* [PATCH v2 0/5] mm: Unconditional per-VMA locks and cleanups
@ 2026-06-10 23:04 Dave Hansen
2026-06-10 23:04 ` [PATCH v2 1/5] mm: Make per-VMA locks available universally Dave Hansen
` (4 more replies)
0 siblings, 5 replies; 7+ messages in thread
From: Dave Hansen @ 2026-06-10 23:04 UTC (permalink / raw)
To: linux-kernel
Cc: Dave Hansen, Alice Ryhl, Andrew Morton, Arve Hjønnevåg,
Carlos Llamas, Christian Brauner, David Ahern, David S. Miller,
Greg Kroah-Hartman, Liam R. Howlett, linux-mm, Lorenzo Stoakes,
netdev, Shakeel Butt, Suren Baghdasaryan, Todd Kjos,
Vlastimil Babka
[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 4328 bytes --]
tl;dr: Make per-VMA locks available in all configs. Simplify some
of the per-VMA lock users now that they can rely on them being
always available.
Binder and networking folks: Your code is the target of the cleanups.
I'm cc'ing you now on v2 because there's emerging consensus on the mm
side that the approach here is sane. I'm not quite sure how this pile
would get merged, but ack/review tags would be appreciated if this
looks good to you.
Longer version:
When working on some x86 shadow stack code, it was a real pain to
avoid causing recursive locking problems with mmap_lock. One way
to avoid those was to avoid mmap_lock and use per-VMA locks instead.
They are great, but they are not available in all configs which
makes them unusable in generic code, or if you want to completely
avoid mmap_lock.
Make per-VMA locks available in all configs. Right now, they are
only available on select architectures when SMP and MMU are enabled.
But all of the primitives that per-VMA locks are built on (RCU, maple
trees, refcounts) work just fine without SMP or MMU.
The only real downside is that making VMAs a wee bit bigger on !MMU
and !SMP builds.
The upside is much cleaner code, lower complexity and less #ifdeffery.
Clean up a binder VMA locking site now that it can rely on per-VMA
locks.
Building on top of universally-available per-VMA locks, introduce a
new helper. Since the new API does not require callers to have a
fallback to mmap_lock, it's much easier to use. Callers can
potentially replace this very common kernel idiom:
mmap_read_lock(mm);
vma = vma_lookup()
// fiddle with vma
mmap_read_unlock(mm);
with:
vma = vma_start_read_unlocked(mm, address);
// fiddle with vma
vma_end_read(vma);
Which avoids mmap_lock entirely in the fast path.
Use that new API for another binder site and one in the TCP code.
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: "Liam R. Howlett" <Liam.Howlett@oracle.com>
Cc: Lorenzo Stoakes <ljs@kernel.org>
Cc: Vlastimil Babka <vbabka@kernel.org>
Cc: Shakeel Butt <shakeel.butt@linux.dev>
Cc: linux-mm@kvack.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Arve Hjønnevåg <arve@android.com>
Cc: Todd Kjos <tkjos@android.com>
Cc: Christian Brauner <christian@brauner.io>
Cc: Carlos Llamas <cmllamas@google.com>
Cc: Alice Ryhl <aliceryhl@google.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: David Ahern <dsahern@kernel.org>
Cc: netdev@vger.kernel.org
Changes from v1:
* Better naming and non-loopy, simpler implementation.
Thanks Suren and Lorenzo!
* Cc networking and binder folks
* Add tags. Thanks reviewers!
* Drop x86 shadow stack changes
arch/arm/Kconfig | 1
arch/arm64/Kconfig | 1
arch/loongarch/Kconfig | 1
arch/powerpc/platforms/powernv/Kconfig | 1
arch/powerpc/platforms/pseries/Kconfig | 1
arch/riscv/Kconfig | 1
arch/s390/Kconfig | 1
arch/x86/Kconfig | 2 -
drivers/android/binder_alloc.c | 43 ++++++++-----------------
fs/proc/internal.h | 2 -
fs/proc/task_mmu.c | 51 ------------------------------
include/linux/mm.h | 12 -------
include/linux/mm_types.h | 7 ----
include/linux/mmap_lock.h | 51 +-----------------------------
kernel/bpf/task_iter.c | 5 ---
kernel/fork.c | 2 -
mm/Kconfig | 13 -------
mm/Kconfig.debug | 1
mm/debug.c | 4 --
mm/init-mm.c | 2 -
mm/memory.c | 2 -
mm/mmap_lock.c | 51 ++++++++++++++++--------------
mm/pagewalk.c | 2 -
mm/rmap.c | 2 -
mm/userfaultfd.c | 55 ---------------------------------
net/ipv4/tcp.c | 31 +++++-------------
rust/kernel/mm.rs | 7 ----
tools/testing/vma/include/dup.h | 4 --
tools/testing/vma/vma_internal.h | 1
29 files changed, 54 insertions(+), 303 deletions(-)
^ permalink raw reply [flat|nested] 7+ messages in thread
* [PATCH v2 1/5] mm: Make per-VMA locks available universally
2026-06-10 23:04 [PATCH v2 0/5] mm: Unconditional per-VMA locks and cleanups Dave Hansen
@ 2026-06-10 23:04 ` Dave Hansen
2026-06-10 23:04 ` [PATCH v2 2/5] binder: Make shrinker rely solely on per-VMA lock Dave Hansen
` (3 subsequent siblings)
4 siblings, 0 replies; 7+ messages in thread
From: Dave Hansen @ 2026-06-10 23:04 UTC (permalink / raw)
To: linux-kernel
Cc: Dave Hansen, Alice Ryhl, Andrew Morton, Arve Hjønnevåg,
Carlos Llamas, Christian Brauner, David Ahern, David S. Miller,
Greg Kroah-Hartman, Liam R. Howlett, linux-mm, Lorenzo Stoakes,
netdev, Shakeel Butt, Suren Baghdasaryan, Todd Kjos,
Vlastimil Babka
[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 27337 bytes --]
From: Dave Hansen <dave.hansen@linux.intel.com>
The per-VMA locks have been around for several years. They've had some
bugs worked out of them and have seen quite wide use. However, they
are still only available when architectures explicitly enable them.
Remove the conditional compilation around the per-VMA locks, making
them available on all architectures and configs.
The approach up to now seemed to be to add ARCH_SUPPORTS_PER_VMA_LOCK
when the architecture started using per-VMA locks in the fault
handler. But, contrary to the naming, the Kconfig option does not
really indicate whether the architecture supports per-VMA locks or
not. It is more of a marker for whether the architecture is likely to
benefit from per-VMA locks.
To me, the most important thing side-effect of universal availability
is letting per-VMA locks be used in SMP=n configs. This lets us use
per-VMA locking in all x86 code without fallbacks.
Overall, this just generally makes the kernel simpler. Just look at
the diffstat. It also opens the door to users that want to use the
per-VMA locks in common code. Doing *that* brings additional
simplifications.
The downside of this is adding some fields to vm_area_struct and
mm_struct. There are likely ways to optimize this, especially for
things like SMP=n configs. For now, do the simplest thing: use the
same implementation everywhere.
