Re: linux-next: manual merge of the bpf-next tree with the vfs-brauner tree

[Daniel Borkmann]

On 6/4/26 2:59 PM, Mark Brown wrote:
> Hi all,
>
> Today's linux-next merge of the bpf-next tree got a conflict in:
>
>    kernel/bpf/inode.c
>
> between commit:
>
>    a146500e11445 ("bpf: Add simple xattr support to bpffs")
>
> from the vfs-brauner tree and commit:
>
>    b93c55b4932dd ("bpf: fix UAF by restoring RCU-delayed inode freeing in bpffs")
>
> from the bpf-next tree.
>
> This merge resolution definitely needs a double check.
>
> I fixed it up (see below) and can carry the fix as necessary. This
> is now fixed as far as linux-next is concerned, but any non trivial
> conflicts should be mentioned to your upstream maintainer when your tree
> is submitted for merging.  You may also want to consider cooperating
> with the maintainer of the conflicting tree to minimise any particularly
> complex conflicts.
>
> diff --cc kernel/bpf/inode.c
> index c3f79b5a2f8c0,188c774a469ca..0000000000000
> --- a/kernel/bpf/inode.c
> +++ b/kernel/bpf/inode.c
> @@@ -842,9 -768,12 +842,13 @@@ static void bpf_destroy_inode(struct in
>    
>    	if (!bpf_inode_type(inode, &type))
>    		bpf_any_put(inode->i_private, type);
>   +	simple_xattrs_free(&opts->xa_cache, &bi->xattrs, NULL);
>    }
>    
> + /*
> +  * Called after RCU grace period - safe to free inode and anything
> +  *  that might be accessed by RCU pathwalk (inode fields, i_link).
> +  */
>    static void bpf_free_inode(struct inode *inode)
>    {
>    	if (S_ISLNK(inode->i_mode))

Ack, looks good to me! The extra comment could probably be dropped (in other
words going with what's in Christian's tree as resolution), but its fine either
way.

Thanks,
Daniel