Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Wed, 13 Mar 2019 16:12:58 -0600
From:   Keith Busch <kbusch@kernel.org>
To:     "Elliott, Robert (Persistent Memory)" <elliott@hpe.com>
Cc:     Arnd Bergmann <arnd@arndb.de>, Jiri Kosina <jikos@kernel.org>,
        Jens Axboe <axboe@kernel.dk>, Omar Sandoval <osandov@fb.com>,
        Hannes Reinecke <hare@suse.de>,
        "linux-block@vger.kernel.org" <linux-block@vger.kernel.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH] floppy: fix harmless clang build warning
Message-ID: <20190313221257.GA15153@localhost.localdomain>
References: <20190313211433.4178148-1-arnd@arndb.de>
 <AT5PR8401MB116964D0C90783A164834C2CAB4A0@AT5PR8401MB1169.NAMPRD84.PROD.OUTLOOK.COM>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <AT5PR8401MB116964D0C90783A164834C2CAB4A0@AT5PR8401MB1169.NAMPRD84.PROD.OUTLOOK.COM>
User-Agent: Mutt/1.9.1 (2017-09-22)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Wed, Mar 13, 2019 at 10:01:43PM +0000, Elliott, Robert (Persistent Memory) wrote:
> > @@ -3899,7 +3899,7 @@ static void __init config_types(void)
> >  	if (!UDP->cmos)
> >  		UDP->cmos = FLOPPY0_TYPE;
> >  	drive = 1;
> > -	if (!UDP->cmos && FLOPPY1_TYPE)
> > +	if (!UDP->cmos)
> >  		UDP->cmos = FLOPPY1_TYPE;
> > 
> >  	/* FIXME: additional physical CMOS drive detection should go here */
> > --
> > 2.20.0
> 
> On x86 it expands to a hardware read, so this would change the control flow.
> 
> #define FLOPPY1_TYPE                                    \
> ({                                                      \
>         unsigned long flags;                            \
>         unsigned char val;                              \
>         spin_lock_irqsave(&rtc_lock, flags);            \
>         val = CMOS_READ(0x10) & 15;                     \
>         spin_unlock_irqrestore(&rtc_lock, flags);       \
>         val;                                            \
> })

The end result is still the same, though. It just doesn't read the
hardware twice when FLOPPY1_TYPE is non-zero.