hugetlb: take hugetlb vma_lock when clearing vma_lock->vma pointer

hugetlb file truncation/hole punch code may need to back out and take
locks in order in the routine hugetlb_unmap_file_folio().  This code could
race with vma freeing as pointed out in [1] and result in accessing a
stale vma pointer.  To address this, take the vma_lock when clearing the
vma_lock->vma pointer.

[1] https://lore.kernel.org/linux-mm/01f10195-7088-4462-6def-909549c75ef4@huawei.com/

[mike.kravetz@oracle.com: address build issues]
Link: https://lkml.kernel.org/r/Yz5L1uxQYR1VqFtJ@monkey
Link: https://lkml.kernel.org/r/20221005011707.514612-3-mike.kravetz@oracle.com
Fixes: "hugetlb: use new vma_lock for pmd sharing synchronization"
Signed-off-by: Mike Kravetz <mike.kravetz@oracle.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Cc: Axel Rasmussen <axelrasmussen@google.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Davidlohr Bueso <dave@stgolabs.net>
Cc: James Houghton <jthoughton@google.com>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Miaohe Lin <linmiaohe@huawei.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Mina Almasry <almasrymina@google.com>
Cc: Muchun Song <songmuchun@bytedance.com>
Cc: Naoya Horiguchi <naoya.horiguchi@linux.dev>
Cc: Pasha Tatashin <pasha.tatashin@soleen.com>
Cc: Peter Xu <peterx@redhat.com>
Cc: Prakash Sangappa <prakash.sangappa@oracle.com>
Cc: Sven Schnelle <svens@linux.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 45e305d182f6be0ccff292cb65b748c95fe5ac96..01f3e36caa6c7c1bc9114d2c0437012419def387 100644 (file)
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -93,6 +93,7 @@ struct mutex *hugetlb_fault_mutex_table ____cacheline_aligned_in_smp;
 static int hugetlb_acct_memory(struct hstate *h, long delta);
 static void hugetlb_vma_lock_free(struct vm_area_struct *vma);
 static void hugetlb_vma_lock_alloc(struct vm_area_struct *vma);
+static void __hugetlb_vma_unlock_write_free(struct vm_area_struct *vma);
 
 static inline bool subpool_is_free(struct hugepage_subpool *spool)
 {
@@ -5188,8 +5189,7 @@ void __unmap_hugepage_range_final(struct mmu_gather *tlb,
         * be asynchrously deleted.  If the page tables are shared, there
         * will be issues when accessed by someone else.
         */
-       hugetlb_vma_unlock_write(vma);
-       hugetlb_vma_lock_free(vma);
+       __hugetlb_vma_unlock_write_free(vma);
 
        i_mmap_unlock_write(vma->vm_file->f_mapping);
 }
@@ -6828,6 +6828,30 @@ void hugetlb_vma_lock_release(struct kref *kref)
        kfree(vma_lock);
 }
 
+void __hugetlb_vma_unlock_write_put(struct hugetlb_vma_lock *vma_lock)
+{
+       struct vm_area_struct *vma = vma_lock->vma;
+
+       /*
+        * vma_lock structure may or not be released as a result of put,
+        * it certainly will no longer be attached to vma so clear pointer.
+        * Semaphore synchronizes access to vma_lock->vma field.
+        */
+       vma_lock->vma = NULL;
+       vma->vm_private_data = NULL;
+       up_write(&vma_lock->rw_sema);
+       kref_put(&vma_lock->refs, hugetlb_vma_lock_release);
+}
+
+static void __hugetlb_vma_unlock_write_free(struct vm_area_struct *vma)
+{
+       if (__vma_shareable_flags_pmd(vma)) {
+               struct hugetlb_vma_lock *vma_lock = vma->vm_private_data;
+
+               __hugetlb_vma_unlock_write_put(vma_lock);
+       }
+}
+
 static void hugetlb_vma_lock_free(struct vm_area_struct *vma)
 {
        /*
@@ -6839,14 +6863,8 @@ static void hugetlb_vma_lock_free(struct vm_area_struct *vma)
        if (vma->vm_private_data) {
                struct hugetlb_vma_lock *vma_lock = vma->vm_private_data;
 
-               /*
-                * vma_lock structure may or not be released, but it
-                * certainly will no longer be attached to vma so clear
-                * pointer.
-                */
-               vma_lock->vma = NULL;
-               kref_put(&vma_lock->refs, hugetlb_vma_lock_release);
-               vma->vm_private_data = NULL;
+               down_write(&vma_lock->rw_sema);
+               __hugetlb_vma_unlock_write_put(vma_lock);
        }
 }
 
@@ -6997,6 +7015,10 @@ void hugetlb_vma_lock_release(struct kref *kref)
 {
 }
 
+static void __hugetlb_vma_unlock_write_free(struct vm_area_struct *vma)
+{
+}
+
 static void hugetlb_vma_lock_free(struct vm_area_struct *vma)
 {
 }