Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
MIME-Version: 1.0
References: <20210714113140.2949995-1-o451686892@gmail.com>
In-Reply-To: <20210714113140.2949995-1-o451686892@gmail.com>
From:   Marco Elver <elver@google.com>
Date:   Wed, 14 Jul 2021 13:56:20 +0200
Message-ID: <CANpmjNMe+wyuFLJ0nOk-4Sr57GgBf6VuHa3hmsVUZYJ1gjuC5A@mail.gmail.com>
Subject: Re: [PATCH] kfence: defer kfence_test_init to ensure that kunit
 debugfs is created
To:     Weizhao Ouyang <o451686892@gmail.com>
Cc:     glider@google.com, dvyukov@google.com, akpm@linux-foundation.org,
        kasan-dev@googlegroups.com, linux-mm@kvack.org,
        linux-kernel@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Wed, 14 Jul 2021 at 13:31, Weizhao Ouyang <o451686892@gmail.com> wrote:
> kfence_test_init and kunit_init both use the same level late_initcall,
> which means if kfence_test_init linked ahead of kunit_init,
> kfence_test_init will get a NULL debugfs_rootdir as parent dentry,
> then kfence_test_init and kfence_debugfs_init both create a debugfs
> node named "kfence" under debugfs_mount->mnt_root, and it will throw
> out "debugfs: Directory 'kfence' with parent '/' already present!" with
> EEXIST. So kfence_test_init should be deferred.
>
> Signed-off-by: Weizhao Ouyang <o451686892@gmail.com>

Tested-by: Marco Elver <elver@google.com>

Thank you.

> ---
>  mm/kfence/kfence_test.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/mm/kfence/kfence_test.c b/mm/kfence/kfence_test.c
> index 7f24b9bcb2ec..942cbc16ad26 100644
> --- a/mm/kfence/kfence_test.c
> +++ b/mm/kfence/kfence_test.c
> @@ -852,7 +852,7 @@ static void kfence_test_exit(void)
>         tracepoint_synchronize_unregister();
>  }
>
> -late_initcall(kfence_test_init);
> +late_initcall_sync(kfence_test_init);
>  module_exit(kfence_test_exit);
>
>  MODULE_LICENSE("GPL v2");
> --
> 2.30.2