Re: [PATCH] mm/mmap: Fix locking issues in vma_expand()

[Liam Howlett <liam.howlett@oracle.com>]

Please disregard this patch, it is missing 1/2 the fix and the fixes tag
in the log message.  The proper fix will be arriving shortly.

Sorry about the noise.

* Liam R. Howlett <Liam.Howlett@oracle.com> [220712 10:41]:
> Hugh noted that the locking was mixed up in vma_expand(), revert to the
> correct order.
> 
> Suggested-by: Hugh Dickins <hughd@google.com>
> Signed-off-by: Liam R. Howlett <Liam.Howlett@oracle.com>
> ---
>  mm/mmap.c | 14 +++++++++-----
>  1 file changed, 9 insertions(+), 5 deletions(-)
> 
> diff --git a/mm/mmap.c b/mm/mmap.c
> index 9eb663cde5c7..001b54fe19b4 100644
> --- a/mm/mmap.c
> +++ b/mm/mmap.c
> @@ -574,7 +574,8 @@ inline int vma_expand(struct ma_state *mas, struct vm_area_struct *vma,
>  		if (next->anon_vma && !vma->anon_vma) {
>  			int error;
>  
> -			vma->anon_vma = next->anon_vma;
> +			anon_vma = next->anon_vma;
> +			vma->anon_vma = anon_vma;
>  			error = anon_vma_clone(vma, next);
>  			if (error)
>  				return error;
> @@ -592,16 +593,19 @@ inline int vma_expand(struct ma_state *mas, struct vm_area_struct *vma,
>  
>  	vma_adjust_trans_huge(vma, start, end, 0);
>  
> +	if (file) {
> +		mapping = file->f_mapping;
> +		root = &mapping->i_mmap;
> +		uprobe_munmap(vma, vma->vm_start, vma->vm_end);
> +		i_mmap_lock_write(mapping);
> +	}
> +
>  	if (anon_vma) {
>  		anon_vma_lock_write(anon_vma);
>  		anon_vma_interval_tree_pre_update_vma(vma);
>  	}
>  
>  	if (file) {
> -		mapping = file->f_mapping;
> -		root = &mapping->i_mmap;
> -		uprobe_munmap(vma, vma->vm_start, vma->vm_end);
> -		i_mmap_lock_write(mapping);
>  		flush_dcache_mmap_lock(mapping);
>  		vma_interval_tree_remove(vma, root);
>  	}
> -- 
> 2.35.1