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Reviewed-by: Suren Baghdasaryan <surenb@google.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: "Liam R. Howlett" <Liam.Howlett@oracle.com>
Cc: Lorenzo Stoakes <ljs@kernel.org>
Cc: Vlastimil Babka <vbabka@kernel.org>
Cc: Shakeel Butt <shakeel.butt@linux.dev>
Cc: linux-mm@kvack.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Arve Hjønnevåg <arve@android.com>
Cc: Todd Kjos <tkjos@android.com>
Cc: Christian Brauner <christian@brauner.io>
Cc: Carlos Llamas <cmllamas@google.com>
Cc: Alice Ryhl <aliceryhl@google.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: David Ahern <dsahern@kernel.org>
Cc: netdev@vger.kernel.org
--
Changes from v1:
* Remove a bunch of left over CONFIG_PER_VMA_LOCKs
* Trim some speculation out of the changelog
---
b/arch/arm/Kconfig | 1
b/arch/arm64/Kconfig | 1
b/arch/loongarch/Kconfig | 1
b/arch/powerpc/platforms/powernv/Kconfig | 1
b/arch/powerpc/platforms/pseries/Kconfig | 1
b/arch/riscv/Kconfig | 1
b/arch/s390/Kconfig | 1
b/arch/x86/Kconfig | 2 -
b/fs/proc/internal.h | 2 -
b/fs/proc/task_mmu.c | 51 ----------------------------
b/include/linux/mm.h | 12 ------
b/include/linux/mm_types.h | 7 ---
b/include/linux/mmap_lock.h | 48 ---------------------------
b/kernel/bpf/task_iter.c | 5 --
b/kernel/fork.c | 2 -
b/mm/Kconfig | 13 -------
b/mm/Kconfig.debug | 1
b/mm/debug.c | 4 --
b/mm/init-mm.c | 2 -
b/mm/memory.c | 2 -
b/mm/mmap_lock.c | 24 -------------
b/mm/pagewalk.c | 2 -
b/mm/rmap.c | 2 -
b/mm/userfaultfd.c | 55 -------------------------------
b/rust/kernel/mm.rs | 7 ---
b/tools/testing/vma/include/dup.h | 4 --
b/tools/testing/vma/vma_internal.h | 1
27 files changed, 1 insertion(+), 252 deletions(-)
diff -puN arch/arm64/Kconfig~unconditional-vma-locks arch/arm64/Kconfig
--- a/arch/arm64/Kconfig~unconditional-vma-locks 2026-06-10 15:57:53.491348630 -0700
+++ b/arch/arm64/Kconfig 2026-06-10 15:57:54.069369179 -0700
@@ -80,7 +80,6 @@ config ARM64
select ARCH_SUPPORTS_INT128 if CC_HAS_INT128
select ARCH_SUPPORTS_NUMA_BALANCING
select ARCH_SUPPORTS_PAGE_TABLE_CHECK
- select ARCH_SUPPORTS_PER_VMA_LOCK
select ARCH_SUPPORTS_HUGE_PFNMAP if TRANSPARENT_HUGEPAGE
select ARCH_SUPPORTS_RT
select ARCH_SUPPORTS_SCHED_SMT
diff -puN arch/arm/Kconfig~unconditional-vma-locks arch/arm/Kconfig
--- a/arch/arm/Kconfig~unconditional-vma-locks 2026-06-10 15:57:53.499348914 -0700
+++ b/arch/arm/Kconfig 2026-06-10 15:57:54.070369215 -0700
@@ -41,7 +41,6 @@ config ARM
select ARCH_SUPPORTS_ATOMIC_RMW
select ARCH_SUPPORTS_CFI
select ARCH_SUPPORTS_HUGETLBFS if ARM_LPAE
- select ARCH_SUPPORTS_PER_VMA_LOCK
select ARCH_SUPPORTS_RT
select ARCH_USE_BUILTIN_BSWAP
select ARCH_USE_CMPXCHG_LOCKREF
diff -puN arch/loongarch/Kconfig~unconditional-vma-locks arch/loongarch/Kconfig
--- a/arch/loongarch/Kconfig~unconditional-vma-locks 2026-06-10 15:57:53.542350439 -0700
+++ b/arch/loongarch/Kconfig 2026-06-10 15:57:54.070369215 -0700
@@ -68,7 +68,6 @@ config LOONGARCH
select ARCH_SUPPORTS_LTO_CLANG_THIN
select ARCH_SUPPORTS_MSEAL_SYSTEM_MAPPINGS
select ARCH_SUPPORTS_NUMA_BALANCING if NUMA
- select ARCH_SUPPORTS_PER_VMA_LOCK
select ARCH_SUPPORTS_RT
select ARCH_SUPPORTS_SCHED_SMT if SMP
select ARCH_SUPPORTS_SCHED_MC if SMP
diff -puN arch/powerpc/platforms/powernv/Kconfig~unconditional-vma-locks arch/powerpc/platforms/powernv/Kconfig
--- a/arch/powerpc/platforms/powernv/Kconfig~unconditional-vma-locks 2026-06-10 15:57:53.544350510 -0700
+++ b/arch/powerpc/platforms/powernv/Kconfig 2026-06-10 15:57:54.070369215 -0700
@@ -17,7 +17,6 @@ config PPC_POWERNV
select PPC_DOORBELL
select MMU_NOTIFIER
select FORCE_SMP
- select ARCH_SUPPORTS_PER_VMA_LOCK
select PPC_RADIX_BROADCAST_TLBIE if PPC_RADIX_MMU
default y
diff -puN arch/powerpc/platforms/pseries/Kconfig~unconditional-vma-locks arch/powerpc/platforms/pseries/Kconfig
--- a/arch/powerpc/platforms/pseries/Kconfig~unconditional-vma-locks 2026-06-10 15:57:53.552350794 -0700
+++ b/arch/powerpc/platforms/pseries/Kconfig 2026-06-10 15:57:54.070369215 -0700
@@ -23,7 +23,6 @@ config PPC_PSERIES
select HOTPLUG_CPU
select FORCE_SMP
select SWIOTLB
- select ARCH_SUPPORTS_PER_VMA_LOCK
select PPC_RADIX_BROADCAST_TLBIE if PPC_RADIX_MMU
default y
diff -puN arch/riscv/Kconfig~unconditional-vma-locks arch/riscv/Kconfig
--- a/arch/riscv/Kconfig~unconditional-vma-locks 2026-06-10 15:57:53.559351043 -0700
+++ b/arch/riscv/Kconfig 2026-06-10 15:57:54.070369215 -0700
@@ -70,7 +70,6 @@ config RISCV
select ARCH_SUPPORTS_LTO_CLANG_THIN
select ARCH_SUPPORTS_MSEAL_SYSTEM_MAPPINGS if 64BIT && MMU
select ARCH_SUPPORTS_PAGE_TABLE_CHECK if MMU
- select ARCH_SUPPORTS_PER_VMA_LOCK if MMU
select ARCH_SUPPORTS_RT
select ARCH_SUPPORTS_SHADOW_CALL_STACK if HAVE_SHADOW_CALL_STACK
select ARCH_SUPPORTS_SCHED_MC if SMP
diff -puN arch/s390/Kconfig~unconditional-vma-locks arch/s390/Kconfig
--- a/arch/s390/Kconfig~unconditional-vma-locks 2026-06-10 15:57:53.571351470 -0700
+++ b/arch/s390/Kconfig 2026-06-10 15:57:54.071369250 -0700
@@ -153,7 +153,6 @@ config S390
select ARCH_SUPPORTS_MSEAL_SYSTEM_MAPPINGS
select ARCH_SUPPORTS_NUMA_BALANCING
select ARCH_SUPPORTS_PAGE_TABLE_CHECK
- select ARCH_SUPPORTS_PER_VMA_LOCK
select ARCH_USE_BUILTIN_BSWAP
select ARCH_USE_CMPXCHG_LOCKREF
select ARCH_USE_SYM_ANNOTATIONS
diff -puN arch/x86/Kconfig~unconditional-vma-locks arch/x86/Kconfig
--- a/arch/x86/Kconfig~unconditional-vma-locks 2026-06-10 15:57:53.577351684 -0700
+++ b/arch/x86/Kconfig 2026-06-10 15:57:54.071369250 -0700
@@ -27,7 +27,6 @@ config X86_64
select ARCH_HAS_GIGANTIC_PAGE
select ARCH_SUPPORTS_MSEAL_SYSTEM_MAPPINGS
select ARCH_SUPPORTS_INT128 if CC_HAS_INT128
- select ARCH_SUPPORTS_PER_VMA_LOCK
select ARCH_SUPPORTS_HUGE_PFNMAP if TRANSPARENT_HUGEPAGE
select HAVE_ARCH_SOFT_DIRTY
select MODULES_USE_ELF_RELA
@@ -1885,7 +1884,6 @@ config X86_USER_SHADOW_STACK
bool "X86 userspace shadow stack"
depends on AS_WRUSS
depends on X86_64
- depends on PER_VMA_LOCK
select ARCH_USES_HIGH_VMA_FLAGS
select ARCH_HAS_USER_SHADOW_STACK
select X86_CET
diff -puN fs/proc/internal.h~unconditional-vma-locks fs/proc/internal.h
--- a/fs/proc/internal.h~unconditional-vma-locks 2026-06-10 15:57:53.579351755 -0700
