Re: [PATCH] serial: atmel: fix compiler warning on address cast

[Russell King - ARM Linux <linux@arm.linux.org.uk>]

On Thu, Oct 08, 2015 at 03:49:54PM +0100, Andre Przywara wrote:
> Hi,
> 
> On 08/10/15 11:38, Alexandre Belloni wrote:
> > On 08/10/2015 at 10:37:49 +0100, Russell King - ARM Linux wrote :
> >> On Thu, Oct 08, 2015 at 11:01:48AM +0200, Alexandre Belloni wrote:
> >>> On 05/10/2015 at 18:00:52 +0100, Andre Przywara wrote :
> >>>> Turning on KVM and LPAE support on top of a multi_v7_defconfig will
> >>>> produce a compiler warning in the Atmel serial driver:
> >>>> drivers/tty/serial/atmel_serial.c: In function 'atmel_verify_port':
> >>>> drivers/tty/serial/atmel_serial.c:2299:6: warning: cast to pointer from integer of different size [-Wint-to-pointer-cast]
> >>>>   if ((void *)port->mapbase != ser->iomem_base)
> >>>>       ^
> >>>>
> >>>> Fix that by using the cast on the right hand side instead, as similar
> >>>> code already does in other drivers.
> >>>>
> >>>> Signed-off-by: Andre Przywara <andre.przywara@arm.com>
> >>> Acked-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
> >>
> >> Hmm, this actually points at a problem, and the cast screems out about it.
> >> Think about it, can you guess what it is?
> 
> Russell, I'd really appreciate if we wouldn't have to play
> "guess-what-I-am-thinking" here. Can't you just say what's bothering you?

Actually, I think it's fine - the interface is just another one of the
weird cases we have in the kernel.

However, for the record, things get typed a certain way for a reason.
Virtual addresses that can be dereferenced are generally pointer like.
Physical addresses are generally integers.

Casting between the two generally indicates a bug, and whenever you see
a cast between the two, it should have special attention paid to it to
make sure that it really is correct.

There are a few famous places in the kernel where casting between an
unsigned long and a virtual address is pretty standard, but they're
the exception to the rule (eg, the argument to a timer, get_free_page()).
This is another example.

Sorry for creating confusion.

-- 
FTTC broadband for 0.8mile line: currently at 9.6Mbps down 400kbps up
according to speedtest.net.