Re: [PATCH v6 12/15] tty: serial: 8250: Correct conversion specifiers

From: Andy Shevchenko
Date: Thu Jan 14 2016 - 04:31:14 EST


On Wed, 2016-01-13 at 22:29 +0100, Anton Wuerfel wrote:
> This patch fixes compiler warnings about wrong conversion specifiersÂ

This patch obviously doesn't _fix_warnings_, rather changes specifiers
to get it consistent with argument.

> used
> in a debug output in 8250_pnp.c. The precise warning is:

Usually there is no need to cite all linesâ

>
> drivers/tty/serial/8250/8250_pnp.c: In function âserial_pnp_probeâ:
> include/linux/dynamic_debug.h:64:16: warning: format â%xâ expects
> argument
> of [...]

OK.

>
> include/linux/dynamic_debug.h:84:2: note: in expansion of macro
> âDEFINE_DYNAMIC_DEBUG_METADATAâ
>
> include/linux/device.h:1179:2: note: in expansion of macro
> âdynamic_dev_dbgâ

Redundant.

>
> drivers/tty/serial/8250/8250_pnp.c:467:2: note: in expansion of macro
> âdev_dbgâ
> Â dev_dbg(&dev->dev,
> Â ^
> include/linux/dynamic_debug.h:64:16: warning: format â%lxâ expects
> argument
> of [...]

OK.

>
> include/linux/dynamic_debug.h:84:2: note: in expansion of macro
> âDEFINE_DYNAMIC_DEBUG_METADATAâ
>
> include/linux/device.h:1179:2: note: in expansion of macro
> âdynamic_dev_dbgâ

Redundant.

>
> drivers/tty/serial/8250/8250_pnp.c:467:2: note: in expansion of macro
> âdev_dbgâ
> Â dev_dbg(&dev->dev,
> Â ^
>
> Those warnings never got triggered, because the command was nested
> in an #ifdef, which is removed by a patch of this series.
>
> Signed-off-by: Anton WÃrfel <anton.wuerfel@xxxxxx>
> Signed-off-by: Phillip Raffeck <phillip.raffeck@xxxxxx>
> Cc: linux-kernel@xxxxxxxxxxxx
> ---
> Âdrivers/tty/serial/8250/8250_pnp.c | 4 ++--
> Â1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/tty/serial/8250/8250_pnp.c
> b/drivers/tty/serial/8250/8250_pnp.c
> index 0c32c59..92f4412 100644
> --- a/drivers/tty/serial/8250/8250_pnp.c
> +++ b/drivers/tty/serial/8250/8250_pnp.c
> @@ -465,8 +465,8 @@ serial_pnp_probe(struct pnp_dev *dev, const
> struct pnp_device_id *dev_id)
> Â return -ENODEV;
> Â
> Â dev_dbg(&dev->dev,
> - Â"Setup PNP port: port %x, mem 0x%lx, irq %d, type
> %d\n",
> - Âuart.port.iobase, uart.port.mapbase,
> + Â"Setup PNP port: port %lx, mem 0x%lx, irq %d, type
> %d\n",
> + Âuart.port.iobase, (unsigned long)uart.port.mapbase,

This approach is not fully correct. In Linux kernel we have %p
extensions to cover some custom types such as resource_size_t, of which
mapbase is.

Thus,

"Setup PNP port: port %lx, mem %pa, irq %d, type %d\n",
uart.port.iobase, &uart.port.mapbase,

looks better.

> Â Âuart.port.irq, uart.port.iotype);
> Â
> Â if (flags & CIR_PORT) {

--
Andy Shevchenko <andriy.shevchenko@xxxxxxxxxxxxxxx>
Intel Finland Oy