+++ b/fs/proc/internal.h 2026-06-10 15:57:54.071369250 -0700
@@ -382,10 +382,8 @@ struct mem_size_stats;
struct proc_maps_locking_ctx {
struct mm_struct *mm;
-#ifdef CONFIG_PER_VMA_LOCK
bool mmap_locked;
struct vm_area_struct *locked_vma;
-#endif
};
struct proc_maps_private {
diff -puN fs/proc/task_mmu.c~unconditional-vma-locks fs/proc/task_mmu.c
--- a/fs/proc/task_mmu.c~unconditional-vma-locks 2026-06-10 15:57:53.594352288 -0700
+++ b/fs/proc/task_mmu.c 2026-06-10 15:57:54.072369286 -0700
@@ -130,8 +130,6 @@ static void release_task_mempolicy(struc
}
#endif
-#ifdef CONFIG_PER_VMA_LOCK
-
static void reset_lock_ctx(struct proc_maps_locking_ctx *lock_ctx)
{
lock_ctx->locked_vma = NULL;
@@ -213,33 +211,6 @@ static inline bool fallback_to_mmap_lock
return true;
}
-#else /* CONFIG_PER_VMA_LOCK */
-
-static inline bool lock_vma_range(struct seq_file *m,
- struct proc_maps_locking_ctx *lock_ctx)
-{
- return mmap_read_lock_killable(lock_ctx->mm) == 0;
-}
-
-static inline void unlock_vma_range(struct proc_maps_locking_ctx *lock_ctx)
-{
- mmap_read_unlock(lock_ctx->mm);
-}
-
-static struct vm_area_struct *get_next_vma(struct proc_maps_private *priv,
- loff_t last_pos)
-{
- return vma_next(&priv->iter);
-}
-
-static inline bool fallback_to_mmap_lock(struct proc_maps_private *priv,
- loff_t pos)
-{
- return false;
-}
-
-#endif /* CONFIG_PER_VMA_LOCK */
-
static struct vm_area_struct *proc_get_vma(struct seq_file *m, loff_t *ppos)
{
struct proc_maps_private *priv = m->private;
@@ -527,8 +498,6 @@ static int pid_maps_open(struct inode *i
PROCMAP_QUERY_VMA_FLAGS \
)
-#ifdef CONFIG_PER_VMA_LOCK
-
static int query_vma_setup(struct proc_maps_locking_ctx *lock_ctx)
{
reset_lock_ctx(lock_ctx);
@@ -581,26 +550,6 @@ static struct vm_area_struct *query_vma_
return vma;
}
-#else /* CONFIG_PER_VMA_LOCK */
-
-static int query_vma_setup(struct proc_maps_locking_ctx *lock_ctx)
-{
- return mmap_read_lock_killable(lock_ctx->mm);
-}
-
-static void query_vma_teardown(struct proc_maps_locking_ctx *lock_ctx)
-{
- mmap_read_unlock(lock_ctx->mm);
-}
-
-static struct vm_area_struct *query_vma_find_by_addr(struct proc_maps_locking_ctx *lock_ctx,
- unsigned long addr)
-{
- return find_vma(lock_ctx->mm, addr);
-}
-
-#endif /* CONFIG_PER_VMA_LOCK */
-
static struct vm_area_struct *query_matching_vma(struct proc_maps_locking_ctx *lock_ctx,
unsigned long addr, u32 flags)
{
diff -puN include/linux/mmap_lock.h~unconditional-vma-locks include/linux/mmap_lock.h
--- a/include/linux/mmap_lock.h~unconditional-vma-locks 2026-06-10 15:57:53.599352466 -0700
+++ b/include/linux/mmap_lock.h 2026-06-10 15:57:54.072369286 -0700
@@ -76,8 +76,6 @@ static inline void mmap_assert_write_loc
rwsem_assert_held_write(&mm->mmap_lock);
}
-#ifdef CONFIG_PER_VMA_LOCK
-
#ifdef CONFIG_LOCKDEP
#define __vma_lockdep_map(vma) (&vma->vmlock_dep_map)
#else
@@ -484,52 +482,6 @@ struct vm_area_struct *lock_next_vma(str
struct vma_iterator *iter,
unsigned long address);
-#else /* CONFIG_PER_VMA_LOCK */
-
-static inline void mm_lock_seqcount_init(struct mm_struct *mm) {}
-static inline void mm_lock_seqcount_begin(struct mm_struct *mm) {}
-static inline void mm_lock_seqcount_end(struct mm_struct *mm) {}
-
-static inline bool mmap_lock_speculate_try_begin(struct mm_struct *mm, unsigned int *seq)
-{
- return false;
-}
-
-static inline bool mmap_lock_speculate_retry(struct mm_struct *mm, unsigned int seq)
-{
- return true;
-}
-static inline void vma_lock_init(struct vm_area_struct *vma, bool reset_refcnt) {}
-static inline void vma_end_read(struct vm_area_struct *vma) {}
-static inline void vma_start_write(struct vm_area_struct *vma) {}
-static inline __must_check
-int vma_start_write_killable(struct vm_area_struct *vma) { return 0; }
-static inline void vma_assert_write_locked(struct vm_area_struct *vma)
- { mmap_assert_write_locked(vma->vm_mm); }
-static inline void vma_assert_attached(struct vm_area_struct *vma) {}
-static inline void vma_assert_detached(struct vm_area_struct *vma) {}
-static inline void vma_mark_attached(struct vm_area_struct *vma) {}
-static inline void vma_mark_detached(struct vm_area_struct *vma) {}
-
-static inline struct vm_area_struct *lock_vma_under_rcu(struct mm_struct *mm,
- unsigned long address)
-{
- return NULL;
-}
-
-static inline void vma_assert_locked(struct vm_area_struct *vma)
-{
- mmap_assert_locked(vma->vm_mm);
-}
-
-static inline void vma_assert_stabilised(struct vm_area_struct *vma)
-{
- /* If no VMA locks, then either mmap lock suffices to stabilise. */
- mmap_assert_locked(vma->vm_mm);
-}
-
-#endif /* CONFIG_PER_VMA_LOCK */
-
static inline void mmap_write_lock(struct mm_struct *mm)
{
__mmap_lock_trace_start_locking(mm, true);
diff -puN include/linux/mm.h~unconditional-vma-locks include/linux/mm.h
--- a/include/linux/mm.h~unconditional-vma-locks 2026-06-10 15:57:53.745357660 -0700
+++ b/include/linux/mm.h 2026-06-10 15:57:54.073369321 -0700
@@ -890,7 +890,6 @@ static inline void vma_numab_state_free(
* These must be here rather than mmap_lock.h as dependent on vm_fault type,
* declared in this header.
*/
-#ifdef CONFIG_PER_VMA_LOCK
static inline void release_fault_lock(struct vm_fault *vmf)
{
if (vmf->flags & FAULT_FLAG_VMA_LOCK)
@@ -906,17 +905,6 @@ static inline void assert_fault_locked(c
else
mmap_assert_locked(vmf->vma->vm_mm);
}
-#else
-static inline void release_fault_lock(struct vm_fault *vmf)
-{
- mmap_read_unlock(vmf->vma->vm_mm);
-}
-
-static inline void assert_fault_locked(const struct vm_fault *vmf)
-{
- mmap_assert_locked(vmf->vma->vm_mm);
-}
-#endif /* CONFIG_PER_VMA_LOCK */
static inline bool mm_flags_test(int flag, const struct mm_struct *mm)
{
diff -puN include/linux/mm_types.h~unconditional-vma-locks include/linux/mm_types.h
--- a/include/linux/mm_types.h~unconditional-vma-locks 2026-06-10 15:57:53.763358300 -0700
+++ b/include/linux/mm_types.h 2026-06-10 15:57:54.074369357 -0700
@@ -959,7 +959,6 @@ struct vm_area_struct {
vma_flags_t flags;
};
-#ifdef CONFIG_PER_VMA_LOCK
/*
* Can only be written (using WRITE_ONCE()) while holding both:
* - mmap_lock (in write mode)
@@ -975,7 +974,7 @@ struct vm_area_struct {
* slowpath.
