vbabka@kernel.org>
Signed-off-by: Brendan Jackman <jackmanb@google.com>
---
include/linux/freetype.h | 70 ++++++++++++++++++++++++++++++++++++++++--------
mm/Kconfig | 3 +++
mm/page_alloc.h | 18 +++++++++++++
3 files changed, 80 insertions(+), 11 deletions(-)
diff --git a/include/linux/freetype.h b/include/linux/freetype.h
index 3b0d44023b6a1..37e88dcccdecc 100644
--- a/include/linux/freetype.h
+++ b/include/linux/freetype.h
@@ -2,6 +2,7 @@
#ifndef _LINUX_FREETYPE_H
#define _LINUX_FREETYPE_H
+#include <linux/log2.h>
#include <linux/types.h>
#include <linux/mmdebug.h>
@@ -64,20 +65,47 @@ static inline bool migratetype_is_mergeable(int mt)
return mt < MIGRATE_PCPTYPES;
}
+enum {
+ /* Defined unconditionally as a hack to avoid a zero-width bitfield. */
+ FREETYPE_UNMAPPED_BIT,
+ NUM_FREETYPE_FLAGS,
+};
+
/*
* A freetype is the identifier for a page freelist. This consists of a
* migratetype, and other bits which encode orthogonal properties of memory.
*/
typedef struct {
- int migratetype;
+ unsigned int migratetype : order_base_2(MIGRATE_TYPES);
+ unsigned int flags : NUM_FREETYPE_FLAGS;
} freetype_t;
+#ifdef CONFIG_PAGE_ALLOC_UNMAPPED
+#define FREETYPE_UNMAPPED BIT(FREETYPE_UNMAPPED_BIT)
+#define NUM_UNMAPPED_FREETYPES 1
+#else
+#define FREETYPE_UNMAPPED 0
+#define NUM_UNMAPPED_FREETYPES 0
+#endif
+
+#define FREETYPE_FLAGS_MASK FREETYPE_UNMAPPED
+
/*
* Return a dense linear index for freetypes that have lists in the free area.
* Return -1 for other freetypes.
*/
static inline int freetype_idx(freetype_t freetype)
{
+ /* For FREETYPE_UNMAPPED, only MIGRATE_UNMOVABLE has an index. */
+ if (freetype.flags & FREETYPE_UNMAPPED) {
+ VM_WARN_ON_ONCE(freetype.flags & ~FREETYPE_UNMAPPED);
+ if (freetype.migratetype != MIGRATE_UNMOVABLE)
+ return -1;
+ return MIGRATE_TYPES;
+ }
+ /* No other flags are supported. */
+ VM_WARN_ON_ONCE(freetype.flags);
+
return freetype.migratetype;
}
@@ -85,33 +113,53 @@ static inline freetype_t freetype_from_idx(unsigned int idx)
{
freetype_t freetype;
- freetype.migratetype = idx;
+ if (idx == MIGRATE_TYPES) {
+ freetype.flags = FREETYPE_UNMAPPED;
+ freetype.migratetype = MIGRATE_UNMOVABLE;
+ } else {
+ VM_WARN_ON_ONCE(idx < 0 || idx > MIGRATE_TYPES);
+ freetype.flags = 0;
+ freetype.migratetype = idx;
+ }
return freetype;
}
-/* No freetype flags actually exist yet. */
-#define NR_FREETYPE_IDXS MIGRATE_TYPES
+/* One for each migratetype, plus one for MIGRATE_UNMOVABLE-FREETYPE_UNMAPPED */
+#define NR_FREETYPE_IDXS (MIGRATE_TYPES + NUM_UNMAPPED_FREETYPES)
static inline unsigned int freetype_flags(freetype_t freetype)
{
- /* No flags supported yet. */
- return 0;
+ return freetype.flags;
}
+#ifdef CONFIG_PAGE_ALLOC_UNMAPPED
+static inline bool freetype_unmapped(freetype_t freetype)
+{
+ return !!(freetype.flags & FREETYPE_UNMAPPED);
+}
+#else
+static inline bool freetype_unmapped(freetype_t freetype)
+{
+ return false;
+}
+#endif
+
static inline bool freetypes_equal(freetype_t a, freetype_t b)
{
- return a.migratetype == b.migratetype;
+ return a.migratetype == b.migratetype && a.flags == b.flags;
}
static inline freetype_t migrate_to_freetype(enum migratetype mt,
unsigned int flags)
{
- freetype_t freetype;
+ freetype_t freetype = {
+ .migratetype = mt,
+ .flags = flags,
+ };
- /* No flags supported yet. */
- VM_WARN_ON_ONCE(flags);
+ VM_WARN_ON_ONCE(flags & ~FREETYPE_FLAGS_MASK);
+ VM_WARN_ON_ONCE(mt >= MIGRATE_TYPES);
- freetype.migratetype = mt;
return freetype;
}
diff --git a/mm/Kconfig b/mm/Kconfig
index adf71788971a2..aa5f041ce1a67 100644
--- a/mm/Kconfig
+++ b/mm/Kconfig
@@ -1530,3 +1530,6 @@ config MERMAP_KUNIT_TEST
source "mm/damon/Kconfig"
endmenu
+
+config PAGE_ALLOC_UNMAPPED
+ bool
diff --git a/mm/page_alloc.h b/mm/page_alloc.h
index 9928aa9012588..fac8e5304bb03 100644
--- a/mm/page_alloc.h
+++ b/mm/page_alloc.h
@@ -56,6 +56,24 @@
* alloc_tag_sub_check().
*/
#define ALLOC_NO_CODETAG 0x1000
+#ifdef CONFIG_PAGE_ALLOC_UNMAPPED
+/*
+ * Allocate pages that aren't present in the direct map. If the caller changes
+ * direct map presence, it must be restored to the previous state before freeing
+ * the page. (This is true regardless of ALLOC_UNMAPPED).
+ *
+ * This uses the mermap (when __GFP_ZERO), so it's only valid to allocate with
+ * this flag where that's valid, namely from process context after the mermap
+ * has been initialised for that process. This also means that the allocator
+ * leaves behind stale TLB entries in the mermap region. The caller is
+ * responsible for ensuring they are flushed as needed.
+ *
+ * This is currently incompatible with __GFP_MOVABLE and __GFP_RECLAIMABLE, but
+ * only because of allocator implementation details, if a usecase arises this
+ * restriction could be dropped.
+ */
+#define ALLOC_UNMAPPED 0x2000
+#endif
/* Flags that allow allocations below the min watermark. */
#define ALLOC_RESERVES (ALLOC_NON_BLOCK|ALLOC_MIN_RESERVE|ALLOC_HIGHATOMIC|ALLOC_OOM)
--
2.54.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v3 17/26] mm: encode freetype flags in pageblock flags
2026-07-26 22:22 [PATCH v3 00/26] mm: Add ALLOC_UNMAPPED and AS_NO_DIRECT_MAP Brendan Jackman
` (15 preceding siblings ...)
2026-07-26 22:22 ` [PATCH v3 16/26] mm: add definitions for allocating unmapped pages Brendan Jackman
@ 2026-07-26 22:22 ` Brendan Jackman
2026-07-26 22:22 ` [PATCH v3 18/26] mm/page_alloc: separate pcplists by freetype flags Brendan Jackman
` (8 subsequent siblings)
25 siblings, 0 replies; 29+ messages in thread
From: Brendan Jackman @ 2026-07-26 22:22 UTC (permalink / raw)
To: Borislav Petkov, Dave Hansen, Peter Zijlstra, Andrew Morton,
David Hildenbrand, Vlastimil Babka, Mike Rapoport, Wei Xu,
Johannes Weiner, Zi Yan, Lorenzo Stoakes
Cc: linux-mm, linux-kernel, x86, rppt, Sumit Garg, Will Deacon,
rientjes, Kalyazin, Nikita, patrick.roy, Itazuri, Takahiro,
Andy Lutomirski, David Kaplan, Thomas Gleixner, Yosry Ahmed,
Patrick Bellasi, Reiji Watanabe, Sean Christopherson,
Brendan Jackman
In preparation for implementing allocation from FREETYPE_UNMAPPED lists.
Since it works nicely with the existing allocator logic, and also offers
a simple way to amortize TLB flushing costs, ALLOC_UNMAPPED will be
implemented by changing mappings at pageblock granularity. Therefore,
encode the mapping state in the pageblock flags.
Also add the necessary logic to record this from a freetype, and
reconstruct a freetype from the pageblock flags.
Acked-by: Vlastimil Babka (SUSE) <vbabka@kernel.org>
Signed-off-by: Brendan Jackman <jackmanb@google.com>
---
include/linux/pageblock-flags.h | 10 ++++++++++
mm/page_alloc.c | 33 ++++++++++++++++++++-------------
2 files changed, 30 insertions(+), 13 deletions(-)
diff --git a/include/linux/pageblock-flags.h b/include/linux/pageblock-flags.h
index 9a6c3ea17684d..b634280050071 100644
--- a/include/linux/pageblock-flags.h
+++ b/include/linux/pageblock-flags.h
@@ -11,6 +11,8 @@
#ifndef PAGEBLOCK_FLAGS_H
#define PAGEBLOCK_FLAGS_H
+#include <linux/freetype.h>
+#include <linux/log2.h>
#include <linux/types.h>
/* Bit indices that affect a whole block of pages */
@@ -18,6 +20,9 @@ enum pageblock_bits {
PB_migrate_0,
PB_migrate_1,
PB_migrate_2,
+ PB_freetype_flags,
+ PB_freetype_flags_end = PB_freetype_flags + NUM_FREETYPE_FLAGS - 1,
+
PB_compact_skip,/* If set the block is skipped by compaction */
#ifdef CONFIG_MEMORY_ISOLATION
@@ -37,6 +42,7 @@ enum pageblock_bits {
#define NR_PAGEBLOCK_BITS (roundup_pow_of_two(__NR_PAGEBLOCK_BITS))
#define PAGEBLOCK_MIGRATETYPE_MASK (BIT(PB_migrate_0)|BIT(PB_migrate_1)|BIT(PB_migrate_2))
+#define PAGEBLOCK_FREETYPE_FLAGS_MASK (((1 << NUM_FREETYPE_FLAGS) - 1) << PB_freetype_flags)
#ifdef CONFIG_MEMORY_ISOLATION
#define PAGEBLOCK_ISO_MASK BIT(PB_migrate_isolate)
@@ -44,6 +50,10 @@ enum pageblock_bits {
#define PAGEBLOCK_ISO_MASK 0
#endif
+#define PAGEBLOCK_FREETYPE_MASK (PAGEBLOCK_MIGRATETYPE_MASK | \
+ PAGEBLOCK_ISO_MASK | \
+ PAGEBLOCK_FREETYPE_FLAGS_MASK)
+
#if defined(CONFIG_HUGETLB_PAGE)
#ifdef CONFIG_HUGETLB_PAGE_SIZE_VARIABLE
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 4965019dc2286..33b7d354e1e03 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -363,11 +363,8 @@ get_pfnblock_bitmap_bitidx(const struct page *page, unsigned long pfn,
unsigned long *bitmap;
unsigned long word_bitidx;
-#ifdef CONFIG_MEMORY_ISOLATION
- BUILD_BUG_ON(NR_PAGEBLOCK_BITS != 8);
-#else
- BUILD_BUG_ON(NR_PAGEBLOCK_BITS != 4);
-#endif
+ /* NR_PAGEBLOCK_BITS must divide word size. */
+ BUILD_BUG_ON(NR_PAGEBLOCK_BITS != 4 && NR_PAGEBLOCK_BITS != 8);
BUILD_BUG_ON(__MIGRATE_TYPE_END > PAGEBLOCK_MIGRATETYPE_MASK);
VM_BUG_ON_PAGE(!zone_spans_pfn(page_zone(page), pfn), page);
@@ -440,19 +437,20 @@ static __always_inline freetype_t
__get_pfnblock_freetype(const struct page *page, unsigned long pfn,
bool ignore_iso)
{
- unsigned long mask = PAGEBLOCK_MIGRATETYPE_MASK | PAGEBLOCK_ISO_MASK;
+ unsigned long mask = PAGEBLOCK_FREETYPE_MASK;
+ enum migratetype migratetype;
+ unsigned int ft_flags;
unsigned long flags;
- enum migratetype mt;
flags = __get_pfnblock_flags_mask(page, pfn, mask);
- mt = flags & PAGEBLOCK_MIGRATETYPE_MASK;
+ ft_flags = (flags & PAGEBLOCK_FREETYPE_FLAGS_MASK) >> PB_freetype_flags;
+ migratetype = flags & PAGEBLOCK_MIGRATETYPE_MASK;
#ifdef CONFIG_MEMORY_ISOLATION
- if (!ignore_iso && flags & BIT(PB_migrate_isolate))
- mt = MIGRATE_ISOLATE;
+ if (!ignore_iso && (flags & BIT(PB_migrate_isolate)))
+ migratetype = MIGRATE_ISOLATE;
#endif
-
- return migrate_to_freetype(mt, 0);
+ return migrate_to_freetype(migratetype, ft_flags);
}
/**
@@ -576,6 +574,15 @@ static void set_pageblock_migratetype(struct page *page,
PAGEBLOCK_MIGRATETYPE_MASK | PAGEBLOCK_ISO_MASK);
}
+static inline void set_pageblock_freetype_flags(struct page *page,
+ unsigned int ft_flags)
+{
+ unsigned int flags = ft_flags << PB_freetype_flags;
+
+ __set_pfnblock_flags_mask(page, page_to_pfn(page), flags,
+ PAGEBLOCK_FREETYPE_FLAGS_MASK);
+}
+
void __meminit init_pageblock_migratetype(struct page *page,
enum migratetype migratetype,
bool isolate)
@@ -599,7 +606,7 @@ void __meminit init_pageblock_migratetype(struct page *page,
flags |= BIT(PB_migrate_isolate);
#endif
__set_pfnblock_flags_mask(page, page_to_pfn(page), flags,
- PAGEBLOCK_MIGRATETYPE_MASK | PAGEBLOCK_ISO_MASK);
+ PAGEBLOCK_FREETYPE_MASK);
}
#ifdef CONFIG_DEBUG_VM
--
2.54.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v3 18/26] mm/page_alloc: separate pcplists by freetype flags
2026-07-26 22:22 [PATCH v3 00/26] mm: Add ALLOC_UNMAPPED and AS_NO_DIRECT_MAP Brendan Jackman
` (16 preceding siblings ...)
2026-07-26 22:22 ` [PATCH v3 17/26] mm: encode freetype flags in pageblock flags Brendan Jackman
@ 2026-07-26 22:22 ` Brendan Jackman
2026-07-26 22:22 ` [PATCH v3 19/26] mm/page_alloc: rename ALLOC_NON_BLOCK back to _HARDER Brendan Jackman
` (7 subsequent siblings)
25 siblings, 0 replies; 29+ messages in thread
From: Brendan Jackman @ 2026-07-26 22:22 UTC (permalink / raw)
To: Borislav Petkov, Dave Hansen, Peter Zijlstra, Andrew Morton,
David Hildenbrand, Vlastimil Babka, Mike Rapoport, Wei Xu,
Johannes Weiner, Zi Yan, Lorenzo Stoakes
Cc: linux-mm, linux-kernel, x86, rppt, Sumit Garg, Will Deacon,
rientjes, Kalyazin, Nikita, patrick.roy, Itazuri, Takahiro,
Andy Lutomirski, David Kaplan, Thomas Gleixner, Yosry Ahmed,
Patrick Bellasi, Reiji Watanabe, Sean Christopherson,
Brendan Jackman
The normal freelists are already separated by this flag, so now update
the pcplists accordingly. This follows the most "obvious" design where
ALLOC_UNMAPPED is supported at arbitrary orders.
