Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::3:7 as permitted sender) client-ip=2620:137:e000::3:7;
MIME-Version: 1.0
References: <20231026192830.21288-1-rdunlap@infradead.org>
In-Reply-To: <20231026192830.21288-1-rdunlap@infradead.org>
From:   Amir Goldstein <amir73il@gmail.com>
Date:   Thu, 26 Oct 2023 22:46:06 +0300
Message-ID: <CAOQ4uxhYiu+ou0SiwYsuSd-YayRq+1=zgUw_2G79L8SxkDQV7g@mail.gmail.com>
Subject: Re: [PATCH] exportfs: handle CONFIG_EXPORTFS=m also
To:     Randy Dunlap <rdunlap@infradead.org>
Cc:     linux-kernel@vger.kernel.org, Chuck Lever <chuck.lever@oracle.com>,
        Jeff Layton <jlayton@kernel.org>, linux-nfs@vger.kernel.org,
        Christian Brauner <brauner@kernel.org>,
        Alexander Viro <viro@zeniv.linux.org.uk>,
        linux-fsdevel@vger.kernel.org, Arnd Bergmann <arnd@arndb.de>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Thu, Oct 26, 2023 at 10:28=E2=80=AFPM Randy Dunlap <rdunlap@infradead.or=
g> wrote:
>
> When CONFIG_EXPORTFS=3Dm, there are multiple build errors due to
> the header <linux/exportfs.h> not handling the =3Dm setting correctly.
> Change the header file to check for CONFIG_EXPORTFS enabled at all
> instead of just set =3Dy.
>
> Fixes: dfaf653dc415 ("exportfs: make ->encode_fh() a mandatory method for=
 NFS export")
> Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
> Cc: Chuck Lever <chuck.lever@oracle.com>
> Cc: Jeff Layton <jlayton@kernel.org>
> Cc: linux-nfs@vger.kernel.org
> Cc: Amir Goldstein <amir73il@gmail.com>
> Cc: Christian Brauner <brauner@kernel.org>
> Cc: Alexander Viro <viro@zeniv.linux.org.uk>
> Cc: linux-fsdevel@vger.kernel.org
>
> ---
>  include/linux/exportfs.h |    2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff -- a/include/linux/exportfs.h b/include/linux/exportfs.h
> --- a/include/linux/exportfs.h
> +++ b/include/linux/exportfs.h
> @@ -314,7 +314,7 @@ extern struct dentry *exportfs_decode_fh
>  /*
>   * Generic helpers for filesystems.
>   */
> -#ifdef CONFIG_EXPORTFS
> +#if IS_ENABLED(CONFIG_EXPORTFS)
>  int generic_encode_ino32_fh(struct inode *inode, __u32 *fh, int *max_len=
,
>                             struct inode *parent);
>  #else

Thanks for the fix, but Arnd reported that this fix could cause a link
issue in some configuration - I did not verify.

I would much rather turn EXPORTFS into a bool config
and avoid the unneeded build test matrix.

See comparison to the amount of code of EXPORTFS
to BUFFER_HEAD and FS_IOMAP code which are bool:

~/src/linux$ wc -l fs/exportfs/*.c
636 fs/exportfs/expfs.c

~/src/linux$ wc -l fs/buffer.c fs/mpage.c
  3164 fs/buffer.c
   685 fs/mpage.c
  3849 total

~/src/linux$ wc -l fs/iomap/*.c
 2002 fs/iomap/buffered-io.c
  754 fs/iomap/direct-io.c
  124 fs/iomap/fiemap.c
   97 fs/iomap/iter.c
  104 fs/iomap/seek.c
  195 fs/iomap/swapfile.c
   13 fs/iomap/trace.c
 3289 total

Thanks,
Amir.