*/
unsigned int vm_lock_seq;
-#endif
+
/*
* A file's MAP_PRIVATE vma can be in both i_mmap tree and anon_vma
* list, after a COW of one of the file pages. A MAP_SHARED vma
@@ -1007,7 +1006,6 @@ struct vm_area_struct {
#ifdef CONFIG_NUMA_BALANCING
struct vma_numab_state *numab_state; /* NUMA Balancing state */
#endif
-#ifdef CONFIG_PER_VMA_LOCK
/*
* Used to keep track of firstly, whether the VMA is attached, secondly,
* if attached, how many read locks are taken, and thirdly, if the
@@ -1050,7 +1048,6 @@ struct vm_area_struct {
#ifdef CONFIG_DEBUG_LOCK_ALLOC
struct lockdep_map vmlock_dep_map;
#endif
-#endif
/*
* For areas with an address space and backing store,
* linkage into the address_space->i_mmap interval tree.
@@ -1249,7 +1246,6 @@ struct mm_struct {
* init_mm.mmlist, and are protected
* by mmlist_lock
*/
-#ifdef CONFIG_PER_VMA_LOCK
struct rcuwait vma_writer_wait;
/*
* This field has lock-like semantics, meaning it is sometimes
@@ -1269,7 +1265,6 @@ struct mm_struct {
* mmap_lock.
*/
seqcount_t mm_lock_seq;
-#endif
#ifdef CONFIG_FUTEX_PRIVATE_HASH
struct mutex futex_hash_lock;
struct futex_private_hash __rcu *futex_phash;
diff -puN kernel/bpf/task_iter.c~unconditional-vma-locks kernel/bpf/task_iter.c
--- a/kernel/bpf/task_iter.c~unconditional-vma-locks 2026-06-10 15:57:53.773358655 -0700
+++ b/kernel/bpf/task_iter.c 2026-06-10 15:57:54.074369357 -0700
@@ -835,11 +835,6 @@ __bpf_kfunc int bpf_iter_task_vma_new(st
BUILD_BUG_ON(sizeof(struct bpf_iter_task_vma_kern) != sizeof(struct bpf_iter_task_vma));
BUILD_BUG_ON(__alignof__(struct bpf_iter_task_vma_kern) != __alignof__(struct bpf_iter_task_vma));
- if (!IS_ENABLED(CONFIG_PER_VMA_LOCK)) {
- kit->data = NULL;
- return -EOPNOTSUPP;
- }
-
/*
* Reject irqs-disabled contexts including NMI. Operations used
* by _next() and _destroy() (vma_end_read, fput, bpf_iter_mmput_async)
diff -puN kernel/fork.c~unconditional-vma-locks kernel/fork.c
--- a/kernel/fork.c~unconditional-vma-locks 2026-06-10 15:57:53.783359011 -0700
+++ b/kernel/fork.c 2026-06-10 15:57:54.074369357 -0700
@@ -1067,9 +1067,7 @@ static void mmap_init_lock(struct mm_str
{
init_rwsem(&mm->mmap_lock);
mm_lock_seqcount_init(mm);
-#ifdef CONFIG_PER_VMA_LOCK
rcuwait_init(&mm->vma_writer_wait);
-#endif
}
static struct mm_struct *mm_init(struct mm_struct *mm, struct task_struct *p,
diff -puN mm/debug.c~unconditional-vma-locks mm/debug.c
--- a/mm/debug.c~unconditional-vma-locks 2026-06-10 15:57:53.785359082 -0700
+++ b/mm/debug.c 2026-06-10 15:57:54.075369392 -0700
@@ -157,17 +157,13 @@ void dump_vma(const struct vm_area_struc
pr_emerg("vma %px start %px end %px mm %px\n"
"prot %lx anon_vma %px vm_ops %px\n"
"pgoff %lx file %px private_data %px\n"
-#ifdef CONFIG_PER_VMA_LOCK
"refcnt %x\n"
-#endif
"flags: %#lx(%pGv)\n",
vma, (void *)vma->vm_start, (void *)vma->vm_end, vma->vm_mm,
(unsigned long)pgprot_val(vma->vm_page_prot),
vma->anon_vma, vma->vm_ops, vma->vm_pgoff,
vma->vm_file, vma->vm_private_data,
-#ifdef CONFIG_PER_VMA_LOCK
refcount_read(&vma->vm_refcnt),
-#endif
vma->vm_flags, &vma->vm_flags);
}
EXPORT_SYMBOL(dump_vma);
diff -puN mm/init-mm.c~unconditional-vma-locks mm/init-mm.c
--- a/mm/init-mm.c~unconditional-vma-locks 2026-06-10 15:57:53.808359899 -0700
+++ b/mm/init-mm.c 2026-06-10 15:57:54.075369392 -0700
@@ -39,10 +39,8 @@ struct mm_struct init_mm = {
.page_table_lock = __SPIN_LOCK_UNLOCKED(init_mm.page_table_lock),
.arg_lock = __SPIN_LOCK_UNLOCKED(init_mm.arg_lock),
.mmlist = LIST_HEAD_INIT(init_mm.mmlist),
-#ifdef CONFIG_PER_VMA_LOCK
.vma_writer_wait = __RCUWAIT_INITIALIZER(init_mm.vma_writer_wait),
.mm_lock_seq = SEQCNT_ZERO(init_mm.mm_lock_seq),
-#endif
.user_ns = &init_user_ns,
#ifdef CONFIG_SCHED_MM_CID
.mm_cid.lock = __RAW_SPIN_LOCK_UNLOCKED(init_mm.mm_cid.lock),
diff -puN mm/Kconfig~unconditional-vma-locks mm/Kconfig
--- a/mm/Kconfig~unconditional-vma-locks 2026-06-10 15:57:53.816360183 -0700
+++ b/mm/Kconfig 2026-06-10 15:57:54.075369392 -0700
@@ -1394,19 +1394,6 @@ config LRU_GEN_STATS
config LRU_GEN_WALKS_MMU
def_bool y
depends on LRU_GEN && ARCH_HAS_HW_PTE_YOUNG
-# }
-
-config ARCH_SUPPORTS_PER_VMA_LOCK
- def_bool n
-
-config PER_VMA_LOCK
- def_bool y
- depends on ARCH_SUPPORTS_PER_VMA_LOCK && MMU && SMP
- help
- Allow per-vma locking during page fault handling.
-
- This feature allows locking each virtual memory area separately when
- handling page faults instead of taking mmap_lock.
config LOCK_MM_AND_FIND_VMA
bool
diff -puN mm/Kconfig.debug~unconditional-vma-locks mm/Kconfig.debug
--- a/mm/Kconfig.debug~unconditional-vma-locks 2026-06-10 15:57:53.820360326 -0700
+++ b/mm/Kconfig.debug 2026-06-10 15:57:54.075369392 -0700
@@ -310,7 +310,6 @@ config DEBUG_KMEMLEAK_VERBOSE
config PER_VMA_LOCK_STATS
bool "Statistics for per-vma locks"
- depends on PER_VMA_LOCK
help
Say Y here to enable success, retry and failure counters of page
faults handled under protection of per-vma locks. When enabled, the
diff -puN mm/memory.c~unconditional-vma-locks mm/memory.c
--- a/mm/memory.c~unconditional-vma-locks 2026-06-10 15:57:53.830360681 -0700
+++ b/mm/memory.c 2026-06-10 15:57:54.076369428 -0700
@@ -6659,7 +6659,6 @@ static vm_fault_t sanitize_fault_flags(s
!is_cow_mapping(vma->vm_flags)))
return VM_FAULT_SIGSEGV;
}
-#ifdef CONFIG_PER_VMA_LOCK
/*
* Per-VMA locks can't be used with FAULT_FLAG_RETRY_NOWAIT because of
* the assumption that lock is dropped on VM_FAULT_RETRY.
