+ mm-handle-userfaults-under-vma-lock.patch added to mm-unstable branch

[Andrew Morton <akpm@linux-foundation.org>]

The patch titled
     Subject: mm: handle userfaults under VMA lock
has been added to the -mm mm-unstable branch.  Its filename is
     mm-handle-userfaults-under-vma-lock.patch

This patch will shortly appear at
     https://git.kernel.org/pub/scm/linux/kernel/git/akpm/25-new.git/tree/patches/mm-handle-userfaults-under-vma-lock.patch

This patch will later appear in the mm-unstable branch at
    git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm

Before you just go and hit "reply", please:
   a) Consider who else should be cc'ed
   b) Prefer to cc a suitable mailing list as well
   c) Ideally: find the original patch on the mailing list and do a
      reply-to-all to that, adding suitable additional cc's

*** Remember to use Documentation/process/submit-checklist.rst when testing your code ***

The -mm tree is included into linux-next via the mm-everything
branch at git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
and is updated there every 2-3 working days

------------------------------------------------------
From: Suren Baghdasaryan <surenb@google.com>
Subject: mm: handle userfaults under VMA lock
Date: Fri, 30 Jun 2023 14:19:57 -0700

Enable handle_userfault to operate under VMA lock by releasing VMA lock
instead of mmap_lock and retrying.  Note that FAULT_FLAG_RETRY_NOWAIT
should never be used when handling faults under per-VMA lock protection
because that would break the assumption that lock is dropped on retry.

Link: https://lkml.kernel.org/r/20230630211957.1341547-7-surenb@google.com
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Acked-by: Peter Xu <peterx@redhat.com>
Cc: Alistair Popple <apopple@nvidia.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Christian Brauner <brauner@kernel.org>
Cc: Christoph Hellwig <hch@lst.de>
Cc: David Hildenbrand <david@redhat.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Davidlohr Bueso <dave@stgolabs.net>
Cc: Hillf Danton <hdanton@sina.com>
Cc: "Huang, Ying" <ying.huang@intel.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Jan Kara <jack@suse.cz>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Josef Bacik <josef@toxicpanda.com>
Cc: Laurent Dufour <ldufour@linux.ibm.com>
Cc: Liam R. Howlett <Liam.Howlett@oracle.com>
Cc: Lorenzo Stoakes <lstoakes@gmail.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Michel Lespinasse <michel@lespinasse.org>
Cc: Minchan Kim <minchan@google.com>
Cc: Pavel Tatashin <pasha.tatashin@soleen.com>
Cc: Punit Agrawal <punit.agrawal@bytedance.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Yu Zhao <yuzhao@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---

 fs/userfaultfd.c   |   34 ++++++++++++++--------------------
 include/linux/mm.h |   24 ++++++++++++++++++++++++
 mm/memory.c        |    9 ---------
 3 files changed, 38 insertions(+), 29 deletions(-)

--- a/fs/userfaultfd.c~mm-handle-userfaults-under-vma-lock
+++ a/fs/userfaultfd.c
@@ -277,17 +277,16 @@ static inline struct uffd_msg userfault_
  * hugepmd ranges.
  */
 static inline bool userfaultfd_huge_must_wait(struct userfaultfd_ctx *ctx,
-					 struct vm_area_struct *vma,
-					 unsigned long address,
-					 unsigned long flags,
-					 unsigned long reason)
+					      struct vm_fault *vmf,
+					      unsigned long reason)
 {
+	struct vm_area_struct *vma = vmf->vma;
 	pte_t *ptep, pte;
 	bool ret = true;
 
-	mmap_assert_locked(ctx->mm);
+	assert_fault_locked(vmf);
 
-	ptep = hugetlb_walk(vma, address, vma_mmu_pagesize(vma));
+	ptep = hugetlb_walk(vma, vmf->address, vma_mmu_pagesize(vma));
 	if (!ptep)
 		goto out;
 
@@ -308,10 +307,8 @@ out:
 }
 #else
 static inline bool userfaultfd_huge_must_wait(struct userfaultfd_ctx *ctx,
-					 struct vm_area_struct *vma,
-					 unsigned long address,
-					 unsigned long flags,
-					 unsigned long reason)
+					      struct vm_fault *vmf,
+					      unsigned long reason)
 {
 	return false;	/* should never get here */
 }
@@ -325,11 +322,11 @@ static inline bool userfaultfd_huge_must
  * threads.
  */
 static inline bool userfaultfd_must_wait(struct userfaultfd_ctx *ctx,
-					 unsigned long address,
-					 unsigned long flags,
+					 struct vm_fault *vmf,
 					 unsigned long reason)
 {
 	struct mm_struct *mm = ctx->mm;
+	unsigned long address = vmf->address;
 	pgd_t *pgd;
 	p4d_t *p4d;
 	pud_t *pud;
@@ -338,7 +335,7 @@ static inline bool userfaultfd_must_wait
 	pte_t ptent;
 	bool ret = true;
 