If necessary, it would be possible to avoid the proliferation of
pcplists by restricting orders that can be allocated from them with this
FREETYPE_UNMAPPED.
On the other hand, there's currently no usecase for movable/reclaimable
unmapped memory, and constraining the migratetype doesn't have any
tricky plumbing implications. So, take advantage of that and assume that
FREETYPE_UNMAPPED implies MIGRATE_UNMOVABLE.
Overall, this just takes the existing space of pindices and tacks
another bank on the end. For !THP this is just 4 more lists, with THP
there is a single additional list for hugepages.
Reviewed-by: Vlastimil Babka (SUSE) <vbabka@kernel.org>
Signed-off-by: Brendan Jackman <jackmanb@google.com>
---
include/linux/mmzone.h | 11 ++++++++++-
mm/page_alloc.c | 37 ++++++++++++++++++++++++++++++++-----
2 files changed, 42 insertions(+), 6 deletions(-)
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 78455285dda7c..2533a228f71fd 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -789,8 +789,17 @@ enum zone_watermarks {
#else
#define NR_PCP_THP 0
#endif
+/*
+ * FREETYPE_UNMAPPED can currently only be used with MIGRATE_UNMOVABLE, so for
+ * those there's no need to encode the migratetype in the pindex.
+ */
+#ifdef CONFIG_PAGE_ALLOC_UNMAPPED
+#define NR_UNMAPPED_PCP_LISTS (PAGE_ALLOC_COSTLY_ORDER + 1 + !!NR_PCP_THP)
+#else
+#define NR_UNMAPPED_PCP_LISTS 0
+#endif
#define NR_LOWORDER_PCP_LISTS (MIGRATE_PCPTYPES * (PAGE_ALLOC_COSTLY_ORDER + 1))
-#define NR_PCP_LISTS (NR_LOWORDER_PCP_LISTS + NR_PCP_THP)
+#define NR_PCP_LISTS (NR_LOWORDER_PCP_LISTS + NR_PCP_THP + NR_UNMAPPED_PCP_LISTS)
/*
* Flags used in pcp->flags field.
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 33b7d354e1e03..30ba61c20209c 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -690,8 +690,25 @@ static void bad_page(struct page *page, const char *reason)
add_taint(TAINT_BAD_PAGE, LOCKDEP_NOW_UNRELIABLE);
}
-static inline unsigned int order_to_pindex(int migratetype, int order)
+static inline unsigned int order_to_pindex(freetype_t freetype, int order)
{
+ int migratetype = free_to_migratetype(freetype);
+
+ VM_BUG_ON(migratetype >= MIGRATE_PCPTYPES);
+ VM_BUG_ON(order > PAGE_ALLOC_COSTLY_ORDER &&
+ (!IS_ENABLED(CONFIG_TRANSPARENT_HUGEPAGE) || order != HPAGE_PMD_ORDER));
+
+ /* FREETYPE_UNMAPPED currently always means MIGRATE_UNMOVABLE. */
+ if (freetype_flags(freetype) & FREETYPE_UNMAPPED) {
+ int order_offset = order;
+
+ VM_BUG_ON(migratetype != MIGRATE_UNMOVABLE);
+ if (order > PAGE_ALLOC_COSTLY_ORDER)
+ order_offset = PAGE_ALLOC_COSTLY_ORDER + 1;
+
+ return NR_LOWORDER_PCP_LISTS + NR_PCP_THP + order_offset;
+ }
+
if (IS_ENABLED(CONFIG_TRANSPARENT_HUGEPAGE)) {
bool movable = migratetype == MIGRATE_MOVABLE;
@@ -704,8 +721,18 @@ static inline unsigned int order_to_pindex(int migratetype, int order)
static inline int pindex_to_order(unsigned int pindex)
{
- int order = pindex / MIGRATE_PCPTYPES;
+ unsigned int unmapped_base = NR_LOWORDER_PCP_LISTS + NR_PCP_THP;
+ int order;
+ if (pindex >= unmapped_base) {
+ order = pindex - unmapped_base;
+ if (IS_ENABLED(CONFIG_TRANSPARENT_HUGEPAGE) &&
+ order > PAGE_ALLOC_COSTLY_ORDER)
+ return HPAGE_PMD_ORDER;
+ return order;
+ }
+
+ order = pindex / MIGRATE_PCPTYPES;
if (IS_ENABLED(CONFIG_TRANSPARENT_HUGEPAGE)) {
if (pindex >= NR_LOWORDER_PCP_LISTS)
order = HPAGE_PMD_ORDER;
@@ -3007,7 +3034,7 @@ static bool free_frozen_page_commit(struct zone *zone,
*/
pcp->alloc_factor >>= 1;
__count_vm_events(PGFREE, 1 << order);
- pindex = order_to_pindex(free_to_migratetype(freetype), order);
+ pindex = order_to_pindex(freetype, order);
list_add(&page->pcp_list, &pcp->lists[pindex]);
pcp->count += 1 << order;
@@ -3545,7 +3572,7 @@ static struct page *rmqueue_pcplist(struct zone *preferred_zone,
* frees.
*/
pcp->free_count >>= 1;
- list = &pcp->lists[order_to_pindex(free_to_migratetype(freetype), order)];
+ list = &pcp->lists[order_to_pindex(freetype, order)];
page = __rmqueue_pcplist(zone, order, freetype, alloc_flags, pcp, list);
pcp_spin_unlock(pcp);
if (page) {
@@ -5428,7 +5455,7 @@ unsigned long alloc_pages_bulk_noprof(gfp_t gfp, int preferred_nid,
goto failed;
/* Attempt the batch allocation */
- pcp_list = &pcp->lists[order_to_pindex(free_to_migratetype(ac.freetype), 0)];
+ pcp_list = &pcp->lists[order_to_pindex(ac.freetype, 0)];
while (nr_populated < nr_pages) {
/* Skip existing pages */
--
2.54.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v3 19/26] mm/page_alloc: rename ALLOC_NON_BLOCK back to _HARDER
2026-07-26 22:22 [PATCH v3 00/26] mm: Add ALLOC_UNMAPPED and AS_NO_DIRECT_MAP Brendan Jackman
` (17 preceding siblings ...)
2026-07-26 22:22 ` [PATCH v3 18/26] mm/page_alloc: separate pcplists by freetype flags Brendan Jackman
@ 2026-07-26 22:22 ` Brendan Jackman
2026-07-26 22:22 ` [PATCH v3 20/26] mm/page_alloc: introduce ALLOC_NOBLOCK Brendan Jackman
` (6 subsequent siblings)
25 siblings, 0 replies; 29+ messages in thread
From: Brendan Jackman @ 2026-07-26 22:22 UTC (permalink / raw)
To: Borislav Petkov, Dave Hansen, Peter Zijlstra, Andrew Morton,
David Hildenbrand, Vlastimil Babka, Mike Rapoport, Wei Xu,
Johannes Weiner, Zi Yan, Lorenzo Stoakes
Cc: linux-mm, linux-kernel, x86, rppt, Sumit Garg, Will Deacon,
rientjes, Kalyazin, Nikita, patrick.roy, Itazuri, Takahiro,
Andy Lutomirski, David Kaplan, Thomas Gleixner, Yosry Ahmed,
Patrick Bellasi, Reiji Watanabe, Sean Christopherson,
Brendan Jackman
Commit 1ebbb21811b7 ("mm/page_alloc: explicitly define how __GFP_HIGH
non-blocking allocations accesses reserves") renamed ALLOC_HARDER to
ALLOC_NON_BLOCK because the former is "a vague description".
However, vagueness is accurate here, this is a vague flag. It is not set
for __GFP_NOMEMALLOC. It doesn't really mean "allocate without blocking"
but rather "allow dipping into atomic reserves, _because_ of the need
not to block".
A later commit will need an alloc flag that really means "don't block
here", so go back to the flag's old name and update the commentary
to try and give it a slightly clearer meaning.
Signed-off-by: Brendan Jackman <jackmanb@google.com>
---
mm/page_alloc.c | 8 ++++----
mm/page_alloc.h | 9 +++++----
2 files changed, 9 insertions(+), 8 deletions(-)
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 30ba61c20209c..fb522a09f2e60 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -3431,7 +3431,7 @@ struct page *rmqueue_buddy(struct zone *preferred_zone, struct zone *zone,
* reserves as failing now is worse than failing a
* high-order atomic allocation in the future.
*/
- if (!page && (alloc_flags & (ALLOC_OOM|ALLOC_NON_BLOCK)))
+ if (!page && (alloc_flags & (ALLOC_OOM|ALLOC_HARDER)))
page = __rmqueue_smallest(zone, order, ft_high);
if (!page) {
@@ -3811,7 +3811,7 @@ bool __zone_watermark_ok(struct zone *z, unsigned int order, unsigned long mark,
* or (GFP_KERNEL & ~__GFP_DIRECT_RECLAIM) do not get
* access to the min reserve.
*/
- if (alloc_flags & ALLOC_NON_BLOCK)
+ if (alloc_flags & ALLOC_HARDER)
min -= min / 4;
}
@@ -4733,7 +4733,7 @@ alloc_flags_nonblocking(gfp_t gfp_mask, unsigned int order)
if (gfp_mask & __GFP_NOMEMALLOC)
return 0;
- alloc_flags |= ALLOC_NON_BLOCK;
+ alloc_flags |= ALLOC_HARDER;
if (order > 0 && (gfp_mask & __GFP_HIGH))
alloc_flags |= ALLOC_HIGHATOMIC;
@@ -4750,7 +4750,7 @@ alloc_flags_slowpath(gfp_t gfp_mask, unsigned int order)
* The caller may dip into page reserves a bit more if the caller
* cannot run direct reclaim, or if the caller has realtime scheduling
* policy or is asking for __GFP_HIGH memory. GFP_ATOMIC requests will
- * set both ALLOC_NON_BLOCK and ALLOC_MIN_RESERVE(__GFP_HIGH).
+ * set both ALLOC_HARDER and ALLOC_MIN_RESERVE(__GFP_HIGH).
*/
if (gfp_mask & __GFP_HIGH)
alloc_flags |= ALLOC_MIN_RESERVE;
diff --git a/mm/page_alloc.h b/mm/page_alloc.h
index fac8e5304bb03..2307b173a459f 100644
--- a/mm/page_alloc.h
+++ b/mm/page_alloc.h
@@ -32,9 +32,10 @@
#define ALLOC_OOM ALLOC_NO_WATERMARKS
#endif
-#define ALLOC_NON_BLOCK 0x10 /* Caller cannot block. Allow access
- * to 25% of the min watermark or
- * 62.5% if __GFP_HIGH is set.
+#define ALLOC_HARDER 0x10 /* Because the caller cannot block,
+ * allow access to 25% of the min
+ * watermark or 62.5% if __GFP_HIGH is
+ * set.
*/
#define ALLOC_MIN_RESERVE 0x20 /* __GFP_HIGH set. Allow access to 50%
* of the min watermark.
@@ -76,7 +77,7 @@
#endif
/* Flags that allow allocations below the min watermark. */
-#define ALLOC_RESERVES (ALLOC_NON_BLOCK|ALLOC_MIN_RESERVE|ALLOC_HIGHATOMIC|ALLOC_OOM)
+#define ALLOC_RESERVES (ALLOC_HARDER|ALLOC_MIN_RESERVE|ALLOC_HIGHATOMIC|ALLOC_OOM)
/*
* Structure for holding the mostly immutable allocation parameters passed
--
2.54.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v3 20/26] mm/page_alloc: introduce ALLOC_NOBLOCK
2026-07-26 22:22 [PATCH v3 00/26] mm: Add ALLOC_UNMAPPED and AS_NO_DIRECT_MAP Brendan Jackman
` (18 preceding siblings ...)
2026-07-26 22:22 ` [PATCH v3 19/26] mm/page_alloc: rename ALLOC_NON_BLOCK back to _HARDER Brendan Jackman
@ 2026-07-26 22:22 ` Brendan Jackman
2026-07-26 22:22 ` [PATCH v3 21/26] mm/page_alloc: implement FREETYPE_UNMAPPED allocations Brendan Jackman
` (5 subsequent siblings)
25 siblings, 0 replies; 29+ messages in thread
From: Brendan Jackman @ 2026-07-26 22:22 UTC (permalink / raw)
To: Borislav Petkov, Dave Hansen, Peter Zijlstra, Andrew Morton,
David Hildenbrand, Vlastimil Babka, Mike Rapoport, Wei Xu,
Johannes Weiner, Zi Yan, Lorenzo Stoakes
Cc: linux-mm, linux-kernel, x86, rppt, Sumit Garg, Will Deacon,
rientjes, Kalyazin, Nikita, patrick.roy, Itazuri, Takahiro,
Andy Lutomirski, David Kaplan, Thomas Gleixner, Yosry Ahmed,
Patrick Bellasi, Reiji Watanabe, Sean Christopherson,
Brendan Jackman
This flag is set unless we can be sure the caller isn't in an atomic
context.
The allocator will soon start needing to call set_direct_map_* APIs
which cannot be called with IRQs off. It will need to do this even
before direct reclaim is possible.
Despite the fact that, in principle, ALLOC_NOBLOCK is distinct from
__GFP_DIRECT_RECLAIM, in order to avoid introducing a GFP flag, just
infer the former based on whether the caller set the latter. This means
that, in practice, ALLOC_NOBLOCK is just !__GFP_DIRECT_RECLAIM, except
that it is not influenced by gfp_allowed_mask. This requires some rather
ugly plumbing to get it into the slowpath without clobbering it.
Call it ALLOC_NOBLOCK in order to try and mitigate confusion vs the
recently-removed ALLOC_NON_BLOCK, which meant something different.
Signed-off-by: Brendan Jackman <jackmanb@google.com>
---
mm/page_alloc.c | 23 +++++++++++++++++++----
mm/page_alloc.h | 1 +
2 files changed, 20 insertions(+), 4 deletions(-)
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index fb522a09f2e60..ecdd3780192e8 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -5286,6 +5286,19 @@ __alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order,
return page;
}
+/* Set alloc flags before applying gfp_allowed_mask. */
+static inline unsigned int init_alloc_flags(gfp_t gfp_mask, unsigned int alloc_flags)
+{
+ /*
+ * If the caller allowed __GFP_DIRECT_RECLAIM, they can't be atomic.
+ * Note this is a separate determination from whether direct reclaim is
+ * actually allowed, it must happen before applying gfp_allowed_mask.