@@ -6668,7 +6667,6 @@ static vm_fault_t sanitize_fault_flags(s
(FAULT_FLAG_VMA_LOCK | FAULT_FLAG_RETRY_NOWAIT)) ==
(FAULT_FLAG_VMA_LOCK | FAULT_FLAG_RETRY_NOWAIT)))
return VM_FAULT_SIGSEGV;
-#endif
return 0;
}
diff -puN mm/mmap_lock.c~unconditional-vma-locks mm/mmap_lock.c
--- a/mm/mmap_lock.c~unconditional-vma-locks 2026-06-10 15:57:53.834360824 -0700
+++ b/mm/mmap_lock.c 2026-06-10 15:57:54.077369463 -0700
@@ -43,9 +43,6 @@ void __mmap_lock_do_trace_released(struc
EXPORT_SYMBOL(__mmap_lock_do_trace_released);
#endif /* CONFIG_TRACING */
-#ifdef CONFIG_MMU
-#ifdef CONFIG_PER_VMA_LOCK
-
/* State shared across __vma_[start, end]_exclude_readers. */
struct vma_exclude_readers_state {
/* Input parameters. */
@@ -431,7 +428,6 @@ fallback:
return vma;
}
-#endif /* CONFIG_PER_VMA_LOCK */
#ifdef CONFIG_LOCK_MM_AND_FIND_VMA
#include <linux/extable.h>
@@ -548,23 +544,3 @@ fail:
return NULL;
}
#endif /* CONFIG_LOCK_MM_AND_FIND_VMA */
-
-#else /* CONFIG_MMU */
-
-/*
- * At least xtensa ends up having protection faults even with no
- * MMU.. No stack expansion, at least.
- */
-struct vm_area_struct *lock_mm_and_find_vma(struct mm_struct *mm,
- unsigned long addr, struct pt_regs *regs)
-{
- struct vm_area_struct *vma;
-
- mmap_read_lock(mm);
- vma = vma_lookup(mm, addr);
- if (!vma)
- mmap_read_unlock(mm);
- return vma;
-}
-
-#endif /* CONFIG_MMU */
diff -puN mm/pagewalk.c~unconditional-vma-locks mm/pagewalk.c
--- a/mm/pagewalk.c~unconditional-vma-locks 2026-06-10 15:57:53.851361429 -0700
+++ b/mm/pagewalk.c 2026-06-10 15:57:54.077369463 -0700
@@ -446,7 +446,6 @@ static inline void process_mm_walk_lock(
static inline void process_vma_walk_lock(struct vm_area_struct *vma,
enum page_walk_lock walk_lock)
{
-#ifdef CONFIG_PER_VMA_LOCK
switch (walk_lock) {
case PGWALK_WRLOCK:
vma_start_write(vma);
@@ -461,7 +460,6 @@ static inline void process_vma_walk_lock
/* PGWALK_RDLOCK is handled by process_mm_walk_lock */
break;
}
-#endif
}
/*
diff -puN mm/rmap.c~unconditional-vma-locks mm/rmap.c
--- a/mm/rmap.c~unconditional-vma-locks 2026-06-10 15:57:54.018367366 -0700
+++ b/mm/rmap.c 2026-06-10 15:57:54.077369463 -0700
@@ -260,11 +260,9 @@ static void check_anon_vma_clone(struct
/* For the anon_vma to be compatible, it can only be singular. */
VM_WARN_ON_ONCE(operation == VMA_OP_MERGE_UNFAULTED &&
!list_is_singular(&src->anon_vma_chain));
-#ifdef CONFIG_PER_VMA_LOCK
/* Only merging an unfaulted VMA leaves the destination attached. */
VM_WARN_ON_ONCE(operation != VMA_OP_MERGE_UNFAULTED &&
vma_is_attached(dst));
-#endif
}
static void maybe_reuse_anon_vma(struct vm_area_struct *dst,
diff -puN mm/userfaultfd.c~unconditional-vma-locks mm/userfaultfd.c
--- a/mm/userfaultfd.c~unconditional-vma-locks 2026-06-10 15:57:54.049368468 -0700
+++ b/mm/userfaultfd.c 2026-06-10 15:57:54.078369499 -0700
@@ -104,7 +104,6 @@ struct vm_area_struct *find_vma_and_prep
return vma;
}
-#ifdef CONFIG_PER_VMA_LOCK
/*
* uffd_lock_vma() - Lookup and lock vma corresponding to @address.
* @mm: mm to search vma in.
@@ -164,34 +163,6 @@ static void uffd_mfill_unlock(struct vm_
vma_end_read(vma);
}
-#else
-
-static struct vm_area_struct *uffd_mfill_lock(struct mm_struct *dst_mm,
- unsigned long dst_start,
- unsigned long len)
-{
- struct vm_area_struct *dst_vma;
-
- mmap_read_lock(dst_mm);
- dst_vma = find_vma_and_prepare_anon(dst_mm, dst_start);
- if (IS_ERR(dst_vma))
- goto out_unlock;
-
- if (validate_dst_vma(dst_vma, dst_start + len))
- return dst_vma;
-
- dst_vma = ERR_PTR(-ENOENT);
-out_unlock:
- mmap_read_unlock(dst_mm);
- return dst_vma;
-}
-
-static void uffd_mfill_unlock(struct vm_area_struct *vma)
-{
- mmap_read_unlock(vma->vm_mm);
-}
-#endif
-
static void mfill_put_vma(struct mfill_state *state)
{
if (!state->vma)
@@ -1672,7 +1643,6 @@ out_success:
return 0;
}
-#ifdef CONFIG_PER_VMA_LOCK
static int uffd_move_lock(struct mm_struct *mm,
unsigned long dst_start,
unsigned long src_start,
@@ -1747,31 +1717,6 @@ static void uffd_move_unlock(struct vm_a
vma_end_read(dst_vma);
}
-#else
-
-static int uffd_move_lock(struct mm_struct *mm,
- unsigned long dst_start,
- unsigned long src_start,
- struct vm_area_struct **dst_vmap,
- struct vm_area_struct **src_vmap)
-{
- int err;
-
- mmap_read_lock(mm);
- err = find_vmas_mm_locked(mm, dst_start, src_start, dst_vmap, src_vmap);
- if (err)
- mmap_read_unlock(mm);
- return err;
-}
-
-static void uffd_move_unlock(struct vm_area_struct *dst_vma,
- struct vm_area_struct *src_vma)
-{
- mmap_assert_locked(src_vma->vm_mm);
- mmap_read_unlock(dst_vma->vm_mm);
-}
-#endif
-
/**
* move_pages - move arbitrary anonymous pages of an existing vma
* @ctx: pointer to the userfaultfd context
diff -puN rust/kernel/mm.rs~unconditional-vma-locks rust/kernel/mm.rs
--- a/rust/kernel/mm.rs~unconditional-vma-locks 2026-06-10 15:57:54.051368539 -0700
+++ b/rust/kernel/mm.rs 2026-06-10 15:57:54.078369499 -0700
@@ -174,7 +174,6 @@ impl MmWithUser {
/// When per-vma locks are disabled, this always returns `None`.
#[inline]
pub fn lock_vma_under_rcu(&self, vma_addr: usize) -> Option<VmaReadGuard<'_>> {
- #[cfg(CONFIG_PER_VMA_LOCK)]
{
// SAFETY: Calling `bindings::lock_vma_under_rcu` is always okay given an mm where
// `mm_users` is non-zero.
@@ -188,12 +187,6 @@ impl MmWithUser {
});
}
}
-
- // Silence warnings about unused variables.
- #[cfg(not(CONFIG_PER_VMA_LOCK))]
- let _ = vma_addr;
-
- None
}
/// Lock the mmap read lock.
diff -puN tools/testing/vma/include/dup.h~unconditional-vma-locks tools/testing/vma/include/dup.h
--- a/tools/testing/vma/include/dup.h~unconditional-vma-locks 2026-06-10 15:57:54.064369001 -0700
+++ b/tools/testing/vma/include/dup.h 2026-06-10 15:57:54.078369499 -0700
@@ -569,7 +569,6 @@ struct vm_area_struct {
vma_flags_t flags;
};
-#ifdef CONFIG_PER_VMA_LOCK
/*
* Can only be written (using WRITE_ONCE()) while holding both:
* - mmap_lock (in write mode)
@@ -585,7 +584,6 @@ struct vm_area_struct {
* slowpath.