-	mmap_assert_locked(mm);
+	assert_fault_locked(vmf);
 
 	pgd = pgd_offset(mm, address);
 	if (!pgd_present(*pgd))
@@ -440,7 +437,7 @@ vm_fault_t handle_userfault(struct vm_fa
 	 * Coredumping runs without mmap_lock so we can only check that
 	 * the mmap_lock is held, if PF_DUMPCORE was not set.
 	 */
-	mmap_assert_locked(mm);
+	assert_fault_locked(vmf);
 
 	ctx = vma->vm_userfaultfd_ctx.ctx;
 	if (!ctx)
@@ -556,15 +553,12 @@ vm_fault_t handle_userfault(struct vm_fa
 	spin_unlock_irq(&ctx->fault_pending_wqh.lock);
 
 	if (!is_vm_hugetlb_page(vma))
-		must_wait = userfaultfd_must_wait(ctx, vmf->address, vmf->flags,
-						  reason);
+		must_wait = userfaultfd_must_wait(ctx, vmf, reason);
 	else
-		must_wait = userfaultfd_huge_must_wait(ctx, vma,
-						       vmf->address,
-						       vmf->flags, reason);
+		must_wait = userfaultfd_huge_must_wait(ctx, vmf, reason);
 	if (is_vm_hugetlb_page(vma))
 		hugetlb_vma_unlock_read(vma);
-	mmap_read_unlock(mm);
+	release_fault_lock(vmf);
 
 	if (likely(must_wait && !READ_ONCE(ctx->released))) {
 		wake_up_poll(&ctx->fd_wqh, EPOLLIN);
--- a/include/linux/mm.h~mm-handle-userfaults-under-vma-lock
+++ a/include/linux/mm.h
@@ -705,6 +705,17 @@ static inline bool vma_try_start_write(s
 	return true;
 }
 
+static inline void vma_assert_locked(struct vm_area_struct *vma)
+{
+	int mm_lock_seq;
+
+	if (__is_vma_write_locked(vma, &mm_lock_seq))
+		return;
+
+	lockdep_assert_held(&vma->vm_lock->lock);
+	VM_BUG_ON_VMA(!rwsem_is_locked(&vma->vm_lock->lock), vma);
+}
+
 static inline void vma_assert_write_locked(struct vm_area_struct *vma)
 {
 	int mm_lock_seq;
@@ -728,6 +739,14 @@ static inline void release_fault_lock(st
 		mmap_read_unlock(vmf->vma->vm_mm);
 }
 
+static inline void assert_fault_locked(struct vm_fault *vmf)
+{
+	if (vmf->flags & FAULT_FLAG_VMA_LOCK)
+		vma_assert_locked(vmf->vma);
+	else
+		mmap_assert_locked(vmf->vma->vm_mm);
+}
+
 struct vm_area_struct *lock_vma_under_rcu(struct mm_struct *mm,
 					  unsigned long address);
 
@@ -748,6 +767,11 @@ static inline void release_fault_lock(st
 	mmap_read_unlock(vmf->vma->vm_mm);
 }
 
+static inline void assert_fault_locked(struct vm_fault *vmf)
+{
+	mmap_assert_locked(vmf->vma->vm_mm);
+}
+
 #endif /* CONFIG_PER_VMA_LOCK */
 
 /*
--- a/mm/memory.c~mm-handle-userfaults-under-vma-lock
+++ a/mm/memory.c
@@ -5410,15 +5410,6 @@ retry:
 	if (!vma_start_read(vma))
 		goto inval;
 
-	/*
-	 * Due to the possibility of userfault handler dropping mmap_lock, avoid
-	 * it for now and fall back to page fault handling under mmap_lock.
-	 */
-	if (userfaultfd_armed(vma)) {
-		vma_end_read(vma);
-		goto inval;
-	}
-
 	/* Check since vm_start/vm_end might change before we lock the VMA */
 	if (unlikely(address < vma->vm_start || address >= vma->vm_end)) {
 		vma_end_read(vma);
_

Patches currently in -mm which might be from surenb@google.com are

swap-remove-remnants-of-polling-from-read_swap_cache_async.patch
mm-add-missing-vm_fault_result_trace-name-for-vm_fault_completed.patch
mm-drop-per-vma-lock-when-returning-vm_fault_retry-or-vm_fault_completed.patch
mm-change-folio_lock_or_retry-to-use-vm_fault-directly.patch
mm-handle-swap-page-faults-under-per-vma-lock.patch
mm-handle-userfaults-under-vma-lock.patch