+ */
+ if (!(gfp_mask & __GFP_DIRECT_RECLAIM))
+ alloc_flags |= ALLOC_NOBLOCK;
+ return alloc_flags;
+}
+
static inline bool prepare_alloc_pages(gfp_t gfp_mask, unsigned int order,
int preferred_nid, nodemask_t *nodemask,
struct alloc_context *ac, gfp_t *alloc_gfp,
@@ -5364,8 +5377,10 @@ unsigned long alloc_pages_bulk_noprof(gfp_t gfp, int preferred_nid,
struct zoneref *z;
struct per_cpu_pages *pcp;
struct list_head *pcp_list;
- struct alloc_context ac;
- unsigned int alloc_flags = ALLOC_WMARK_LOW;
+ struct alloc_context ac = {
+ .alloc_flags = init_alloc_flags(gfp, ALLOC_DEFAULT),
+ };
+ unsigned int alloc_flags = ac.alloc_flags | ALLOC_WMARK_LOW;
int nr_populated = 0, nr_account = 0;
/*
@@ -5584,9 +5599,9 @@ struct page *__alloc_frozen_pages_noprof(gfp_t gfp, unsigned int order,
struct page *page;
gfp_t alloc_gfp; /* The gfp_t that was actually used for allocation */
struct alloc_context ac = {
- .alloc_flags = alloc_flags,
+ .alloc_flags = init_alloc_flags(gfp, alloc_flags),
};
- unsigned int fastpath_alloc_flags = alloc_flags;
+ unsigned int fastpath_alloc_flags = ac.alloc_flags;
/* Other flags could be supported later if needed. */
if (WARN_ON(alloc_flags & ~(ALLOC_NOLOCK | ALLOC_NO_CODETAG)))
diff --git a/mm/page_alloc.h b/mm/page_alloc.h
index 2307b173a459f..02db12c9a1dc2 100644
--- a/mm/page_alloc.h
+++ b/mm/page_alloc.h
@@ -75,6 +75,7 @@
*/
#define ALLOC_UNMAPPED 0x2000
#endif
+#define ALLOC_NOBLOCK 0x4000 /* Caller may be atomic */
/* Flags that allow allocations below the min watermark. */
#define ALLOC_RESERVES (ALLOC_HARDER|ALLOC_MIN_RESERVE|ALLOC_HIGHATOMIC|ALLOC_OOM)
--
2.54.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v3 21/26] mm/page_alloc: implement FREETYPE_UNMAPPED allocations
2026-07-26 22:22 [PATCH v3 00/26] mm: Add ALLOC_UNMAPPED and AS_NO_DIRECT_MAP Brendan Jackman
` (19 preceding siblings ...)
2026-07-26 22:22 ` [PATCH v3 20/26] mm/page_alloc: introduce ALLOC_NOBLOCK Brendan Jackman
@ 2026-07-26 22:22 ` Brendan Jackman
2026-07-26 22:22 ` [PATCH v3 22/26] mm: Minimal KUnit tests for some new page_alloc logic Brendan Jackman
` (4 subsequent siblings)
25 siblings, 0 replies; 29+ messages in thread
From: Brendan Jackman @ 2026-07-26 22:22 UTC (permalink / raw)
To: Borislav Petkov, Dave Hansen, Peter Zijlstra, Andrew Morton,
David Hildenbrand, Vlastimil Babka, Mike Rapoport, Wei Xu,
Johannes Weiner, Zi Yan, Lorenzo Stoakes
Cc: linux-mm, linux-kernel, x86, rppt, Sumit Garg, Will Deacon,
rientjes, Kalyazin, Nikita, patrick.roy, Itazuri, Takahiro,
Andy Lutomirski, David Kaplan, Thomas Gleixner, Yosry Ahmed,
Patrick Bellasi, Reiji Watanabe, Sean Christopherson,
Brendan Jackman
Currently FREETYPE_UNMAPPED allocs will always fail because, although the
lists exist to hold them, there is no way to actually create an unmapped
page block. This commit adds one, and also the logic to map it back
again when that's needed.
Doing this at pageblock granularity ensures that the pageblock flags can
be used to infer which freetype a page belongs to. It also provides nice
batching of TLB flushes, and also avoids creating too much unnecessary
TLB fragmentation in the physmap.
There are some functional requirements for flipping a block:
- Unmapping requires a TLB shootdown, meaning IRQs must be enabled.
- Updating the pagetables might require allocating a pagetable to break
down a huge page. This would deadlock if the zone lock was held.
This makes allocations that need to change sensitivity _somewhat_
similar to those that need to fallback to a different migratetype. But,
the locking requirements mean that this can't just be squashed into the
existing "fallback" allocator logic, instead a new allocator path just
for this purpose is needed.
The new path is assumed to be much cheaper than the really heavyweight
stuff like compaction and reclaim. But at present it is treated as less
desirable than the mobility-related "fallback" and "stealing" logic.
This might turn out to need revision (in particular, maybe it's a
problem that __rmqueue_steal(), which causes fragmentation, happens
before __rmqueue_direct_map()), but that should be treated as a subsequent
optimisation project.
Adding alloc_flags to gfp_freetype() requires moving it to
mm/page_alloc.h so it can refer to ALLOC_UNMAPPED. It was already only
used in internal mm code.
Now that unmapped pageblocks actually exist, exclude them from
migration. Migrating unmapped pages via the mermap should be possible
but that's something to be added later when needed.
Signed-off-by: Brendan Jackman <jackmanb@google.com>
---
mm/Kconfig | 7 +-
mm/compaction.c | 8 ++-
mm/page_alloc.c | 214 +++++++++++++++++++++++++++++++++++++++++++++++++-------
mm/page_alloc.h | 13 +++-
mm/page_owner.c | 6 +-
5 files changed, 215 insertions(+), 33 deletions(-)
diff --git a/mm/Kconfig b/mm/Kconfig
index aa5f041ce1a67..6a89cbf8e8c6e 100644
--- a/mm/Kconfig
+++ b/mm/Kconfig
@@ -1527,9 +1527,10 @@ config MERMAP_KUNIT_TEST
If unsure, say N.
+config PAGE_ALLOC_UNMAPPED
+ bool
+ depends on !HIGHMEM
+
source "mm/damon/Kconfig"
endmenu
-
-config PAGE_ALLOC_UNMAPPED
- bool
diff --git a/mm/compaction.c b/mm/compaction.c
index 67b01af024e17..c9eb3947ffc79 100644
--- a/mm/compaction.c
+++ b/mm/compaction.c
@@ -1393,6 +1393,9 @@ static bool suitable_migration_source(struct compact_control *cc,
block_ft = get_pageblock_freetype(page);
block_mt = free_to_migratetype(block_ft);
+ if (freetype_unmapped(get_pageblock_freetype(page)))
+ return false;
+
/*
* CMA pages can only be taken by ALLOC_CMA requests. For anybody
* else, vacating a CMA block consumes free pages the caller
@@ -1444,6 +1447,9 @@ static bool suitable_migration_target(struct compact_control *cc,
return false;
}
+ if (freetype_unmapped(get_pageblock_freetype(page)))
+ return false;
+
if (cc->ignore_block_suitable)
return true;
@@ -2588,7 +2594,7 @@ compact_zone(struct compact_control *cc, struct capture_control *capc)
INIT_LIST_HEAD(&cc->freepages[order]);
INIT_LIST_HEAD(&cc->migratepages);
- cc->freetype = gfp_freetype(cc->gfp_mask);
+ cc->freetype = gfp_freetype(cc->gfp_mask, cc->alloc_flags);
if (!is_via_compact_memory(cc->order)) {
ret = compaction_suit_allocation_order(cc->zone, cc->order,
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index ecdd3780192e8..ce06ed0d65d8d 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -34,6 +34,7 @@
#include <linux/folio_batch.h>
#include <linux/memory_hotplug.h>
#include <linux/nodemask.h>
+#include <linux/set_memory.h>
#include <linux/vmstat.h>
#include <linux/fault-inject.h>
#include <linux/compaction.h>
@@ -1001,6 +1002,26 @@ static void change_pageblock_range(struct page *pageblock_page,
}
}
+/*
+ * Can pages of these two freetypes be combined into a single higher-order free
+ * page?
+ */
+static inline bool can_merge_freetypes(freetype_t a, freetype_t b)
+{
+ if (freetypes_equal(a, b))
+ return true;
+
+ if (!migratetype_is_mergeable(free_to_migratetype(a)) ||
+ !migratetype_is_mergeable(free_to_migratetype(b)))
+ return false;
+
+ /*
+ * Mustn't "just" merge pages with different freetype flags, changing
+ * those requires updating pagetables.
+ */
+ return freetype_flags(a) == freetype_flags(b);
+}
+
/*
* Freeing function for a buddy system allocator.
*
@@ -1069,9 +1090,7 @@ static inline void __free_one_page(struct page *page,
buddy_ft = get_pfnblock_freetype(buddy, buddy_pfn);
buddy_mt = free_to_migratetype(buddy_ft);
- if (migratetype != buddy_mt &&
- (!migratetype_is_mergeable(migratetype) ||
- !migratetype_is_mergeable(buddy_mt)))
+ if (!can_merge_freetypes(freetype, buddy_ft))
goto done_merging;
}
@@ -1088,7 +1107,9 @@ static inline void __free_one_page(struct page *page,
/*
* Match buddy type. This ensures that an
* expand() down the line puts the sub-blocks
- * on the right freelists.
+ * on the right freelists. Freetype flags are
+ * already set correctly because of
+ * can_merge_freetypes().
*/
change_pageblock_range(buddy, order, migratetype);
}
@@ -1409,7 +1430,7 @@ static __always_inline bool __free_pages_prepare(struct page *page,
{
int bad = 0;
bool skip_kasan_poison = should_skip_kasan_poison(page);
- bool init = want_init_on_free();
+ bool init = want_init_on_free() && !freetype_unmapped(get_pageblock_freetype(page));
bool compound = PageCompound(page);
struct folio *folio = page_folio(page);
@@ -1900,14 +1921,19 @@ static inline bool should_skip_kasan_unpoison(gfp_t flags)
return flags & __GFP_SKIP_KASAN;
}
-static inline bool should_skip_init(gfp_t flags)
+static inline bool should_skip_init(gfp_t gfp, unsigned int alloc_flags)
{
/* Don't skip, if hardware tag-based KASAN is not enabled. */
if (!kasan_hw_tags_enabled())
return false;
+#ifdef CONFIG_PAGE_ALLOC_UNMAPPED
+ if (alloc_flags & ALLOC_UNMAPPED)
+ return true;
+#endif
+
/* For hardware tag-based KASAN, skip if requested. */
- return (flags & __GFP_SKIP_ZERO);
+ return (gfp & __GFP_SKIP_ZERO);
}
inline void post_alloc_hook(struct page *page, unsigned int order,
@@ -1915,7 +1941,7 @@ inline void post_alloc_hook(struct page *page, unsigned int order,
{
const bool zero_tags = gfp_flags & __GFP_ZEROTAGS;
bool init = !want_init_on_free() && want_init_on_alloc(gfp_flags) &&
- !should_skip_init(gfp_flags);
+ !should_skip_init(gfp_flags, alloc_flags);
int i;
set_page_private(page, 0);
@@ -3344,7 +3370,7 @@ int __isolate_free_page(struct page *page, unsigned int order)
* Only change normal pageblocks (i.e., they can merge
* with others)
*/
- if (migratetype_is_mergeable(free_to_migratetype(ft)))
+ if (can_merge_freetypes(old_ft, new_ft))
move_freepages_block(zone, page, old_ft, new_ft);
}
}
@@ -3400,6 +3426,127 @@ static inline void zone_statistics(struct zone *preferred_zone, struct zone *z,
#endif
}
+#ifdef CONFIG_PAGE_ALLOC_UNMAPPED
+/* Try to allocate a page by mapping/unmapping a block from the direct map. */
+static inline struct page *
+__rmqueue_direct_map(struct zone *zone, unsigned int request_order,
+ unsigned int alloc_flags, freetype_t freetype)
+{
+ unsigned int ft_flags_other = freetype_flags(freetype) ^ FREETYPE_UNMAPPED;
+ freetype_t ft_other = migrate_to_freetype(free_to_migratetype(freetype),
+ ft_flags_other);
+ bool want_mapped = !(freetype_flags(freetype) & FREETYPE_UNMAPPED);
+ enum rmqueue_mode rmqm = RMQUEUE_NORMAL;
+ unsigned long irq_flags;
+ int nr_pageblocks, nr_freed;
+ struct page *page;
+ int alloc_order;
+ int err;
+
+ if (freetype_idx(ft_other) < 0)
+ return NULL;
+
+ /*
+ * Might need a TLB shootdown. Even if IRQs are on this isn't
+ * safe if the caller holds a lock (in case the other CPUs need that
+ * lock to handle the shootdown IPI).
+ */
+ if (alloc_flags & ALLOC_NOBLOCK)
+ return NULL;
+
+ if (!can_set_direct_map() || alloc_flags & ALLOC_NOLOCK)
+ return NULL;
+
+ lockdep_assert(!irqs_disabled() || unlikely(early_boot_irqs_disabled));
+
+ /*
+ * Need to [un]map a whole pageblock (otherwise it might require
+ * allocating pagetables). First allocate it.
+ */
+ alloc_order = max(request_order, pageblock_order);
+ nr_pageblocks = 1 << (alloc_order - pageblock_order);
+ spin_lock_irqsave(&zone->lock, irq_flags);
+ /* First try a block that already has the right migratetype. */
+ page = __rmqueue(zone, alloc_order, ft_other, alloc_flags, &rmqm);
+ if (!page) {
+ /* Fallback to changing a block's migratetype. */
+ rmqm = RMQUEUE_CLAIM;
+ page = __rmqueue(zone, alloc_order, ft_other, alloc_flags, &rmqm);
+ }
+ spin_unlock_irqrestore(&zone->lock, irq_flags);
+ if (!page)
+ return NULL;
+
+ /*
+ * Now that IRQs are on it's safe to do a TLB shootdown, and now that we
+ * released the zone lock it's possible to allocate a pagetable if
+ * needed to split up a huge page.
+ *
+ * Note that modifying the direct map may need to allocate pagetables.
+ * What about unbounded recursion? Here are the assumptions that make it
+ * safe:
+ *
+ * - The direct map starts out fully mapped at boot. (This is not really
+ * an "assumption" as it's in direct control of page_alloc.c).
+ *
+ * - Once pages in the direct map are broken down, they are not
+ * re-aggregated into larger pages again.
+ *
+ * - Pagetables are never allocated with ALLOC_UNMAPPED.
+ *
+ * Under these assumptions, a pagetable might need to be allocated while
+ * _unmapping_ stuff from the direct map during an ALLOC_UNMAPPED
+ * allocation. But, the allocation of that pagetable never requires
+ * allocating a further pagetable.