*/
unsigned int vm_lock_seq;
-#endif
/*
* A file's MAP_PRIVATE vma can be in both i_mmap tree and anon_vma
@@ -618,10 +616,8 @@ struct vm_area_struct {
#ifdef CONFIG_NUMA_BALANCING
struct vma_numab_state *numab_state; /* NUMA Balancing state */
#endif
-#ifdef CONFIG_PER_VMA_LOCK
/* Unstable RCU readers are allowed to read this. */
refcount_t vm_refcnt;
-#endif
/*
* For areas with an address space and backing store,
* linkage into the address_space->i_mmap interval tree.
diff -puN tools/testing/vma/vma_internal.h~unconditional-vma-locks tools/testing/vma/vma_internal.h
--- a/tools/testing/vma/vma_internal.h~unconditional-vma-locks 2026-06-10 15:57:54.066369072 -0700
+++ b/tools/testing/vma/vma_internal.h 2026-06-10 15:57:54.078369499 -0700
@@ -15,7 +15,6 @@
#include <stdlib.h>
#define CONFIG_MMU
-#define CONFIG_PER_VMA_LOCK
#ifdef __CONCAT
#undef __CONCAT
_
^ permalink raw reply [flat|nested] 7+ messages in thread
* [PATCH v2 2/5] binder: Make shrinker rely solely on per-VMA lock
2026-06-10 23:04 [PATCH v2 0/5] mm: Unconditional per-VMA locks and cleanups Dave Hansen
2026-06-10 23:04 ` [PATCH v2 1/5] mm: Make per-VMA locks available universally Dave Hansen
@ 2026-06-10 23:04 ` Dave Hansen
2026-06-10 23:04 ` [PATCH v2 3/5] mm: Add RCU-based VMA lookup helper that waits for writers Dave Hansen
` (2 subsequent siblings)
4 siblings, 0 replies; 7+ messages in thread
From: Dave Hansen @ 2026-06-10 23:04 UTC (permalink / raw)
To: linux-kernel
Cc: Dave Hansen, Alice Ryhl, Andrew Morton, Arve Hjønnevåg,
Carlos Llamas, Christian Brauner, David Ahern, David S. Miller,
Greg Kroah-Hartman, Liam R. Howlett, linux-mm, Lorenzo Stoakes,
netdev, Shakeel Butt, Suren Baghdasaryan, Todd Kjos,
Vlastimil Babka
[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 4366 bytes --]
From: Dave Hansen <dave.hansen@linux.intel.com>
tl;dr: lock_vma_under_rcu() is already a trylock. No need to do both
it and mmap_read_trylock().
Long Version:
== Background ==
Historically, binder used an mmap_read_trylock() in its shrinker code.
This ensures that reclaim is not blocked on an mmap_lock. Commit
95bc2d4a9020 ("binder: use per-vma lock in page reclaiming") added
support for the per-VMA lock, but left mmap_read_trylock() as a
fallback.
This was presumably because the per-VMA locking can fail for several
reasons and most (all?) lock_vma_under_rcu() callers have a fallback
to mmap_read_trylock().
== Problem ==
The fallback is not worth the complexity here. lock_vma_under_rcu() is
essentially already a non-blocking trylock. The main reason it fails
is also the reason mmap_read_trylock() fails: something is holding
mmap_write_lock().
The only remedy for a collision with mmap_write_lock() is to wait,
which this code can not do. So the "fallback" after
lock_vma_under_rcu() failure is not really a fallback: it is really
likely to just be retrying in vain. That retry in an of itself isn't
horrible. But it adds complexity.
== Solution ==
Now that per-VMA locks are universally available, lock_vma_under_rcu()
will not persistently fail. Rely on it alone and simplify the code.
Full disclosure: I originally tried to do this with
lock_vma_under_rcu_wait(), but it did not fit well with the mmap_lock
trylock semantics. Claude caught this in a review and suggested the
approach in this path. It seemed sane to me. So, Suggesed-by: Claude,
I guess.
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Reviewed-by: Suren Baghdasaryan <surenb@google.com>
Acked-by: Lorenzo Stoakes <ljs@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: "Liam R. Howlett" <Liam.Howlett@oracle.com>
Cc: Vlastimil Babka <vbabka@kernel.org>
Cc: Shakeel Butt <shakeel.butt@linux.dev>
Cc: linux-mm@kvack.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Arve Hjønnevåg <arve@android.com>
Cc: Todd Kjos <tkjos@android.com>
Cc: Christian Brauner <christian@brauner.io>
Cc: Carlos Llamas <cmllamas@google.com>
Cc: Alice Ryhl <aliceryhl@google.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: David Ahern <dsahern@kernel.org>
Cc: netdev@vger.kernel.org
--
Changes from v1:
* Move forward even if 'vma' is NULL in binder_alloc_free_page().
This can happen if the VMA is unmapped (Sashiko).
* Rename goto label to be more accurate for new lock scheme
---
b/drivers/android/binder_alloc.c | 26 +++++++++-----------------
1 file changed, 9 insertions(+), 17 deletions(-)
diff -puN drivers/android/binder_alloc.c~binder-try-vma-lock drivers/android/binder_alloc.c
--- a/drivers/android/binder_alloc.c~binder-try-vma-lock 2026-06-10 15:57:55.274412018 -0700
+++ b/drivers/android/binder_alloc.c 2026-06-10 15:57:55.277412124 -0700
@@ -1142,7 +1142,6 @@ enum lru_status binder_alloc_free_page(s
struct vm_area_struct *vma;
struct page *page_to_free;
unsigned long page_addr;
- int mm_locked = 0;
size_t index;
if (!mmget_not_zero(mm))
@@ -1151,15 +1150,12 @@ enum lru_status binder_alloc_free_page(s
index = mdata->page_index;
page_addr = alloc->vm_start + index * PAGE_SIZE;
- /* attempt per-vma lock first */
+ /*
+ * Attempt per-vma lock. This is essentially a
+ * "trylock". It can fail even if the VMA exists
+ * for 'page_addr'.
+ */
vma = lock_vma_under_rcu(mm, page_addr);
- if (!vma) {
- /* fall back to mmap_lock */
- if (!mmap_read_trylock(mm))
- goto err_mmap_read_lock_failed;
- mm_locked = 1;
- vma = vma_lookup(mm, page_addr);
- }
if (!mutex_trylock(&alloc->mutex))
goto err_get_alloc_mutex_failed;
@@ -1188,13 +1184,11 @@ enum lru_status binder_alloc_free_page(s
zap_vma_range(vma, page_addr, PAGE_SIZE);
trace_binder_unmap_user_end(alloc, index);
+
+ vma_end_read(vma);
}
mutex_unlock(&alloc->mutex);
- if (mm_locked)
- mmap_read_unlock(mm);
- else
- vma_end_read(vma);
mmput_async(mm);
binder_free_page(page_to_free);
@@ -1203,11 +1197,9 @@ enum lru_status binder_alloc_free_page(s
err_invalid_vma:
mutex_unlock(&alloc->mutex);
err_get_alloc_mutex_failed:
- if (mm_locked)
- mmap_read_unlock(mm);
- else
+ if (vma)
vma_end_read(vma);
-err_mmap_read_lock_failed:
+err_vma_lock_failed:
mmput_async(mm);
err_mmget:
return LRU_SKIP;
_
^ permalink raw reply [flat|nested] 7+ messages in thread
* [PATCH v2 3/5] mm: Add RCU-based VMA lookup helper that waits for writers
2026-06-10 23:04 [PATCH v2 0/5] mm: Unconditional per-VMA locks and cleanups Dave Hansen
2026-06-10 23:04 ` [PATCH v2 1/5] mm: Make per-VMA locks available universally Dave Hansen
2026-06-10 23:04 ` [PATCH v2 2/5] binder: Make shrinker rely solely on per-VMA lock Dave Hansen
@ 2026-06-10 23:04 ` Dave Hansen
2026-06-10 23:40 ` Dave Hansen
2026-06-10 23:04 ` [PATCH v2 4/5] binder: Remove mmap_lock fallback Dave Hansen
2026-06-10 23:04 ` [PATCH v2 5/5] tcp: Remove mmap_lock fallback path Dave Hansen
4 siblings, 1 reply; 7+ messages in thread
From: Dave Hansen @ 2026-06-10 23:04 UTC (permalink / raw)
To: linux-kernel
Cc: Dave Hansen, Alice Ryhl, Andrew Morton, Arve Hjønnevåg,
Carlos Llamas, Christian Brauner, David Ahern, David S. Miller,
Greg Kroah-Hartman, Liam R. Howlett, linux-mm, Lorenzo Stoakes,
netdev, Shakeel Butt, Suren Baghdasaryan, Todd Kjos,
Vlastimil Babka
[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 4423 bytes --]
From: Dave Hansen <dave.hansen@linux.intel.com>
== Background ==
There are basically two parallel ways to look up a VMA: the
traditional way, which is protected by mmap_lock, and the RCU-based
per-VMA lock way which is based on RCU and refcounts.
