Thorvald reported a WARNING [1]. And the root cause is below race:
CPU 1 CPU 2
fork hugetlbfs_fallocate
dup_mmap hugetlbfs_punch_hole
i_mmap_lock_write(mapping);
vma_interval_tree_insert_after -- Child vma is visible through i_mmap tree.
i_mmap_unlock_write(mapping);
hugetlb_dup_vma_private -- Clear vma_lock outside i_mmap_rwsem!
i_mmap_lock_write(mapping);
hugetlb_vmdelete_list
vma_interval_tree_foreach
hugetlb_vma_trylock_write -- Vma_lock is cleared.
tmp->vm_ops->open -- Alloc new vma_lock outside i_mmap_rwsem!
hugetlb_vma_unlock_write -- Vma_lock is assigned!!!
i_mmap_unlock_write(mapping);
hugetlb_dup_vma_private() and hugetlb_vm_op_open() are called outside
i_mmap_rwsem lock while vma lock can be used in the same time. Fix this
by deferring linking file vma until vma is fully initialized. Those vmas
should be initialized first before they can be used.
Reported-by: Thorvald Natvig <[email protected]>
Closes: https://lore.kernel.org/linux-mm/20240129161735.6gmjsswx62o4pbja@revolver/T/ [1]
Fixes: 8d9bfb260814 ("hugetlb: add vma based lock for pmd sharing")
Signed-off-by: Miaohe Lin <[email protected]>
---
kernel/fork.c | 33 +++++++++++++++++----------------
1 file changed, 17 insertions(+), 16 deletions(-)
diff --git a/kernel/fork.c b/kernel/fork.c
index 84de5faa8c9a..99076dbe27d8 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -714,6 +714,23 @@ static __latent_entropy int dup_mmap(struct mm_struct *mm,
} else if (anon_vma_fork(tmp, mpnt))
goto fail_nomem_anon_vma_fork;
vm_flags_clear(tmp, VM_LOCKED_MASK);
+ /*
+ * Copy/update hugetlb private vma information.
+ */
+ if (is_vm_hugetlb_page(tmp))
+ hugetlb_dup_vma_private(tmp);
+
+ /*
+ * Link the vma into the MT. After using __mt_dup(), memory
+ * allocation is not necessary here, so it cannot fail.
+ */
+ vma_iter_bulk_store(&vmi, tmp);
+
+ mm->map_count++;
+
+ if (tmp->vm_ops && tmp->vm_ops->open)
+ tmp->vm_ops->open(tmp);
+
file = tmp->vm_file;
if (file) {
struct address_space *mapping = file->f_mapping;
@@ -730,25 +747,9 @@ static __latent_entropy int dup_mmap(struct mm_struct *mm,
i_mmap_unlock_write(mapping);
}
- /*
- * Copy/update hugetlb private vma information.
- */
- if (is_vm_hugetlb_page(tmp))
- hugetlb_dup_vma_private(tmp);
-
- /*
- * Link the vma into the MT. After using __mt_dup(), memory
- * allocation is not necessary here, so it cannot fail.
- */
- vma_iter_bulk_store(&vmi, tmp);
-
- mm->map_count++;
if (!(tmp->vm_flags & VM_WIPEONFORK))
retval = copy_page_range(tmp, mpnt);
- if (tmp->vm_ops && tmp->vm_ops->open)
- tmp->vm_ops->open(tmp);
-
if (retval) {
mpnt = vma_next(&vmi);
goto loop_out;
--
2.33.0
On 4/10/2024 2:14 AM, Miaohe Lin wrote:
> Thorvald reported a WARNING [1]. And the root cause is below race:
>
> CPU 1 CPU 2
> fork hugetlbfs_fallocate
> dup_mmap hugetlbfs_punch_hole
> i_mmap_lock_write(mapping);
> vma_interval_tree_insert_after -- Child vma is visible through i_mmap tree.
> i_mmap_unlock_write(mapping);
> hugetlb_dup_vma_private -- Clear vma_lock outside i_mmap_rwsem!
> i_mmap_lock_write(mapping);
> hugetlb_vmdelete_list
> vma_interval_tree_foreach
> hugetlb_vma_trylock_write -- Vma_lock is cleared.
> tmp->vm_ops->open -- Alloc new vma_lock outside i_mmap_rwsem!
> hugetlb_vma_unlock_write -- Vma_lock is assigned!!!
> i_mmap_unlock_write(mapping);
>
> hugetlb_dup_vma_private() and hugetlb_vm_op_open() are called outside
> i_mmap_rwsem lock while vma lock can be used in the same time. Fix this
> by deferring linking file vma until vma is fully initialized. Those vmas
> should be initialized first before they can be used.
>
> Reported-by: Thorvald Natvig <[email protected]>
> Closes: https://lore.kernel.org/linux-mm/20240129161735.6gmjsswx62o4pbja@revolver/T/ [1]
> Fixes: 8d9bfb260814 ("hugetlb: add vma based lock for pmd sharing")
> Signed-off-by: Miaohe Lin <[email protected]>
> ---
> kernel/fork.c | 33 +++++++++++++++++----------------
> 1 file changed, 17 insertions(+), 16 deletions(-)
>
> diff --git a/kernel/fork.c b/kernel/fork.c
> index 84de5faa8c9a..99076dbe27d8 100644
> --- a/kernel/fork.c
> +++ b/kernel/fork.c
> @@ -714,6 +714,23 @@ static __latent_entropy int dup_mmap(struct mm_struct *mm,
> } else if (anon_vma_fork(tmp, mpnt))
> goto fail_nomem_anon_vma_fork;
> vm_flags_clear(tmp, VM_LOCKED_MASK);
> + /*
> + * Copy/update hugetlb private vma information.
> + */
> + if (is_vm_hugetlb_page(tmp))
> + hugetlb_dup_vma_private(tmp);
> +
> + /*
> + * Link the vma into the MT. After using __mt_dup(), memory
> + * allocation is not necessary here, so it cannot fail.
> + */
> + vma_iter_bulk_store(&vmi, tmp);
> +
> + mm->map_count++;
> +
> + if (tmp->vm_ops && tmp->vm_ops->open)
> + tmp->vm_ops->open(tmp);
> +
> file = tmp->vm_file;
> if (file) {
> struct address_space *mapping = file->f_mapping;
> @@ -730,25 +747,9 @@ static __latent_entropy int dup_mmap(struct mm_struct *mm,
> i_mmap_unlock_write(mapping);
> }
>
> - /*
> - * Copy/update hugetlb private vma information.
> - */
> - if (is_vm_hugetlb_page(tmp))
> - hugetlb_dup_vma_private(tmp);
> -
> - /*
> - * Link the vma into the MT. After using __mt_dup(), memory
> - * allocation is not necessary here, so it cannot fail.
> - */
> - vma_iter_bulk_store(&vmi, tmp);
> -
> - mm->map_count++;
> if (!(tmp->vm_flags & VM_WIPEONFORK))
> retval = copy_page_range(tmp, mpnt);
>
> - if (tmp->vm_ops && tmp->vm_ops->open)
> - tmp->vm_ops->open(tmp);
> -
> if (retval) {
> mpnt = vma_next(&vmi);
> goto loop_out;
Looks good.
Reviewed-by: Jane Chu <[email protected]>
-jane