+ */
+ err = set_direct_map_valid_noflush(page,
+ nr_pageblocks << pageblock_order, want_mapped);
+ if (err == -ENOMEM || WARN_ONCE(err, "err=%d\n", err)) {
+ set_direct_map_valid_noflush(page,
+ nr_pageblocks << pageblock_order, !want_mapped);
+ spin_lock_irqsave(&zone->lock, irq_flags);
+ /* Important: free using _old_ freetype. */
+ __free_one_page(page, page_to_pfn(page), zone,
+ alloc_order, ft_other, FPI_SKIP_REPORT_NOTIFY);
+ spin_unlock_irqrestore(&zone->lock, irq_flags);
+ return NULL;
+ }
+
+ if (want_mapped) {
+ /* Exposing formerly-protected data; scrub it. */
+ clear_highpages_kasan_tagged(page, nr_pageblocks << pageblock_order);
+ } else {
+ unsigned long start = (unsigned long)page_address(page);
+ unsigned long end = start + (nr_pageblocks << (pageblock_order + PAGE_SHIFT));
+
+ flush_tlb_kernel_range(start, end);
+ }
+
+ for (int i = 0; i < nr_pageblocks; i++) {
+ struct page *block_page = page + (pageblock_nr_pages * i);
+
+ set_pageblock_freetype_flags(block_page, freetype_flags(freetype));
+ }
+
+ if (request_order >= alloc_order)
+ return page;
+
+ /* Free any remaining pages in the block. */
+ spin_lock_irqsave(&zone->lock, irq_flags);
+ nr_freed = expand(zone, page, request_order, alloc_order, freetype);
+ account_freepages(zone, nr_freed, free_to_migratetype(freetype));
+ spin_unlock_irqrestore(&zone->lock, irq_flags);
+
+ return page;
+}
+#else /* CONFIG_PAGE_ALLOC_UNMAPPED */
+static inline struct page *__rmqueue_direct_map(struct zone *zone, unsigned int request_order,
+ unsigned int alloc_flags, freetype_t freetype)
+{
+ return NULL;
+}
+#endif /* CONFIG_PAGE_ALLOC_UNMAPPED */
+
static __always_inline
struct page *rmqueue_buddy(struct zone *preferred_zone, struct zone *zone,
unsigned int order, unsigned int alloc_flags,
@@ -3433,13 +3580,15 @@ struct page *rmqueue_buddy(struct zone *preferred_zone, struct zone *zone,
*/
if (!page && (alloc_flags & (ALLOC_OOM|ALLOC_HARDER)))
page = __rmqueue_smallest(zone, order, ft_high);
-
- if (!page) {
- spin_unlock_irqrestore(&zone->lock, flags);
- return NULL;
- }
}
spin_unlock_irqrestore(&zone->lock, flags);
+
+ /* Try changing direct map, now we've released the zone lock */
+ if (!page)
+ page = __rmqueue_direct_map(zone, order, alloc_flags, freetype);
+ if (!page)
+ return NULL;
+
} while (check_new_pages(page, order));
/*
@@ -3660,6 +3809,8 @@ static void reserve_highatomic_pageblock(struct page *page, int order,
return;
ft_high = freetype_with_migrate(ft, MIGRATE_HIGHATOMIC);
+ if (freetype_idx(ft_high) < 0)
+ return;
if (order < pageblock_order) {
if (move_freepages_block(zone, page, ft, ft_high) == -1)
return;
@@ -3975,13 +4126,15 @@ alloc_flags_nofragment(struct zone *zone, gfp_t gfp_mask)
}
/* Must be called after current_gfp_context() which can change gfp_mask */
-static inline unsigned int alloc_flags_cma(gfp_t gfp_mask)
+static inline unsigned int alloc_flags_cma(gfp_t gfp_mask, unsigned int alloc_flags)
{
#ifdef CONFIG_CMA
- if (free_to_migratetype(gfp_freetype(gfp_mask)) == MIGRATE_MOVABLE)
- return ALLOC_CMA;
+ if (free_to_migratetype(gfp_freetype(gfp_mask, alloc_flags)) == MIGRATE_MOVABLE)
+ alloc_flags |= ALLOC_CMA;
#endif
- return ALLOC_DEFAULT;
+ alloc_flags |= ALLOC_DEFAULT;
+
+ return alloc_flags;
}
/*
@@ -4770,7 +4923,7 @@ alloc_flags_slowpath(gfp_t gfp_mask, unsigned int order)
} else if (unlikely(rt_or_dl_task(current)) && in_task())
alloc_flags |= ALLOC_MIN_RESERVE;
- alloc_flags |= alloc_flags_cma(gfp_mask);
+ alloc_flags = alloc_flags_cma(gfp_mask, alloc_flags);
if (defrag_mode)
alloc_flags |= ALLOC_NOFRAGMENT;
@@ -5085,7 +5238,7 @@ __alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order,
reserve_flags = __gfp_pfmemalloc_flags(gfp_mask);
if (reserve_flags)
- alloc_flags = alloc_flags_cma(gfp_mask) | reserve_flags |
+ alloc_flags = alloc_flags_cma(gfp_mask, alloc_flags) | reserve_flags |
ac->alloc_flags | (alloc_flags & ALLOC_KSWAPD);
/*
@@ -5307,7 +5460,11 @@ static inline bool prepare_alloc_pages(gfp_t gfp_mask, unsigned int order,
ac->highest_zoneidx = gfp_zone(gfp_mask);
ac->zonelist = node_zonelist(preferred_nid, gfp_mask);
ac->nodemask = nodemask;
- ac->freetype = gfp_freetype(gfp_mask);
+ ac->freetype = gfp_freetype(gfp_mask, *alloc_flags);
+
+ /* Not implemented yet. */
+ if (freetype_flags(ac->freetype) & FREETYPE_UNMAPPED && gfp_mask & __GFP_ZERO)
+ return false;
if (cpusets_enabled()) {
*alloc_gfp |= __GFP_HARDWALL;
@@ -5331,7 +5488,7 @@ static inline bool prepare_alloc_pages(gfp_t gfp_mask, unsigned int order,
should_fail_alloc_page(gfp_mask, order))
return false;
- *alloc_flags |= alloc_flags_cma(gfp_mask);
+ *alloc_flags = alloc_flags_cma(gfp_mask, *alloc_flags);
/* Dirty zone balancing only done in the fast path */
ac->spread_dirty_pages = (gfp_mask & __GFP_WRITE);
@@ -5590,6 +5747,16 @@ static inline bool alloc_nolock_allowed(void)
static const gfp_t gfp_nolock = __GFP_NOWARN | __GFP_ZERO | __GFP_NOMEMALLOC |
__GFP_COMP;
+/*
+ * ALLOC_ flags other mm callers are allowed to set. More could be trivially
+ * added here if needed.
+ */
+#ifdef CONFIG_PAGE_ALLOC_UNMAPPED
+#define SUPPORTED_INPUT_ALLOC_FLAGS ALLOC_NOLOCK | ALLOC_NO_CODETAG | ALLOC_UNMAPPED
+#else
+#define SUPPORTED_INPUT_ALLOC_FLAGS ALLOC_NOLOCK | ALLOC_NO_CODETAG
+#endif
+
/*
* This is the 'heart' of the zoned buddy allocator.
*/
@@ -5603,8 +5770,7 @@ struct page *__alloc_frozen_pages_noprof(gfp_t gfp, unsigned int order,
};
unsigned int fastpath_alloc_flags = ac.alloc_flags;
- /* Other flags could be supported later if needed. */
- if (WARN_ON(alloc_flags & ~(ALLOC_NOLOCK | ALLOC_NO_CODETAG)))
+ if (WARN_ON(alloc_flags & ~(SUPPORTED_INPUT_ALLOC_FLAGS)))
return NULL;
if (!alloc_order_allowed(gfp, order, alloc_flags))
diff --git a/mm/page_alloc.h b/mm/page_alloc.h
index 02db12c9a1dc2..5fa8fc527347c 100644
--- a/mm/page_alloc.h
+++ b/mm/page_alloc.h
@@ -308,9 +308,10 @@ static inline bool free_area_empty(struct free_area *area, freetype_t freetype)
#define GFP_MOVABLE_MASK (__GFP_RECLAIMABLE|__GFP_MOVABLE)
#define GFP_MOVABLE_SHIFT 3
-static inline freetype_t gfp_freetype(const gfp_t gfp_flags)
+static inline freetype_t gfp_freetype(const gfp_t gfp_flags, unsigned int alloc_flags)
{
unsigned int migratetype;
+ unsigned int ft_flags = 0;
VM_WARN_ON((gfp_flags & GFP_MOVABLE_MASK) == GFP_MOVABLE_MASK);
BUILD_BUG_ON((1UL << GFP_MOVABLE_SHIFT) != ___GFP_MOVABLE);
@@ -327,7 +328,15 @@ static inline freetype_t gfp_freetype(const gfp_t gfp_flags)
>> GFP_MOVABLE_SHIFT;
}
- return migrate_to_freetype(migratetype, 0);
+#ifdef CONFIG_PAGE_ALLOC_UNMAPPED
+ if (alloc_flags & ALLOC_UNMAPPED) {
+ if (WARN_ON_ONCE(migratetype != MIGRATE_UNMOVABLE))
+ migratetype = MIGRATE_UNMOVABLE;
+ ft_flags |= FREETYPE_UNMAPPED;
+ }
+#endif
+
+ return migrate_to_freetype(migratetype, ft_flags);
}
#undef GFP_MOVABLE_MASK
#undef GFP_MOVABLE_SHIFT
diff --git a/mm/page_owner.c b/mm/page_owner.c
index ed7fcc0d70f3f..e86d2028d26c6 100644
--- a/mm/page_owner.c
+++ b/mm/page_owner.c
@@ -527,7 +527,7 @@ void pagetypeinfo_showmixedcount_print(struct seq_file *m,
page_owner = get_page_owner(page_ext);
page_mt = free_to_migratetype(
- gfp_freetype(page_owner->gfp_mask));
+ gfp_freetype(page_owner->gfp_mask, ALLOC_DEFAULT));
if (pageblock_mt != page_mt) {
if (is_migrate_cma(pageblock_mt))
count[MIGRATE_MOVABLE]++;
@@ -626,7 +626,7 @@ print_page_owner(char __user *buf, size_t count, unsigned long pfn,
/* Print information relevant to grouping pages by mobility */
pageblock_mt = get_pageblock_migratetype(page);
- page_mt = free_to_migratetype(gfp_freetype(page_owner->gfp_mask));
+ page_mt = free_to_migratetype(gfp_freetype(page_owner->gfp_mask, ALLOC_DEFAULT));
ret += scnprintf(kbuf + ret, count - ret,
"PFN 0x%lx type %s Block %lu type %s Flags %pGp\n",
pfn,
@@ -686,7 +686,7 @@ void __dump_page_owner(const struct page *page)
page_owner = get_page_owner(page_ext);
gfp_mask = page_owner->gfp_mask;
- mt = free_to_migratetype(gfp_freetype(gfp_mask));
+ mt = free_to_migratetype(gfp_freetype(gfp_mask, ALLOC_DEFAULT));
if (!test_bit(PAGE_EXT_OWNER, &page_ext->flags)) {
pr_alert("page_owner info is not present (never set?)\n");
--
2.54.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v3 22/26] mm: Minimal KUnit tests for some new page_alloc logic
2026-07-26 22:22 [PATCH v3 00/26] mm: Add ALLOC_UNMAPPED and AS_NO_DIRECT_MAP Brendan Jackman
` (20 preceding siblings ...)
2026-07-26 22:22 ` [PATCH v3 21/26] mm/page_alloc: implement FREETYPE_UNMAPPED allocations Brendan Jackman
@ 2026-07-26 22:22 ` Brendan Jackman
2026-07-26 22:22 ` [PATCH v3 23/26] mm: Split out NR_FREE_PAGES_BLOCKS_[UN]MAPPED Brendan Jackman
` (3 subsequent siblings)
25 siblings, 0 replies; 29+ messages in thread
From: Brendan Jackman @ 2026-07-26 22:22 UTC (permalink / raw)
To: Borislav Petkov, Dave Hansen, Peter Zijlstra, Andrew Morton,
David Hildenbrand, Vlastimil Babka, Mike Rapoport, Wei Xu,
Johannes Weiner, Zi Yan, Lorenzo Stoakes
Cc: linux-mm, linux-kernel, x86, rppt, Sumit Garg, Will Deacon,
rientjes, Kalyazin, Nikita, patrick.roy, Itazuri, Takahiro,
Andy Lutomirski, David Kaplan, Thomas Gleixner, Yosry Ahmed,
Patrick Bellasi, Reiji Watanabe, Sean Christopherson,
Brendan Jackman
Add a simple smoke test for ALLOC_UNMAPPED that tries to exercise
flipping pageblocks between mapped/unmapped state.
Also add some basic tests for some freelist-indexing helpers.
Simplest way to run these on x86:
tools/testing/kunit/kunit.py run --arch=x86_64 "page_alloc.*" \
--kconfig_add CONFIG_MERMAP=y --kconfig_add CONFIG_PAGE_ALLOC_UNMAPPED=y
Signed-off-by: Brendan Jackman <jackmanb@google.com>
---
mm/Kconfig | 7 ++
mm/Makefile | 1 +
mm/internal.h | 6 ++
mm/page_alloc.c | 11 +-
mm/tests/page_alloc_kunit.c | 243 ++++++++++++++++++++++++++++++++++++++++++++
5 files changed, 265 insertions(+), 3 deletions(-)
diff --git a/mm/Kconfig b/mm/Kconfig
index 6a89cbf8e8c6e..bdd3e083520c0 100644
--- a/mm/Kconfig
+++ b/mm/Kconfig
@@ -1531,6 +1531,13 @@ config PAGE_ALLOC_UNMAPPED
bool
depends on !HIGHMEM
+config PAGE_ALLOC_KUNIT_TEST
+ tristate "KUnit tests for the page allocator" if !KUNIT_ALL_TESTS
+ depends on KUNIT
+ default KUNIT_ALL_TESTS
+ help
+ Builds KUnit tests for the page allocator.
+
source "mm/damon/Kconfig"
endmenu
diff --git a/mm/Makefile b/mm/Makefile
index 3507cb7ed6a9a..12366f8c0f537 100644
--- a/mm/Makefile
+++ b/mm/Makefile
@@ -149,3 +149,4 @@ obj-$(CONFIG_LAZY_MMU_MODE_KUNIT_TEST) += tests/lazy_mmu_mode_kunit.o
obj-$(CONFIG_MEM_ALLOC_PROFILING) += alloc_tag.o
obj-$(CONFIG_MERMAP) += mermap.o
obj-$(CONFIG_MERMAP_KUNIT_TEST) += tests/mermap_kunit.o
+obj-$(CONFIG_PAGE_ALLOC_KUNIT_TEST) += tests/page_alloc_kunit.o
diff --git a/mm/internal.h b/mm/internal.h
index fada318d13541..089c53bf9a336 100644
--- a/mm/internal.h
+++ b/mm/internal.h
@@ -1695,4 +1695,10 @@ int __apply_to_page_range(struct mm_struct *mm, unsigned long addr,
unsigned long size, pte_fn_t fn,
void *data, unsigned int flags);
+#if IS_ENABLED(CONFIG_KUNIT)
+unsigned int order_to_pindex(freetype_t freetype, int order);
+int pindex_to_order(unsigned int pindex);
+bool pcp_allowed_order(unsigned int order);
+#endif /* IS_ENABLED(CONFIG_KUNIT) */
+
#endif /* __MM_INTERNAL_H */
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index ce06ed0d65d8d..ac2f6190117ae 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -56,6 +56,7 @@
#include <linux/cacheinfo.h>
#include <linux/pgalloc_tag.h>
#include <asm/div64.h>
+#include <kunit/visibility.h>
#include "internal.h"
#include "mm_init.h"
#include "page_alloc.h"
@@ -465,6 +466,7 @@ freetype_t get_pfnblock_freetype(const struct page *page, unsigned long pfn)
{
return __get_pfnblock_freetype(page, pfn, false);
}
+EXPORT_SYMBOL_IF_KUNIT(get_pfnblock_freetype);
/**
@@ -691,7 +693,7 @@ static void bad_page(struct page *page, const char *reason)
add_taint(TAINT_BAD_PAGE, LOCKDEP_NOW_UNRELIABLE);
}
-static inline unsigned int order_to_pindex(freetype_t freetype, int order)
+VISIBLE_IF_KUNIT inline unsigned int order_to_pindex(freetype_t freetype, int order)
{
int migratetype = free_to_migratetype(freetype);
@@ -719,8 +721,9 @@ static inline unsigned int order_to_pindex(freetype_t freetype, int order)
return (MIGRATE_PCPTYPES * order) + migratetype;
}
+EXPORT_SYMBOL_IF_KUNIT(order_to_pindex);
-static inline int pindex_to_order(unsigned int pindex)
+VISIBLE_IF_KUNIT int pindex_to_order(unsigned int pindex)
{
unsigned int unmapped_base = NR_LOWORDER_PCP_LISTS + NR_PCP_THP;
int order;
@@ -741,8 +744,9 @@ static inline int pindex_to_order(unsigned int pindex)
return order;
}
+EXPORT_SYMBOL_IF_KUNIT(pindex_to_order);
-static inline bool pcp_allowed_order(unsigned int order)
+VISIBLE_IF_KUNIT inline bool pcp_allowed_order(unsigned int order)
{
if (order <= PAGE_ALLOC_COSTLY_ORDER)
return true;
@@ -752,6 +756,7 @@ static inline bool pcp_allowed_order(unsigned int order)
#endif
return false;
}
+EXPORT_SYMBOL_IF_KUNIT(pcp_allowed_order);
/*
* Higher-order pages are called "compound pages". They are structured thusly:
diff --git a/mm/tests/page_alloc_kunit.c b/mm/tests/page_alloc_kunit.c
new file mode 100644
index 0000000000000..b7f900a93cbcf
--- /dev/null
+++ b/mm/tests/page_alloc_kunit.c
@@ -0,0 +1,243 @@
+// SPDX-License-Identifier: GPL-2.0-only
+#include <linux/gfp.h>
+#include <linux/kernel.h>
+#include <linux/mermap.h>
+#include <linux/mm_types.h>
+#include <linux/mm.h>
+#include <linux/pgtable.h>
+#include <linux/set_memory.h>
+#include <linux/sched/mm.h>
+#include <linux/types.h>
+#include <linux/vmalloc.h>
+
+#include <kunit/resource.h>
+#include <kunit/test.h>
+
+#include "internal.h"
+#include "page_alloc.h"
+
+struct free_pages_ctx {
+ unsigned int order;
+ struct list_head pages;
+};
+
+static inline void action_many__free_pages(void *context)
+{
+ struct free_pages_ctx *ctx = context;
+ struct page *page, *tmp;
+
+ list_for_each_entry_safe(page, tmp, &ctx->pages, lru)
+ __free_pages(page, ctx->order);
+}
+
+/*
+ * Allocate a bunch of pages with the same order and GFP/alloc_flags,
+ * transparently take care of error handling and cleanup. Does this all via a
+ * single KUnit resource, i.e. has a fixed memory overhead.
+ */
+static inline struct free_pages_ctx *
+do_many_alloc_pages(struct kunit *test, unsigned int alloc_flags,
+ unsigned int order, unsigned int count)
+{
+ struct free_pages_ctx *ctx = kunit_kzalloc(
+ test, sizeof(struct free_pages_ctx), GFP_KERNEL);
+ gfp_t gfp = GFP_KERNEL | __GFP_THISNODE;
+
+ KUNIT_ASSERT_NOT_NULL(test, ctx);
+ INIT_LIST_HEAD(&ctx->pages);
+ ctx->order = order;
+
+ for (int i = 0; i < count; i++) {
+ struct page *page = __alloc_pages(gfp, order, numa_node_id(),
+ NULL, alloc_flags);
+
+ if (!page) {
+ struct page *page, *tmp;
+
+ list_for_each_entry_safe(page, tmp, &ctx->pages, lru)
+ __free_pages(page, order);
+
+ KUNIT_FAIL_AND_ABORT(test,
+ "Failed to alloc order %d page (GFP *%pG) iter %d",
+ order, &gfp, i);
+ }
+ list_add(&page->lru, &ctx->pages);
+ }
+
+ KUNIT_ASSERT_EQ(test,
+ kunit_add_action_or_reset(test, action_many__free_pages, ctx), 0);
+ return ctx;
+}
+
+#ifdef CONFIG_PAGE_ALLOC_UNMAPPED
+
+/* Do some allocations that force the allocator to map/unmap some blocks. */
+static void test_alloc_map_unmap(struct kunit *test)
+{
+ unsigned long page_majority;
+ struct free_pages_ctx *ctx;
+ struct page *page;
+
+ kunit_attach_mm();
+ mermap_mm_prepare(current->mm);
+
+ /* No cleanup here - assuming kthread "belongs" to this test. */
+ set_cpus_allowed_ptr(current, cpumask_of_node(numa_mem_id()));
+
+ /*
+ * First allocate more than half of the memory in the node as
+ * unmapped. Assuming the memory starts out mapped, this should
+ * exercise the unmap.
+ */
+ page_majority = (node_present_pages(numa_mem_id()) / 2) + 1;
+ ctx = do_many_alloc_pages(test, ALLOC_UNMAPPED, 0, page_majority);
+
+ /* Check pages are unmapped */
+ list_for_each_entry(page, &ctx->pages, lru) {
+ freetype_t ft = get_pfnblock_freetype(page, page_to_pfn(page));
+
+ /*
+ * Logically it should be an EXPECT, but that would
+ * cause heavy log spam on failure so use ASSERT for
+ * concision.
+ */
+ KUNIT_ASSERT_FALSE(test, kernel_page_present(page));
+ KUNIT_ASSERT_TRUE(test, freetype_flags(ft) & FREETYPE_UNMAPPED);
+
+ cond_resched();
+ }
+
+ /*
+ * Now free them again and allocate the same amount without
+ * ALLOC_UNMAPPED. This will exercise the mapping logic.
+ */
+ kunit_release_action(test, action_many__free_pages, ctx);
+ ctx = do_many_alloc_pages(test, ALLOC_DEFAULT, 0, page_majority);
+
+ /* Check pages are mapped. */
+ list_for_each_entry(page, &ctx->pages, lru) {
+ KUNIT_ASSERT_TRUE(test, kernel_page_present(page));
+ cond_resched();
+ }
+}
+
+#endif /* CONFIG_PAGE_ALLOC_UNMAPPED */
+
+static void __test_pindex_helpers(struct kunit *test, unsigned long *bitmap,
+ int mt, unsigned int ftflags, unsigned int order)
+{
+ freetype_t ft = migrate_to_freetype(mt, ftflags);
+ unsigned int pindex;
+ int got_order;
+
+ if (!pcp_allowed_order(order))
+ return;
+
+ if (mt >= MIGRATE_PCPTYPES)
+ return;
+
+ if (freetype_idx(ft) < 0)
+ return;
+
+ pindex = order_to_pindex(ft, order);
+
+ KUNIT_ASSERT_LT_MSG(test, pindex, NR_PCP_LISTS,
+ "invalid pindex %d (order %d mt %d flags %#x)",
+ pindex, order, mt, ftflags);
+ KUNIT_EXPECT_TRUE_MSG(test, test_bit(pindex, bitmap),
+ "pindex %d reused (order %d mt %d flags %#x)",
+ pindex, order, mt, ftflags);
+
+ /*
+ * For THP, two migratetypes map to the same pindex,
+ * just manually exclude one of those cases.
+ */
+ if (!(IS_ENABLED(CONFIG_TRANSPARENT_HUGEPAGE) &&
+ order == HPAGE_PMD_ORDER &&
+ mt == min(MIGRATE_UNMOVABLE, MIGRATE_RECLAIMABLE)))
+ clear_bit(pindex, bitmap);
+
+ got_order = pindex_to_order(pindex);
+ KUNIT_EXPECT_EQ_MSG(test, order, got_order,
+ "roundtrip failed, got %d want %d (pindex %d mt %d flags %#x)",
+ got_order, order, pindex, mt, ftflags);
+}
+
+/* This just checks for basic arithmetic errors. */
+static void test_pindex_helpers(struct kunit *test)
+{
+ DECLARE_BITMAP(bitmap, NR_PCP_LISTS);
+
+ /* Bit means "pindex not yet used". */
+ bitmap_fill(bitmap, NR_PCP_LISTS);
+
+ for (unsigned int order = 0; order < NR_PAGE_ORDERS; order++) {
+ for (int mt = 0; mt < MIGRATE_TYPES; mt++) {
+ __test_pindex_helpers(test, bitmap, mt, 0, order);
+ if (FREETYPE_UNMAPPED)
+ __test_pindex_helpers(test, bitmap, mt,
+ FREETYPE_UNMAPPED, order);
+ }
+ }
+
+ KUNIT_EXPECT_TRUE_MSG(test, bitmap_empty(bitmap, NR_PCP_LISTS),
+ "unused pindices: %*pbl", NR_PCP_LISTS, bitmap);
+}
+
+static void __test_freetype_idx(struct kunit *test, unsigned int order,
+ int migratetype, unsigned int ftflags,
+ unsigned long *bitmap)
+{
+ freetype_t ft = migrate_to_freetype(migratetype, ftflags);
+ int idx = freetype_idx(ft);
+
+ if (idx == -1)
+ return;
+ KUNIT_ASSERT_GE(test, idx, 0);
+ KUNIT_ASSERT_LT(test, idx, NR_FREETYPE_IDXS);
+
+ KUNIT_EXPECT_LT_MSG(test, idx, NR_PCP_LISTS,
+ "invalid idx %d (order %d mt %d flags %#x)",
+ idx, order, migratetype, ftflags);
+ KUNIT_EXPECT_TRUE(test, freetypes_equal(freetype_from_idx(idx), ft));
+ clear_bit(idx, bitmap);
+}
+
+static void test_freetype_idx(struct kunit *test)
+{
+ unsigned long bitmap[bitmap_size(NR_FREETYPE_IDXS)];
+
+ /* Bit means "pindex not yet used". */
+ bitmap_fill(bitmap, NR_FREETYPE_IDXS);
+
+ for (unsigned int order = 0; order < NR_PAGE_ORDERS; order++) {
+ for (int mt = 0; mt < MIGRATE_TYPES; mt++) {
+ __test_freetype_idx(test, order, mt, 0, bitmap);
+ if (FREETYPE_UNMAPPED)
+ __test_freetype_idx(test, order, mt,
+ FREETYPE_UNMAPPED, bitmap);
+ }
+ }
+
+ KUNIT_EXPECT_TRUE_MSG(test, bitmap_empty(bitmap, NR_FREETYPE_IDXS),
+ "unused idxs: %*pbl", NR_FREETYPE_IDXS, bitmap);
+}
+
+static struct kunit_case test_cases[] = {
+#ifdef CONFIG_PAGE_ALLOC_UNMAPPED
+ KUNIT_CASE(test_alloc_map_unmap),
+#endif
+ KUNIT_CASE(test_pindex_helpers),
+ KUNIT_CASE(test_freetype_idx),
+ {}
+};
+
+static struct kunit_suite test_suite = {
+ .name = "page_alloc",
+ .test_cases = test_cases,
+};
+
+kunit_test_suite(test_suite);
+
+MODULE_LICENSE("GPL");
+MODULE_IMPORT_NS("EXPORTED_FOR_KUNIT_TESTING");
--
2.54.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v3 23/26] mm: Split out NR_FREE_PAGES_BLOCKS_[UN]MAPPED
2026-07-26 22:22 [PATCH v3 00/26] mm: Add ALLOC_UNMAPPED and AS_NO_DIRECT_MAP Brendan Jackman
` (21 preceding siblings ...)
2026-07-26 22:22 ` [PATCH v3 22/26] mm: Minimal KUnit tests for some new page_alloc logic Brendan Jackman
@ 2026-07-26 22:22 ` Brendan Jackman
2026-07-26 22:22 ` [PATCH v3 24/26] mm/page_alloc: always direct compact for unmapped allocs Brendan Jackman
` (2 subsequent siblings)
25 siblings, 0 replies; 29+ messages in thread
From: Brendan Jackman @ 2026-07-26 22:22 UTC (permalink / raw)
To: Borislav Petkov, Dave Hansen, Peter Zijlstra, Andrew Morton,
David Hildenbrand, Vlastimil Babka, Mike Rapoport, Wei Xu,
Johannes Weiner, Zi Yan, Lorenzo Stoakes
Cc: linux-mm, linux-kernel, x86, rppt, Sumit Garg, Will Deacon,
rientjes, Kalyazin, Nikita, patrick.roy, Itazuri, Takahiro,
Andy Lutomirski, David Kaplan, Thomas Gleixner, Yosry Ahmed,
Patrick Bellasi, Reiji Watanabe, Sean Christopherson,
Brendan Jackman
In order to infer whether compaction is likely to enable an allocation
that maps/unmaps a pageblock, we need to know how many fully-free
pageblocks of each type there are.
Signed-off-by: Brendan Jackman <jackmanb@google.com>
---
include/linux/mmzone.h | 5 ++++-
include/linux/vmstat.h | 10 ++++++++++
mm/compaction.c | 5 ++---
mm/page_alloc.c | 13 ++++++++++---
mm/vmscan.c | 49 ++++++++++++++++++++++++++++---------------------
mm/vmstat.c | 3 ++-
6 files changed, 56 insertions(+), 29 deletions(-)
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 2533a228f71fd..e063a59bbc883 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -181,7 +181,10 @@ enum numa_stat_item {
enum zone_stat_item {
NR_FREE_PAGES,
- NR_FREE_PAGES_BLOCKS,
+ /* Number of free pages in entirely free pageblocks, with direct map */
+ NR_FREE_PAGES_BLOCKS_MAPPED,
+ /* Ditto, without direct map */
+ NR_FREE_PAGES_BLOCKS_UNMAPPED,
NR_ZONE_LRU_BASE, /* Used only for compaction and reclaim retry */
NR_ZONE_INACTIVE_ANON = NR_ZONE_LRU_BASE,
NR_ZONE_ACTIVE_ANON,
diff --git a/include/linux/vmstat.h b/include/linux/vmstat.h
index 5b31d8e7ae405..debd593756149 100644
--- a/include/linux/vmstat.h
+++ b/include/linux/vmstat.h
@@ -224,6 +224,16 @@ static inline unsigned long zone_page_state(struct zone *zone,
return x;
}
+/*
+ * Approx number of pages in entirely free pageblocks. Due to races this could
+ * actually return a value more than the number of pages in the zone.
+ */
+static inline unsigned long zone_free_pages_blocks(struct zone *zone)
+{
+ return zone_page_state(zone, NR_FREE_PAGES_BLOCKS_MAPPED) +
+ zone_page_state(zone, NR_FREE_PAGES_BLOCKS_UNMAPPED);
+}
+
/*
* More accurate version that also considers the currently pending
* deltas. For that we need to loop over all cpus to find the current
diff --git a/mm/compaction.c b/mm/compaction.c
index c9eb3947ffc79..ed12d2fc6fad3 100644
--- a/mm/compaction.c
+++ b/mm/compaction.c
@@ -2353,8 +2353,7 @@ static enum compact_result __compact_finished(struct compact_control *cc)
if (__zone_watermark_ok(cc->zone, cc->order,
high_wmark_pages(cc->zone),
cc->highest_zoneidx, cc->alloc_flags,
- zone_page_state(cc->zone,
- NR_FREE_PAGES_BLOCKS)))
+ zone_free_pages_blocks(cc->zone)))
return COMPACT_SUCCESS;
return COMPACT_CONTINUE;
@@ -2538,7 +2537,7 @@ compaction_suit_allocation_order(struct zone *zone, unsigned int order,
unsigned long watermark;
if (kcompactd && defrag_mode)
- free_pages = zone_page_state(zone, NR_FREE_PAGES_BLOCKS);
+ free_pages = zone_free_pages_blocks(zone);
else
free_pages = zone_page_state(zone, NR_FREE_PAGES);
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index ac2f6190117ae..d12ce84662ab7 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -866,6 +866,13 @@ static inline void account_freepages(struct zone *zone, int nr_pages,
zone->nr_free_highatomic + nr_pages);
}
+static inline enum zone_stat_item free_pages_blocks_stat(freetype_t ft)
+{
+ if (freetype_flags(ft) & FREETYPE_UNMAPPED)
+ return NR_FREE_PAGES_BLOCKS_UNMAPPED;
+ return NR_FREE_PAGES_BLOCKS_MAPPED;
+}
+
/* Used for pages not on another list */
static inline void __add_to_free_list(struct page *page, struct zone *zone,
unsigned int order, freetype_t freetype,
@@ -890,7 +897,7 @@ static inline void __add_to_free_list(struct page *page, struct zone *zone,
area->nr_free++;
if (order >= pageblock_order && !is_migrate_isolate(free_to_migratetype(freetype)))
- __mod_zone_page_state(zone, NR_FREE_PAGES_BLOCKS, nr_pages);
+ __mod_zone_page_state(zone, free_pages_blocks_stat(freetype), nr_pages);
}
/*
@@ -926,7 +933,7 @@ static inline void move_to_free_list(struct page *page, struct zone *zone,
is_migrate_isolate(old_mt) != is_migrate_isolate(new_mt)) {
if (!is_migrate_isolate(old_mt))
nr_pages = -nr_pages;
- __mod_zone_page_state(zone, NR_FREE_PAGES_BLOCKS, nr_pages);
+ __mod_zone_page_state(zone, free_pages_blocks_stat(new_ft), nr_pages);
}
}
@@ -954,7 +961,7 @@ static inline void __del_page_from_free_list(struct page *page, struct zone *zon
zone->free_area[order].nr_free--;
if (order >= pageblock_order && !is_migrate_isolate(free_to_migratetype(freetype)))
- __mod_zone_page_state(zone, NR_FREE_PAGES_BLOCKS, -nr_pages);
+ __mod_zone_page_state(zone, free_pages_blocks_stat(freetype), -nr_pages);
}
static inline void del_page_from_free_list(struct page *page, struct zone *zone,
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 566c4e837c7d5..5789c39a0a729 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -6935,6 +6935,28 @@ static bool pgdat_watermark_boosted(pg_data_t *pgdat, int highest_zoneidx)
return false;
}
+/*
+ * Helper to get *FREE_PAGES* zone stats with accuracy heuristics.
+ *
+ * When there is a high number of CPUs in the system, the cumulative error from
+ * the vmstat per-cpu cache can blur the line between the watermarks. In that
+ * case, be safe and get an accurate snapshot.
+ *
+ * TODO: NR_FREE_PAGES_BLOCKS_* move in steps of pageblock_nr_pages, while the
+ * vmstat pcp threshold is limited to 125. On many configurations that counter
+ * won't actually be per-cpu cached. But keep things simple for now; revisit
+ * when somebody cares.
+ */
+static inline unsigned long get_free_pages_stat(struct zone *zone,
+ enum zone_stat_item item)
+{
+ unsigned long free_pages = zone_page_state(zone, item);
+
+ if (zone->percpu_drift_mark && free_pages < zone->percpu_drift_mark)
+ return zone_page_state_snapshot(zone, item);
+ return free_pages;
+}
+
/*
* Returns true if there is an eligible zone balanced for the request order
* and highest_zoneidx
@@ -6950,7 +6972,6 @@ static bool pgdat_balanced(pg_data_t *pgdat, int order, int highest_zoneidx)
* meet watermarks.
*/
for_each_managed_zone_pgdat(zone, pgdat, i, highest_zoneidx) {
- enum zone_stat_item item;
unsigned long free_pages;
if (sysctl_numa_balancing_mode & NUMA_BALANCING_MEMORY_TIERING)
@@ -6968,26 +6989,12 @@ static bool pgdat_balanced(pg_data_t *pgdat, int order, int highest_zoneidx)
* has dropped order, simply ensure there are enough
* base pages for compaction, wake kcompactd & sleep.
*/
- if (defrag_mode && order)
- item = NR_FREE_PAGES_BLOCKS;
- else
- item = NR_FREE_PAGES;
-
- /*
- * When there is a high number of CPUs in the system,
- * the cumulative error from the vmstat per-cpu cache
- * can blur the line between the watermarks. In that
- * case, be safe and get an accurate snapshot.
- *
- * TODO: NR_FREE_PAGES_BLOCKS moves in steps of
- * pageblock_nr_pages, while the vmstat pcp threshold
- * is limited to 125. On many configurations that
- * counter won't actually be per-cpu cached. But keep
- * things simple for now; revisit when somebody cares.
- */
- free_pages = zone_page_state(zone, item);
- if (zone->percpu_drift_mark && free_pages < zone->percpu_drift_mark)
- free_pages = zone_page_state_snapshot(zone, item);
+ if (defrag_mode && order) {
+ free_pages = get_free_pages_stat(zone, NR_FREE_PAGES_BLOCKS_UNMAPPED) +
+ get_free_pages_stat(zone, NR_FREE_PAGES_BLOCKS_MAPPED);
+ } else {
+ free_pages = get_free_pages_stat(zone, NR_FREE_PAGES);
+ }
if (__zone_watermark_ok(zone, order, mark, highest_zoneidx,
0, free_pages))
diff --git a/mm/vmstat.c b/mm/vmstat.c
index cb57714539fb5..f5ab6ab641c6d 100644
--- a/mm/vmstat.c
+++ b/mm/vmstat.c
@@ -1200,7 +1200,8 @@ const char * const vmstat_text[] = {
/* enum zone_stat_item counters */
#define I(x) (x)
[I(NR_FREE_PAGES)] = "nr_free_pages",
- [I(NR_FREE_PAGES_BLOCKS)] = "nr_free_pages_blocks",
+ [I(NR_FREE_PAGES_BLOCKS_MAPPED)] = "nr_free_pages_blocks_mapped",
+ [I(NR_FREE_PAGES_BLOCKS_UNMAPPED)] = "nr_free_pages_blocks_unmapped",
[I(NR_ZONE_INACTIVE_ANON)] = "nr_zone_inactive_anon",
[I(NR_ZONE_ACTIVE_ANON)] = "nr_zone_active_anon",
[I(NR_ZONE_INACTIVE_FILE)] = "nr_zone_inactive_file",
--
2.54.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v3 24/26] mm/page_alloc: always direct compact for unmapped allocs
2026-07-26 22:22 [PATCH v3 00/26] mm: Add ALLOC_UNMAPPED and AS_NO_DIRECT_MAP Brendan Jackman
` (22 preceding siblings ...)
2026-07-26 22:22 ` [PATCH v3 23/26] mm: Split out NR_FREE_PAGES_BLOCKS_[UN]MAPPED Brendan Jackman
@ 2026-07-26 22:22 ` Brendan Jackman
2026-07-26 22:22 ` [PATCH v3 25/26] mm: plumb alloc flags into some alloc funcs Brendan Jackman
2026-07-26 22:22 ` [PATCH v3 26/26] mm: add fast path for AS_NO_DIRECT_MAP Brendan Jackman
25 siblings, 0 replies; 29+ messages in thread
From: Brendan Jackman @ 2026-07-26 22:22 UTC (permalink / raw)
To: Borislav Petkov, Dave Hansen, Peter Zijlstra, Andrew Morton,
David Hildenbrand, Vlastimil Babka, Mike Rapoport, Wei Xu,
Johannes Weiner, Zi Yan, Lorenzo Stoakes
Cc: linux-mm, linux-kernel, x86, rppt, Sumit Garg, Will Deacon,
rientjes, Kalyazin, Nikita, patrick.roy, Itazuri, Takahiro,
Andy Lutomirski, David Kaplan, Thomas Gleixner, Yosry Ahmed,
Patrick Bellasi, Reiji Watanabe, Sean Christopherson,
Brendan Jackman
This is the minimal solution for ensuring that compaction can service
unmapped allocations. Without this, it's possible for compaction to just
check watermarks and see plenty of free pages, without being aware of
the direct map state, and thereby cause an ALLOC_UNMAPPED allocation to
fail unnecessarily.
Instead, with this change, promote compact_order to pageblock order for
unmapped allocations, much like defrag_mode. Then, check specifically in
compaction for the presence of wholly mapped blocks that can be unmapped
once direct compact is complete.
This all takes advantage of a major simplification: since unmapped
blocks are currently always unmovable, this can be asymmetric. There is
never a need to promote a !ALLOC_UNMAPPED allocation to compacting at
pageblock_order, because compaction would be trying to generate a
currently-unmapped block to map; that will always fail because it would
require migrating unmapped pages, which is not supported at the moment.
Signed-off-by: Brendan Jackman <jackmanb@google.com>
---
mm/compaction.c | 22 ++++++++++++++++++----
mm/page_alloc.c | 9 +++++++++
2 files changed, 27 insertions(+), 4 deletions(-)
diff --git a/mm/compaction.c b/mm/compaction.c
index ed12d2fc6fad3..fe1aaf293bbce 100644
--- a/mm/compaction.c
+++ b/mm/compaction.c
@@ -2531,12 +2531,25 @@ bool compaction_zonelist_suitable(struct alloc_context *ac, int order,
static enum compact_result
compaction_suit_allocation_order(struct zone *zone, unsigned int order,
int highest_zoneidx, unsigned int alloc_flags,
- bool async, bool kcompactd)
+ bool unmapped, bool async, bool kcompactd)
{
unsigned long free_pages;
unsigned long watermark;
- if (kcompactd && defrag_mode)
+ /*
+ * When trying to generate an unmapped block, check the counter for
+ * direct-mapped blocks specifically, since we'll need to unmap the
+ * whole block to service the allocation.
+ *
+ * Why doesn't this apply to the other way around too? (Mightn't we need
+ * to _map_ a whole block, to service a !ALLOC_UNMAPPED allocation?) No,
+ * because of a likely-temporary simplification: currently, unmapped
+ * blocks never contain movable pages, so compaction isn't going to free
+ * up one of those.
+ */
+ if (unmapped)
+ free_pages = zone_page_state(zone, NR_FREE_PAGES_BLOCKS_MAPPED);
+ else if (kcompactd && defrag_mode)
free_pages = zone_free_pages_blocks(zone);
else
free_pages = zone_page_state(zone, NR_FREE_PAGES);
@@ -2599,6 +2612,7 @@ compact_zone(struct compact_control *cc, struct capture_control *capc)
ret = compaction_suit_allocation_order(cc->zone, cc->order,
cc->highest_zoneidx,
cc->alloc_flags,
+ freetype_unmapped(cc->freetype),
cc->mode == MIGRATE_ASYNC,
!cc->direct_compaction);
if (ret != COMPACT_CONTINUE)
@@ -3084,7 +3098,7 @@ static bool kcompactd_node_suitable(pg_data_t *pgdat)
ret = compaction_suit_allocation_order(zone,
pgdat->kcompactd_max_order,
highest_zoneidx, alloc_flags,
- false, true);
+ false, false, true);
if (ret == COMPACT_CONTINUE)
return true;
}
@@ -3127,7 +3141,7 @@ static void kcompactd_do_work(pg_data_t *pgdat)
ret = compaction_suit_allocation_order(zone,
cc.order, zoneid, cc.alloc_flags,
- false, true);
+ false, false, true);
if (ret != COMPACT_CONTINUE)
continue;
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index d12ce84662ab7..5f1dea7eee15b 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -827,6 +827,9 @@ compaction_capture(struct capture_control *capc, struct page *page,
capc_mt != MIGRATE_MOVABLE)
return false;
+ if (freetype_flags(freetype) != freetype_flags(capc->freetype))
+ return false;
+
if (migratetype != capc_mt)
trace_mm_page_alloc_extfrag(page, capc->order, order,
capc_mt, migratetype);
@@ -4523,6 +4526,12 @@ __alloc_pages_direct_compact(gfp_t gfp_mask, unsigned int order,
if ((alloc_flags & ALLOC_NOFRAGMENT) &&
free_to_migratetype(ac->freetype) != MIGRATE_MOVABLE)
compact_order = max(order, pageblock_order);
+ /*
+ * Unmapped allocations benefit from compaction even at order 0, because the
+ * allocator will actually grab a whole block.
+ */
+ if (freetype_flags(ac->freetype) & FREETYPE_UNMAPPED)
+ compact_order = max(order, pageblock_order);
if (!compact_order)
return NULL;
--
2.54.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v3 25/26] mm: plumb alloc flags into some alloc funcs
2026-07-26 22:22 [PATCH v3 00/26] mm: Add ALLOC_UNMAPPED and AS_NO_DIRECT_MAP Brendan Jackman
` (23 preceding siblings ...)
2026-07-26 22:22 ` [PATCH v3 24/26] mm/page_alloc: always direct compact for unmapped allocs Brendan Jackman
@ 2026-07-26 22:22 ` Brendan Jackman
2026-07-26 22:22 ` [PATCH v3 26/26] mm: add fast path for AS_NO_DIRECT_MAP Brendan Jackman
25 siblings, 0 replies; 29+ messages in thread
From: Brendan Jackman @ 2026-07-26 22:22 UTC (permalink / raw)
To: Borislav Petkov, Dave Hansen, Peter Zijlstra, Andrew Morton,
David Hildenbrand, Vlastimil Babka, Mike Rapoport, Wei Xu,
Johannes Weiner, Zi Yan, Lorenzo Stoakes
Cc: linux-mm, linux-kernel, x86, rppt, Sumit Garg, Will Deacon,
rientjes, Kalyazin, Nikita, patrick.roy, Itazuri, Takahiro,
Andy Lutomirski, David Kaplan, Thomas Gleixner, Yosry Ahmed,
Patrick Bellasi, Reiji Watanabe, Sean Christopherson,
Brendan Jackman
A subsequent patch will want to set ALLOC_UNMAPPED in the allocator APIs
used by filemap.c. To keep that patch from getting overwhelming,
split out the arg plumbing into a separate one.
No functional change intended.
Signed-off-by: Brendan Jackman <jackmanb@google.com>
---
mm/filemap.c | 2 ++
mm/khugepaged.c | 2 +-
mm/mempolicy.c | 31 ++++++++++++++++++++-----------
mm/mempolicy.h | 24 ++++++++++++++++++++++++
mm/migrate.c | 2 +-
mm/page_alloc.c | 6 +++---
mm/page_alloc.h | 2 +-
7 files changed, 52 insertions(+), 17 deletions(-)
diff --git a/mm/filemap.c b/mm/filemap.c
index 341c7c5b28bc7..bf62fee570d8b 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -53,6 +53,8 @@
#include <asm/tlbflush.h>
#include "internal.h"
+#include "mempolicy.h"
+#include "page_alloc.h"
#define CREATE_TRACE_POINTS
#include <trace/events/filemap.h>
diff --git a/mm/khugepaged.c b/mm/khugepaged.c
index 27e8f3077e80f..89ce6bcbc376b 100644
--- a/mm/khugepaged.c
+++ b/mm/khugepaged.c
@@ -1243,7 +1243,7 @@ static enum scan_result alloc_charge_folio(struct folio **foliop, struct mm_stru
int node = collapse_find_target_node(cc);
struct folio *folio;
- folio = __folio_alloc(gfp, order, node, &cc->alloc_nmask);
+ folio = __folio_alloc(gfp, order, node, &cc->alloc_nmask, ALLOC_DEFAULT);
if (!folio) {
*foliop = NULL;
if (is_pmd_order(order))
diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index 5720f7f54d942..73aeda071f5d7 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -119,6 +119,7 @@
#include <linux/memory.h>
#include "internal.h"
+#include "mempolicy.h"
#include "page_alloc.h"
/* Internal flags */
@@ -2406,7 +2407,8 @@ bool mempolicy_in_oom_domain(struct task_struct *tsk,
}
static struct page *alloc_pages_preferred_many(gfp_t gfp, unsigned int order,
- int nid, nodemask_t *nodemask)
+ int nid, nodemask_t *nodemask,
+ unsigned int alloc_flags)
{
struct page *page;
gfp_t preferred_gfp;
@@ -2420,10 +2422,10 @@ static struct page *alloc_pages_preferred_many(gfp_t gfp, unsigned int order,
preferred_gfp = gfp | __GFP_NOWARN;
preferred_gfp &= ~(__GFP_DIRECT_RECLAIM | __GFP_NOFAIL);
page = __alloc_frozen_pages_noprof(preferred_gfp, order, nid, nodemask,
- ALLOC_DEFAULT);
+ alloc_flags);
if (!page)
page = __alloc_frozen_pages_noprof(gfp, order, nid, NULL,
- ALLOC_DEFAULT);
+ alloc_flags);
return page;
}
@@ -2435,11 +2437,12 @@ static struct page *alloc_pages_preferred_many(gfp_t gfp, unsigned int order,
* @pol: Pointer to the NUMA mempolicy.
* @ilx: Index for interleave mempolicy (also distinguishes alloc_pages()).
* @nid: Preferred node (usually numa_node_id() but @mpol may override it).
+ * @alloc_flags: ALLOC_* flags.
*
* Return: The page on success or NULL if allocation fails.
*/
static struct page *alloc_pages_mpol(gfp_t gfp, unsigned int order,
- struct mempolicy *pol, pgoff_t ilx, int nid)
+ struct mempolicy *pol, pgoff_t ilx, int nid, unsigned int alloc_flags)
{
nodemask_t *nodemask;
struct page *page;
@@ -2447,7 +2450,7 @@ static struct page *alloc_pages_mpol(gfp_t gfp, unsigned int order,
nodemask = policy_nodemask(gfp, pol, ilx, &nid);
if (pol->mode == MPOL_PREFERRED_MANY)
- return alloc_pages_preferred_many(gfp, order, nid, nodemask);
+ return alloc_pages_preferred_many(gfp, order, nid, nodemask, alloc_flags);
if (IS_ENABLED(CONFIG_TRANSPARENT_HUGEPAGE) &&
/* filter "hugepage" allocation, unless from alloc_pages() */
@@ -2471,7 +2474,7 @@ static struct page *alloc_pages_mpol(gfp_t gfp, unsigned int order,
*/
page = __alloc_frozen_pages_noprof(
gfp | __GFP_THISNODE | __GFP_NORETRY, order,
- nid, NULL, ALLOC_DEFAULT);
+ nid, NULL, alloc_flags);
if (page || !(gfp & __GFP_DIRECT_RECLAIM))
return page;
/*
@@ -2483,7 +2486,7 @@ static struct page *alloc_pages_mpol(gfp_t gfp, unsigned int order,
}
}
- page = __alloc_frozen_pages_noprof(gfp, order, nid, nodemask, ALLOC_DEFAULT);
+ page = __alloc_frozen_pages_noprof(gfp, order, nid, nodemask, alloc_flags);
if (unlikely(pol->mode == MPOL_INTERLEAVE ||
pol->mode == MPOL_WEIGHTED_INTERLEAVE) && page) {
@@ -2499,11 +2502,11 @@ static struct page *alloc_pages_mpol(gfp_t gfp, unsigned int order,
return page;
}
-struct folio *folio_alloc_mpol_noprof(gfp_t gfp, unsigned int order,
- struct mempolicy *pol, pgoff_t ilx, int nid)
+struct folio *__folio_alloc_mpol_noprof(gfp_t gfp, unsigned int order,
+ struct mempolicy *pol, pgoff_t ilx, int nid, unsigned int alloc_flags)
{
struct page *page = alloc_pages_mpol(gfp | __GFP_COMP, order, pol,
- ilx, nid);
+ ilx, nid, alloc_flags);
if (!page)
return NULL;
@@ -2511,6 +2514,12 @@ struct folio *folio_alloc_mpol_noprof(gfp_t gfp, unsigned int order,
return page_rmappable_folio(page);
}
+struct folio *folio_alloc_mpol_noprof(gfp_t gfp, unsigned int order,
+ struct mempolicy *pol, pgoff_t ilx, int nid)
+{
+ return __folio_alloc_mpol_noprof(gfp, order, pol, ilx, nid, ALLOC_DEFAULT);
+}
+
/**
* vma_alloc_folio - Allocate a folio for a VMA.
* @gfp: GFP flags.
@@ -2555,7 +2564,7 @@ struct page *alloc_frozen_pages_noprof(gfp_t gfp, unsigned order)
pol = get_task_policy(current);
return alloc_pages_mpol(gfp, order, pol, NO_INTERLEAVE_INDEX,
- numa_node_id());
+ numa_node_id(), ALLOC_DEFAULT);
}
/**
diff --git a/mm/mempolicy.h b/mm/mempolicy.h
new file mode 100644
index 0000000000000..92d0bb3640962
--- /dev/null
+++ b/mm/mempolicy.h
@@ -0,0 +1,24 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * mm-internal API for mempolicy.c. Public API lives in
+ * include/linux/mempolicy.h.
+ */
+#ifndef __MM_MEMPOLICY_H
+#define __MM_MEMPOLICY_H
+
+#include <linux/gfp.h>
+#include <linux/mempolicy.h>
+#include "page_alloc.h"
+
+#ifdef CONFIG_NUMA
+struct folio *__folio_alloc_mpol_noprof(gfp_t gfp, unsigned int order,
+ struct mempolicy *pol, pgoff_t ilx, int nid, unsigned int alloc_flags);
+#else
+static inline struct folio *__folio_alloc_mpol_noprof(gfp_t gfp, unsigned int order,
+ struct mempolicy *pol, pgoff_t ilx, int nid, unsigned int alloc_flags)
+{
+ return __folio_alloc_noprof(gfp, order, numa_node_id(), NULL, alloc_flags);
+}
+#endif
+
+#endif
diff --git a/mm/migrate.c b/mm/migrate.c
index 0c772571610d3..583e74cd179c0 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -2233,7 +2233,7 @@ struct folio *alloc_migration_target(struct folio *src, unsigned long private)
if (is_highmem_idx(zidx) || zidx == ZONE_MOVABLE)
gfp_mask |= __GFP_HIGHMEM;
- return __folio_alloc(gfp_mask, order, nid, mtc->nmask);
+ return __folio_alloc(gfp_mask, order, nid, mtc->nmask, ALLOC_DEFAULT);
}
#ifdef CONFIG_NUMA_MIGRATION
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 5f1dea7eee15b..f39b6af3a6b73 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -5888,7 +5888,7 @@ struct page *alloc_pages_node_noprof(int nid, gfp_t gfp_mask, unsigned int order
EXPORT_SYMBOL(alloc_pages_node_noprof);
struct folio *__folio_alloc_noprof(gfp_t gfp, unsigned int order, int preferred_nid,
- nodemask_t *nodemask)
+ nodemask_t *nodemask, unsigned int alloc_flags)
{
struct page *page;
@@ -5898,13 +5898,13 @@ struct folio *__folio_alloc_noprof(gfp_t gfp, unsigned int order, int preferred_
warn_if_node_offline(preferred_nid, gfp);
page = __alloc_pages_noprof(gfp | __GFP_COMP, order,
- preferred_nid, nodemask, ALLOC_DEFAULT);
+ preferred_nid, nodemask, alloc_flags);
return page_rmappable_folio(page);
}
struct folio *folio_alloc_node_noprof(gfp_t gfp, unsigned int order, int nid)
{
- return __folio_alloc_noprof(gfp, order, nid, NULL);
+ return __folio_alloc_noprof(gfp, order, nid, NULL, ALLOC_DEFAULT);
}
EXPORT_SYMBOL(folio_alloc_node_noprof);
diff --git a/mm/page_alloc.h b/mm/page_alloc.h
index 5fa8fc527347c..be83a974edc2b 100644
--- a/mm/page_alloc.h
+++ b/mm/page_alloc.h
@@ -279,7 +279,7 @@ struct page *__alloc_pages_noprof(gfp_t gfp, unsigned int order, int preferred_n
#define __alloc_pages(...) alloc_hooks(__alloc_pages_noprof(__VA_ARGS__))
struct folio *__folio_alloc_noprof(gfp_t gfp, unsigned int order, int preferred_nid,
- nodemask_t *nodemask);
+ nodemask_t *nodemask, unsigned int alloc_flags);
#define __folio_alloc(...) alloc_hooks(__folio_alloc_noprof(__VA_ARGS__))
extern void zone_pcp_reset(struct zone *zone);
--
2.54.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v3 26/26] mm: add fast path for AS_NO_DIRECT_MAP
2026-07-26 22:22 [PATCH v3 00/26] mm: Add ALLOC_UNMAPPED and AS_NO_DIRECT_MAP Brendan Jackman
` (24 preceding siblings ...)
2026-07-26 22:22 ` [PATCH v3 25/26] mm: plumb alloc flags into some alloc funcs Brendan Jackman
@ 2026-07-26 22:22 ` Brendan Jackman
25 siblings, 0 replies; 29+ messages in thread
From: Brendan Jackman @ 2026-07-26 22:22 UTC (permalink / raw)
To: Borislav Petkov, Dave Hansen, Peter Zijlstra, Andrew Morton,
David Hildenbrand, Vlastimil Babka, Mike Rapoport, Wei Xu,
Johannes Weiner, Zi Yan, Lorenzo Stoakes
Cc: linux-mm, linux-kernel, x86, rppt, Sumit Garg, Will Deacon,
rientjes, Kalyazin, Nikita, patrick.roy, Itazuri, Takahiro,
Andy Lutomirski, David Kaplan, Thomas Gleixner, Yosry Ahmed,
Patrick Bellasi, Reiji Watanabe, Sean Christopherson,
Brendan Jackman
Currently AS_NO_DIRECT_MAP maps and unmaps pages on-demand, which means
there is no control over the batching of TLB flushes. Where it's
supported, this patch switches it to use ALLOC_UNMAPPED instead. This
means that pages are provided by the allocator that are already absent
from the direct map, so TLB flushes are maximally amortised.
The tricky bit is that those pages need to be zeroed. The solution comes
from the mermap, which allows zeroing the pages before mapping them into
userspace, without mapping them in the direct map.
Any system that can set_direct_map_*() can also use ALLOC_UNMAPPED, but
the mermap is arch-specific and not supported everywhere, so the slow
path needs to stick around as a fallback for those systems.
Zeroing logic needs to differ between the fast and slow path:
- The fastpath must use the mermap while the slowpath can just zero via
the direct map.
- The existing usecase for AS_NO_DIRECT_MAP zeroes folios before freeing
them. That's because they've been restored to the direct map so they
are now vulnerable to sidechannel attacks and other read
vulnerabilities. That re-mapping doesn't happen in the ALLOC_UNMAPPED
case so zeroing isn't needed there.
The upshot of all the above is that it becomes very inconvenient for
AS_NO_DIRECT_MAP's _users_ to zero the pages. With ALLOC_UNMAPPED it's
not possible for the page allocator itself to zero the pages. So, this
leaves the page cache in the middle as the natural place to handle
zeroing.
Rather than create any complex semantics, this patch just tacks
the currently-desired zeroing logic on as a side-effect
AS_NO_DIRECT_MAP. Later, if users arise that have different zeroing
requirements, more AS_ flags could be added, or this logic could be
shuffled elsewhere to afford the necessary flexibility.
Signed-off-by: Brendan Jackman <jackmanb@google.com>
---
include/linux/mermap.h | 2 +
include/linux/pagemap.h | 38 ++++++++++++-----
mm/Kconfig | 14 ++++++-
mm/filemap.c | 108 ++++++++++++++++++++++++++++++++++++++++++------
mm/internal.h | 39 +++++++++++++++++
mm/mermap.c | 37 +++++++++++++++++
mm/page_alloc.c | 6 +++
mm/page_alloc.h | 3 ++
mm/secretmem.c | 6 ---
9 files changed, 222 insertions(+), 31 deletions(-)
diff --git a/include/linux/mermap.h b/include/linux/mermap.h
index 5457dcb8c9789..762a23b909f9c 100644
--- a/include/linux/mermap.h
+++ b/include/linux/mermap.h
@@ -28,6 +28,8 @@ static inline void *mermap_addr(struct mermap_alloc *alloc)
return (void *)alloc->base;
}
+void mermap_clear_folio(struct folio *folio);
+
/*
* arch_mermap_flush_tlb() is called before a part of the local CPU's mermap
* region is remapped to a new address. No other CPU is allowed to _access_ that
diff --git a/include/linux/pagemap.h b/include/linux/pagemap.h
index 011f6e34859cc..16a41dca15c15 100644
--- a/include/linux/pagemap.h
+++ b/include/linux/pagemap.h
@@ -211,7 +211,12 @@ enum mapping_flags {
AS_WRITEBACK_MAY_DEADLOCK_ON_RECLAIM = 9,
AS_KERNEL_FILE = 10, /* mapping for a fake kernel file that shouldn't
account usage to user cgroups */
- AS_NO_DIRECT_MAP = 11, /* Folios in the mapping are not in the direct map */
+ AS_NO_DIRECT_MAP = 11, /* Folios in the mapping are not in the direct map.
+ They are also zeroed before being added to the mapping, but
+ this is separate from and incompatible with __GFP_ZERO. They
+ are also zeroed again on removal if they get restored to the
+ direct map. */
+ AS_MERMAP_STALE = 12, /* A folio in the mapping may require a mermap TLB flush. */
/* Bits 16-25 are used for FOLIO_ORDER */
AS_FOLIO_ORDER_BITS = 5,
AS_FOLIO_ORDER_MIN = 16,
@@ -362,20 +367,13 @@ static inline gfp_t mapping_gfp_constraint(const struct address_space *mapping,
return mapping_gfp_mask(mapping) & gfp_mask;
}
-/*
- * This is non-atomic. Only to be used before the mapping is activated.
- * Probably needs a barrier...
- */
-static inline void mapping_set_gfp_mask(struct address_space *m, gfp_t mask)
-{
- m->gfp_mask = mask;
-}
-
static inline void mapping_set_no_direct_map(struct address_space *mapping)
{
WARN_ON(!can_set_direct_map());
/* folio_zap_direct_map() doesn't support large folios. */
WARN_ON(mapping_max_folio_order(mapping));
+ /* Incompatible with ALLOC_UNMAPPED. */
+ WARN_ON(mapping_gfp_mask(mapping) & __GFP_ZERO);
set_bit(AS_NO_DIRECT_MAP, &mapping->flags);
}
@@ -389,6 +387,26 @@ static inline bool vma_has_no_direct_map(const struct vm_area_struct *vma)
return vma->vm_file && mapping_no_direct_map(vma->vm_file->f_mapping);
}
+/*
+ * This is non-atomic. Only to be used before the mapping is activated.
+ * Probably needs a barrier...
+ */
+static inline void mapping_set_gfp_mask(struct address_space *m, gfp_t mask)
+{
+ WARN_ON(mask & __GFP_ZERO && mapping_no_direct_map(m));
+ m->gfp_mask = mask;
+}
+
+static inline void mapping_set_mermap_stale(struct address_space *m)
+{
+ set_bit(AS_MERMAP_STALE, &m->flags);
+}
+
+static inline bool mapping_grab_mermap_stale(struct address_space *m)
+{
+ return test_and_clear_bit(AS_MERMAP_STALE, &m->flags);
+}
+
/*
* There are some parts of the kernel which assume that PMD entries
* are exactly HPAGE_PMD_ORDER. Those should be fixed, but until then,
diff --git a/mm/Kconfig b/mm/Kconfig
index bdd3e083520c0..7f01d90e10ef0 100644
--- a/mm/Kconfig
+++ b/mm/Kconfig
@@ -1341,6 +1341,7 @@ config SECRETMEM
default y
bool "Enable memfd_secret() system call" if EXPERT
depends on ARCH_HAS_SET_DIRECT_MAP
+ select PAGECACHE_NO_DIRECT_MAP
help
Enable the memfd_secret() system call with the ability to create
memory areas visible only in the context of the owning process and
@@ -1507,12 +1508,19 @@ config MM_LOCAL_REGION
bool
depends on ARCH_SUPPORTS_MM_LOCAL_REGION
+# This doesn't directly enable any code, it just exists to combine other
+# features - anything that needs AS_NO_DIRECT_MAP should "select" this config
+# and, wherever the platform supports it, that will change the default of those
+# other features which make AS_NO_DIRECT_MAP more efficient.
+config PAGECACHE_NO_DIRECT_MAP
+ bool
+
config ARCH_SUPPORTS_MERMAP
bool
select ARCH_SUPPORTS_MM_LOCAL_REGION
config MERMAP
- bool
+ def_bool PAGECACHE_NO_DIRECT_MAP
depends on ARCH_SUPPORTS_MERMAP
select MM_LOCAL_REGION
@@ -1528,7 +1536,9 @@ config MERMAP_KUNIT_TEST
If unsure, say N.
config PAGE_ALLOC_UNMAPPED
- bool
+ # It doesn't actually depend on the mermap, but with current usecases
+ # one isn't useful without the other.
+ def_bool MERMAP
depends on !HIGHMEM
config PAGE_ALLOC_KUNIT_TEST
diff --git a/mm/filemap.c b/mm/filemap.c
index bf62fee570d8b..db66cd6466330 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -14,6 +14,7 @@
#include <linux/compiler.h>
#include <linux/dax.h>
#include <linux/fs.h>
+#include <linux/mermap.h>
#include <linux/set_memory.h>
#include <linux/sched/signal.h>
#include <linux/uaccess.h>
@@ -242,13 +243,56 @@ static void filemap_free_folio(const struct address_space *mapping,
folio_put_refs(folio, folio_nr_pages(folio));
}
-#ifdef CONFIG_ARCH_HAS_SET_DIRECT_MAP
+/* Fast version: pages are already unmapped, but need zeroing. */
+#if defined(CONFIG_MERMAP)
+static inline int prep_add_unmapped_folio(struct address_space *mapping,
+ struct folio *folio)
+{
+ int err;
+
+ if (!mapping_no_direct_map(mapping))
+ return 0;
+
+ err = mermap_mm_prepare(current->mm);
+ if (err)
+ return err;
+
+ mermap_clear_folio(folio);
+ mapping_set_mermap_stale(mapping);
+ return 0;
+}
+
+static inline void prep_remove_unmapped_folio(struct address_space *mapping,
+ struct folio *folio_ignored)
+{
+ if (!mapping_no_direct_map(mapping))
+ return;
+
+ /* Folio is not going back in the direct map so no need to zero it here. */
+
+ mapping_check_flush_mermap(mapping);
+}
+
+static inline void prep_remove_unmapped_batch(struct address_space *mapping,
+ struct folio_batch *fbatch)
+{
+ prep_remove_unmapped_folio(mapping, NULL);
+}
+/* Slow version: zap and flush direct map on-demand. */
+#elif defined(CONFIG_ARCH_HAS_SET_DIRECT_MAP)
static inline int prep_add_unmapped_folio(struct address_space *mapping,
struct folio *folio)
{
if (!mapping_no_direct_map(mapping))
return 0;
+ /*
+ * Note under this configuration, we could have just allocated with
+ * __GFP_ZERO. But for consistency with the ALLOC_UNMAPPED version it's
+ * forbidden, so zero manually.
+ */
+ folio_zero_segment(folio, 0, folio_size(folio));
+
return folio_zap_direct_map(folio);
}
@@ -259,6 +303,7 @@ static inline void prep_remove_unmapped_folio(struct address_space *mapping,
return;
folio_restore_direct_map(folio);
+ folio_zero_segment(folio, 0, folio_size(folio));
}
static inline void prep_remove_unmapped_batch(struct address_space *mapping,
@@ -267,26 +312,31 @@ static inline void prep_remove_unmapped_batch(struct address_space *mapping,
if (!mapping_no_direct_map(mapping))
return;
- for (int i = 0; i < folio_batch_count(fbatch); i++)
- folio_restore_direct_map(fbatch->folios[i]);
+ for (int i = 0; i < folio_batch_count(fbatch); i++) {
+ struct folio *folio = fbatch->folios[i];
+
+ folio_restore_direct_map(folio);
+ folio_zero_segment(folio, 0, folio_size(folio));
+ }
}
+/* AS_NO_DIRECT_MAP unsupported. */
#else
static inline int prep_add_unmapped_folio(struct address_space *mapping, struct folio *folio)
{
- VM_WARN_ON(mapping_no_direct_map(mapping));
+ VM_WARN_ON(!IS_ENABLED(CONFIG_PAGE_ALLOC_UNMAPPED) && mapping_no_direct_map(mapping));
return 0;
}
static inline void prep_remove_unmapped_folio(struct address_space *mapping,
struct folio *folio)
{
- VM_WARN_ON(mapping_no_direct_map(mapping));
+ VM_WARN_ON(!IS_ENABLED(CONFIG_PAGE_ALLOC_UNMAPPED) && mapping_no_direct_map(mapping));
}
static inline void prep_remove_unmapped_batch(struct address_space *mapping,
struct folio_batch *fbatch)
{
- VM_WARN_ON(mapping_no_direct_map(mapping));
+ VM_WARN_ON(!IS_ENABLED(CONFIG_PAGE_ALLOC_UNMAPPED) && mapping_no_direct_map(mapping));
}
#endif
@@ -1055,31 +1105,53 @@ int filemap_add_folio(struct address_space *mapping, struct folio *folio,
EXPORT_SYMBOL_GPL(filemap_add_folio);
#ifdef CONFIG_NUMA
-struct folio *filemap_alloc_folio_noprof(gfp_t gfp, unsigned int order,
- struct mempolicy *policy)
+static inline
+struct folio *__filemap_alloc_folio_noprof(gfp_t gfp, unsigned int order,
+ struct mempolicy *policy, unsigned int alloc_flags)
{
int n;
struct folio *folio;
if (policy)
- return folio_alloc_mpol_noprof(gfp, order, policy,
- NO_INTERLEAVE_INDEX, numa_node_id());
+ return __folio_alloc_mpol_noprof(gfp, order, policy,
+ NO_INTERLEAVE_INDEX, numa_node_id(), alloc_flags);
if (cpuset_do_page_mem_spread()) {
unsigned int cpuset_mems_cookie;
do {
cpuset_mems_cookie = read_mems_allowed_begin();
n = cpuset_mem_spread_node();
- folio = folio_alloc_node_noprof(gfp, order, n);
+ folio = __folio_alloc_node_noprof(gfp, order, n, alloc_flags);
} while (!folio && read_mems_allowed_retry(cpuset_mems_cookie));
return folio;
}
- return folio_alloc_noprof(gfp, order);
+
+ if ((gfp & __GFP_THISNODE))
+ return __folio_alloc_noprof(gfp, order, numa_node_id(), NULL, alloc_flags);
+
+ return __folio_alloc_mpol_noprof(gfp, order, get_task_policy(current),
+ NO_INTERLEAVE_INDEX, numa_node_id(), alloc_flags);
+}
+
+struct folio *filemap_alloc_folio_noprof(gfp_t gfp, unsigned int order,
+ struct mempolicy *policy)
+{
+ return __filemap_alloc_folio_noprof(gfp, order, policy, ALLOC_DEFAULT);
}
EXPORT_SYMBOL(filemap_alloc_folio_noprof);
+#else
+static inline
+struct folio *__filemap_alloc_folio_noprof(gfp_t gfp, unsigned int order,
+ struct mempolicy *policy, unsigned int alloc_flags)
+{
+ return __folio_alloc_noprof(gfp, order, numa_node_id(), NULL, alloc_flags);
+}
#endif
+#define __filemap_alloc_folio(...) \
+ alloc_hooks(__filemap_alloc_folio_noprof(__VA_ARGS__))
+
/*
* filemap_invalidate_lock_two - lock invalidate_lock for two mappings
*
@@ -1986,6 +2058,15 @@ void *filemap_get_entry(struct address_space *mapping, pgoff_t index)
return folio;
}
+static inline unsigned int mapping_alloc_flags(struct address_space *mapping)
+{
+#ifdef CONFIG_PAGE_ALLOC_UNMAPPED
+ if (IS_ENABLED(CONFIG_MERMAP) && mapping_no_direct_map(mapping))
+ return ALLOC_UNMAPPED;
+#endif
+ return ALLOC_DEFAULT;
+}
+
/**
* __filemap_get_folio_mpol - Find and get a reference to a folio.
* @mapping: The address_space to search.
@@ -2074,7 +2155,8 @@ struct folio *__filemap_get_folio_mpol(struct address_space *mapping,
err = -ENOMEM;
if (order > min_order)
alloc_gfp |= __GFP_NORETRY | __GFP_NOWARN;
- folio = filemap_alloc_folio(alloc_gfp, order, policy);
+ folio = __filemap_alloc_folio(alloc_gfp, order, policy,
+ mapping_alloc_flags(mapping));
if (!folio)
continue;
diff --git a/mm/internal.h b/mm/internal.h
index 089c53bf9a336..eea26ee57f765 100644
--- a/mm/internal.h
+++ b/mm/internal.h
@@ -255,6 +255,43 @@ static inline int mmap_file(struct file *file, struct vm_area_struct *vma)
return err;
}
+#ifdef CONFIG_MERMAP
+static inline void mapping_check_flush_mermap(struct address_space *mapping)
+{
+ /*
+ * Note this flush is hugely over-aggressive: only the mermap region
+ * needs flushing, but assume that flushing the whole address space is
+ * faster. Also only certain mm's (most of the time, just current->mm)
+ * actually have stale entries, but assume the benefit of tracking that
+ * would be minimal.
+ *
+ * Probably more important: the TLB has likely already been flushed
+ * anyway for unrelated reasons since the mermap got used. If/when this
+ * is shown to matter, the simplistic address space flag will need to be
+ * replaced with something more deeply integrated into other TLB
+ * flushing logic to enable proper amortisation.
+ */
+ if (mapping_grab_mermap_stale(mapping))
+ flush_tlb_all();
+}
+
+/*
+ * VMA is being closed, i.e. mm might be losing logical access to the contents.
+ * For AS_NO_DIRECT_MAP, the folios were mermapped so stale TLB entries need to
+ * be removed to prevent CPU sidechannel leaks.
+ */
+static inline void vma_check_flush_mermap(struct vm_area_struct *vma)
+{
+ if (!vma->vm_file || !vma->vm_file->f_mapping ||
+ !mapping_no_direct_map(vma->vm_file->f_mapping))
+ return;
+
+ mapping_check_flush_mermap(vma->vm_file->f_mapping);
+}
+#else
+static inline void vma_check_flush_mermap(struct vm_area_struct *vma) { }
+#endif
+
/*
* If the VMA has a close hook then close it, and since closing it might leave
* it in an inconsistent state which makes the use of any hooks suspect, clear
@@ -262,6 +299,8 @@ static inline int mmap_file(struct file *file, struct vm_area_struct *vma)
*/
static inline void vma_close(struct vm_area_struct *vma)
{
+ vma_check_flush_mermap(vma);
+
if (vma->vm_ops && vma->vm_ops->close) {
vma->vm_ops->close(vma);
diff --git a/mm/mermap.c b/mm/mermap.c
index 2bead38eadfe8..2ac1d59c1b1cd 100644
--- a/mm/mermap.c
+++ b/mm/mermap.c
@@ -336,3 +336,40 @@ void mermap_mm_teardown(struct mm_struct *mm)
free_percpu(mm->mermap.cpu);
}
+
+/*
+ * Zero a folio via the mermap.
+ *
+ * This should be decoupled from the mermap implementation; it could be moved
+ * outside mermap.c if a better place arises to put it.
+ */
+void mermap_clear_folio(struct folio *folio)
+{
+ unsigned int numpages = folio_nr_pages(folio);
+ struct page *page = folio_page(folio, 0);
+ void *mermap;
+
+ BUILD_BUG_ON(IS_ENABLED(CONFIG_HIGHMEM));
+
+ /* Fast path: single mapping (may fail under preemption). */
+ scoped_guard(migrate) {
+ mermap = mermap_get(page, numpages << PAGE_SHIFT, PAGE_KERNEL_NOGLOBAL);
+ if (mermap) {
+ void *buf = kasan_reset_tag(mermap_addr(mermap));
+
+ for (int i = 0; i < numpages; i++)
+ clear_page(buf + (i << PAGE_SHIFT));
+ mermap_put(mermap);
+ return;
+ }
+ }
+
+ /* Slow path, map each page individually (always succeeds). */
+ for (int i = 0; i < numpages; i++) {
+ scoped_guard(preempt) {
+ mermap = mermap_get_reserved(page + i, PAGE_KERNEL_NOGLOBAL);
+ clear_page(kasan_reset_tag(mermap_addr(mermap)));
+ mermap_put(mermap);
+ }
+ }
+}
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index f39b6af3a6b73..4f923a7a18459 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -5902,6 +5902,12 @@ struct folio *__folio_alloc_noprof(gfp_t gfp, unsigned int order, int preferred_
return page_rmappable_folio(page);
}
+struct folio *__folio_alloc_node_noprof(gfp_t gfp, unsigned int order, int nid,
+ unsigned int alloc_flags)
+{
+ return __folio_alloc_noprof(gfp, order, nid, NULL, alloc_flags);
+}
+
struct folio *folio_alloc_node_noprof(gfp_t gfp, unsigned int order, int nid)
{
return __folio_alloc_noprof(gfp, order, nid, NULL, ALLOC_DEFAULT);
diff --git a/mm/page_alloc.h b/mm/page_alloc.h
index be83a974edc2b..c4bf3a83c68d5 100644
--- a/mm/page_alloc.h
+++ b/mm/page_alloc.h
@@ -282,6 +282,9 @@ struct folio *__folio_alloc_noprof(gfp_t gfp, unsigned int order, int preferred_
nodemask_t *nodemask, unsigned int alloc_flags);
#define __folio_alloc(...) alloc_hooks(__folio_alloc_noprof(__VA_ARGS__))
+struct folio *__folio_alloc_node_noprof(gfp_t gfp, unsigned int order, int nid,
+ unsigned int alloc_flags);
+
extern void zone_pcp_reset(struct zone *zone);
extern void zone_pcp_disable(struct zone *zone);
extern void zone_pcp_enable(struct zone *zone);
diff --git a/mm/secretmem.c b/mm/secretmem.c
index c043c53687d95..798f1766bdc27 100644
--- a/mm/secretmem.c
+++ b/mm/secretmem.c
@@ -111,14 +111,8 @@ static int secretmem_migrate_folio(struct address_space *mapping,
return -EBUSY;
}
-static void secretmem_free_folio(struct folio *folio)
-{
- folio_zero_segment(folio, 0, folio_size(folio));
-}
-
static const struct address_space_operations secretmem_aops = {
.dirty_folio = noop_dirty_folio,
- .free_folio = secretmem_free_folio,
.migrate_folio = secretmem_migrate_folio,
};
--
2.54.0
^ permalink raw reply related [flat|nested] 29+ messages in thread