== Problem ==
The mmap_lock one is more straightforward to use but it has a big
disadvantage in that it can not be mixed with page faults since those
can take mmap_lock for read, which can deadlock when mixed with page
faults. For example:
mmap_read_lock(mm);
// Another thread does mmap_write_lock().
// New mmap_lock readers are blocked.
vma = vma_lookup(mm, address);
// This deadlocks on mmap_read_lock() if it faults:
copy_from_user(address);
mmap_read_unlock(mm);
The RCU one can be mixed with faults, but it is not available in all
configs, so all RCU users need to be able to fall back to the
traditional way.
== Solution ==
Add a variant of the RCU-based lookup that waits for writers. This is
basically the same as the existing RCU-based lookup, but it also takes
mmap_lock for read and waits for writers to finish before returning
the VMA. This has some advantages:
1. Callers do not need to have a fallback path for when they
collide with writers.
2. It can be used in contexts where page faults can happen because
it can take the mmap_lock for read but never *holds* it.
3. Its fast path does not require taking mmap_lock for read.
Basically, when applied correctly, this approach results in faster
*and* simpler code.
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: "Liam R. Howlett" <Liam.Howlett@oracle.com>
Cc: Lorenzo Stoakes <ljs@kernel.org>
Cc: Vlastimil Babka <vbabka@kernel.org>
Cc: Shakeel Butt <shakeel.butt@linux.dev>
Cc: linux-mm@kvack.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Arve Hjønnevåg <arve@android.com>
Cc: Todd Kjos <tkjos@android.com>
Cc: Christian Brauner <christian@brauner.io>
Cc: Carlos Llamas <cmllamas@google.com>
Cc: Alice Ryhl <aliceryhl@google.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: David Ahern <dsahern@kernel.org>
Cc: netdev@vger.kernel.org
--
Changes from v1:
* Add a comment explaining that this can not be mixed with other
per-VMA lock or mmap_lock users. It is prone to deadlocks if so.
* Add a FIXME about making the mmap_read_lock() killable
* Add more chaneglog bits about the possibility for an infinite goto
loop.
* Adopt vma_start_read_unlocked() implementation from Lorenzo
---
b/include/linux/mmap_lock.h | 3 +++
b/mm/mmap_lock.c | 27 +++++++++++++++++++++++++++
2 files changed, 30 insertions(+)
diff -puN include/linux/mmap_lock.h~lock-vma-under-rcu-wait include/linux/mmap_lock.h
--- a/include/linux/mmap_lock.h~lock-vma-under-rcu-wait 2026-06-10 15:57:55.828431712 -0700
+++ b/include/linux/mmap_lock.h 2026-06-10 15:57:55.834431925 -0700
@@ -257,6 +257,9 @@ static inline bool vma_start_read_locked
return vma_start_read_locked_nested(vma, 0);
}
+struct vm_area_struct *vma_start_read_unlocked(struct mm_struct *mm,
+ unsigned long address);
+
static inline void vma_end_read(struct vm_area_struct *vma)
{
vma_refcount_put(vma);
diff -puN mm/mmap_lock.c~lock-vma-under-rcu-wait mm/mmap_lock.c
--- a/mm/mmap_lock.c~lock-vma-under-rcu-wait 2026-06-10 15:57:55.831431819 -0700
+++ b/mm/mmap_lock.c 2026-06-10 16:02:50.723860779 -0700
@@ -338,6 +338,33 @@ inval:
return NULL;
}
+/*
+ * Find the VMA covering 'address' and lock it for reading. Waits for writers to
+ * finish if the VMA is being modified. Returns NULL if there is no VMA covering
+ * 'address'.
+ *
+ * Use only in code paths where no mmap_lock and no VMA lock is held.
+ *
+ * The fast path does not take mmap_lock.
+ */
+struct vm_area_struct *vma_start_read_unlocked(struct mm_struct *mm,
+ unsigned long address)
+{
+ struct vm_area_struct *vma;
+
+ /* Fast path: return stable VMA covering 'address': */
+ vma = lock_vma_under_rcu(mm, address);
+ if (vma)
+ return vma;
+
+ /* Slow path: preclude VMA writers by getting mmap read lock. */
+ guard(rwsem_read)(&mm->mmap_lock);
+ if (!vma_start_read_locked(vma))
+ return NULL;
+
+ return vma;
+}
+
static struct vm_area_struct *lock_next_vma_under_mmap_lock(struct mm_struct *mm,
struct vma_iterator *vmi,
unsigned long from_addr)
_
^ permalink raw reply [flat|nested] 7+ messages in thread
* [PATCH v2 4/5] binder: Remove mmap_lock fallback
2026-06-10 23:04 [PATCH v2 0/5] mm: Unconditional per-VMA locks and cleanups Dave Hansen
` (2 preceding siblings ...)
2026-06-10 23:04 ` [PATCH v2 3/5] mm: Add RCU-based VMA lookup helper that waits for writers Dave Hansen
@ 2026-06-10 23:04 ` Dave Hansen
2026-06-10 23:04 ` [PATCH v2 5/5] tcp: Remove mmap_lock fallback path Dave Hansen
4 siblings, 0 replies; 7+ messages in thread
From: Dave Hansen @ 2026-06-10 23:04 UTC (permalink / raw)
To: linux-kernel
Cc: Dave Hansen, Alice Ryhl, Andrew Morton, Arve Hjønnevåg,
Carlos Llamas, Christian Brauner, David Ahern, David S. Miller,
Greg Kroah-Hartman, Liam R. Howlett, linux-mm, Lorenzo Stoakes,
netdev, Shakeel Butt, Suren Baghdasaryan, Todd Kjos,
Vlastimil Babka
[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 2041 bytes --]
From: Dave Hansen <dave.hansen@linux.intel.com>
Previously, the per-VMA locking could fail in the face of writers
which necessitate a fallback to mmap_lock. The new
vma_start_read_unlocked() will wait for writers instead of failing.
Use the new helper. Wait for writers. Remove the fallback to mmap_lock.
