MIME-Version: 1.0
References: <20211227164317.4146918-1-schnelle@linux.ibm.com>
 <20211227164317.4146918-31-schnelle@linux.ibm.com> <YcrIHxTDipVNUuCA@kroah.com>
 <b9d0c0b88ef66f9beb51a880e765177670a76394.camel@linux.ibm.com> <Ycw6kXhd+NV0GMWc@kroah.com>
In-Reply-To: <Ycw6kXhd+NV0GMWc@kroah.com>
From:   Arnd Bergmann <arnd@kernel.org>
Date:   Thu, 30 Dec 2021 11:19:05 -0500
Message-ID: <CAK8P3a1p6_=s4XcJiJSkuZJBrCpwCXQ1MX6EadY00KdypVQisg@mail.gmail.com>
Subject: Re: [RFC 30/32] /dev/port: don't compile file operations without CONFIG_DEVPORT
To:     Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc:     Niklas Schnelle <schnelle@linux.ibm.com>,
        Bjorn Helgaas <bhelgaas@google.com>,
        John Garry <john.garry@huawei.com>,
        Nick Hu <nickhu@andestech.com>,
        Greentime Hu <green.hu@gmail.com>,
        Vincent Chen <deanbo422@gmail.com>,
        Paul Walmsley <paul.walmsley@sifive.com>,
        Palmer Dabbelt <palmer@dabbelt.com>,
        Albert Ou <aou@eecs.berkeley.edu>, Guo Ren <guoren@kernel.org>,
        Arnd Bergmann <arnd@arndb.de>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        linux-arch <linux-arch@vger.kernel.org>,
        linux-pci <linux-pci@vger.kernel.org>,
        linux-riscv <linux-riscv@lists.infradead.org>,
        linux-csky@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Wed, Dec 29, 2021 at 5:38 AM Greg Kroah-Hartman
<gregkh@linuxfoundation.org> wrote:
> > > > -static const struct file_operations __maybe_unused port_fops = {
> > > > +#ifdef CONFIG_DEVPORT
> > > > +static const struct file_operations port_fops = {
> > > >   .llseek         = memory_lseek,
> > > >   .read           = read_port,
> > > >   .write          = write_port,
> > > >   .open           = open_port,
> > > >  };
> > > > +#endif
> > >
> > > Why is this #ifdef needed if it is already __maybe_unused?
> >
> > Because read_port() calls inb() and write_port() calls outb() they
> > wouldn't compile once these are no longer defined. Then however the
> > read_port/write_port symbols in the struct initialization above
> > couldn't be resolved.
> >
> > >
> > > In looking closer, this change could be taken now as the use of this
> > > variable already is behind this same #ifdef statement, right?
> >
> > Yes
>
> Great, feel free to send this individually, not as a RFC patch, and I
> will be glad to queue it up.

I think this patch should contain the 'depends on HAS_IOPORT' that
is currently added in a different patch (char: impi, tpm: depend on
HAS_IOPORT).

However, we can't merge that version until HAS_IOPORT is actually
added to the kernel.

      Arnd