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Acked-by: Lorenzo Stoakes <ljs@kernel.org>
Reviewed-by: Suren Baghdasaryan <surenb@google.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: "Liam R. Howlett" <Liam.Howlett@oracle.com>
Cc: Vlastimil Babka <vbabka@kernel.org>
Cc: Shakeel Butt <shakeel.butt@linux.dev>
Cc: linux-mm@kvack.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Arve Hjønnevåg <arve@android.com>
Cc: Todd Kjos <tkjos@android.com>
Cc: Christian Brauner <christian@brauner.io>
Cc: Carlos Llamas <cmllamas@google.com>
Cc: Alice Ryhl <aliceryhl@google.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: David Ahern <dsahern@kernel.org>
Cc: netdev@vger.kernel.org
---
b/drivers/android/binder_alloc.c | 17 +++++------------
1 file changed, 5 insertions(+), 12 deletions(-)
diff -puN drivers/android/binder_alloc.c~binder-vma-waiter drivers/android/binder_alloc.c
--- a/drivers/android/binder_alloc.c~binder-vma-waiter 2026-06-10 15:57:56.419452721 -0700
+++ b/drivers/android/binder_alloc.c 2026-06-10 15:57:56.423452863 -0700
@@ -259,21 +259,14 @@ static int binder_page_insert(struct bin
struct vm_area_struct *vma;
int ret = -ESRCH;
- /* attempt per-vma lock first */
- vma = lock_vma_under_rcu(mm, addr);
- if (vma) {
- if (binder_alloc_is_mapped(alloc))
- ret = vm_insert_page(vma, addr, page);
- vma_end_read(vma);
+ vma = vma_start_read_unlocked(mm, addr);
+ if (!vma)
return ret;
- }
- /* fall back to mmap_lock */
- mmap_read_lock(mm);
- vma = vma_lookup(mm, addr);
- if (vma && binder_alloc_is_mapped(alloc))
+ if (binder_alloc_is_mapped(alloc))
ret = vm_insert_page(vma, addr, page);
- mmap_read_unlock(mm);
+
+ vma_end_read(vma);
return ret;
}
_
^ permalink raw reply [flat|nested] 7+ messages in thread
* [PATCH v2 5/5] tcp: Remove mmap_lock fallback path
2026-06-10 23:04 [PATCH v2 0/5] mm: Unconditional per-VMA locks and cleanups Dave Hansen
` (3 preceding siblings ...)
2026-06-10 23:04 ` [PATCH v2 4/5] binder: Remove mmap_lock fallback Dave Hansen
@ 2026-06-10 23:04 ` Dave Hansen
4 siblings, 0 replies; 7+ messages in thread
From: Dave Hansen @ 2026-06-10 23:04 UTC (permalink / raw)
To: linux-kernel
Cc: Dave Hansen, Alice Ryhl, Andrew Morton, Arve Hjønnevåg,
Carlos Llamas, Christian Brauner, David Ahern, David S. Miller,
Greg Kroah-Hartman, Liam R. Howlett, linux-mm, Lorenzo Stoakes,
netdev, Shakeel Butt, Suren Baghdasaryan, Todd Kjos,
Vlastimil Babka
[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 2966 bytes --]
From: Dave Hansen <dave.hansen@linux.intel.com>
Previously, the per-VMA locking could fail in the face of writers
which necessitates a fallback to mmap_lock. The new
lock_vma_under_rcu_wait() will wait for writers instead of failing.
Use the new helper. Wait for writers. Remove the fallback to mmap_lock.
This really is a nice cleanup. It removes the need to pass the lock
state back and forth to find_tcp_vma().
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Acked-by: Lorenzo Stoakes <ljs@kernel.org>
Reviewed-by: Suren Baghdasaryan <surenb@google.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: "Liam R. Howlett" <Liam.Howlett@oracle.com>
Cc: Vlastimil Babka <vbabka@kernel.org>
Cc: Shakeel Butt <shakeel.butt@linux.dev>
Cc: linux-mm@kvack.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Arve Hjønnevåg <arve@android.com>
Cc: Todd Kjos <tkjos@android.com>
Cc: Christian Brauner <christian@brauner.io>
Cc: Carlos Llamas <cmllamas@google.com>
Cc: Alice Ryhl <aliceryhl@google.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: David Ahern <dsahern@kernel.org>
Cc: netdev@vger.kernel.org
---
b/net/ipv4/tcp.c | 31 +++++++++----------------------
1 file changed, 9 insertions(+), 22 deletions(-)
diff -puN net/ipv4/tcp.c~ipv4-tcp-vma-waiter net/ipv4/tcp.c
--- a/net/ipv4/tcp.c~ipv4-tcp-vma-waiter 2026-06-10 15:57:56.972472379 -0700
+++ b/net/ipv4/tcp.c 2026-06-10 15:57:56.976472521 -0700
@@ -2171,27 +2171,18 @@ static void tcp_zc_finalize_rx_tstamp(st
}
static struct vm_area_struct *find_tcp_vma(struct mm_struct *mm,
- unsigned long address,
- bool *mmap_locked)
+ unsigned long address)
{
- struct vm_area_struct *vma = lock_vma_under_rcu(mm, address);
+ struct vm_area_struct *vma = vma_start_read_unlocked(mm, address);
- if (vma) {
- if (vma->vm_ops != &tcp_vm_ops) {
- vma_end_read(vma);
- return NULL;
- }
- *mmap_locked = false;
- return vma;
- }
+ if (!vma)
+ return NULL;
- mmap_read_lock(mm);
- vma = vma_lookup(mm, address);
- if (!vma || vma->vm_ops != &tcp_vm_ops) {
- mmap_read_unlock(mm);
+ if (vma->vm_ops != &tcp_vm_ops) {
+ vma_end_read(vma);
return NULL;
}
- *mmap_locked = true;
+
return vma;
}
@@ -2212,7 +2203,6 @@ static int tcp_zerocopy_receive(struct s
u32 seq = tp->copied_seq;
u32 total_bytes_to_map;
int inq = tcp_inq(sk);
- bool mmap_locked;
int ret;
zc->copybuf_len = 0;
@@ -2237,7 +2227,7 @@ static int tcp_zerocopy_receive(struct s
return 0;
}
- vma = find_tcp_vma(current->mm, address, &mmap_locked);
+ vma = find_tcp_vma(current->mm, address);
if (!vma)
return -EINVAL;
@@ -2319,10 +2309,7 @@ static int tcp_zerocopy_receive(struct s
zc, total_bytes_to_map);
}
out:
- if (mmap_locked)
- mmap_read_unlock(current->mm);
- else
- vma_end_read(vma);
+ vma_end_read(vma);
/* Try to copy straggler data. */
if (!ret)
copylen = tcp_zc_handle_leftover(zc, sk, skb, &seq, copybuf_len, tss);
_
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH v2 3/5] mm: Add RCU-based VMA lookup helper that waits for writers
2026-06-10 23:04 ` [PATCH v2 3/5] mm: Add RCU-based VMA lookup helper that waits for writers Dave Hansen
@ 2026-06-10 23:40 ` Dave Hansen
0 siblings, 0 replies; 7+ messages in thread
From: Dave Hansen @ 2026-06-10 23:40 UTC (permalink / raw)
To: Dave Hansen, linux-kernel
Cc: Alice Ryhl, Andrew Morton, Arve Hjønnevåg,
Carlos Llamas, Christian Brauner, David Ahern, David S. Miller,
Greg Kroah-Hartman, Liam R. Howlett, linux-mm, Lorenzo Stoakes,
netdev, Shakeel Butt, Suren Baghdasaryan, Todd Kjos,
Vlastimil Babka
On 6/10/26 16:04, Dave Hansen wrote:
> + /* Slow path: preclude VMA writers by getting mmap read lock. */
> + guard(rwsem_read)(&mm->mmap_lock);
> + if (!vma_start_read_locked(vma))
> + return NULL;
Welp, I actually ran and tested this, but it's got a big bug. The slow
path is broken. It needs:
/* Slow path: preclude VMA writers by getting mmap read lock. */
guard(rwsem_read)(&mm->mmap_lock);
+ vma = vma_lookup(mm, address);
if (!vma_start_read_locked(vma))
return NULL;
Because VMA is NULL in slow path otherwise. So it'll definitely need a
v3 or a fixup before it goes anywhere.
^ permalink raw reply [flat|nested] 7+ messages in thread
end of thread, other threads:[~2026-06-10 23:40 UTC | newest]
Thread overview: 7+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-06-10 23:04 [PATCH v2 0/5] mm: Unconditional per-VMA locks and cleanups Dave Hansen
2026-06-10 23:04 ` [PATCH v2 1/5] mm: Make per-VMA locks available universally Dave Hansen
2026-06-10 23:04 ` [PATCH v2 2/5] binder: Make shrinker rely solely on per-VMA lock Dave Hansen
2026-06-10 23:04 ` [PATCH v2 3/5] mm: Add RCU-based VMA lookup helper that waits for writers Dave Hansen
2026-06-10 23:40 ` Dave Hansen
2026-06-10 23:04 ` [PATCH v2 4/5] binder: Remove mmap_lock fallback Dave Hansen
2026-06-10 23:04 ` [PATCH v2 5/5] tcp: Remove mmap_lock fallback path Dave Hansen
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox